The Shared Mind
Converging Evidence in Language and Communication Research (CELCR) Over the past decades, linguists h...
24 downloads
445 Views
20MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
The Shared Mind
Converging Evidence in Language and Communication Research (CELCR) Over the past decades, linguists have taken a broader view of language and are borrowing methods and findings from other disciplines such as cognition and computer sciences, neurology, biology, sociology, psychology, and anthropology. This development has enriched our knowledge of language and communication, but at the same time it has made it difficult for researchers in a particular field of language studies to be aware of how their findings might relate to those in other (sub-)disciplines. CELCR seeks to address this problem by taking a cross-disciplinary approach to the study of language and communication. The books in the series focus on a specific linguistic topic and offer studies pertaining to this topic from different disciplinary angles, thus taking converging evidence in language and communication research as its basic methodology.
Editors Marjolijn H. Verspoor University of Groningen
Wilbert Spooren
Vrije Universiteit Amsterdam
Advisory Board Walter Daelemans
Leo Noordman
Cliff Goddard
Martin Pütz
University of Antwerp University of New England
Tilburg University University of Koblenz-Landau
Roeland van Hout
Radboud University Nijmegen
Volume 12 The Shared Mind. Perspectives on intersubjectivity Edited by Jordan Zlatev, Timothy P. Racine, Chris Sinha and Esa Itkonen
The Shared Mind Perspectives on intersubjectivity
Edited by
Jordan Zlatev Lund University, Copenhagen Business School
Timothy P. Racine Simon Fraser University
Chris Sinha University of Portsmouth
Esa Itkonen University of Turku
John Benjamins Publishing Company Amsterdam / Philadelphia
8
TM
The paper used in this publication meets the minimum requirements of American National Standard for Information Sciences – Permanence of Paper for Printed Library Materials, ansi z39.48-1984.
Library of Congress Cataloging-in-Publication Data The shared mind : perspectives on intersubjectivity / edited by Jordan Zlatev ... [et al.]. p. cm. (Converging Evidence in Language and Communication Research, issn 1566-7774 ; v. 12) Includes bibliographical references and index. 1. Intersubjectivity. 2. Language and languages. 3. Communication. 4. Evolution. I. Zlatev, Jordan. P107.S535 2008 401--dc22 isbn 978 90 272 3900 6 (Hb; alk. paper)
2008015388
© 2008 – John Benjamins B.V. No part of this book may be reproduced in any form, by print, photoprint, microfilm, or any other means, without written permission from the publisher. John Benjamins Publishing Co. · P.O. Box 36224 · 1020 me Amsterdam · The Netherlands John Benjamins North America · P.O. Box 27519 · Philadelphia pa 19118-0519 · usa
Table of contents
Foreword: Shared minds and the science of fiction: Why theories will differ Colwyn Trevarthen 1. Intersubjectivity: What makes us human? Jordan Zlatev, Timothy P. Racine, Chris Sinha and Esa Itkonen
vii
1
Part I. Development 2. Understanding others through primary interaction and narrative practice Shaun Gallagher and Daniel D. Hutto 3. The neuroscience of social understanding John Barresi and Chris Moore 4. Engaging, sharing, knowing: Some lessons from research in autism Peter Hobson and Jessica A. Hobson 5. Coming to agreement: Object use by infants and adults Cintia Rodríguez and Christiane Moro
17 39
67 89
6. The role of intersubjectivity in the development of intentional communication Ingar Brinck
115
7. Sharing mental states: Causal and definitional issues in intersubjectivity Noah Susswein and Timothy P. Racine
141
Part II. Evolution 8. What is the nature of the gestural communication of great apes? Simone Pika
165
vi
The Shared Mind
9. The heterochronic origins of explicit reference David A. Leavens, William D. Hopkins and Kim A. Bard
187
10. The co-evolution of intersubjectivity and bodily mimesis Jordan Zlatev
215
11. First communions: Mimetic sharing without theory of mind Daniel D. Hutto
245
Part III. Language 12. The central role of normativity in language and linguistics Esa Itkonen
279
13. Intersubjectivity and the architecture of the language system Arie Verhagen
307
14. Intersubjectivity in interpreted interactions: The interpreter’s role in co-constructing meaning Terry Janzen and Barbara Shaffer
333
15. Language and the signifying object: From convention to imagination Chris Sinha and Cintia Rodríguez
357
Author index Subject index
379 383
Foreword
Shared minds and the science of fiction Why theories will differ Colwyn Trevarthen
It is a pleasure to respond to these essays on the collective story-making of culture: the experience of reality that human beings create together because they are motivated from birth to experiment with the exchange of fantasies and to find meaning in them. Human consciousness has the special gift of imaginative travel through times and places, and it grows through communication of intentions and interests. Language and the practical tools of our society enrich the products of the game, but its causes are in the movements and preferences of embodied minds, minds that have evolved to act in sympathy and to share history and invention, whatever may turn out to be the topic or task. Our common knowledge and perception of ourselves as knowers of meaningful facts depends upon, and grows from, our innate capacity for intersubjectivity. We know, of course, how involved we are with one another’s intentions, thoughts and feelings, and that much of this intimacy in experience cannot be carried in words. The mental life of others is, as Stein Bråten says, “felt immediately” (Bråten 1998). And yet mind science and its ambitious extension in brain science have, and still mainly do, regard us as single heads processing information, storing it up in memory for reprocessing, and transferring it symbolically. Even when we are granted a body that moves, it is a robot that struggles to know other minds by a hopeless effort of “theorizing” or “simulation”. Such unsympathetic entities are science fictions. We need a science of the imaginative fictions persons so easily share. The authors of this book accept that human life and culture is incomprehensible without intersubjective processes – so, the question mark of the title of the editors’ introduction in Chapter 1 must be rhetorical, and ironic. It is added, perhaps, because our experimental psychology has inherited and largely still pays homage to a scholastic philosophy of minds as separate experience-registering systems that act, and think, on what they alone perceive. But in the real sociable world every act we make, every feeling, has as much power to move others as it
viii Colwyn Trevarthen
has to move our self. With compassion we can see causes of actions in another, even causes that they themselves fail to comprehend or control. All teaching and therapy, indeed all cooperative activities, depend on this sympathetic insight into motive impulses and emotions in human moving. I believe that all the inventions of culture, including the evolving languages that distinguish our different societies, and the arts and technologies that are necessary instruments of communal life and treasures of our history, grow from the ability that every young infant has to enter into the co-creation of a proto-conversational narrative with an entranced parent. Our stories of meaning are built on mimetic skills we have inherited from highly sociable animal ancestors, but we are born with new motives for fictional elaboration of rituals. Even our personality, the “who” we are and the narrative of what we have done and known, grants us the role of one protagonist in a social drama where significant others live as supportive allies or contentious rivals (Trevarthen 1993, 1998a, 1998b). Thus we become companions or aliens in relation to a meaningful world. Shared minds create all we know. Reading this book we sense the authors are glad to be free of a prison built of ideas that are unaware and unsympathetic of how we really live. They present an antithesis to the computational or representational mind, and seek to define the special human mind, which is not just conscious and rational, but has a unique intersubjective awareness that makes up explanations of a shared and artificial world – a mind that builds cultures with power to change nature. Given the exploratory nature of the topic, inevitably, they come to somewhat different conclusions. Because several authors make generous reference to my research on communication in infancy, and the theory of Innate Intersubjectivity I was rash enough to propose 30 years ago, I feel I should explain the particular scientific experience that supported the project, and the influence of teachers and colleagues who were ahead of me in the story. I was trained as a student of biology to master ways of observing in detail how plants grow and how animals move in adaptive ways. My undergraduate teachers were plant ecologists, physiologists and ethologists. My PhD research was on the experimental neuropsychology of visual consciousness in monkeys with Roger Sperry, who had proposed in 1952 that perceiving must be understood as information picked up to guide moving – that the science of consciousness or mind in the brain should begin by asking how the brain moves the body in intelligent ways (Sperry 1952). My experiments with split-brain monkeys proved that willing to do something can indeed determine what a brain sees. I began work with infants in 1967, in collaboration in with Jerome Bruner, who wished to examine infant cognition and learning in a different way from Piaget’s experiments on infants’ object concepts; Berry Brazelton, who was pioneering more sensitive and responsive paediatric care for newborns and their mothers (Brazelton 1979); and Martin Richards, an ethologist of mammalian
Foreword
aternal behaviour. Our aim was to observe what came about, rather than experm iment with a priori hypotheses about infant perception or cognition – to record in as complete detail as possible what could be seen and heard when a mother and infant were communicating, and to compare it with what the baby would do when oriented to an inanimate object. We saw complex conversation-like engagements in which both infant and mother exhibited intuitive competence for sharing their impulses, and we realised that there was no science to explain this. At about the same time two other persons – anthropologist and linguist Mary Catherine Bateson, and developmental psychiatrist Daniel Stern – were discovering the same phenomena and attributing them to innate motivations of an intersubjective kind (Bateson 1979; Stern 1971). We were, without knowing one another, exploring out of the psychological box, free to observe the cleverness in infants and their companions and free to speculate about their significance for human relationships and for cultural learning. All of us were entranced by the infants’ rhythmic sympathy with a parent’s attempts to communicate, and their joint inventiveness. Through the 1970s, using film and television to record and patiently microanalyse, I charted age-related changes in the play and attributed them to innate motives, the development of new sensory and motor competences in the infant, and sensitive intuitive support from the mother (Trevarthen 1974). I took the term ‘intersubjectivity’ from an inspiring article Joanna Ryan wrote on the development of ‘communicative competence’ before language (Ryan 1974), and her comparison of the infant’s tactics with those Jürgen Habermas had defined as the intersubjective functions or ‘dialogic universals’ through which conversational exchanges and cooperative meaning-making are regulated in society (Habermas 1970). At the same time Jerome Bruner led a neo-Vygotskian transformation of developmental and educational theory that gave primacy to collaborative learning in meaningful tasks (Bruner 1968, 1990). Children gain the skills and language of their culture, and learn how to manage the material world in cooperative ways, by way of their will to share purposes, interests and objects (Sinha and Rodríguez this volume). One of my young colleagues, Penelope Hubley, making a careful longitudinal study of mother-infant companionship in the early 1970s, observed changes from proto-conversations of two-month-olds, through play in games, first of the body, then with shared interest in objects, to the remarkable transformation of the infant’s motives at 9 months when the baby became a different kind of partner in intent participation (Hubley and Trevarthen 1979). The enjoyment of playful rituals by six-month olds in games with expressive gestures or toys, enjoying at a new ritualised level the teasing meta-communication that Gregory Bateson had identified as the critical element in animal play (Bateson 1955), was replaced by a more serious intent to do “work” with objects that had some potential for
ix
Colwyn Trevarthen
ractical use, which others would acknowledge. Guided by a companion’s shifting p focus of interest, and by exhortations to complete a little project set by movements of intention, the baby became a self-confident partner – a co-worker. At the same age, about 40 weeks after a full term birth, the baby was a self-possessed and self-conscious performer of many new rituals of social expression. The mutual understanding established in previous months and practiced in games was transformed into what Michael Halliday called proto-linguistic “acts of meaning”: vocal and gestural signals of things that might be named (Halliday 1975). We called it Secondary Intersubjectivity. The relevance to cultural learning of this trajectory in growth of the infant’s mind was clear, as was the primary importance of the “mutual attention” with a familiar companion. Strangers were too uncomprehending to be trusted in such first steps to a conventional world. And sensitive experimental studies by Lynne Murray proved how important contingent and respectful attention and sympathy of feelings was for the infant to build meaning in another’s company (Murray and Trevarthen 1985). True, as Susswein and Racine (this volume) say, my account of the developments, from Primary Intersubjectivity, through Games of the Person and Games with Objects to Secondary Intersubjectivity, which we continued beyond the first use of words to the Imaginative Play of 2 and 3 year olds, was descriptive. Yes, it was a taxonomy of stages in behaviour, but it was meant to be more than that. It implied and explored a theory, and I sought many kinds of evidence for the “causes” of change, especially causes in the infant’s growing mind. I was convinced that the only useful explanation was one that assumed that the fundamental adaptations of body and brain for intersubjectivity were innate, as were the direction and stages of developmental change through the early years, and the learning that was so obviously assisted by the companionship of the parent (Trevarthen 1979, 1989). True to my biological principles, and starting with a theory of how neural systems could generate motives, I looked for explanations in the ontogeny of the brain and body of the child, and for correlations with known age-related changes in brain anatomy and function. It was not difficult to collect evidence that an embryogenetic specification of the motor and sensory functions in somatotopic (body mapping) arrays was essential, as were the theories of experience anticipating motor images of Sperry (1952) and Bernstein (1967). At first we did not have a clue how the intersubjective transfer of these intentional images could be mediated, or what the evident emotional regulations were, but in the last two decades brain science has come some way to closing that gap (cf. Gallese 2005; Pankepp 2005; Barresi and Moore this volume). One discovery of major significance for any theory of the causal factors or processes of intersubjectivity, whether of humans or animals, is that the rhythmic
Foreword
timing and modulation of energy in moving is a “code or principle of conduct” that makes motives share-able. A breakthrough in the exploration of human communication before language has come from the demonstration of its special polyrhythmic “musicality” (Trevarthen 1999, 2008). The science of time in the mind or “biochronology” is discovering how impulses regulating the pace and harmony of moving pass from actor to perceiver of action. How, as Ellen Dissnayake claims (Dissnayake 2000), the temporal arts originate in intrinsic dynamic processes, already exquisitely present in a newborn baby, that keep the voluntary and conscious self whole and in well-being, and make them public for sharing in intimacy. Mimesis, which I see as richer already in the neonate than Zlatev (this volume) does (because I am sure that self and other are distinct negotiants in the newborn baby’s mind) is the parent of linguistic narrative, as Merlin Donald (2001) proposes, and ‘musical semantics’ as defined by Ole Kühl sets the stage for reference with symbols (Kühl 2007). I think these are the natural foundations for Itkonen’s “normative practices” that keep languages, and other cultural creations, coherent, productive and changing (Itkonen this volume). The story is new and there is plenty of room for different plots, but we have an open prospect and a sense of adventure. The science of the shared mind looks like the best game in town.
References Barresi, J. and Moore, C. this volume. “The neuroscience of social understanding.” Bateson, G. 1955. “A theory of play and fantasy.” Psychiatric Research Reports, Series A 2: 39–51 Bateson, M.C. 1979. “The epigenesis of conversational interaction: A personal account of research development.” In Before Speech: The Beginning of Human Communication, M. Bullowa (ed.), 63–77. London: Cambridge University Press. Bernstein, N. 1967. Coordination and Regulation of Movements. New York: Pergamon. Bråten, S. 1998. “Intersubjective communion and understanding: Development and perturbation.” In Intersubjective Communication and Emotion in Early Ontogeny, S. Bråten (ed.), 372–382. Cambridge: Cambridge University Press. Brazelton, T.B. 1979. “Evidence of communication during neonatal behavioural assessment.” In Before Speech: The Beginning of Human Communication, M. Bullowa (ed.), 79–88. London, Cambridge University Press. Bruner, J.S. 1968. Processes of Cognitive Growth: Infancy. (Heinz Werner Lectures, 1968) Worcester, Mass: Clark University Press with Barri Publishers. Bruner, J.S. 1990. Acts of Meaning. Cambridge, Mass.: Harvard University Press. Dissanayake, E. 2000. Art and Intimacy: How the Arts Began. University of Washington Press, Seattle and London. Donald, M. 2001. A Mind So Rare: The Evolution of Human Consciousness. New York, NY and London, England: Norton.
xi
xii Colwyn Trevarthen
Gallese, V. 2005. “Embodied simulation: From neurons to phenomenal experience.” Phenomenology and the Cognitive Sciences 4: 23–48. Habermas, J. 1970. “Towards a theory of communicative competence.” Recent Sociology Vol. 12: 115–148, London: Macmillan. Halliday, M.A.K. 1975. Learning How to Mean: Explorations in the Development of Language. London: Edward Arnold. Hubley, P. and Trevarthen, C. 1979. “Sharing a task in infancy.” In Social Interaction During Infancy: New Directions for Child Development, 4, I. Uzgiris (ed.), 57–80. San Francisco: Jossey-Bass. Itkonen, E. this volume. “The central role of normativity for language and linguistics.” Kühl, O. 2007. Musical Semantics. (European Semiotics: Language, Cognition and Culture. No. 7). Bern: Peter Lang. Murray, L. and Trevarthen, C. 1985. “Emotional regulation of interactions between two- montholds and their mothers.” In Social Perception in Infants, T.M. Field and N.A. Fox (eds), 177– 197. Norwood, NJ: Ablex. Panksepp, J. (2005). “On the embodied neural nature of core emotional affects.” Journal of Consciousness Studies 12: 158–84. Ryan, J. 1974. “Early language development: Towards a communicational analysis.” In The Integration of a Child into a Social World, M.P.M. Richards (ed.), 185–213. London: Cambridge University Press. Sinha, C. and Rodríguez, C. this volume. “Language and the signifying object: From convention to imagination.” Sperry, R. W. 1952. “Neurology and the mind-brain problem.” American Scientist 40: 291–312. Stern, D.N. 1971. “A micro-analysis of mother-infant interaction: Behaviors regulating social contact between a mother and her three-and-a-half-month-old twins.” Journal of American Academy of Child Psychiatry 10: 501–517. Susswein, N and Racine, T.P. this volume. “Sharing mental states: Causal and definitional issues in intersubjectivity” Trevarthen, C. 1974. “Conversations with a two-month-old.” New Scientist 2 May: 230–235. Trevarthen, C. 1979. “Instincts for human understanding and for cultural cooperation: Their development in infancy.” In Human Ethology, M. von Cranach, K. Foppa, W. Lepenies and D. Ploog (eds), 530–571. Cambridge: Cambridge University Press. Trevarthen, C. 1989. “Motives for culture in young children – their natural development through communication.” In The Nature of Culture (Proceedings of the International and Interdisciplinary Symposium, Ruhr Universität, Bochum, October 7–11, 1986), W. Koch (ed.), 80–119. Bochum: Brockmeyer. Trevarthen, C. 1993. “The self born in intersubjectivity: An infant communicating.” In The Perceived Self: Ecological and Interpersonal Sources of Self-Knowledge, U. Neisser (ed.), 121– 173. New York: Cambridge University Press. Trevarthen, C. 1998a. “The concept and foundations of infant intersubjectivity.” In Intersubjective Communication and Emotion in Early Ontogeny, S. Bråten (ed.), 15–46. Cambridge: Cambridge University Press. Trevarthen, C. 1998b. “The nature of motives for human consciousness.” Psychology: The Journal of the Hellenic Psychological Society (Special Issue: “The Place of Psychology in Contemporary Sciences”, Part 2. Guest Editor, T. Velli) 4(3): 187–221.
Foreword xiii
Trevarthen, C. 1999. “Musicality and the Intrinsic Motive Pulse: Evidence from human psychobiology and infant communication.” Musicae Scientiae, Special Issue, 1999–2000, “Rhythms, musical narrative, and the origins of human communication”, 157–213. Liège: European Society for the Cognitive Sciences of Music. Trevarthen, C. 2008. “The musical art of infant conversation: Narrating in the time of sympathetic experience, without rational interpretation, before words.” In Musicae Scientiae, Special Issue “Narrative in music and interaction”. (in press), M. Imberty & M. Gratier (eds). Liège: European Society for the Cognitive Sciences of Music. Zlatev, J. this volume. “The co-evolution of intersubjectivity and bodily mimesis.”
chapter 1
Intersubjectivity What makes us human? Jordan Zlatev, Timothy P. Racine, Chris Sinha and Esa Itkonen 1.
Introduction
The title of this book, The Shared Mind, conforms to a linguistic schema – “The X Mind” – that has become common within the interdisciplinary fields of cognitive science and consciousness studies. The current volume thus stands in a line of succession from The Embodied Mind (Varela, Thompson and Rosch 1992), The Discursive Mind (Harré and Gillet 1994), The Conscious Mind (Chalmers 1996), The Extended Mind (Clark and Chalmers 1998) and The Social Mind (Valsiner and van der Veer 2000). Like some of its predecessors, The Shared Mind advances an anti-thesis to the “classical” Computational (Jackendoff 1987) or Representational Mind (Fodor 1987), with its oft-criticised neglect of the role of the body, phenomenal experience, social interaction and culture. At the same time, the present volume advances a position (or rather, a set of related positions) that has not been sufficiently explored by its predecessors. Non-human animals also have an “embodied mind”, and there are no good reasons to deny that at least birds and mammals also have a “conscious mind” (Edelman 1992). However, although other species may have varying degrees of awareness, they do not seem be fully aware of the subjectivity of others. And whereas human beings go on to engage in discursive practices and rely on material and symbolic culture, both of which have powerful formative effects on the human mind, something more ontogenetically and phylogenetically basic seems required to be able to benefit from these central aspects of human social life. This foundation seems to be provided by a uniquely human capacity for intersubjectivity. In the simplest terms, intersubjectivity is understood by the authors represented in this book as the sharing of experiential content (e.g., feelings, perceptions, thoughts, and linguistic meanings) among a plurality of subjects. Although some non-human species manifest some aspects of the capacity or capacities that make up intersubjectivity, they appear to lack others. On the other hand, no human being is entirely devoid of the human intersubjective potential – even though they
Jordan Zlatev et al.
may be delayed or challenged in the expression of some of its manifestations, such as is the case for people with autism. These considerations underlie our bold contentions that the human mind is quintessentially a shared mind and that intersubjectivity is at the heart of what makes us human.
2.
Intersubjectivity vs. “Theory of mind”
The hitherto dominant approach in psychology, cognitive science and philosophy has been to analyze what has come to be known as social cognition in terms of a “theory of mind” (or “mentalizing”) that purportedly solves the philosophical and developmental problem of “other minds”. Consider, for example, the title of a recent volume with an apparently similar theme to the present one: Other minds: How humans bridge the divide between self and others (Malle and Hodges 2005). Despite the important empirical findings and hypotheses generated by the Theory of Mind (ToM) approach, it is our contention that its framing of the research question has significantly obscured rather than clarified what needs to be explained. The basic assumptions of the ToM approach can be formulated as follows: – There is a primary separation between the self and (the minds of) others. – The individual must bridge this separation either by some form of “theory” or “simulation” of the other’s mind, a process that is more or less fallible. – The main “bodily” structures that are directly relevant for the process are those innate or acquired “modules” engaged in the inferential or simulation processes. – Cognition develops essentially “from the inside out”, with innate or acquired cognitive skills being eventually transferred or projected onto others for the purpose of explaining and predicting their behaviour. From such a point of departure, it is unsurprising that there appears to be not only a divide, but a veritable gulf between self and others, one that is so wide that it is doubtful whether it could ever truly be bridged. Such a pessimistic assessment of the human condition is hard to justify – how, if it were so, would young children be able to coordinate their basic activities with others, and eventually acquire a shared public language? How could we account for such universal forms of human experience as mutual affection and sympathy? In contrast to the four claims . In stating this we are aware that profound and multiple intellectual impairments may raise empirical questions about this claim, but we make it as a generalization with a fundamental theoretical status. We also stress that, even given cases of empirical doubt, our claim does not imply that such individuals should be thought of as not having the status of human beings.
Intersubjectivity: What makes us human?
listed above, the contributors to the present volume broadly agree on the following propositions: – Human beings are primordially connected in their subjectivity, rather than functioning as monads who need to “infer” that others are also endowed with experiences and mentalities that are similar to their own. – The sharing of experiences is not only, not even primarily, on a cognitive level, but also (and more basically) on the level of affect, perceptual processes and conative (action-oriented) engagements. – Such sharing and understanding is based on embodied interaction (e.g., empathic perception, imitation, gesture and practical collaboration). – Crucial cognitive capacities are initially social and interactional and are only later understand in private or representational terms. The main precursors and originators of these views in the last century were Husserl, Vygotsky and Wittgenstein. Husserl, the founder of phenomenology, has only recently been properly understood in the Anglo-Saxon world to be concerned not with the nature of private experience, but with structures of experience which give us a common life-world, serving as the pre-condition of any objectivity (Zahavi 2003; Moran 2005). Furthermore, he was the first modern thinker to emphasize the role of the body for the emotional tone and the perceptual richness of the life-world, and for our transparent relations with others (cf. Gallagher 2005). For example, he stated: I do not first constitute my things and my world solipsistically, then grasp by empathy the other ‘I’ which too grasps itself solipsistically as constituting its world, and then and only then, the constituted unity of both are to be identified; my self unity (Sinneinheit) exists because of the facts that the foreign multiplicity is not different from mine, it is eo ipso the same…” (Husserliana 14: 10, translated and quoted by Moran 2005: 225)
Other scholars such as Merleau-Ponty (1962), Scheler (1954) and Schutz (1966) continued this tradition and developed complementary accounts of intersubjectivity (cf. Zahavi 2001) whose common theme is that the basic forms of understanding others are not inferential, but rather direct (cf. the chapter by Gallagher and Hutto). Scheler stresses the implications of this for accounts of perception in a way that is reminiscent of J. J. Gibson’s (1979) ecological psychology: For we certainly believe ourselves to be directly acquainted with another person’s joy in his laughter, with his sorrow and pain in tears, with his shame in blushing, with his entreaty in his outstretched hands … And with the tenor of his thoughts in the sound of his words. If anyone tells me that this is not “perception”, for it
Jordan Zlatev et al.
cannot be so, in view of the fact that a perception is simply a “complex of physical sensations” … I would beg him to turn aside from such questionable theories and address himself to the phenomenological facts. (Scheler 1954, cited in Gallagher 2005: 228)
Compare Scheler also to the later Wittgenstein (1980: §570), who similarly attempted to dispel the myth of the isolated subject: “We see emotion.” – As opposed to what? – We do not see facial contortions and make the inference that he is feeling joy, grief, boredom. We describe a face immediately as sad, radiant, bored, even when we are unable to give any other description of the features.
Through enigmatic aphorisms such as “Nothing is hidden” and “Understanding is not a mental phenomenon”, Wittgenstein highlighted the essential dependence of thinking on public criteria, and concentrating on the linguistic aspect of this issue, he rendered the notion of a “private language” self-contradictory (cf. the chapter by Itkonen). Wittgenstein also drew attention to the fact that body, mind and behaviour are different aspects of the unity that we call persons. Thus, although such aspects of persons are non-identical and we therefore cannot reduce one to the other, Wittgenstein argued that they are necessarily and, hence conceptually, related and that we typically talk about one via the other. Vygotsky was a more multi-faceted thinker, creatively combining philosophy, psychology, literature, primatology and education. From a broadly Marxist perspective (though he was accused of “idealism” by the guardians of Soviet ideological orthodoxy), he famously asserted the general principle of the primacy of social interaction in the development of what he considered to be the specifically human “higher mental functions”, such as memory, reasoning and language: Every function in the child’s cultural development appears twice: first, on the social level, and later, on the individual level; first between people (interpsychological), and then inside the child (intrapsychological). This applies equally to voluntary attention, to logical memory, and to the formation of concepts. All the higher functions originate as actual relations between human individuals. (Vygotsky 1978: 57)
Like Wittgenstein, the crucial social semiotic mediational “tool” for Vygotsky was language, but he also considered the role of other semiotic resources such as artifacts and gestures (cf. the chapters by Rodríguez and Moro; Sinha and Rodríguez) in the child’s cultural development.
3.
Intersubjectivity: What makes us human?
Perspectives
Although there is a good deal of coherence between the positions of the phenomenologists, Wittgenstein and Vygotsky (as well as those of other classic theorists who feature in the discussions in the following chapters, such as Durkheim, Mead and Bateson) with respect to what they reject – that is, the notion of a monadic, individual mind, ultimately incapable of reaching out beyond its confines to the world and others – there are important differences between the positions that they advocate. In a similar vein, while all the authors represented here agree on the crucial role of intersubjectivity in human communication and consciousness of self and other, they offer (as the subtitle of this volume suggests) different answers to questions such as the following: – What is or are the precise sense or senses of the term “intersubjectivity”, at what level of organization does it exist, and how does it relate to other notions of the shared mind such as “common knowledge” (cf. the chapters by Itkonen and Sinha and Rodríguez). – More specifically, should we understand the term “intersubjectivity” as pertaining primarily to a mental or inter-mental capacity, or to the actual instances of participatory practice that both depend upon, and are instrumental in developing, this human capacity? Are these merely different aspects or emphases, or do they constitute fundamentally different perspectives (cf. the chapter by Susswein & Racine)? – To what extent is there a species-specific, biological basis for the human capacity for intersubjectivity per se, and to what extent is it the consequence of social, ecological and cultural factors (cf. the chapters by Leavens, Hopkins and Bard, Rodríguez and Moro and Sinha and Rodríguez)? – To what extent is human intersubjectivity brought about by language, and what might be the prerequisite conditions for developing or evolving a capacity for language (cf. the chapters by Gallagher and Hutto, Hutto and Zlatev)? – Does intersubjectivity involve an irreducibly mental aspect that is accessible to consciousness, or is this an (over-) attribution based on manifest behaviour (cf. the chapters by Brinck and Leavens, Hopkins and Bard)? – What aspects of human intersubjectivity (e.g., the mutual understanding between two subjects that they are attending to the same object) might play a causal role in guiding action and which are definitional rather than causal (cf. the chapter by Susswein and Racine)?
Jordan Zlatev et al.
Because they are addressed within the chapters that follow, we will not attempt to answer these questions here. But we wish to highlight the following points of (possible) disagreement between some of the authors, to which the reader may wish to pay special attention: – The chapters by Gallagher and Hutto, Barresi and Moore and Zlatev adopt stage-based accounts of the development (and evolution) of intersubjectivity, while Brinck argues for more continuous development, involving partially independent capacities. – The chapters by Pika, Zlatev and Verhagen focus on both continuities and discontinuities between animal and human intersubjectivity, while Leavens, Hopkins and Bard find support for a strong form of continuity, prior to the emergence of language. – Hobson and Hobson base their account of autism on an impairment of a specifically human capacity for identification with others, while Brinck seeks an account in which more simple skills and developmental patterns combine and interact in order to yield more general cognitive and emotional endowments. – Susswein and Racine argue that “intersubjectivity” is primarily a taxonomic term, used to group together certain kinds of social interactions which we by definition take to involve one or another form of “experiential sharing”, rather than a term denoting hidden mental or neurological processes (in contradistinction to e.g., Barresi and Moore). – Finally, while most authors adopt a definition of intersubjectivity such as the sharing and understanding of experiential content, Sinha and Rodríguez conclude the volume by stressing the primacy of participatory engagement, and the need to extend “inter-mentality” to encompass “inter-corporeality” and “inter-objectivity”. In another sense of the word ‘perspectives’, this volume brings together approaches and insights from a variety of disciplines: philosophy, linguistics, primatology, evolutionary theory, neuroscience and typical and atypical human development. The authors do not limit themselves to their disciplinary confines, a fact that strengthens the dialogic aspect of the book. Nevertheless, for the sake of perspicuity, we have organised the contributions thematically into three parts dealing primarily with Development, Evolution and Language respectively.
4.
Intersubjectivity: What makes us human?
Overview of the chapters
It is appropriate that Part I should focus on ontogenetic development, since it is largely through the path-breaking work of Colwyn Trevarthen (1979) on primary (dyadic) and secondary (triadic) intersubjectivity in the first year of life, and Daniel Stern (1985) on the “interpersonal world of the infant”, that the concept of intersubjectivity has emerged as a key issue for the contemporary sciences of the human mind. Departing from Trevarthen’s seminal work, Gallagher and Hutto present an account of the progressive emergence of intersubjective skills in childhood, arguing that understanding others requires neither a “theory theory” nor a “simulation theory” of mind, but is made possible by a sensitivity to bodily movements, gazes, facial expressions and, in secondary intersubjectivity, interactions in pragmatic contexts. In reply to claims that a theory of mind is necessary to account for the more sophisticated interactions of older children and adults involving the concepts of folk psychology, Gallagher and Hutto propose the Narrative Practice Hypothesis, according to which it is through direct encounters with stories about reasons for acting in interactive contexts with caregivers that children become familiar with the core structure of folk psychology. Gallagher and Hutto demonstrate how combining insights from the phenomenological and the Wittgensteinian traditions can yield a productive and novel account of a range of empirical findings, consistent with recent developments in neuroscience. Barresi and Moore elaborate on this neuroscientific theme, going much further than the now familiar (but rarely explanatory) references to “mirror neurons”. They present an up-dated version of their theory of social understanding called Intentional Relations Theory, consisting of four levels, through which first-person and third-person information is “matched”, yielding both self-other equivalence and differentiation. Interestingly, the two models of the development of intersubjectivity presented in these two chapters, while formulated independently, appear to be largely compatible. At the same time, the relationship between intentional relations and narrative would merit further consideration, since the latter does not seem to play a crucial role in the model offered by Barresi and Moore. Finally, Barresi and Moore outline an application of their stage or level-based model to autism, suggesting that it is precisely the inability to combine proprioceptive and sensorimotor information about the self with exteroceptive information about others that is at the core of autism spectrum disorders.
Jordan Zlatev et al.
Autism is further discussed by Hobson and Hobson, who highlight the pivotal significance of the human propensity to identify with other persons, which they suggest is compromised in children with autism. In reviewing a number of their recent studies, the authors argue that in order to share experiences, typically developing children are psychologically linked to the other person and at the same time differentiated. Hobson and Hobson argue further that these early forms of sharing, and the varieties of communication that they support, provide the foundations for the conceptual understanding of self and other that emerges around the middle of the second year of life. Methodologically, the authors show that our human capacity for reflective intersubjectivity is necessary for analysts to be able to make their judgments in the coding of “overt behaviour”, and that such rating is itself a “second-person” phenomenon, rather than a matter of detached observation. Rodríguez and Moro also adopt the use of a qualitative, “clinical” method in analyzing parent-infant triadic interactions. They draw on Vygotsky and Wittgenstein to show that long before a child is able to produce his or her first symbolic and ostensive uses of objects, the first pointing gestures and the first words, the adult acts with the child as a “symbol maker”. That is, the adult produces ostensive actions with objects, points to them to make clear his or her intentions, using them in a canonical manner and talking to the child almost constantly. Rodríguez and Moro argue that this implies a long process and a variety of levels of intersubjective adult-baby “agreement” on the use of objects, which serves as a precondition for subsequent social and cognitive development. On a more general level they argue against the assumption (basic to ToM approaches) of the “opacity” of social reality. Brinck proposes a conceptual clarification of intersubjectivity as “the sharing of experiences”, suggesting that it divides into a multitude of sub-concepts depending on the understanding of the terms “sharing” and “experiences”. She distinguishes, following Stern (1985), three kinds of shared experiences: emotion, attention, and intention. However, Brinck argues for a more external, behaviourbased way of defining them (thus implicitly disagreeing methodologically with Hobson and Hobson). Brinck argues that different combinations of these forms of intersubjectivity enable corresponding intentionally communicative behaviours, providing a novel explanation of why intentional communication first appears at the end of the first year, despite the fact that its “ingredients” are manifest much earlier. It is, she claims, the ability to decontextualize and combine the various intersubjective skills in a flexible manner that underlies the emergence of intentional communication, rather than this being the culmination of a stage model. Susswein and Racine take a more deflationary approach than the preceding chapters, exploring the distinction between causal and definitional issues in order
Intersubjectivity: What makes us human?
to distinguish between explanations of what an organism is doing and how they are doing it, as well as between different types of causal explanation. They argue that intersubjectivity is a taxonomic rather than a causal explanatory concept, i.e., a technical concept used to classify interactive behaviours and abilities rather than to denote vehicles or causes of those behaviours and abilities. They critically examine the idea that intersubjective engagement involves the sharing of mental states, and argue that the role of mental states and experience in intersubjective engagement is often misconstrued. Finally they apply this approach to human activity to distinguish reflective and practical understanding and consider the meaning of “declarative” pointing in early childhood. Part II of the volume deals with the capacities for intentional communication and intersubjectivity of non-human primates, and especially of our closest relatives in the animal kingdom, the Great apes; and more generally, with the question of the evolution of human intersubjectivity. Comparative psychology is a field rife with controversies, with persistent disagreement between those who emphasize discontinuities and those who argue for (strong) continuity between non-human and human capacities. This debate is reflected by the diverging positions taken by the first two chapters in this section. Pika compares the communicative gestures of bonobos, chimpanzees, gorillas and orangutans and shows that Great apes have multifaceted gestural repertoires. She demonstrates that these gestures are performed in multiple contexts and are used flexibly, but unlike those of human children seem to be learned mainly via an individual learning process. Being intentional and referential acts, the gestures of the Great apes display at least a nascent understanding of intentionality, and can plausibly have served as a stepping stone in the evolution of language. On the other hand, Pika also emphasizes the differences between human and ape gestures, suggesting an innate bias for human cultural learning. In contrast, in analyzing the pointing gesture, Leavens, Hopkins and Bard argue against such an innate bias on the basis of the fact that although both captive and wild apes are sampled from the same gene pool, captive apes spontaneously point without overt training whereas wild apes almost never point. The authors review empirical evidence of the development of pointing in apes and in human children, highlighting the significance of species-typical motor development, culturally-specific patterns of child restraint, and stages of cognitive development for the development what they call “explicit reference”. Their major claim is that a capacity for explicit reference emerges in our nearest living relatives when they experience similar circumstances to those of human children, involving the rich emotional contours of human affectivity. Zlatev distinguishes between five evolutionary (and developmental) levels of intersubjectivity and suggests that the latter has co-evolved with bodily mimesis:
10
Jordan Zlatev et al.
the use of the body for communicative and representational purposes. He reviews evidence from primatology to suggest that feral and captive apes are at least to a degree capable of the first two levels (involving e.g., empathy, shared attention and imitation), but not of the third level, “triadic mimesis”, which involves an understanding of communicative intentions. Enculturated, language-trained apes, on the other hand, show some aspects of triadic mimesis, suggesting how our predecessors could have bootstrapped themselves to this level without language and may have inherited a biological bias for it. The emergence of language, on the other hand, opened the way to higher levels, allowing (consistently with the proposal of Gallagher and Hutto in Part I) the understanding of “beliefs” and the use of folk psychology. Hutto develops a similar evolutionary scenario in more detail, arguing that an innate “theory of mind” mechanism is neither necessary nor sufficient to account for the evolutionary course of hominid social interaction. Such a purported “module”, he argues, fails to explain the remarkable technical advances and the imitative capacities of Homo ergaster/erectus lying somewhere between those of apes and modern humans. Hutto considers the evidence instead for a Mimetic Ability Hypothesis and invokes the notion of re-enactive imagination in determining what this ancient adaptation was likely to have involved. He argues that this hypothesis provides a better explanation of the kinds of intersubjectivity that would have been necessary for the development of language, thereby undercutting the strongest argument for positing the existence of innate theory of mind modules, namely the support they lend to intention-based semantics. Part III takes up and develops the theme of the relationship between intersubjectivity and language. It is widely acknowledged that language requires intersubjectivity, most straightforwardly because one needs to know what another is referring to in order to learn a language. But this consensus conceals a good deal of debate concerning the nature of language and its relation to thought and consciousness, and the extent to which language might stretch and reshape our basic intersubjective capacities. Itkonen opens this section by focusing on the central role of normativity for language and linguistics, an issue discussed in brief by Zlatev in Part II, as well as the other contributors of this section. He supports his argument by Wittgenstein’s “private language argument”, showing that language is impossible without public criteria of correctness. Itkonen further presents and defends the “ontology of the social” as third-level common knowledge, and suggests a dialectical synthesis between “collectivism” and “individualism”: common knowledge consist of mental states in particular configurations, being therefore both based on and irreducible to (individual) consciousness. Finally, he spells out a number of ramifications of his arguments for theoretical and empirical linguistics and concludes by
Intersubjectivity: What makes us human?
inpointing a priori (cf. “devout”: Honderich 2006) physicalism (“there is nothing p but matter and energy fields”) as the roots for the anti-normative bias in linguistics and related fields. Verhagen argues that intersubjectivity is systemically encoded in natural languages, contrasting this view with the standard “informational” conception of linguistic communication. He reviews constructions at different levels of grammatical organization to show that linguistically coded relations of intersubjective coordination exhibit the specific character of being “rhetorical” or “argumentative”. He argues that although these forms may result in cooperation, their immediate function is to influence another person’s mind and to make discourse proceed in a particular direction. Because of this, he suggests that language is to some extent analogous with other animal communication systems, which also involve the management and assessment of other organisms, notably conspecifics. On the other hand, he argues for an important discontinuity, in that such management and assessment is oriented in human beings to other minds, rather than to immediate behavior. Janzen and Shaffer take this theme into a different context, noting that because interlocutors make constant assumptions about what information is active within each other’s consciousness, this process creates an interesting challenge in third-party interpretation from one language to another. Their major point is that when an interpreter is introduced into a discourse event this affects the nature of the interchange, since the interpreter will also make assumptions about each of the interlocutors’ knowledge states. They use examples from ASL-English interpretation and discuss the notion of “expansions”, which have been claimed to be grammatically required in ASL. The authors persuasively argue against this claim. Rather they suggest that understanding contextualization as a successful discourse strategy provides a more appropriate approach to the expression of shared and non-shared knowledge and discuss the implications of contextualization for the notion of intersubjectivity in general. Sinha and Rodríguez end the book where this introductory chapter began, by contrasting the approach to the human mind (and social cognition) based on intersubjectivity with that of “theory of mind”, arguing for the advantages of the former. By combining Durkheim’s notion of social facts with that of Searle (1995) they argue for the irreducibility of the social to the individual, in a way that converges with the chapter by Itkonen. They also argue against a purely mentalistic conception of intersubjectivity, as a property of the “unmediated mind”. Sinha and Rodríguez point out the need to consider inter-corporeality, which extends beyond the body to encompass “inter-objectivity”. Objects, they claim, are not only referents of language, but signifiers in their own right, and it is through participatory engagement with the social and material world that children enter the
11
12
Jordan Zlatev et al.
realm of language. Their chapter concludes by connecting with Gallagher and Hutto’s Narrative Practice Hypothesis, stressing the importance of narrativity in the construction of both complex human cognition and shared social cultural identity.
5.
Conclusion
As is obvious from these summaries, the “perspectives” expressed in these chapters do not converge on a single, univocal notion of intersubjectivity, but rather point to a complex phenomenon, or a set of related phenomena, in which experiential, behavioural, genetic and neural processes and levels are interwoven in both potentiating and actualizing “what it means to be human”. We hope that this introductory chapter has conveyed to the reader our enthusiasm in working on this interdisciplinary and “intersubjective” project. The increasing body of research on social cognition in developmental and comparative psychology, and the prefixing of the term “social” to previously rather individual-oriented fields such as cognitive linguistics and cognitive neuroscience, reflect a changing intellectual context in which we hope that The Shared Mind will make a significant contribution to rethinking some of the fundamental questions of our fields. Such a rethinking is an essential, but radically challenging enterprise. The conceptual difficulties encountered by the dominant tradition in the cognitive sciences in attempting to explain the nature of “social cognition”, language and communication are not accidental. They stem from the epistemological and methodological individualism inherited from the “possessive individualist” cast of Western culture (and capitalism), and the dominant position accorded in this tradition to natural science and technology vis-a-vis the humanities and social sciences (Macintyre 1997; Taylor 1989). Three of us have collaborated for over 15 years, and our conversations have often revolved around the need for a coherent theoretical statement addressing the centrality of intersubjectivity and normativity for linguistics, psychology and cognitive science. The press of other work, and the difficulty of integrating the various perspectives, has led to the repeated postponement of that venture. In the meantime, however, both the need for, and the timeliness of, a book like the present volume has become increasingly evident. Tim Racine joined the group at the Jean Piaget Society meeting in Vancouver in the summer of 2005, where the idea for the book emerged from two symposia in which half of the authors represented in the volume participated. Reading (and writing) a large number of drafts of the chapters of this book, and actively
Intersubjectivity: What makes us human?
c omparing the points of view of the entire volume, helped us, as editors, realize how much our agreement outweighs whatever differences remain. Although this book is a “polyphonic” enterprise, the voices of the different chapters do not always blend harmonically. This is only to be expected in addressing such a quintessentialy interdisciplinary topic as intersubjectivity. In this introductory chapter, we have attempted to show the reader both how a focus on intersubjectivity offers a different (and we believe more productive) perspective to social cognition than the “theory of mind” approach, and to highlight some of the controversies within this approach, thereby contributing to defining prospects for further empirical and conceptual investigations. Without meaning to seem unduly naïve, we offer this volume to the reader as a source for reflection on human nature, and on the possibilities for good and ill that are potentiated by the Shared Mind in our shared world.
References Chalmers, D. 1996. The Conscious Mind. Oxford: Oxford University Press. Clark, A. and Chalmers, D.J. 1998. “The extended mind.” Analysis 58: 10–23. Edelman, G. 1992. Bright Air, Brilliant Fire: On the Matter of the Mind. New York: Basic Books. Fodor, J.A. 1987. Psychosemantics; The Problem of Meaning in the Philosophy of Mind. Cambridge, MA: MIT Press. Gallagher, S. 2005. How the Body Shapes the Mind. Oxford: Oxford University Press. Gibson, J.J. 1979. The Ecological Approach to Visual Perception. Boston, Houghton Mifflin. Harré, R. and Gillet, G. 1994. The Discursive Mind. London: Sage Publications. Honderich, T. 2006. “Radical externalism.” Journal of Consciousness Studies 13 (7–8): 2–13. Jackendoff, R. 1987. Consciousness and the Computational Mind. Cambridge, Mass.: MIT Press. Merleau-Ponty, M. 1962 [1945]. Phenomenology of Perception. London: Routledge and Kegan Paul. MacIntyre, A. 1997. After Virtue. London: Duckworth. Malle, B.F. & Hodges, S.D. (eds.). 2005. Other Minds. New York: Guilford Press. Moran, D. 2005. Edmund Husserl: Founder of Phenomenology. Cambridge: Polity Press. Searle, J.R. 1995. The Construction of Social Reality. New York, NY: The Free Press. Scheler, M. 1954 [1913]. The Nature of Sympathy (translated by P. Heath). Hamden, CT: Archon Books. Schutz, A. 1966. Collected Papers III: Studies in Phenomenological Philosophy. The Hague: Martinus Nijhoff. Stern, D. 1985. The Interpersonal World of the Infant. New York, NY: Basic Books. Taylor, C. 1989. Sources of the Self: The Making of Modern Identity. Cambridge: Cambridge University press. Trevarthen, C. 1979. “Communication and cooperation in early infancy. A description of primary intersubjectivity.” In Before Speech: The Beginning of Human Communication, M. Bullowa (ed.), 99–136. London: Cambridge University Press.
13
14
Jordan Zlatev et al.
Valsiner, J. and van der Veer, R. 2000. The Social Mind: Construction of the Idea. Cambridge, UK: Cambridge University Press. Varela, F., Thompson, E. and Rosch, E. 1992. The Embodied Mind: Cognitive Science and Human Experience. Cambridge, MA: MIT Press. Vygotsky, L. 1978. Mind in Society. The Development of Higher Psychological Processes. Cambridge, MA: Harvard University Press. Wittgenstein, L. 1980. Remarks on the Philosophy of Psychology, Volume 2 (translated by C.G. Luckhardt and M.A.E. Aue). Oxford: Blackwell. Zahavi, D. 2001. “Beyond empathy. Phenomenological approaches to intersubjectivity.” Journal of Consciousness Studies 8: 151–167. Zahavi, D. 2003. Husserl’s Phenomenology. Stanford, CA: Stanford University Press.
part i
Development
chapter 2
Understanding others through primary interaction and narrative practice Shaun Gallagher and Daniel D. Hutto We argue that theory-of-mind (ToM) approaches, such as “theory theory” and “simulation theory”, are both problematic and not needed. They account for neither our primary and pervasive way of engaging with others nor the true basis of our folk psychological understanding, even when narrowly construed. Developmental evidence shows that young infants are capable of grasping the purposeful intentions of others through the perception of bodily movements, gestures, facial expressions etc. Trevarthen’s notion of primary intersubjectivity can provide a theoretical framework for understanding these capabilities and his notion of secondary intersubjectivity shows the importance of pragmatic contexts for infants starting around one year of age. The recent neuroscience of resonance systems (i.e., mirror neurons, shared representations) also supports this view. These ideas are worked out in the context of an embodied “Interaction Theory” of social cognition. Still, for more sophisticated intersubjective interactions in older children and adults, one might argue that some form of ToM is required. This thought is defused by appeal to narrative competency and the Narrative Practice Hypothesis (or NPH). We propose that repeated encounters with narratives of a distinctive kind is the normal route through which children acquire an understanding of the forms and norms that enable them to make sense of actions in terms of reasons. A potential objection to this hypothesis is that it presupposes ToM abilities. Interaction Theory is deployed once again to answer this by providing an alternative approach to understanding basic narrative competency and its development.
1.
Introduction
Our intention in this chapter is to explicate an account of how we come to understand others, without appealing to the dominant theory-of-mind (ToM) approaches of “theory theory” (e.g., Leslie 1987; Gopnik 1993) or “simulation theory” (e.g., Gordon 1986; Goldman 2002). We have elsewhere provided good reasons to
18
Shaun Gallagher and Daniel D. Hutto
doubt that either of these theories can give an accurate or adequate account of our everyday intersubjective abilities for understanding the intentions and the behaviours of other persons (see Gallagher 2001, 2004, 2006, 2007a, 2007b; Hutto 2004, 2005, 2006a, 2007a, 2007b, 2008). We will briefly summarize that critique here, but our main purpose is to set out a more positive account of just these everyday intersubjective abilities and show that they are not reducible (or inflatable) to the mind-reading or mentalizing described by approaches to social cognition which presume a “theory-of-mind”. This positive account involves three kinds of processes which together are sufficient to deliver the nuanced adult capacity for understanding (as well as for mis-understanding) others. These processes include (1) intersubjective perceptual processes, (2) pragmatically contextualized comprehension, and (3) narrative competence. We argue on the basis of evidence from developmental psychology that the capacity for understanding others is, on average, well established by the time the child reaches four or five years of age, and that it continues to be enriched on the basis of further experience as we become mature adults.
2.
A brief critique of the dominant approaches to social cognition
Theory theory (TT) and simulation theory (ST), the standard and dominant approaches to social cognition, share the important supposition that when we attempt to understand the actions of others, we do so by making sense of them in terms of their mental processes to which we have no direct access. That is, we attempt to “mind read” their beliefs, desires, and intentions, and such mind reading or mentalizing is our primary and pervasive way of understanding their behaviour. Furthermore, both TT and ST characterize social cognition as a process of explaining or predicting what another person has done or will do. TT claims that we explain another person’s behaviour by appealing to either an innate or acquired “theory” of how people behave in general, a theory that is framed in terms of mental states (e.g., beliefs and desires) causing or motivating behaviour. ST claims that we have no need for a theory like this, because we have a model, namely, our own mind, that we can use to simulate the other person’s mental states. We model others’ beliefs and desires as if we were in their situation. Claims that such theory or simulation processes are explicit (conscious) are dubious from a phenomenological point of view. That is, if in fact such processes are primary, pervasive, and explicit, they should show up in our experience – in
Primary interaction and narrative practice
the way that we experience others – and they rarely do. The phenomenological critique also rejects the idea, clearly found in TT, that our everyday dealings with others involve an observational, third-person stance toward them – observing them and trying to come up with explanations of their behaviour. Rather, our everyday encounters with others tend to be second-person and interactive. Claims that the processes described by TT or ST are implicit (or not explicitly conscious) run into a different set of objections. In the case of TT, there is no evidence that such processes are implicit, or even clarity about what precisely that means. Moreover, although TT appeals to false-belief experiments, such experiments are set up to test for explicit rather than implicit theory-of-mind processes (Gallagher 2001) – subjects are asked to explicitly consider the meanings of an observed third-party’s behaviour. Implicit approaches to ST appeal to the neuroscience of mirror neurons and shared representations (cf. Barresi and Moore this volume), but there is no justification for calling these subpersonal processes “simulation”, since according to ST, simulation involves the instrumental use of a first-person model to form third-person “as if ” or “pretend” mental states. In subpersonal processes, (1) there is no first- or third-person (activation of mirror neurons, for example, are considered to be “neutral” in regard to who the agent is) (see e.g., deVignemont 2004; Gallese 2005; Hurley 2005; Jeannerod and Pacherie 2004); (2) nothing (or no one) is using a model; and (3) neuronal processes cannot pretend. As vehicles neurons either fire or they don’t. More importantly, in terms of relevant content, if they are neutral with respect to first- and third-person, pretence in just these terms (I pretend to be you) is not possible. In effect, simulation, as defined by ST, is a personal-level concept that cannot be legitimately applied to subpersonal processes. . This is not to deny that in some circumstances, for example, in observing puzzling cases of another person’s behaviour, we may in fact explicitly appeal to theory or employ simulation. The claim here is simply that most of our everyday interactions are not of this sort. Puzzling cases are the exception. . Goldman and Sripada (2005: 208), acknowledging the discrepancy between the ST definition of simulation and the working of subpersonal mirror processes, propose a minimal definition of simulation: “Applied to mindreading, a minimally necessary condition [for simulation] is that the state ascribed to the target is ascribed as a result of the attributor’s instantiating, undergoing, or experiencing, that very state. In the case of successful simulation, the experienced state matches that of the target”. If this is a necessary condition, it cannot be a sufficient one, because on this minimal definition and without something further, it’s not clear what would motivate me to ascribe the state that I was undergoing to someone else. Furthermore, if this were as automatic as mirror neurons firing, then it would seem that we would not be able to attribute a state different from our own to someone else. But we do this all the time. Practically speaking, this proposal also raises puzzles about interacting with more than one other person.
19
20 Shaun Gallagher and Daniel D. Hutto
In addition to these phenomenological and logical objections to TT and ST, there is good evidence from developmental psychology that our ability to understand others emerges much earlier than TT or ST would predict. An objection can also be raised against the idea that a general theory (folk psychology) would have sufficient explanatory power to explain the particularities of a large diversity of behaviours found in everyday life, or that it could be very reliable in the face of multiple possibilities for motivation. Similarly it has been objected that running a first-person simulation routine, that is, a process that is based on one’s own mental states, seems inadequate to explain the diversity of behaviours found in the world. These objections throw doubt on TT and ST approaches. The question, however, is whether there is a positive account that can avoid these objections. We turn now to the construction of that alternative account, in three parts: intersubjective perception, pragmatically contextualized comprehension, and narrative competency.
3.
Intersubjective perception and interaction
Long before the child reaches the age of four, the capacities for human interaction and intersubjective understanding are already accomplished in certain embodied practices – practices that are emotional, sensory-motor, perceptual, and nonconceptual. These practices include proto-mimesis (Zlatev, this volume), imitation, the parsing of perceived intentions (Baldwin, Baird, Saylor and Clark 2001), emotional interchange (Hobson 2004), and generally the processes that fall under the heading of primary intersubjectivity (Trevarthen 1979). These embodied practices constitute our primary access for understanding others, and they continue to do so even after we attain our more sophisticated abilities in this regard (Gallagher 2001). In most intersubjective situations, that is, in situations of social interaction, we have a direct perceptual understanding of another person’s intentions because their intentions are explicitly expressed in their embodied actions and their expressive behaviors. This understanding does not require us to postulate or infer a belief or a desire hidden away in the other person’s mind. What we might Is it possible to simulate the neural/mental/emotional states of two other people at the same time if in fact our simulations must be such that we instantiate, undergo, or experience, those two (possibly very different) states? (see Gallagher 2007b). We suggest that these issues would also have to be addressed by Barresi and Moore (this volume) in order to clarify their proposal for a matching system.
Primary interaction and narrative practice
r eflectively or abstractly call their belief or desire is expressed directly in their actions and behaviours. This phenomenologically direct understanding is likely made possible by the above mentioned complex neuronal processes described as the mirror neuron system(s) and shared representations. In contrast to interpreting these neuronal resonance processes as implicit simulations, which on the functional level would involve cognitive processes over and above the perception of action, Gallagher (2005, in press) has argued that they in fact instantiate a form of enactive social perception. A primary, perceptual sense of others is already implicit in the behaviour of the newborn. In neonate imitation, which depends not only on a contrast, in some sense, between self and non-self, and a proprioceptive sense of one’s own body, but also a responsiveness to the fact that the other is of the same sort as oneself (Bermúdez 1996; Gallagher 1996; Gallagher and Meltzoff 1996), infants are able to distinguish between inanimate objects and people. The fact that they imitate only human faces (see Legerstee 1991; Johnson 2000; Johnson et al. 1998) suggests that infants are able to parse the surrounding environment into those entities that perform human actions (people) and those that do not (things) (Meltzoff and Brooks 2001). An intermodal tie between a proprioceptive sense of one’s body and the face that one sees is already functioning at birth. For the infant, the other person’s body presents opportunities for action and expressive behaviour – opportunities that it can pursue through imitation. There is, in this case, a common bodily intentionality that is shared by the perceiving subject and the perceived other. From early infancy humans, and perhaps some animals (see e.g., the studies by Myowa-Yamakoshi 2001; Myowa-Yamakoshi et al. 2004; also cited by Zlatev this volume) have capabilities for primary-intersubjective interaction with others. The early capabilities that contribute to primary intersubjectivity constitute an immediate, non-mentalizing mode of interaction. Infants, notably without the intervention of theory or simulation, are able to see bodily movement as goal-directed intentional movement, and to perceive other persons as agents. This does not require advanced cognitive abilities; rather, it is a perceptual capacity that is “fast, automatic, irresistible and highly stimulus-driven” (Scholl and Tremoulet 2000: 299). Evidence for this early, non-mentalizing interpretation of the intentional actions of others can be found in numerous studies. Baldwin and colleagues, for example, have shown that infants at 10–11 months are able to parse some kinds of continuous action according to intentional boundaries (Baldwin and Baird 2001; Baldwin et al. 2001). The infant follows the other person’s eyes, and perceives various movements of the head, the mouth, the hands, and more general body movements as meaningful, goal-directed movements. Such perceptions give the infant, by the end of the first year of life, a non-conceptual,
21
22
Shaun Gallagher and Daniel D. Hutto
action-based understanding of the intentions and dispositions of other persons which does not involve inferences about beliefs or desires understood as mental states (Allison, Puce and McCarthy 2000; Baldwin 1993; Johnson 2000; Johnso, Slaughter and Carey 1998). Primary intersubjectivity also includes affective coordination between the gestures and expressions of the infant and those of caregivers with whom they interact. Infants “vocalize and gesture in a way that seems ‘tuned’ [affectively and temporally] to the vocalizations and gestures of the other person” (Gopnik and Meltzoff 1997: 131). Infants at 5 to 7 months detect correspondences between visual and auditory information that specify the expression of emotions (Walker 1982). The perception of emotion in the movement of others, however, does not involve taking a theoretical stance or creating a simulation of some inner state. It is a perceptual experience of embodied comportment (Bertenthal, Proffitt and Cutting 1984; Moore, Hobson and Lee 1997). This kind of perception-based understanding, therefore, is not a form of mind-reading. In seeing the actions and expressive movements of the other person one already sees their meaning; no inference to a hidden set of mental states (beliefs, desires, etc.) is necessary. The capabilities involved in primary intersubjectivity suggest that before we are in a position to wonder what the other person believes or desires, we already have specific perceptual understanding about what they feel, whether they are attending to us or not, whether their intentions are friendly or not, and so forth. There is, in primary intersubjectivity, a common bodily intentionality that is shared across the perceiving subject and the perceived other. As Gopnik and Meltzoff indicate, “we innately map the visually perceived motions of others onto our own kinesthetic sensations” (1997: 129), and the evidence from recent research on mirror neurons and resonance systems in social neuroscience supports this. Thus, before we are in a position to theorize, simulate, explain or predict mental states in others, we are already in a position to interact with and to understand others in terms of their expressions, gestures, intentions, and emotions, and how they act toward ourselves and others. Furthermore, primary intersubjectivity is not primary simply in developmental terms. Rather it remains primary across all face-to-face intersubjective experiences, and it underpins those developmentally later, and occasional, practices that may involve explaining or predicting mental states in others (see e.g., Stern’s (1985) idea of a “layered model” in which
. In citing Gopnik and Meltzoff ’s claim about the necessity for innate mappings we are not thereby endorsing their theory-theoretic construal of what this involves. Indeed, much of the evidence developed by Meltzoff and cited by Gopnik and Meltzoff supports the idea of a strong intersubjective perceptual capacity in the infant.
Primary interaction and narrative practice
evelopmentally primary understandings are not “superseded” but remain and d operate in parallel to more advanced ones).
4.
Pragmatic intersubjectivity
If human faces are especially salient, even for the youngest infants, or if we continue to be capable of perceptually grasping the meaning of the other’s expressions and intentional movements, such face-to-face interaction does not exhaust the possibilities of intersubjective understanding. Expressions, intonations, gestures, and movements, along with the bodies that manifest them, do not float freely in the air; we find them in the world, and infants soon start to notice how others interact with the world. When infants begin to tie actions to pragmatic contexts, they enter into what Trevarthen calls ‘secondary intersubjectivity’. Around the age of 1 year, infants go beyond the person-to-person immediacy of primary intersubjectivity, and enter into contexts of shared attention – shared situations – in which they learn what things mean and what they are for (see Trevarthen and Hubley 1978). Behaviour representative of joint attention begins to develop around 9–14 months (Phillips, Baron-Cohen and Rutter 1992). In such interactions the child looks to the body and the expressive movement of the other to discern the intention of the person or to find the meaning of some object. The child can understand that the other person wants food or intends to open the door; that the other can see him (the child) or is looking at the door. This is not taking an intentional stance, i.e., treating the other as if they had desires or beliefs hidden away in their minds; rather, the intentionality is perceived in the embodied actions of others. They begin to see that another’s movements and expressions often depend on meaningful and pragmatic contexts and are mediated by the surrounding world. Others are not given (and never were given) primarily as objects that we encounter cognitively, or in need of explanation. We perceive them as agents whose actions are framed in pragmatic contexts. It follows that there is not one uniform way in which we relate to others, but that our relations are mediated through the various pragmatic circumstances of our encounters. Indeed, we are caught up in such pragmatic circumstances, and are already existing in reference to others, from the very beginning (consider for example the infant’s dependency on others . Of course, the fact that another’s feelings can be hidden is completely consistent with expressivism of this sort. As Wittgenstein says “One can say He is hiding his feelings. But that means that it is not a priori they are always hidden” (Wittgenstein 1992: 35e). The point is that our initial, basic engagements with others are not estranged, even if sophisticated creatures like us are capable of hiding or faking their emotions.
23
24
Shaun Gallagher and Daniel D. Hutto
for nourishment), even if it takes some time to sort out which agents provide sustenance, and which ones are engaged in other kinds of activities. As we noted, children do not simply observe others; they are not passive observers. Rather they interact with others and in doing so they develop further capabilities in the contexts of those interactions. If the capacities of primary intersubjectivity, like the detection of intentions in expressive movement and eye direction, are sufficient to enable the child to recognize dyadic relations between the other and the self, or between the other and the world, something more is added to this in secondary intersubjectivity. As noted, in joint attention, beginning around 9–14 months, the child alternates between monitoring the gaze of the other and what the other is gazing at, checking to verify that they are continuing to look at the same thing. Indeed, the child also learns to point at approximately this same time. Eighteen-month-old children comprehend what another person intends to do with an instrument in a specific context. They are able to re-enact to completion the goal-directed behaviour that someone else fails to complete. Thus, the child, on seeing an adult who tries to manipulate a toy and who appears frustrated about being unable to do so, quite readily picks up the toy and shows the adult how to do it (Meltzoff 1995; Meltzoff and Brooks 2001). Our understanding of the actions of others occurs on the highest, most appropriate pragmatic level possible. That is, we understand actions at the most relevant pragmatic (intentional, goal-oriented) level, ignoring possible subpersonal or lower-level descriptions, and also ignoring interpretations in terms of beliefs, desires, or hidden mental states. Rather than making an inference to what the other person is intending by starting with bodily movements, and moving thence to the level of mental events, we see actions as meaningful in the context of the physical and intersubjective environment. If, in the vicinity of a loose board, I see you reach for a hammer and nail, I know what your intentions are as much from the hammer, nail, and loose board as from anything that I observe about your bodily expression or postulate in your mind. We interpret the actions of others in terms of their goals and intentions set in contextualized situations, rather than abstractly in terms of either their muscular performance or their beliefs. The environment, the situation, or the pragmatic context is never perceived neutrally (without meaning), either in regard to our own possible actions, or in regard to the actions and possibilities of others. As Gibson’s theory of affordances . Our understanding of the performance of mimes who work without props depends on their excellent ability to express intentions in their movements, but also on our familiarity with contexts. The mime’s talent for expressive movements is clearly demonstrated in contrast to what we often experience in the game of charades or pantomime when we haven’t a clue about what the player is trying to represent.
Primary interaction and narrative practice
(e.g., Gibson 1979) suggests, we see things in relation to their possible uses, and therefore never as a disembodied observer. Likewise, our perception of the other person, as another agent, is never of an entity existing outside of a situation, but rather of an agent in a pragmatic context that throws light on the intentions (or possible intentions) of that agent. Theory-of-mind approaches, which involve theory (as an application of folk psychology) or simulation, and which focus on the acquisition of the concept of mental states (like belief) around age 3 or 4 years, miss some basic and important capacities for social cognition. Yet, the acknowledgement of capabilities for understanding others that define primary and secondary intersubjectivity – the embodied, sensory-motor (emotion informed) capabilities that enable us to perceive the intentions of others (from birth onward), and the perceptual and action capabilities that enable us to understand others in the pragmatically contextualized situations of everyday life (from 12–18 months onward) – is not sufficient to address what are clearly new developments around the ages of 2, 3 and 4 years. The “elephant in the room” around the age of 2 years is, of course, language. But if language development itself is something that depends on some of the capabilities of primary and secondary intersubjectivity, language also carries these capabilities forward and puts them into service in much more sophisticated social contexts (on this point, from a different perspective, also see Zlatev this volume). Do children, upon passing explicit false-belief tests, acquire the final conceptual component needed for their mature understanding of reasons, as is the pervasive claim in the theory-of-mind literature? Or does their newfound understanding of false belief simply equate to a capacity to recognize that the other (whether Maxie, or Sally-Ann, or Snoopy, etc.) has a divergent point of view from their own, and no more? And, what lies at the root of this sort of understanding? Is this sort of mastery of the concept of belief a natural consequence of the maturation of theory-of-mind modules, grounded in introspective acts of ostensive denotation or the product of extensive, evidence-based theorizing on their part? We propose that none of these proposals hold up well under close scrutiny (see Hutto 2008: Chs. 9 and 10). If so it is more plausible to think that an understanding of divergent cognitive perspectives is the result of children beginning to participate in conversations of the kind that require recognition of conflicting points of view. This sort of activity can be seen as a natural extension of those forms of imaginative pretend play that require children to occupy different character roles and adopt personas that are different to their own (Hutto 2008: Ch. 7). A child’s initial understanding of the concept of belief is likely to depend on many things but it is notable that many false-belief tests are presented in the form of a narrative and could be interpreted as tests for a certain level of narrative competency. It also worth observing that the strongest data concerning successful
25
26 Shaun Gallagher and Daniel D. Hutto
false-belief performance stems from experiments conducted almost entirely on European and American subjects, whose early lives are awash with folk psychological narratives encountered in fairy tales, children books, comic books, television and films (Richner and Nicolopoulou 2001: 408; Nelson 2003: 22). The form, content and focus of the stories and storytelling practices are much the same in these cultures. Indeed, they even share many of the same canonical ‘texts’. Even more important, we must ask, what role does this mature understanding of false-belief play in the lives of children? And, what drives its development and facilitates its incorporation into larger explanatory schemas of explicitly making sense of actions in terms of reasons (in which attributions of belief plays an important but nevertheless limited part)? In addressing these questions it is vital to be aware, as Carpendale and Lewis (2004: 91) stress, that: Proponents of the dominant theories have been notably quiet about what happens in development after the child’s fifth birthday. However research that explores whether 5-year-olds can use simple false belief knowledge to make inferences about their own and other’s perspectives finds that they singularly fail to do so.
5.
Making sense of reasons
The ability and motivation to use one’s knowledge of false belief in a wider explanatory context, it seems, is late-developing. It comes into play only after children gain an explicit, practical mastery of the concept of belief. This suggests that false belief understanding is not the crowning moment in their early understanding of other minds; children must develop further still if they are to make sense of actions in terms of reasons. What does this involve? Let’s focus on an example. Someone might ask: Why is Laura going to India? If I don’t really know Laura, and if I’ve never heard her say why she is going to India, then I may attempt to get at her reasons in the third-person. This is surely something we do regularly. This sort of speculative attempt at folk-psychological explanation might run as follows. Laura is a young, American college student. Why do young American college students travel to India? Laura, like many young American college students, may believe that India is a romantic place and that she can learn about Eastern meditation practices there and have an adventure. So Laura might desire to go to India for such reasons. One reaches this conclusion by calling on background knowledge – general knowledge or beliefs about what American college students tend to think and value as well as one’s knowledge and beliefs about widely held beliefs about India. The attributed reason may be correct
Primary interaction and narrative practice
or incorrect in Laura’s case, but lacking detailed information about Laura, one is forced to appeal to generalizations informed by knowledge of an impersonal sort. Two things are worthy of note. First, this kind of speculation is not likely to be very reliable in most interesting cases. Second, there is no obvious reason to think that the background knowledge or beliefs in question is theoretical. To say that one is operating with theories about India and theories about the belief-forming tendencies of American students in such cases is surely to stretch the notion of theory beyond reasonable limits. Let’s modify the example slightly. If I know Laura, but do not know precisely why she is going to India, I will be able to make a more informed guess about her reasons. Laura is the kind of person who really wants to help children in the third world, so that is probably why she is going to India. I will have learnt this about her from my previous exchanges with her or on the basis of what others have told me about her. In this case too, my attribution is knowledge-based but the knowledge in question this time is particular and personal. Although, again, hardly theoretical my attribution remains speculative and suppositional. Here’s a third case. Knowing Laura I may already know her reason for going to India or I might get at it by a much more reliable means. I may know why she is going because she may have already told me so. If not, I could always ask her. Of course, she may be lying or self-deceived, but even acknowledging those possibilities direct conversation is undeniably the most secure route to her reasons. It is important to stress that in each of these cases the capacity to understand why Laura acted (or might have acted), and our ability to digest these answers is framed by the activity of checking to see if her reason, as it were, makes sense. Guessing at or learning of a person’s reason is only a small part of the story of our everyday understanding of why others act. It is also necessary to situate and evaluate reasons in wider contexts and against certain normative assumptions. Would it make sense for anyone go to India for that sort of reason? In particular, does it make sense for Laura to go? Is doing so in line with her character, her larger ambitions, her existing projects, or her history? What does it say about her? Does it make her a generous person, an idealist or merely naïve? Understanding reasons for action demands more than simply knowing which beliefs and desires have moved a person to act. To understand intentional action requires contextualizing these, both in terms of cultural norms and the peculiarities of a particular person’s history or values. In this light, reasons for acting are best thought of as “the elements of a possible storyline” (Velleman 2000: 28). As such, making explicit a person’s narrative is the medium for understanding and evaluating reasons and making sense of
27
28
Shaun Gallagher and Daniel D. Hutto
actions. Such narratives allow us to understand a person’s ‘rationale’ when this is not immediately obvious. Sometimes there is a need to frame and justify our reasons but more often than not, when all proceeds normally there is simply no need. This does not imply that in such cases we quietly grasp and deploy a set of explicit generalizations about how others will act. Rather, it is through shared training about the roles and rules of our common world that I learn how I ought to behave in various circumstances, and at the same time I learn how you ought to behave as well, ceteris paribus. Knowledge of what I ought to do in certain circumstances supplies a handy guide to the likely behaviour of others, in so far as they do not step out of line. Such learning does not take the form of internalizing explicit rules (at least not as a set of theoretical propositions), nor does it depend on our applying ones that are somehow already built-in subpersonally. Rather our expectations of others results from our becoming accustomed to local norms, coming to embody them, as it were, through habit and practice. This, we suggest, and not the wielding of theoretical generalizations, is the crucial backdrop against which we make sense of reasons for action via narratives of the folk psychological variety.
6.
The narrative practice hypothesis
How do we get this sort of complex and nuanced understanding of why people do what they do? People do not wear their reasons for action on their sleeves and they cannot be readily discerned or understood by deploying the kind of embodied heuristics described earlier in this paper. We suggest that the pervasive presence of narrative in our daily lives, and the development of specific kinds of narrative competency, can provide a more parsimonious alternative to theory or simulation approaches, and a better way to account for the more nuanced understandings (and mis-understandings) we have of others. Competency with different kinds of narratives enables us to understand others in a variety of ways. Distinctive kinds of narrative encounters are what first allow us to develop our folk psychological competence. Hutto calls this “the narrative practice hypothesis”. It claims that “children normally achieve [folk psychological] understanding by engaging in story-telling practices, with the support of others. The stories about those who act for reasons – i.e., folk psychological narratives – are the foci of this practice. Stories of this special kind provide the crucial training set needed for understanding reasons” (Hutto 2007b: 53). Accordingly, children acquire their skilled competence in understanding reasons by being exposed to and by engaging with narratives when appropriately and actively supported by their care givers. For example, in acts of storytelling, such
Primary interaction and narrative practice
active support takes the form of children being prompted to answer certain questions and by having their attention directed at particular events. In the case of folk psychological narratives this will normally involve jointly attending to mentalistic terms such as “wish”, “believe” and “know” and discussing what the story characters know, feel and want. During this process children learn how these states of mind behave in relation to each other and other terms in the psychological family. Importantly, these attitudes exist in a wider context such that children learn how and why these attitudes matter to the protagonists of such stories. Time and time again reasons for acting, of different types and complexity, are put on show in this way. By attending to enough of these exemplars, it is possible for children to develop an implicit practical understanding of how to make sense of persons as those who act for reasons. This is nothing like fashioning the concepts of the attitudes by means of theorizing or having a core theory about how they interrelate. Coming to understand what it is to act for a reason – to understand folk psychologically – requires being trained by means of a specific kind of narrative practice. They can achieve this because even simple folk psychological narratives, like their more sophisticated cousins “represent the moment by moment experiences of fictional minds, as well as the coloration that those experiences acquire from the characters’ broader cognitive and emotional stances towards situations and events” (Herman 2007: 147). This proposal is consistent with a number of recent empirical studies that have established that there are important links between narrative abilities and our capacity to understand others (Astington 1990; Dunn 1991; Feldman, Bruner, Renderer and Spitzer 1990; Lewis 1994, Lewis, Freeman, Hagestadt and Douglas 1994; Nelson 2007, Peterson and McCabe 1994). Exposure to stories is a critical determiner of folk-psychological abilities and it has been shown that this relation is stronger than mere correlation. Apparently narrative training causally influences what are considered to be basic ‘theory-of-mind’ skills for the better (Guajardo and Watson 2002). Controlled studies have shown that narrative training is responsible for improving performances on false belief tasks. Thus, it has been concluded that narrative is an effective tool for “at least modest improvements in children’s theory of mind development” (Guajardo and Watson 2002: 320). Similarly, it has been observed that “frequent conversations about the mind can accelerate growth of a ToM” (Garfield, Peterson and Perry 2001: 513). A complementary idea is that other kinds of narrative competencies enable a less mediated interpretation of the other’s actions and intentions, that is, without the mediation of folk psychology. After all, folk psychological explanation is just one kind of narrative practice. We argue here that how we go about developing a nuanced understanding of others may involve one or both of these
29
30
Shaun Gallagher and Daniel D. Hutto
paths – employing a narrative-informed folk psychology, and/or a less mediated narrative practice – and which one is appropriate will depend on the context.
7.
Folk psychological and other kinds of narratives
What are narratives? This is a tricky question and providing a good answer to it is beyond the scope of this paper. A very minimal definition will suffice for our purposes. Larmarque tells us that for something to be a narrative “at least two events must be depicted in a narrative and there must be some more or less loose, albeit non-logical relation between the events. Crucially, there is a temporal dimension in narrative” (Lamarque 2004: 394; see also Lamarque and Olsen 1994: 225). This neutral characterisation easily lends itself to the idea that there are different types of narratives and that these can be classified by such common features as their constituents and subject matter. Folk psychological narratives – as exemplified by Little Red Riding Hood – are distinguished by being about agents who act for reasons. Importantly, narratives of this kind can play their special role in development by being the objects of joint attention in early learning. That is the core claim of the NPH. In this light it should be emphasised that, as social cognizers, we do not use folk psychological narratives nearly as often as the tradition supposes. They are not, for example, the basis of all interpersonal interaction. On the contrary, they generally only come into play in those cases in which the actions of others deviate from what is normally expected in such a way that we encounter difficulty understanding them. In such cases the other’s actions become noticeable, falling into the spotlight for special attention and explanation – and potentially, explanations of a specific sort that involve understanding the other’s reasons for taking the particular action – where this is not in some way obvious or already known. Folk psychology is needed only in rare cases where we are not already familiar with the other person’s story, or are perplexed by another’s actions. For “When things ‘are as they should be’, the narratives of folk psychology are unnecessary” (Bruner 1990: 40). Appeal to folk psychology may come into play when culturally-based expectations are violated. For the most part, well-rehearsed patterns of behaviour and coordination dominate. By and large, we get by without having to make any folk psychological attributions at all and without seeking explications from others because most everyday social interaction takes place in normal (and normalized) environments. Again, we can learn a great deal from developmental psychology. Around the age of two, children are in secure possession of “an early intentional understanding of persons having internal goals and wants that differ from person to person”
Primary interaction and narrative practice
(Wellman and Phillips 2001: 130; Bartsch and Wellman 1995). Young children are somewhat practiced in understanding things as other people understand them in pragmatic contexts, and when the capacities associated with primary and secondary intersubjectivity are combined with several other newly acquired capacities, young children are ready to understand things and people in emerging narrative structures. And in this context it must be acknowledged that many other kinds of narratives – those of the non-folk psychological variety – can take us a long way to the understanding we seek, without resorting to the folk psychological framework per se (or at least without always having to do so). We learn to make sense of persons (others as well as ourselves) in dramatic and narrative ways as young children. When children listen to stories, or play-act (and the same applies to adults who are exposed to parables, plays, myths, novels, etc.) they become familiarized with sets of characters and with a range of ordinary or extra-ordinary situations, and the sorts of actions appropriate to them, all of which helps to shape their expectations. An education in narratives of many sorts – even of the more general and less personal variety – provides knowledge of what actions are acceptable and in what circumstances, what sort of events are important and noteworthy, what can account for action, and what kind of explanations constitute the giving of good reasons. Moreover, children are well supported in this process. Typically, they are provided with running commentaries on stories that teach them not only which actions are suited to particular situations but also which reasons for acting are acceptable and which are not. It is by absorbing such standards that we first learn how to judge an action’s appropriateness (though, of course, in time such standards are sometimes questioned and overturned). Quite generally, stories – real or fictional – teach us what others can expect from us, but just as importantly, what we can expect from others in certain situations. This is not just coming to know what others ought to (and thus are likely to) do, but what they ought to (and thus are likely to) think and feel, as indexed to the sort of people they are. Narratives provide an important source of guidance for staking out the boundaries of what is acceptable and what is not. Through them we learn the norms associated with social roles that pervade our everyday environments – shops, restaurants, homes and theatres.
. There are “two aspects of children’s narrative activity which are too often treated in mutual isolation: the discursive exposition of narratives in storytelling and their enactments in pretend play” (see Richner & Nicolopoulou 2001: 408). “Children’s first narrative productions occur in action, in episodes of symbolic play by groups of peers, accompanied by – rather than solely through – language. Play is an important developmental source of narrative” (Nelson 2003: 28).
31
32
Shaun Gallagher and Daniel D. Hutto
Engaging with narratives is not a passive affair: it presupposes a wide range of emotive and interactive abilities. To appreciate such stories children must initially be capable, at least to some degree, of imaginative identification and of responding emotively, just as they do in basic social engagements. In this respect “conversations about written and oral stories are natural extensions of children’s earlier experiences with the sharing of event structures” (Guajardo and Watson 2002: 307). Through them children discover why characters act as they do in particular cases, becoming accustomed to standard scripts – scenarios, characters, plots, etc. The kind of emotional resonance that one finds already in infancy, in primary intersubjectivity, seems to play an important role in gaining narrative competency. Decety and Chaminade (2003) have shown this connection as it plays out in the brain. In their fMRI study, subjects were presented with a series of video clips showing actors telling sad and neutral stories, as if they had personally experienced them. The stories were told with either congruent or incongruent motor expression of emotion. Subjects were then asked to rate the mood of the actor and how likeable they found that person. Watching sad stories versus neutral stories was associated with increased processing activity in emotion related structures (including the amygdala and parieto-frontal areas, predominantly in the right hemisphere). These areas were not activated when the narrator showed incongruent facial expressions. The reasonable hypothesis is that conflict between what we sense as the emotional state of the other person, simply on the basis of seeing their faces and actions, and the narrative content they present, is disruptive to understanding. Whatever is going on in the brain correlates not simply to features of action and expression (and the subjectivity of the other person) but to the larger story, the scene, the circumstance of the other person, and how features of action and expression match or fail to match those circumstances. If the emotional character of the other person is not in character with the narrative framework – with the story that I could tell about her and her circumstances – it is difficult to understand that person, the story, or both.
8.
Narrative competency and “landscape of consciousness”
We have argued that the abilities for intersubjective interaction and understanding that start with primary and secondary intersubjectivity, develop along a route that in most ordinary cases exploits narrative competency rather than the procedures, subpersonal or explicit, associated with traditional theory-of-mind accounts. This should provide the means of staving off a common worry about the NPH. Janet Astington (1990) has argued that acquiring narrative competency requires having a theory of mind. Citing Bruner’s concept of the landscape of
Primary interaction and narrative practice
consciousness (“what those involved in the action know, think, or feel, or do not know, think, or feel” [Bruner 1986: 14]), she suggests that to understand narrative we need access to the characters’ minds, and to have the latter requires us to have a theory-of-mind. But Bruner himself offers good experimental evidence against the necessity of the landscape of consciousness (LC) for understanding narratives. Feldman, Bruner et al. (1990), in a study of narrative comprehension in adults, presented two different versions of the same story to two groups, respectively. The first and original story mentioned the mental states of the characters as the story develops, and so was rich in LC. The second story was the very same story stripped of mental terms, leaving only the landscape of actions (LA). The results showed no significant differences (1) in subjects using reader-related mental verbs when they recount the LC narrative; (2) in recounting the facts of the stories – “the retellings were virtually indistinguishable”; (3) in recounting the order of events; and (4) when providing a meaning summary (gist) for the story “there is no version difference in the kind of gist given.” A likely explanation of these results is that the structure of these person-narratives, as revealed explicitly in basic plots, can be identified, responded to and described on several levels and ways. Often this happens all at once. But not everyone is equally proficient at this. It is possible to be alive to the major events in a drama without always being able to decipher, with full clarity or perhaps not at all, the reasons why a protagonist will have acted. It is thus possible to have some sense of what is going on in an unfolding drama without understanding it in toto (this is apparently a common experience for those first encountering Shakespearean plays). What is important is that seeking a narrative understanding of the other’s reasons is not a matter of characterizing the other’s ‘inner’ life – if this is understood as a series of causally efficacious mental states. What we are attempting to understand is much richer, it is the other’s reasons as they figure against the larger history and set of projects, and that is best captured in a narrative form. Coming to understand another’s reasons should not be understood as designating their discrete ‘mental states’ but their attitudes and responses as whole situated persons. I encounter the other person, not abstracted from their circumstances, but in the middle of something that has a beginning and that is going somewhere. I see them in the framework of a story in which either I have a part to play or I don’t. The narrative is not primarily about what is ‘going on inside their heads’; it’s about the events going on in the world around them, the world that we share with them,
. For further discussion of the distinction between properly folk psychological narratives and those dramatic re-enactments which only involve intentional attitudes, yet which share the same basic formats (see Hutto 2006).
33
34
Shaun Gallagher and Daniel D. Hutto
and in their lives and the way they understand and respond to such events. Crucially, coming to appreciate the other’s story – to see why they are doing what they are doing – does not require a capacity for mentalizing inferences or simulations. Our understanding of others is ordinarily not based on attempts to get into their heads; typically we do not need to access a “landscape of consciousness” since we already have access to a “landscape of action” which is constituted by their embodied actions and the rich worldly contexts within which they act – contexts that operate as scaffolds for the meaning and significance of actions and expressive movements.
9.
Conclusions
In this chapter we have argued that there is no need to appeal to standard theoryof-mind and simulative explanations of how we understand others as the basis for making sense of them folk psychologically. What begins as perceptual and emotional resonance processes in early infancy, which allow us to pick up the feelings and intentions of others from their movements, gestures, and facial expressions, feeds into the development of a more nuanced understanding of how and why people act as they do, found in our ability to frame their actions, and our own, in narrative ways. Our everyday abilities for intersubjective engagement and . This is not to deny that some narratives are more psychological than others – those of James Joyce or Dostoyevsky, as Jordan Zlatev suggests (private correspondence). Luckily Joyce, Dostoyevsky and other novelists put us in the heads of their characters and we do not have to theorize or simulate our way in there. The NPH does not deny that human beings are complicated psychological creatures, or that the psychological lives of Stephen Dedalus or Raskolnikov are not fascinating in ways that outstrip an understanding in folk psychological terms. The issue is how we come to understand people in our everyday interactions with them. . The idea that narrative understanding does not rest on or presuppose ToM abilities per se (including simulation capacities that involve making belief/desire predictions and explanations) is in line with Greg Currie’s (2007) recent claim that our skills in comprehending narratives involve the adoption of frameworks through which we identify with (and are effectively ‘asked to’ take on) certain personas, which can be understood as embodied ‘stances’ that particular narratives invite us to adopt. The activity of framework adoption is quite distinct from understanding a story’s content – as detailed in its plot or fabula. As Currie characterizes it, adoption or attention to a narrative framework activates our subpersonal mechanisms for imitative and emotional responding – thus it is something that engages us viscerally. He contrasts this with the idea that attention to narrative framework involves developing a ‘theory’ (even if a not very explicit one) about the persona embedded in narrative; although he does not wholly reject the latter proposal since he acknowledges it may have a role when it comes to communicating about narratives.
Primary interaction and narrative practice
interaction are, in the later stages of childhood, transformed by encounters with narratives. It is exposure to these complex objects of joint attention – and not facility with theoretical knowledge or simulative routines – that is responsible for the development of sophisticated folk psychological abilities and understanding; abilities which remain importantly in play in our adult life.
References Allison, T., Puce, Q. and McCarthy, G. 2000. “Social perception from visual cues: role of the STS region.” Trends in Cognitive Science 4 (7): 267–278. Astington, J. 1990. “Narrative and the child’s theory of mind.” In Narrative Thought and Narrative Language, B.K. Britton and A.D. Pellegrini (eds.), 151–71. Hillsdale, New Jersey: Erlbaum. Baldwin, D.A. 1993. “Infants’ ability to consult the speaker for clues to word reference.” Journal of Child Language 20: 395–418. Baldwin, D.A. and Baird, J.A. 2001. “Discerning intentions in dynamic human action.” Trends in Cognitive Science 5 (4): 171–78. Baldwin, D.A., Baird, J.A., Saylor, M.M. and Clark, M.A. 2001. “Infants parse dynamic action.” Child Development 72 (3): 708–17. Barresi, J. and Moore, C. this volume. “The neuroscience of social understanding.” Bartsch, K. and Wellman, H. 1995. Children Talk About the Mind. New York: Oxford University Press. Bermúdez, J.L. 1996. “The moral significance of birth.” Ethics 106: 378–403. Bertenthal, B.I., Proffitt, D.R. and Cutting, J.E. 1984. “Infant sensitivity to figural coherence in biomechanical motions.” Journal of Experimental Child Psychology 37: 213–30. Bruner, J. 1986. Actual Minds, Possible Worlds. Cambridge, MA: Harvard University Press. Bruner, J. 1990. Acts of Meaning. Cambridge, MA: Harvard University Press. Carpendale, J.I.M. and Lewis, C. 2004. “Constructing an understanding of the mind: The development of children’s social understanding within social interaction.” Behavioural and Brain Sciences 27: 79–151. Currie, G. 2007. “Framing narratives.” In Narrative and Understanding Persons, D.D. Hutto (ed.), 17–42. Cambridge: Cambridge University Press. deVignemont, F. 2004. “The co-consciousness hypothesis.” Phenomenology and the Cognitive Sciences 3 (1): 97–114. Decety, J. and Chaminade, T. 2003. “Neural correlates of feeling sympathy.” Neuropsychologia 41: 127–128. Dunn, J. 1991. “Understanding others: Evidence from naturalistic studies of children”. In Natural Theories of Mind, A Whiten (ed.), 51–61. Oxford: Blackwell Feldman, C.F., Bruner, J., Renderer, B. and Spitzer, S. 1990. “Narrative comprehension.” In Narrative Thought and Narrative Language, B.K. Britton and A.D. Pellegrini (eds.), 1–78. Hillsdale, NJ: Lawrence Erlbaum Associates. Gallagher, S. 1996. “The moral significance of primitive self-consciousness.” Ethics 107: 129– 140.
35
36
Shaun Gallagher and Daniel D. Hutto
Gallagher, S. 2001. “The practice of mind: Theory, simulation or primary interaction?” Journal of Consciousness Studies 8(5–7): 83–108. Gallagher, S. 2004. “Understanding interpersonal problems in autism: Interaction theory as an alternative to theory of mind.” Philosophy, Psychiatry, and Psychology 11 (3): 199–217. Gallagher, S. 2005. “Phenomenological contributions to a theory of social cognition.” Husserl Studies 21: 95–110. Gallagher, S. 2006. “The narrative alternative to theory of mind.” In Radical Enactivism: Intentionality, Phenomenology, and Narrative, R. Menary (ed.), 223–29. Amsterdam: John Benjamins. Gallagher S. 2007a. “Logical and phenomenological arguments against simulation theory.” In Folk Psychology Re-Assessed, D. D. Hutto and M Ratcliffe (eds.), 63–78. Dordrecht: Springer. Gallagher, S. 2007b. “Simulation trouble.” Social Neuroscience 2 (3–4): 353–65. Gallagher, S. in press. “Direct perception in the intersubjective context”. Consciousness and Cognition. Gallagher, S. and Meltzoff, A. 1996. “The earliest sense of self and others: Merleau-Ponty and recent developmental studies.” Philosophical Psychology 9: 213–236. Garfield, J.L., Peterson, C.C. and Perry T. 2001. “Social cognition, language acquisition and the development of the theory of mind.” Mind and Language 16: 494–541 Gallese, V. 2005. “‘Being like me’: Self-other identity, mirror neurons and empathy.” In Perspectives on Imitation, S.L. Hurley and N. Chater (eds.), 101–118. Cambridge, MA: MIT Press. Gibson, J.J. 1979. The Ecological Approach to Visual Perception. Boston, Houghton Mifflin. Goldman, A. I. 2002. Simulation theory and mental concepts. In Simulation and Knowledge of Action, J. Dokic and J. Proust (eds.), 1–19. Amsterdam: John Benjamins. Goldman, A. and Sripada, C.S. 2005. “Simulationist models of face-based emotion recognition.” Cognition 94: 193–213. Gopnik, A. 1993. “How we know our minds: The illusion of first-person knowledge of intentionality.” Behavioral and Brain Sciences 16: 1–14. Gopnik, A. and Meltzoff, A. 1997. Words, Thoughts, and Theories. Cambridge, MA: MIT Press. Gordon, R. 1986. “Folk psychology as simulation.” Mind and Language 1: 158–171. Guajardo, N.R. and Watson, A. 2002. “Narrative discourse and theory of mind development.” The Journal of Genetic Psychology 163: 305–25. Herman, D. 2007. “Cognition, emotion and consciousness.” In The Cambridge Companion to Narrative, D. Herman (ed.), 245–259. Cambridge: Cambridge University Press. Hurley, S.L. 2005. “Active perception and perceiving action: The shared circuits model.” In Perceptual Experience, T. Gendler and J. Hawthorne (eds.), New York: Oxford University Press. Hutto, D.D. 2004. “The limits of spectatorial folk psychology.” Mind and Language 19: 548–73. Hutto, D.D. 2005. “Knowing what?: Radical versus conservative enactivism.” Phenomenology and the Cognitive Sciences 4(4): 389–405. Hutto, D.D. 2006. “Narrative practice and understanding reasons: Reply to Gallagher.” Radical Enactivism: Intentionality, Phenomenology, and Narrative, R. Menary (ed.), 231–247. Amsterdam: John Benjamins. Hutto, D.D. 2007a. “Folk psychology without theory or simulation.” In Folk Psychology Reassessed, D.D. Hutto and M. Ratcliffe (eds.), 115–135. Dordrecht: Springer.
Primary interaction and narrative practice
Hutto, D.D. 2007b. “The narrative practice hypothesis.” In Narrative and Understanding Persons, D.D. Hutto (ed.), 43–68. Cambridge: Cambridge University Press. Hutto, D.D. 2008. Folk Psychological Narratives: The Sociocultural Basis of Understanding Reasons. Cambridge, MA: MIT Press. Jeannerod, M. and Pacherie, E. 2004. “Agency, simulation, and self-identification.” Mind and Language 19 (2): 113–46. Johnson, S.C. 2000. “The recognition of mentalistic agents in infancy.” Trends in Cognitive Science 4: 22–28. Johnson, S. Slaughter, V. and Carey, S. 1998. “Whose gaze will infants follow? The elicitation of gaze-following in 12-month-old infants.” Developmental Science 1: 233–38. Lamarque P., 2004. “On not expecting too much from narrative.” Mind and Language 19: 393–408. Lamarque P. and Olsen S. 1994. Truth, Fiction and Literature. Oxford: Oxford University Press. Legerstee, M. 1991. “The role of person and object in eliciting early imitation.” Journal of Experimental Child Psychology 51: 423–33. Leslie, A.M. 1987. “Pretense and representation: The origins of ‘theory of mind’.” Psychological Review 94: 412–426. Lewis, C. 1994. “Episodes, events and narratives in the child’s understanding of mind.” In Children’s Early Understanding of the Mind, C. Lewis and P. Mitchell (eds.), 457–478. Hillsdale, New Jersey: Erlbaum Lewis, C., Freeman, N.H., Hagestadt, C. and Douglas, H. 1994. “Narrative access and production in preschooler’s false belief reasoning.” Cognitive Development 9: 397–424 Meltzoff, A.N. 1995. “Understanding the intentions of others: Re-enactment of intended acts by 18-month-old children.” Developmental Psychology 31: 838–50. Meltzoff, A.N. and Brooks, R. 2001. “‘Like Me’ as a building block for understanding other minds: Bodily acts, attention, and intention.” In Intentions and Intentionality: Foundations of Social Cognition, B. Malle, L.J. Moses and D.A. Baldwin (eds.), 171–191. Cambridge, MA: MIT Press. Moore, D. G., Hobson, R. P. and Lee, A. 1997. “Components of person perception: An investigation with autistic, non-autistic retarded and typically developing children and adolescents.” British Journal of Developmental Psychology 15: 401–423. Myowa-Yamakoshi, M. 2001. “Evolutionary foundation and development of imitation.” In Primate Origins of Human Cognition and Behavior, T. Matsuzawa (ed.), 349–367. Dordrecht: Springer. Myowa-Yamakoshi, M., Tomonaga, M., Tanaka, M. and Matsuzawa, T. 2004. “Imitation in neonatal chimpanzees (Pan troglodytes).” Developmental Science 7 (4): 437–42. Nelson, K. 2003. “Narrative and the emergence of a consciousness of Self.” In Narrative and Consciousness, G. D. Fireman, T. E. J. McVay and O. Flanagan (eds.), 17–36. Oxford: Oxford University Press. Nelson, K. 2007. Young Minds in Social Worlds. Cambridge, MA: Harvard University Press Peterson, C. and McCabe, A. 1994. “A social interactionist account of developing decontextualised narrative skill.” Developmental Psychology 30: 937–48 Phillips, W., Baron-Cohen, S. and Rutter, M. 1992. “The role of eye-contact in the detection of goals: Evidence from normal toddlers, and children with autism or mental handicap.” Development and Psychopathology 4: 375–83.
37
38
Shaun Gallagher and Daniel D. Hutto
Richner E.S. and Nicolopoulou A. 2001. “The narrative construction of differing conceptions of the person in the development of young children’s social understanding.” Early Education and Development 12: 393–432. Scholl, B.J. and Tremoulet, P.D. 2000. “Perceptual causality and animacy.” Trends in Cognitive Sciences 4 (8): 299–309. Stern, D. 1985. The Interpersonal World of the Infant: A View from Psychoanalysis and Developmental Psychology. New York: Basic Books. Trevarthen, C. 1979. “Communication and cooperation in early infancy. A description of primary intersubjectivity.” In Before Speech: The Beginning of Human Communication, M. Bullowa (ed.), 99–136. London: Cambridge University Press. Trevarthen, C. and Hubley, P. 1978. “Secondary intersubjectivity: Confidence, confiding and acts of meaning in the first year.” In Action, Gesture and Symbol: The Emergence of Language, A. Lock (ed.), 183–229. London: Academic Press. Velleman, J.D. 2000. The Possibility of Practical Reason. Oxford: Oxford University Press Walker, A.S. 1982. “Intermodal perception of expressive behaviors by human infants.” Journal of Experimental Child Psychology 33: 514–35. Wellman H. and Phillips A. 2001. “Developing intentional understandings. In Intentions and Intentionality, B. Malle, L. J. Moses and D.A. Baldwin (eds.), 125–148. Cambridge, MA: MIT Press. Wittgenstein L. 1992. Last Writings on the Philosophy of Psychology Volume 2: The Inner and the Outer. Cambridge: Blackwell. Zlatev, J. this volume. “The co-evolution of intersubjectivity and bodily mimesis.”
chapter 3
The neuroscience of social understanding John Barresi and Chris Moore How do we understand and engage with the purposeful, emotional and mental activities of other people and how does this knowledge develop? What can recent work on mirror neurons in monkeys and human beings teach us about how the brain supports social understanding? According to Intentional Relations Theory (Barresi and Moore 1996), the understanding of the self-other equivalence requires concurrent knowledge of mind from both a first- and a third-person point of view and that any mental concept must directly match and link these two ways of knowing it. In this chapter we will argue that Intentional Relations Theory is consistent with and can help interpret recent neurophysiological findings on “mirror neurons” that fire equivalently for intentional relations (i.e., object-directed actions, emotions, and mental activities) of self and other.
1.
Introduction
Human beings, like many other social animals, spend an enormous amount of time engaged in activities that require quick adjustments to socially transmitted information. By observing others we learn to adapt effectively to changes in the environment as well as to the actions and reactions of our social peers. How do we do it? To what extent do we need to understand the mental processes governing our own and others’ actions or can we function socially based on simple mechanisms by which we come to share psychological states with others, without understanding them? In other words, to what extent does a skill at mind sharing function as a form of social understanding well before we come to a level of mind understanding? Furthermore, how do these two capacities – mind-sharing and mind-understanding – relate to each other? In the Theory of Mind (ToM) approach to social understanding emphasis is placed on sophisticated abilities to understand mental states – in particular the ability to attribute representational mental states such as beliefs to self and other. It is the ability to attribute false beliefs that is taken as a hallmark of the specifically human form of mentalistic social understanding that characterizes a “theory of
40 John Barresi and Chris Moore
mind”. However, social understanding is a more general phenomenon that occurs both in many social species that seem to have no ToM of this kind, and in children well before the late preschool period when the understanding of false beliefs develops. Indeed, an early form of social understanding is evidenced essentially from birth as neonates show a particular sensitivity to human social stimuli. We suggest that the kinds of social sensitivity observed in infants as well as in many social animals should be seen as forms of non-reflective social understanding, dependent on an array of mechanisms that yield an ability to share mental states with others without necessarily recognizing that those shared mental states are in fact attributable to individual agents. A satisfactory account of the development of social understanding will require an explanation of how these original mechanisms that enable early social responsiveness combine with later developing skills to yield more sophisticated forms of intersubjectivity. In parallel, such an account must specify how engaging in shared understanding or shared mental activities with others facilitates the later more individualistic understanding of mind. In the present chapter we will approach these problems with a focus on recent findings in the neuroscience of social understanding. With the discovery in monkeys of pre-motor “mirror neurons” that respond to the actions of others as well as to their own motor plans, there is reason to believe that even monkeys somehow understand actions of both self and others in a similar object-directed way. But should such a “common code” between perception and action be treated merely as an instance supporting the common coding hypothesis (Prinz 1997; Knoblich and Jordan 2002) or a more elaborate understanding for what we have called action intentional relations (Barresi and Moore 1996)? Even if it seems unlikely that monkeys represent these actions as full-blown mental events involving conscious intentions of the other, distinct from their own, it is still a question of how simple is their understanding here and how it connects to more elaborate forms of social understanding. Perhaps their understanding occurs more simply as sharing in the goal-directed nature of the activity of the other by entering into a comparable goal-directed pre-motor state, while not themselves engaging in the activity. Such a sub-personal level of understanding of the action of another would in effect convert it into a first-person representation of one’s own actions, but it would not yet represent that action as what we call an intentional relation, involving a representation of an agent as well as the object-directed action. Nevertheless, such sub-personal matching between goal-directed actions of self and other provides a basis for eventual understanding of full blown intentional relations that can be applied to agent-oriented actions directed at objects at a personal level, whether of self or other. We believe that this is the way that these phenomena should be understood and that this matching between aspects of the observing monkey’s intentional relations (IRs) and the IRs of others provides evidence for the matching
The neuroscience of social understanding
hypothesis that we have previously proposed as the basis of social understanding. We believe that our general account, which we have named Intentional Relations Theory (IRT) is superior to alternative accounts of the origins and development of intersubjectivity, and in the present chapter will bring neuroscientific evidence involving humans as well as monkeys to support our position.
2.
The matching problem of social understanding and three approaches to intersubjectivity
A fundamental aspect of human social understanding is what we have previously referred to as ‘self-other equivalence’. Human beings understand self and other to be essentially the same kind of thing – namely a human agent or person that can engage in a variety of intentional relations with objects or states of affairs. This aspect of human social understanding is quite obvious and passes unnoticed in commonsense psychology and yet it hides a significant epistemic problem. How can we attribute the same meaning to actions of other individuals that we attribute to our own actions when the third-person information that we have of the actions of others is radically different from the first-person information that we have of our own actions? The information we get about others’ actions is apparently information about the overt aspects of behavior, while the object towards which the action is directed is often not obvious (or even opaque in the case of mental states such as beliefs). In contrast the information we get about our own actions is apparently information about our orientations towards the objects and events we witness or imagine but does not typically include information about ourselves as the actor or agent being so oriented. So how are these qualitatively different forms of information recognized to be tokens of the same type – expressions of intentional relations between an agent (self or other) and some object or state of affairs? In the recent history of research on social understanding, there are three fundamentally different answers to this question. According to the ‘theory theory’ (TT) approach humans have innately, or acquire early in development, a ToM mechanism that can be applied uniformly to self and other based purely on inference from behavior (e.g., Gopnik 1993; Leslie 1987). Self-other equivalence in this account is based on the fact that one can interpret one’s own behavior in the same way that one can interpret the behavior of others. For instance, consider an example of what we have called an emotional intentional relation – the case of love. Since love is a public concept, whose main criterion of application is supposedly based on behavior, a person can know when she or another person is in love by noticing the same kinds of behavior of self and other directed toward the object of love.
41
42
John Barresi and Chris Moore
In contrast, Simulation theorists (ST) would take a different view from TT on how a person knows about her own love versus another person’s love (e.g., Goldman 1992; Gordon 1986; Humphrey 1984; Harris 1989). On their view, love may have some behavioral consequences that can be used to identify it in another person, but it is fundamentally a subjective mental state, and without a personal appreciation of the “feeling state” that usually goes with the overt behavior, we cannot truly understand love as a psychological state. We understand love “directly” in our own case, but only indirectly and by simulation in the case of another person. We must imagine what someone else feels when we observe their behavior in context (e.g., around the object of love), in order to understand the psychological, intentional, and subjective meaning of their behavior. In our own case, our behavior is a consequence of this subjective state, so no inference is necessary from our own behavior to the mental state that we are in. Although we need to reflect on these states to categorize them, we do not need knowledge of comparable states in other people to form these categories and concepts. A third kind of theory invokes the notion of matching or sharing attitudes or psychological states between self and other and is represented in a range of different accounts (e.g., Gallagher and Hutto this volume; Gallese, Keysers and Rizzolatti 2004; Hobson 1991, 1998; Hobson and Hobson this volume; Wilson and Knoblich 2005; Zlatev this volume). Although the various theories in this third group can all be considered to invoke some form of intersubjectivity – understood widely as involving matched or shared mental states between or among individuals – they vary on the extent to which they provide an account of the foundations or the origins of intersubjectivity and on the processes by which infants are hypothesized to move from forms of intersubjective sharing of mental states to understanding that self and other are persons or selves that might have distinct mental states. Several of these theories (e.g., Gallagher and Hutto this volume; Hobson and Hobson this volume) invoke Trevarthen’s (e.g., Trevarthen and Hubley 1978) concepts of “primary intersubjectivity” and “secondary intersubjectivity” to describe early phases of development. However, while the capacity for mind-sharing is evident in these forms of intersubjectivity, what isn’t clear is how the infant moves from sharing mental states with others to understanding mental phenomena as distinct and possibly different in self and other. In the case of Gallagher and Hutto, this latter form of understanding is thought to rely on the acquisition of language and of the differentiating roles of self and other in situated narratives, some of which involve folk psychological terms. Our own Intentional Relations Theory (Barresi 2001, 2004; Barresi and Moore 1996; Moore 1999, 2006, 2007) does not differ substantially from these accounts in its interpretation of the early phases of development of social understanding that involve mind-sharing through processes that produce interpersonal matching
The neuroscience of social understanding
of self and other. However, it differs from these other accounts, as well as from ToM accounts, in explicitly addressing the genesis of the recognition of self-other equivalence and difference, as involving a developmental shift from mind-sharing to mind-understanding. The key notion in IRT is that the first-person information that we have about our own IRs (e.g., the “feeling” of love for someone) is distinctly different from the third-person information that we have about the IRs of others (e.g., another’s “behavior” toward the object of love), and that in order to develop uniform concepts or representations of IRs that can be applied equally, but distinctly, to self and other, we need to match these two types of information in a single concept or form of knowledge that contains both types of information. In Barresi and Moore (1996) we posited an “intentional schema” to integrate this multimodal combination of first- and third-person information initially derived from self and other. On this view, being in love should not be defined primarily as a private, subjective experience, as in the ST view, nor as a mental intentional state that can be inferred from behavior, as in the TT view, but as an embodied IR between the agent and object, that, in the case of love, involves both feelings and concomitant behavioral expressions. Moreover, in learning the concept of love or any other IR, it is supposed that we must learn both the first-person, “inner” aspect, of the IR, as well as the third-person, “outer” aspect; otherwise, we fail to have the concept. For instance, one can be in love, say for the first time, without knowing it, because all one knows about love is the outer aspect, and one does not recognize this outer aspect in one’s feelings for another until one’s concomitant behavior is pointed out to one. Of course, love in our culture is primarily a social concept and learned to a large extent through language. But other more basic IRs, like fearing, seeing, or picking up are more fundamental, and may be understood to some extent by an organism without the mediation of language. In the rest of the chapter, we consider in more detail Intentional Relations Theory and specifically the issue of how 1st and 3rd person information about intentional relations are integrated. We go on to review the neuroscientific findings that support this approach to social understanding. We then consider autism as a case of failure to integrate 1st and 3rd person information in the understanding of self and other.
3.
Matching of 1st and 3rd person information and their integration
In Barresi and Moore (1996) we developed a model of social understanding that focused on the origins of understanding of IRs. We distinguished 4 levels of understanding IRs and used these levels to interpret both developmental and phylogenetic differences in social understanding (cf. Zlatev this volume, for a
43
44 John Barresi and Chris Moore
similar multilevel model). At level 1, the organism represents the activities of self and other in distinctly different ways and neither in terms of IRs. We suggested that most animals typically operate at this level and it may also characterize social understanding in certain forms of psychopathology such as autism. We will return to consider this level and the case of autism in Section 5. In the rest of this section we review 3 levels of social understanding in which first and third person information about IRs are integrated. We devote most attention to how such integration is possible in the first place.
3.1
Interactive routes to matching
In order to understand IRs at all, the organism must be able to combine first person information about IRs with third person information about IRs into integrated representations involving an agent, an intentional relation and an object that can be equally applied both to the IRs of self and the IRs of others or to the joint activity of self and other. This combination occurs at level 2 of our model when there is matched first- and third-person information about intentional action available to the organism. There are various ways in which such matching can come about. Our suggestion is that matching occurs normally in human development when infant and mother engage in interactions, initially dyadic and later triadic. These interactions are typically patterned in such a way that the infant and mother both express and experience similar psychological activity. For example, in dyadic interactions, infant and mother may smile and vocalize in close synchrony. Whether the synchrony between an infant and adult in interactions of this sort is based on innate contagious mechanisms, or occurs through a form of mimicry initiated at first by the adult, it seems clear that there is a matching in such cases, where first-person information about self can be experienced concurrently with matched third-person information about the other. We believe that in such early dyadic communicative interactions the infant acquires integrated knowledge of first- and third-person aspects of emotional expressions, though not yet of intentional relations involving those expressions directed at objects. Dyadic interactions do not revolve around objects so the intentionality of the shared psychological activity is at best implicit. However, in the triadic interactions that develop at about 9 months of age, the patterned interaction is now object-focused so that both infant and mother may share psychological activity to a particular object – they may look at the same object or produce similar object-directed actions through imitation. We have argued that such interactive experiences are crucial for the development of understanding IRs because it is
The neuroscience of social understanding
in these interactions that the infant’s first-person experience of her own objectdirected psychological activity is coordinated reliably with their corresponding third-person experience of the mother’s object-directed psychological activity. Reliable coordination of the available first- and third-person information allows the construction of representations of intentional activity that integrate both forms of information and are thereby applicable to the joint activity of self and other, and subsequently with further development to individual activities of either self or other.
3.2 Noninteractive routes to matching Although dyadic and triadic interactions provide the normal context for the sharing of psychological activity in human development, it is probably not necessary for there to be joint engagement of either dyadic or triadic kinds for a degree of matching of intentional relations to occur. For instance, as indicated earlier, research on monkeys seems to show that they can represent the goal-directed actions of another organism in the same manner as they represent their own actions (Gallese et al. 1996; Rizzolatti et al. 1996). The pre-motor ‘mirror’ neurons mediating these representations fire in the planning and execution of the monkey’s own actions, but also in perceiving comparable goal-directed actions in another animate being. While we do not wish to exclude the possibility of innate forms of matching between self and other, for instance in emotional expressive domains where unlearned forms of mimicry may be the basic mechanism for matching, in the case of action understanding a learning mechanism needs to be involved. Matching between perception and action may come about because for certain forms of psychological activity such as object-directed reaching, the organism gains information about its own action via more than one perceptual modality (Keysers and Perrett 2004). When a monkey reaches for objects, it is reliably provided with both visual and proprioceptive information about its own reaching, and an integrated multimodal representation of the action will result. Then vision may mediate the connection to the action of others. The same multimodal representation will later be activated by only the relevant visual information and thereby can be applied to the experience of seeing another organism perform the action. Vision here serves as a third-person ‘bridging’ modality that can be applied to both self and other, thus linking the strictly first-person information of proprioception to the available third-person information about goal-directedness. In this way a representation of action that is similarly applicable to the actions of both self and other may be achieved. However, it should be noted that
45
46 John Barresi and Chris Moore
all that is involved here is the understanding of the action, per se, not of an agent performing the action. Thus the representation is at a sub-personal rather than at a personal, or agent, level of representation. Hence, an organism does not here understand intentional relations involving agents, but only sub-personal actions directed at objects. There is evidence that such a process may also operate in early human development. Woodward (1998) has shown that infants are able to recognize the goal-directed reaches of others at about the same time as they themselves engage in visually guided reaching. Importantly, teaching infants to make object-directed reaches at an early age is correlated with their representation of similar reaching actions of another person (Sommerville, Woodward, and Needham 2005). Thus, at least for simple actions, it seems that learning to succeed at an action, which involves coordination of first-person (e.g., proprioceptive) and typically third-person (e.g., visual) information of one’s own action, may be correlated with representing the similar actions of others.
3.3 Sub-personal and interpersonal forms of understanding IRs It will be recognized that the latter route to representations of actions that are equally applicable to self and other will only serve for those actions, such as manual reaching, for which the same perceptual information is available for both self and other. It is in such circumstances that a common code for the perception and production of action can bear fruit both in monkeys and humans, with a sub-personal level of understanding of goal-directed actions. However, in the understanding of intentional relations more is required. The difference between the human case and the cases of monkeys is that the dyadic and triadic interactive contexts of early human development provide multiple instances in which there are richly elaborated structures of shared intentional relations. For example, in a typical episode of a joint attentional (triadic) interaction, there may be shared emotional experience (e.g., smiling), shared object-directed action (e.g., object exchange) and shared epistemic activity (e.g., gaze following). These interactive structures therefore provide not just experiences in which a particular, simply observable, action intentional relation is shared but experiences in which a variety of different yet complementary intentional relations of various types are shared. As a result, there is the opportunity for infants to acquire complex representations of intentional activity that combine and integrate the first-person information pertaining to their own activity and the third-person information pertaining to the activity of others across a range of intentional relations. This difference between the human and animal cases, such as monkeys, is important because it may explain why humans step onto the path of development that leads ultimately to an agent level
The neuroscience of social understanding
form of social understanding, whereas monkeys appear not to. To see why, it is important to examine whether the earliest forms of integrated representations of intentional relations are recognized to be at a ‘personal’ or at a ‘subpersonal’ level. Some authors (e.g., Tomasello 1999) have argued that the phenomena of triadic interactions arising at about 9 months signal the development of a concept of an ‘intentional agent’ that can be applied equally to self and other agents. However, a plausible alternative is that concepts of intentionality are initially acquired in a more piecemeal way. For example, Woodward and her colleagues’ research (for a review see Woodward 2005) has shown that infants represent the object-directedness of different actions at different points in development. Whereas reaching is represented as object-directed before 6 months, gaze is not represented as object-directed until the end of the first year. Furthermore, when such intentional relations are first being acquired, the acquisition does not appear to be correlated so that infants who represent gaze as object-directed may not represent pointing as object-directed and vice versa. To explain this pattern of results, Moore (2006) proposed the notion of ‘intentional islands’ (cf. Tomasello 1992, on ‘verb islands’ in language acquisition), whereby intentional representations start out as separate sub-personal ‘islands’ relevant to particular object-directed actions and are only gradually integrated into more complex concepts at a personal level relevant to goal-directed agents. We suggest that it is the richly structured patterns of intentional relations that occur in triadic interactions, which allows the generation of the more complex representations of goal-directed agents. In contrast, while other animals such as monkeys may acquire sub-personal integrated representations of object-directed actions, such as reaching, without experience of rich combinations of shared intentional relations, they do not proceed to construct representations of goal-directed agents.
. Great apes provide evidence that they stepped onto a new path similar to, but not the same as, our own. Chimpanzees, and probably other apes, engage in intense social interactions that promote an understanding of other’s actions on an individual level, through what Zlatev (this volume), ascribes to dyadic mimesis and which, we (Barresi and Moore 1996) originally hypothesized was associated with their general imitative ability. Recent research suggests that the evolutionary path taken here may be different from our own in that while learning in dyadic interactions between infant and mother chimpanzees involves an apprenticeship relationship (Matsuzawa 2007) human dyadic and triadic relationships between human infant and adults is much more intensely communicative and collaborative (Tomasello et al. 2005). A consequence of this latter form of interaction results in what Zlatev calls triadic mimesis, which is roughly similar to level 2 interactions transforming to level 3 interactions in our own model.
47
48 John Barresi and Chris Moore
3.4 Individualistic understanding of IRs So far we have advanced from a sub-personal understanding of the simple actions of self and other that do not explicitly code for agent to the capacity for understanding shared IRs evident at level 2 of our model. This sharing entails the existence of representations of IRs that are interpersonal, though probably not explicitly represented as interpersonal. Rather the interrelated and similar IRs of self and other are understood using a uniform representational form that codes for the concurrent identity between first-person information of self and third-person information of the other. But it is not yet the case that agents are recognized to be individual centres of intentional activity. The next level of understanding IRs (Barresi and Moore 1996) requires the ability to reflect on, or imagine IRs as properties of individual agents. According to IRT this requires the use of imagination to fill in the third-person information for IRs of self and first-person information for IRs of others. Without this ability it would not be possible to represent diversity of intentional relations across self and other when the same object is involved. In the developmental account given in Barresi and Moore (1996), children attain level 3 of understanding IRs during the second year of life. A variety of phenomena evidence this change (see Moore 2007). On the one hand the child becomes capable of recognizing the self as an individual agent as seen by phenomena such as mirror self-recognition. On the other hand, children become able to appreciate that others may have a different intentional orientation to an object from the self. For example, 18-month-olds understand that someone else may like something that they do not and vice versa (Repacholi and Gopnik 1997) and they understand that they may see something that someone else does not and vice versa (Moll and Tomasello 2005). At this point in development, therefore, children are able to attribute some forms of mental states, those exhibited in present activities, to individual agents, both self and other. This level of understanding goes beyond mind-sharing toward a conceptual understanding of individuals as embodied agents with points of views that may differ from each other. In some respects our account here is similar to the simulation account. However, whereas ST proposes that we simulate the mental state of the other through imaginative substitution of our own mental states, we here suggest that only the first-person aspect of the intentional relation of the other requires imaginative construction, as the third-person aspect is pragmatically available in the situation. Moreover, we suggest that at this same time the infant acquires the skill to understand its own intentional relations by imagining the third-person aspect that goes with the currently first-person experience of the intentional relation, something the ST does not even attempt to explain. Our account also differs from Gallagher and Hutto, since we do not think that language
The neuroscience of social understanding
alone mediates the conceptual development that occurs at this time, which allows one to distinguish one’s own from the other’s embodied mental states. Indeed, their narrative interpretation of how children distinguish mental states of self and other, seems to focus on only representational mental states such as false beliefs, a capacity for which we provide a separate account in the next section.
3.5 Representation of mental agents In the fourth year, pre-school children achieve yet another level of social understanding, when they can imagine both first- and third-person properties of a mental state. This results in children developing knowledge of mental representation as such, which allows them to show evidence of the conceptual understanding of mind seen in traditional ToM tasks. However, according to IRT, the levels of intentional understanding at which there is an understanding of individual minds derive from previous shared intentional activities where first- and third-person information originally became associated. It is the derivation from shared psychological activity that enables the concepts of mind that humans have, yielding notions like love having both internal bases involving feelings and external bases involving behavior. All levels of social understanding which depend originally on the integration of first- and third-person information are held to be different from Level 1 forms of understanding of self and other, which rely separately on firstperson information alone to understand self and third-person information alone to understand others. Consideration of level 1 will become important later in the chapter when we discuss autism (see Section 5). We turn now to research on the neuroscience of social understanding to see to what extent there is support for the model of social understanding we have outlined here. We should note, however, that whereas the evidence from neuroscience indicates a particular pattern of brain organization underlying social understanding in adult human beings as well as nonhuman primates, there is of course no guarantee that the same organization exists at all earlier stages of human development.
4.
Neuroscience and social understanding
In reviewing research on the neuroscience of social understanding, we will organize the initial review into sections dealing with action IRs, emotion IRs, and epistemic IRs, respectively. In these sections our concern will be to identify brain regions and processes that deal primarily with first- and third-person information separately, from areas where first- and third-person information meet and
49
50
John Barresi and Chris Moore
where their integration makes possible relatively uniform application of these representations to both self and others. Where first- and third-person information is separated we would expect them to apply differentially to self and other, with first-person information tending to apply mostly to self and third-person information mostly to other. Where they are integrated, the question becomes how we use this integrated information to distinguish between self and other. We will also identify regions in which lower level perceptual processing can be distinguished from higher level metacognitive processing. Finally, we identify research indicating that first- and third-person information is sometimes represented independently, in particular in the case of autistic individuals. Figure 1 depicts essential compontents of IRT along with possible anatomical correlates that will be described in subsequent sections of this chapter.
4.1 Action intentional relations Since the discovery of mirror neurons in the premotor cortex in monkeys that respond to the goal-directed actions of others (Rizzolatti et al. 1996), studies have investigated whether evidence can be found for similar neural structures in humans. A standard paradigm used in a number of these studies is to compare an observation condition, where participants watch the activity of another person, an execution condition where participants perform the action on cue, and an imitation condition, where participants perform the action that they observe another person perform. Transitory Magnetic Stimulation (TMS) studies affecting processing in the relevant neural systems have attempted either to facilitate/produce actions in observation conditions, or to interfere with actions in action or imitation conditions (see Iacoboni 2005, for a review). Taken together these studies affirm that premotor and parietal cortices in humans show mirror properties similar to those in individual neurons of monkeys. Both of these areas are active when performing the actions or observing the actions of others, and more active than in either of these conditions when these actions are both observed and imitated. In contrast to the additional activation found in these two regions (premotor and parietal) when imitating compared to mere observing, a third region, the Superior Temporal Sulcus (STS), tends to show the same level of activation in both observation and imitation conditions but is inactive in the action-only condition. Iacoboni, Kaplan and Wilson (in press) have proposed a model incorporating IRT in accounting for these findings. They propose that the STS provides thirdperson visual information of the action that is being performed. This information is transferred to the Posterior Parietal, where it is matched with first-person information on the kinesthetic, kinematic and somatosensory properties that might go
Figure 1. Essential components of Intentional Relations Theory and possible anatomical correlates. Third-person information involves exterior senses, which tend to apply more to others than to self; first-person information involves action intentions and interior senses, which tend to apply more to self than to others. Intentional schemas are posited to involve multimodal association areas where first- and third-person information get integrated. Although our main focus is on object-directed intentional relations, non-object directed integration are expected to occur at body-schema levels as well. The central site for the integration of first- and third-person information involving agents in intentional relations is hypothesized to be the temporal-parietal junction (TPJ) and/or inferior parietal (IP), which is hypothesized to involve an egocentric or first-person representation of the agent in space in the right hemisphere and a connected allocentric or third-person representation of the agent in the left. Second order, or reflective representations of intentional relations are hypothesized to occur in the prefrontal cortex. The directions of arrows represent the dominant direction of information processing, though feedback and other connections also occur between anatomical regions both within and between boxes of the model.
The neuroscience of social understanding 51
52
John Barresi and Chris Moore
with the action – information provided by internal first-person sources of information integrated in the inferior parietal (see Figure 1). This matched representation of embodied action is then forwarded to the pre-motor area where alternative action plans can be compared to this input. This feed forward mechanism, is then matched to information being fed back from alternative pre-motor plans, and an interpretation is made, in the inferior parietal, between alternative interpretations. In their model, both the pre-motor area and the inferior parietal areas involve matching between first- and third-person properties and so are attributed to involve integration of first- and third-person information by intentional schemas. One way to conceive of the relationship between these two areas is that the inferior parietal (and/or nearby Temporal/Parietal Junction – TPJ) provides an egocentric, body-centered representation of the source of action of an agent-inworld, while the pre-motor area represents the goal or object of the action. Both require matching of first- and third-person information and together provide a full representation of the action intentional relation. From the point of view of IRT, the more important area of integration of first- and third-person information is the inferior parietal or TPJ, rather than the pre-motor area, particularly as this area seems to reappear on complex ToM tasks, and may be crucial for distinguishing self and other as intentional agents. Whereas mirror neurons in the pre-motor area may be insensitive to the difference between self and other and focus mainly on the goals of actions, something that monkeys and young infants can represent, we would hypothesize that left and right parietal regions represent agents in intentional relations, and might be used to distinguish self from other as intentional agents. Studies by Decety and his colleagues (see Decety and Grezes 2006 for a review) provide support for the idea that the TPJ is the locus of a body-centered integration of first- and third-person information that applies both to self and to other but that may also be used to distinguish self from other. In these studies, imitations of other-by-self or self-byother are compared. The general finding is that TPJ (they include studies citing inferior parietal as well as posterior STS) is more active on the right side when other imitates self, but more active on the left side, when self imitates other. One way to interpret this difference is that left TPJ is more active, when a third-person representation of a human body in space is more dominant than a first-person representation, and that the reverse is true for right TPJ. In other words, when the participant is the original source of the action, the right hemisphere is dominant and when the participant is imitating the other, the left hemisphere is dominant. More typically, we would suggest that when left TPJ is dominant, another person is being represented, where third-person information is perceived but first-person information is imagined (what might be called an allocentric representation of a person in space). However, when right TPJ is dominant, it is the self that
The neuroscience of social understanding
is typically being represented, where first-person information is perceived and third-person information imagined (what might be called an egocentric representation). Independent support for this idea comes from studies of brain damage on these two sides. As we shall see, damage at the left TPJ is found to be associated with failure at false belief tasks involving representations of others, whereas other studies have demonstrated that damage at the right TPJ is associated with spatial neglect, a distortion of egocentric or first-person perspective of space (see Halligan, Fink, Marshall and Vallar 2003). Furthermore, damage at the TJP (or IP) has recently been shown to produce autoscopic hallucinations – seeing oneself – with right-sided damage associated with a non-egocentric out-of-body experience of self, and left-sided damage associated with an egocentric seeing of one’s double (Blanke and Mohr 2005). Taken together, these findings support Iacoboni et al.’s application of the IRT to their imitation studies, and their attribution of our notion of intentional schema to the inferior parietal, as they provide independent evidence that the inferior parietal or TPJ is the main center for an integrated representation of a person in space, whether it is self or other. But these findings also highlight how we can distinguish self from other through the source of information that drives the representation, third-person if it is other and first-person if it is self. These findings also provide a basis for connecting the more complex human activities involved in traditional false belief tasks, which have been shown also to require representations involving the TPJ and more mundane actions that are investigated in imitation tasks. However, in considering imitative tasks, it should be noted that imitation of novel actions requires skills that do not appear in monkeys, and only appear in humans in a full blown state during the second year of life, when the infant is forming its concept of an intentional agent. Indeed, two-year-olds find it particularly fascinating to engage in mutual imitation, where they take turns leading and following each other in novel intentional actions, in a manner analogous to the Decety studies. This play behavior can be interpreted as working out possibilities made available at this time by developments in the use of the intentional schema, both to understand self and other individually and to discriminate self from other even in contexts, where both actors are performing similar actions.
4.2 Affective and motivational intentional relations Typically, when dealing with action IRs, first-person information directly involves motor plans, proprioception, and kinesthetic feedback, while third-person information directly involves visual and auditory information. The integration of these
53
54
John Barresi and Chris Moore
sources of information yields representations of a body acting in space with these first- and third-person resources integrated into a representation that can be applied to self or other, possibly through the use of vision and audition as bridging modalities that provide information about actions of self as well as other. Even so, there is a residual motor component, including a readiness to act (see, e.g., Ramnani and Miall 2004), as well as the sense of agency previously discussed that tends to distinguish self from other. When it comes to affective and motivational IRs, the focus is more on sensation than on action. So, the distinction between first-person information and third-person information and their integration, will tend to focus more on internal states within the body rather than on external appearances and expressions. Research involving such affective intentional relations has been consistent in showing the importance of integrated somatic representations of internal feeling states of a person whether such representations are applied to self or other. Generalizing such representations of internal states to another person occurs even when there is no social judgment involved in the task and where the participant merely observes the other. Recent research on pain has been particularly revealing. With respect to pain in self and other, single cells in the Cingulate Cortex (CC) have been found to respond not only to own pain, but also to the appearance of pain in another (Hutchinson et al. 1999). This response occurred even though no instructions to empathize were involved. In an fMRI study of empathy for another’s pain, where again no instructions to empathize were involved, Singer, et al. (2004) had female participants and their partners receive mild shocks following a signal which indicated who was to receive the shock. The participants could see the hands of self and other as well as the signals while they were in the magnetic resonance chamber. It was found that certain primary somatosensory areas responded only to pain in self, but that the Anterior Insula, and the CC responded to the shock signal and anticipated pain both in self and in other. It has been hypothesized by Craig (2003) and Damasio (1999) that the anterior portion of the Insula, particularly on the right side, is a recently evolved region of the brain that represents a “feeling self ”. This region and the CC may both be involved in conscious representation of pain, in contrast to the primary sensory cortex, which may measure the intensity and sensory quality of the pain stimulus, but which may not always contribute to consciousness of pain. Part of the evidence for the distinction is that placebo effects, where perception of pain is induced, produce activations in Anterior Insula but not in the primary sensory areas (Wager et al. 2004). It seems then that, like mirror regions in the pre-motor and parietal areas, this ‘feeling self ’ level of representation of pain is responsive, not only to one’s own feeling of pain, but to the expressed, or merely inferred, pain of another person.
The neuroscience of social understanding
What the Singer et al. (2004) study seems to show is that the areas involved in conscious perception or feeling of one’s own pain, are also active for the anticipated pain of another. Without instructions to do so, the participants seem to participate empathically in the anticipated pain of the other, thus sharing in it, and presumably being aware of their pain by sharing in it. Further support for this interpretation comes from the fact that dispositional measures of empathic ability were obtained in this study and a correlation between degree of dispositional empathy and degree of activation in the Insula and CC for the observeother condition was found. Therefore, not only does the third-person perception of the other’s behavioral situation apparently result in a conceptual understanding of the feeling state of the other, but it actually induces a comparable feeling state in the observer, which may be the ground upon which conceptual understanding is based. The degree to which this internal feeling state is induced seems to depend on the capacity for empathy, or sympathetic imagination, of the observer. However, as a subsequent study shows (Singer et al. 2006), it also depends on how one feels about the other person. If one has reason to like the other, then there is a stronger tendency to show an empathic response to the other’s pain, than if one has reason to dislike the other person. In the latter case, men, but not women, were shown not to have this empathic response to the other’s pain, but instead showed evidence of personal pleasure at seeing the other in pain. So the story here is fairly complex. Unlike the action mirror system in the pre-motor area, which seems to depend only on attention to the activity of the other, the degree of identification with or caring for the other may matter in representing the feeling states of the other in the same mode as one’s own feeling states. In the original Singer, et al. (2004) study, as well as in similar studies on observing touch (Keysers et al. 2004), and disgust (Wicker et al. 2003) in others, primary sensory areas could be used to provide first-person information that distinguished between self and other. However, subsequent research on observation of localized pain inducing stimuli on another person raises the issue of whether primary sensory areas are immune to empathically induced responses. For instance, Avenanti et al. (2005) had participants observe needles being pierced into the hand of another person and found TMS motor cortex induced inhibitory responses of hand muscles that matched those that would have occurred in their own case. Based on this and other findings, Singer and Frith (2005) have suggested that whether one is attending to – or imagining – the emotional response of the other person or the sensory quality of the pain may be what distinguishes these two kinds of results. The implication of this is that to the extent that one can project oneself into the particular situation and experiential state of the other to that extent will one tend to display a matching embodied state. According to IRT, it is the fact that one has at one’s disposal this personal shared experiential base upon
55
56
John Barresi and Chris Moore
which to understand the state of the other person that one succeeds in accurately imagining that state. But to elicit such an internal state that typically applies to self when observing another, a matching must occur between the expressed state of the other and one’s own associated experience of being in a comparable state, or be elicited by attending to the situation that the other is in as if it were shared. In the case of an expressed state this requires matching of first-person information about the appropriate internal state to third-person information about expressed state. So motor aspects of the behavior of others may be a mediating factor in situations where we have no direct personal experience of emotional responses in those situations, or where we would respond differently from the other person. Several other studies conducted by Iacoboni and his colleagues indicate that mirroring of expressed affective states may be an important basis for understanding emotions in others. In these studies fMRI brain imaging of participants occurred either while they were engaged in observing or imitating a variety of emotional expressions depicted in photos (Carr et al. 2003; Dapretto et al. 2006). In the study reported by Carr et al. (2003), observing and imitating emotional expressions in others activated regions involved in those emotional expressions for self, in particular the amygdala and insula were involved, but also the pre-motor area and STS. Again these results can be interpreted as eliciting from third-person information (STS) the matching first-person action information necessary to understand the internal state of the other individual. Because the observation and imitation condition had similar pre-motor findings to action studies, this suggests that implicit if not explicit matching of emotional expression is involved in emotional empathy, which may feed into the representation of the feeling self in the insula. So far we have seen that matching between first- and third-person information seems to occur when observing another person’s affective state, and it may not require active use of imagination to feel and understand another’s affective IRs in that a form of affective sharing may occur directly in response to the situation or the other’s expression. Indeed, from a phylogenetic as well as developmental perspective contagion of emotional states from one organism to another is the original basis of emotional sharing (cf. Zlatev this volume). However, as we have argued, sharing a psychological state is not the same as understanding that state. Other evidence suggests that understanding affective states in the sense of attributing emotions and other affective states to individuals as well as discriminating one’s own from another’s emotional state, likely requires frontal activity, and occurs later in human development. It appears necessary to have the involvement of frontal areas, in particular, the Medial Prefrontal Cortex (MPFC), in order to reflect on and understand the mental state as either one’s own, or another’s.
The neuroscience of social understanding
The role of the MPFC in understanding at a reflective level pain states in self and other is highlighted in another recent study directly comparing imagination of self and other in pain as compared to damage to a manikin figure (Jackson et al. 2006). While in a magnet, participants viewed images of arms and legs apparently from a first-person perspective in situations likely to be painful or neutral. They were told to imagine the body part as their own, or another person’s, or that of a manikin. In line with the notion that the MPFC is involved in representing second order IRs, there was a strong response in this region only for the humans, but not for the manikin. In addition, there was differential activation in the posterior cingulate, which responded to pain in self and other, and to the inferior parietal. As in previous studies the insula and ACC were responsive to both self and other in a comparison between pain and non-pain conditions. But differences between self and other also occurred. The comparison between self and other found several regions of difference, indicating different routes to representing the same pain state in self and other, and the ability to distinguish between our own and another’s pain. Taken together the results on emotional processing show that matching can occur not only in the motor system where actions or expressions of others are mimicked, perhaps subpersonally, but that feeling states that are connected to those expressions in ourselves are often also active when observing others or in inferring their emotional states in conditions where sympathetic contagion or empathy might be elicited. These internal feeling states are then processed further in frontal areas when we are attempting to understand the emotional state of the other as distinguished from our own emotional response. Both the matching in the premotor area and in the feeling self can be viewed as first-person aspects of emotional IRs, while the visual expressions can be viewed as third-person aspects. However, for second-order representations of these IRs, frontal activity is necessary.
4.3 False belief and complex social inference tasks A considerable amount of research has been devoted to establishing the brain basis of the understanding of the more complex intentional relations characteristic of “theory of mind”. The focus of studies using ToM tasks is on determining brain regions functionally involved in the interpretation of complex stories of social interaction that are visually or verbally presented and in attributing mental states to individuals in these stories. Two brain regions have been shown to be most active in brain imaging studies using various techniques, when compared to control conditions involving comparable processing of non-ToM stimuli: (1) The Temporal/Parietal Junction (TPJ; including neighboring Superior Temporal regions
57
58
John Barresi and Chris Moore
i ncorporating STS as well as Inferior Parietal regions, cf. Decety and Grezes 2006); (2) The Medial Prefrontal Cortex (MPFC). The TPJ is believed to be an area in which complex visual stimuli, often involving biological motion and social interaction, are analyzed or represented perceptually and semantically. In the section on action IRs our discussion of the TPJ focused only on intentional actions of a single agent, but the TPJ is also crucial for social interactions and for interpreting more complex mental states than actions. Hence, in terms of IRT the TPJ, at least on the left side, can be understood as representing the third-person information about IRs of one or more organisms, involved in simple or complex object and interpersonal interactions. For instance, even in monkeys this area has individual neurons that are sensitive to eye direction of a person being observed by the monkey and the congruence with the person’s behavior involving another object, with their direction of gaze. Comparable findings with humans, involving more complex IRs, for instance, involving intentions, have been made using fMRI (see, e.g., Pelphery et al. 2004). So this region can pick up epistemic as well as action IRs and is also involved in emotion IRs, involving multiple agents. The second region of importance for the ToM tasks is the MPFC. This region appears to be important for “decoupling” (Leslie 1994), or creating second order representations of IRs that can be attributed to individuals. Reflective or conceptual understanding of the intentionality of the behavior seems an important activity for this region. Indeed, merely noting a stimulus as an act of an intentional agent rather than a machine seems sufficient to involve this region (Ramani and Miall 2004). But this region has a number of other functions of a metacognitive, or executive, sort, and there appear to be subregions with specialized functions, some of which we will consider shortly. Some recent elegant research using simple false belief tasks presented in stories and in videos, along with a number of important controls, to brain damaged patients with frontal and/or temporal-parietal lesions (Apperly, Samson, Chiavarino and Humphreys 2004; Samson, Apperly, Chiavarino and Humphreys 2004; Samson, Apperly, Kathirgamanathan, and Humphreys 2005) has provided evidence in partial congruence with these imaging studies. They found that damage to the left TPJ produces a fairly specific deficit in false belief reasoning about others, but that damage in the frontal regions does not. So it appears that a functional TPJ at least on the left (no tested patients had right TPJ damage) is necessary for false belief reasoning. By contrast, it appears that the impact of brain damage in frontal regions is less specific and more diverse, including effects on performance on tasks involving executive function but not on ToM tasks. Indeed, in one of their patients with frontal damage, there was evidence that problems occurred only on false belief tasks that required the inhibition of first-person knowledge of the real location but not on false belief tasks for which the participant
The neuroscience of social understanding
did not have knowledge of the real location (Samson et al. 2005). This result is congruent with other findings which suggest that executive function associated with frontal activity may be necessary to differentiate between mental states of self and other, and thus for attributing distinct mental states to individuals. In these circumstances a single mental state that is shared between self and other that might be used in cases of passive observation or empathic responding, will not be sufficient for mental state attribution. If we consider just the two main regions involved in research with complex ToM tasks, these results fit well with what we would expect based on the “theory theory” (TT) approach to social understanding. The TPJ provides third-person behavioral analysis of animate activity or apparently animate activity, while the MPFC decouples or represents abstractly IRs, presumably in a theoretical or conceptual format. That the same behavioral analysis and conceptual representation could be applied to self and other is suggested by the fact that the MPFC shows overlap in activity for a variety of tasks involving self and other (e.g., see Decety and Sommerville 2003, for a summary of this research). It is possible that, in line with TT, TPJ analyzes and represents animate activity and IRs based mostly on visual or third-person information. As such the matching problem may not arise if the IRs of self and other are both analyzed in a behavioristic (or third-person) mode. MPFC could then provide “decoupled” (second order) representations of intentional relations of agents, whether they are of self or other (or jointly self and other). However, the fact that the MPFC (and perhaps the TPC, particularly on the right side) is activated in cases of self-representation that seem not to be based entirely on third-person information about the self suggests that integrated representations involving both first- and third-person information of the kind postulated by IRT are involved. Furthermore, the frontal region and other regions along the midline have been postulated to be part of a system for representation and regulation of self (Northoff and Bermpohl 2004). So, perhaps, the MPFC generates a second order representation of another’s mental states, through prior association between a third-person behavioral analysis mainly from the left TPJ that applies more often to another person than to self and a simulation of firstperson components of mental states found in the rest of the typically right-sided self-system. This latter interpretation is consistent with studies showing differential responses for self and other in high level processing of social stimuli (e.g., Lou et al. 2004). The main conclusion to be drawn from these studies is that complex ToM tasks involve two main regions of the brain, a posterior one associated with perceptual representation of IRs and an anterior one associated with metarepresentation of these perceptual representations. Furthermore, there is a good deal of
59
60 John Barresi and Chris Moore
overlap between the regions involved in representing self and other. Nevertheless, differences that occur suggest a mapping of third-person information typical of what we have from others to first-person information more typically associated with self. While the need for perceptual and metarepresentational processes for understanding individual IRs in complex ToM tasks is congruent with TT, the overlap between self and other, and use of first-person information as well as third-person information in these tasks fits better with the IRT approach to social understanding.
5.
Level 1 understanding of intentional relations – the case of autism
Finally, it is worth mentioning some imaging research that supports the notion that representations of intentional relations can occur in distinct forms. Dapretto et al. (2006) studied high functioning autistic children and matched controls using the same imitation task as in the Carr et al. (2003) study mentioned earlier. However, in addition to observing and imitating emotional expressions, the autistic participants were measured on severity of autism, using several standardized scales. The behavioral findings were that the autistic participants were as able to imitate emotional expressions as other children, but the imaging findings suggested that the means that they used were different. The typically developing children replicated the results of the adult study, where mirror neuron pre-motor and insula areas were involved in observation and imitation of emotions, along with other areas. But in autistic children these mirror neuron areas were not as involved, and degree of involvement of these areas during imitation was inversely related to severity of autism in the social domain. Furthermore, other areas, the left anterior parietal and the right visual association areas, were more involved for autistic than for typical children. It was suggested that these latter areas served as an alternative route to imitation in this group instead of the usual one involving the mirror neuron system. These results, combined with other findings, support the notion put forward by Barresi and Moore (1996) that the main reason why autistic people have difficulty in ToM tasks as well as emotion understanding and imitation is that they do not match and integrate first- and third-person information through an intermodal intentional schema, hence that they acquire and deploy independent first-person (or egocentric) and third-person (or allocentric) theories of mind. At the time that we wrote our article we had no idea how the notion of intentional schema might relate to brain activity. However, with the discovery of mirror neurons at about the same time, we, as well as others (e.g., Iacoboni et al. 2007) have
The neuroscience of social understanding
been able to make the connection. The Dapretto et al. study probably provides the best confirmation for the view that it is the lack of matching of these two types of information through an intentional schema that is at the heart of problems in social understanding of autistic individuals. The inability to readily transform third-person perceptions into first-person matching experiences, as well as to make the reverse mapping, and thus to engage in mind sharing, makes it difficult for autistic individuals to make sense of mind, because of the absence of a direct connection between the two necessary, inseparably tied aspects of all mental phenomena, an externally available bodily expressive component, and an internally available feeling component. As a result of this deficiency in ability to share mind with others, they lose interest in other people, and have difficulty learning from them. Eventually, if they do attempt to reflect on and understand mind in self and others, they form two radically different accounts: on the one hand they develop rather complex TT-like accounts of mind from a third-person view of their own and other people’s behavior; and on the other hand they overgeneralize in apparent simulation their own egocentric first-person perspective to others (cf. Frith and de Vignemont 2005). Because of lack of mind sharing during infancy and beyond, they are faced with intractable problems in understanding mind beyond those that appear as purely third-person TT types, or purely first-person ST types, instead of integrated theories where matching of first- and third-person information is involved as we have proposed in IRT.
6.
Conclusion
Recent discoveries in the neuroscience of social understanding have opened a new window through which to evaluate theories of social understanding. In the present chapter we have primarily examined our own intentional relations theory (Barresi and Moore 1996) in light of these new discoveries. IRT has three important elements. First, it postulates a distinction between first- and third-person information pertaining to intentional relations, as well as a requirement that both forms of information be combined in order to generate representations of intentional action that are shared between, or equally applicable to, self and others. Second, it postulates that a distinction may be made between a level of social understanding at which first- and third-person information are integrated without being attributable to individual agents and more complex levels of social understanding at which integrated representations are recognized to be properties of individual agents. In human ontogeny (and possibly in phylogeny), the latter levels of social understanding are founded on the former level. Third, it postulates that under
61
62
John Barresi and Chris Moore
certain conditions first- and third-person information about intentional relations may be processed separately so that the activities of self and other are represented independently. In humans, such a condition is seen in autism. The neuroscience of social understanding shows that integrating first- and third-person information through matching these two types of information occurs in the understanding of action, emotion, as well as epistemic IRs of self and other. The fact that matching between first- and third-person aspects of IRs for self and other occurs immediately on-line for a variety of IRs is congruent with the notion that both aspects are necessary to fully extract the meaning of these activities. Such matching occurs early on in life, though this process of mindsharing does not develop into understanding individual minds until later in development. On our account, it is only through processes that bring about shared psychological states between individuals early on, and provide the initial basis for social understanding, that later development of our usual understanding of individual minds becomes a possibility. Although TT might account for some instances of theories of mind generated purely from behavior, it is only in autistic individuals were we see exaggerated “theories” of this type. However, in autistic individuals there is evidence of a failure in mapping first- and third-person information from very early on in life, which prevents shared mental activity in dyadic interactions. ST does better than TT in accounting for a variety of phenomena involving emotional empathy, and understanding epistemic states. But it cannot account, without special pleading, for matching phenomena involved in action understanding. Again, autistic individuals provide a window into the problem. They can generalize either first- or third-person representations separately from self to other or the reverse. However, because they did not initially engage in shared mental life with others, they have problems understanding the meaning of social activity when the integration of both first- and third-person information is involved. Without prior matching and integrating these two types of information in earlier shared mental activity associated with dyadic and triadic interactions, the concepts that they generate based either on behavior alone or internal states alone are diminished when compared to our usual understanding of IRs of self and other. Thus, we believe that matching theories like IRT provide the best account of how we come to understand our own as well as other minds.
The neuroscience of social understanding
References Apperly, I.A., Samson D., Chiavarino C. and Humphreys G.W. 2004. “Frontal and left temporoparietal contributions to theory of mind: Neuropsychological evidence from a false belief task with reduced language and executive demands.” Journal of Cognitive Neuroscience 16: 1773–84. Avenanti, A., Bueti, D., Galati, G. and Aglioti, S.M. 2005. “Transcranial magnetic stimulation highlights the sensorimotor side of empathy for pain.” Nature Neuroscience 8: 955–960. Barresi, J. 2001. “Extending self-consciousness into the future.” In The Self in Time: Developmental Perspectives, C. Moore and K. Lemmon (eds.), 141–161. Mahwah, NJ: Lawrence Erlbaum Associates. Barresi, J. 2004. “Intentional relations and divergent perspectives in social understanding.” In Ipseity and Alterity: Interdisciplinary Approaches to Intersubjectivity, S. Gallagher and S. Watson (eds.), 74–99. Rouen: Presses Universitaires de Rouen. Barresi, J. and Moore, C. 1996. “Intentional relations and social understanding.” Behavioral and Brain Sciences 19: 107–154. Blanke, O. and Mohr, C. 2005. “Out-of-body experience, heautoscopy, and autoscopic hallucination of neurological origin Implications for neurocognitive mechanisms of corporeal awareness and self consciousness.” Brain Research Reviews 50: 184–199. Carr, L., Iacoboni, M., Dubeau, M.C., Mazziotta, J.C. and Lenzi, G.L. 2003. “Neural mechanisms of empathy in humans: A relay from neural systems for imitation to limbic areas.” Proceedings of the National Academy of Science, USA 100: 5497–5502. Craig, A.D. 2003. “Interoception: The sense of the physiological condition of the body.” Current Opinion in Neurobiology 13: 500–505. Damasio, A.R. 1999. The Feeling of What Happens: Body and Emotion in the Making of Consciousness. New York: Harcourt Brace. Dapretto, M., Davies, M.S., Pfeifer, J.H., Scott, A.A., Sigman, M., Bookheimer S.Y. and Iacoboni, M. 2006. “Understanding emotions in others: mirror neuron dysfunction in children with Autism Spectrum Disorder.” Nature Neuroscience 9: 28–30. Decety, J. and Grezes, J. 2006. “The power of simulation: Imagining one’s own and other’s behavior.” Brain Research 1079: 4–14. Decety, J. and Sommerville, J. 2003. “Shared representations between self and other: A social cognitive neuroscience view.” Trends in Cognitive Sciences 7: 527–533. Frith, U. and de Vignemont, F. 2005. “Egocentrism, allocentrism, and Asperger syndrome.” Consciousness and Cognition 14: 719–738. Gallagher, S. and Hutto, D.D. this volume. “Understanding others through primary interaction and narrative practice.” Gallese, V., Keysers, C. and Rizzolatti, G. 2004. “A unifying view of the basis of social cognition.” Trends in Cognitive Science 8: 396–403. Gallese, V., Fadiga, L., Fogassi, L. and Rizzolatti, G. 1996. “Action recognition in the premotor cortex.” Brain 119: 593–609. Goldman, A. 1992. “In defense of the simulation theory.” Mind and Language 7: 104–119. Gopnik, A. 1993. “How we know our minds: The illusion of first-person knowledge of intentionality.” Behavioral and Brain Sciences 16: 1–14. Gordon, R. 1986. “Folk psychology as simulation.” Mind and Language 1: 158–171.
63
64 John Barresi and Chris Moore
Halligan, P.W., Fink, G.R., Marshall, J.C. and Vallar, G. 2003. “Spatial cognition: Evidence from visual neglect.” Trends in Cognitive Sciences 7: 125–133. Harris, P. 1989. Children and Emotion. Oxford: Basil Blackwell. Hobson R.P. 1991. “Against the theory of ‘theory of mind’.” British Journal of Developmental Psychology 9: 33–51. Hobson, R.P. 1998. “The intersubjective foundations of thought.” In Intersubjective Communication and Emotion in Early Ontogeny, S. Braten (ed.), 283–296. Cambridge: Cambridge University Press. Hobson, R.P. 2002. The Cradle of Thought. Exploring the Origins of Thinking. London: Macmillan. Hobson, R.P. and Hobson, J. this volume. “Engaging, sharing, knowing: Some lessons from research in autism.” Humphrey, N. 1984. Consciousness Regained. Oxford: Oxford University Press. Hutchison, W.D., Davis, K.D., Lozano, A.M., Tasker, R.R. and Dostrovsky, J.O. 1999. “Pain-related neurons in the human cingulate cortex.” Nature Neuroscience 2: 403–405. Iacoboni, M. 2005. “Understanding others: Imitation, language, empathy.” In Perspectives on Imitation: From Neuroscience to Social Science, S. Hurley, and N. Chater (eds.), 77–99. Cambridge, MA: MIT Press. Iacoboni, M., Kaplan, J. and Wilson, S. in press. “A neural architecture for imitation and intentional relations.” In Imitation and Social Learning in Robots, Humans and Animals: Behavioural, Social and Communicative Dimensions, C. Nehaniv and K. Dautenhahn, (eds.). Cambridge, UK: Cambridge University Press. Jackson, P.L., Brunet, E., Meltzoff, A.N. and Decety, J. 2006. “Empathy examined through the neural mechanisms involved in imagining how I feel versus how you feel pain.” Neuropsychologia 44: 752–761. Keysers, C. and Perrett, D.I. 2004. “The neural correlates of social perception: A Hebbian network perspective.” Trends in Cognitive Sciences 8: 501–507. Keysers, C., Wicker, B., Gazzola, V., Anton, J., Fogassi, L. and Gallese, V. 2004. “A touching sight: SII/PV activation during the observation and experience of touch.” Neuron 42: 335–346. Knoblich, G. and Jordan, J.S. 2002. “The mirror system and joint action.” In Mirror Neurons and the Evolution of Brain and Language, M.I. Stamenov and V. Gallese (eds.), 115–124. Amsterdam: John Benjamins. Leslie, A.M. 1987. “Pretense and representation: The origins of ‘theory of mind’.” Psychological Review 94: 412–426. Leslie, A.M. 1994. “ToMM, ToBy, and Agency: Core architecture and domain specificity.” In Mapping the Mind: Domain Specificity in Cognition and Culture, L.A. Hirschfeld and S.A. Gelman (eds.), 119–148. New York: Cambridge University Press. Lou, H.C., Luber, B., Crupain, M., Keenan, J. P., Nowak, M., Kjaer, T.W., Sackeim, H.A. and Lisanby S.H. 2004. “Parietal cortex and representation of the mental Self.” Proceedings of the National Academy of Science, USA 101: 6827–6832. Matsuzawa, T. 2007. Comparative cognitive development. Developmental Science 10: 97–103. Moll, H. and Tomasello, M. 2005. “12- and 18-month-old infants follow gaze to spaces behind barriers.” Developmental Science 7: F1–F9. Moore, C. 1999. “Intentional relations and triadic interaction.” In Developing Theories of Intention, P. D. Zelazo, J. W. Astington and D. R. Olson (eds.), 43–62. Mahwah, NJ: Lawrence Erlbaum Associates.
The neuroscience of social understanding
Moore, C. 2006. “Representing intentional relations and acting intentionally in infancy: Current insights and open questions.” In Human Body Perception from the Inside Out, G. Knoblich, I. Thornton, M. Grosjean and M. Shiffrar (eds.), 427–442. New York: Oxford University Press. Moore, C. 2007. “Understanding self and other in the second year.” In Transitions in Early Socioemotional Development: The Toddler Years, C.A. Brownell and C.B. Kopp (eds.), 43–65. New York: Guilford Press. Northoff, G. and Bermpohl, F. 2004. “Cortical midline structures and the self.” Trends in Cognitive Sciences 8: 102–7. Prinz, W. 1997. “Perception and action planning.” European Journal of Cognitive Psychology 9: 129–154. Ramnani, N., and Miall, C.R. 2004. “A system in the human brain for predicting the actions of others.” Nature Neuroscience 7: 85–90. Repacholi, B.M. and Gopnik, A. 1997. “Early reasoning about desires: Evidence from 14- and 18-month-olds.” Developmental Psychology 33: 12–21. Rizzolatti, G., Fadiga, L., Gallese,V. and Fogassi, L. 1996. “Premotor cortex and the recognition of motor actions.” Cognitive Brain Research 3: 131–141. Samson D., Apperly I.A., Chiavarino C. and Humphreys G.W. 2004. “Left temporoparietal junction is necessary for representing someone else’s belief.” Nature Neuroscience 7: 499–500. Samson, D., Apperly, I. A., Kathirgamanathan, U., and Humphreys, G.W. 2005. “Seeing it my way: A case of a selective deficit in inhibiting self-perspective.” Brain 128: 1102–1111. Sebanz, N. and Frith, C. 2004. “Beyond simulation? Neural mechanisms for predicting the actions of others.” Nature Neuroscience 7: 5–6. Singer, T. and Frith, C. 2005. “The painful side of empathy.” Nature Neuroscience 8: 845–846. Singer, T., Seymour, B., O’Doherty, J., Kaube, H., Dolan, R. and Frith, C. 2004. “Empathy for pain involves the affective but not sensory components of pain.” Science 303: 1157–1162. Singer, T., Seymour, B., O’Doherty, J.P., Stephan, K.E., Dolan, R.J. and Frith, C.D. 2006. “Empathic neural responses are modulated by the perceived fairness of others.” Nature 439: 466–469. Sommerville, J.A., Woodward, A.L. and Needham, A. 2005. “Action experience alters 3-monthold infants’ perception of others’ actions.” Cognition 96: B1–B11. Swanson, D., Apperly, I.A., Kathirgamanathan, U. and Humphreys G.W. 2005. “Seeing it my way: A case of a selective deficit in inhibiting self-perspective.” Brain 128: 1102–1111. Tomasello, M. 1992. First Verbs: A Case Study of Early Grammatical Development. New York: Cambridge University Press. Tomasello, M. 1999. The Cultural Origins of Human Cognition. Cambridge, MA: Harvard University Press. Tomasello, M., Carpenter, M., Call, J., Behne, T. & Moll, H. 2005. Understanding and sharing intentions: The origins of cultural cognition. Behavioral and Brain Sciences 28: 675–691. Trevarthen, C. & Hubley, P. 1978. Secondary intersubjectivity: Confidence, confiding and acts of meaning in the first year. In Action, Gesture, and Symbol: The Emergence of Language. A. Lock (ed.), 183–229. New York: Academic Press. Wager, T.D., Rilling, J.K., Smith, E.E., Sokolik, A., Casey, K.L., Davidson, R.J., Kosslyn, S.M., Rose, R.M. and Cohen, J.D. 2004. “Placebo-Induced Changes in fMRI in the Anticipation and Experience of Pain.” Science 303: 1162–1167.
65
66 John Barresi and Chris Moore
Wicker, B., Keysers, C., Plailly, J., Royet, J.-P., Gallese, V. and Rizzolatti, G. 2003. “Both of us disgusted in my insula: The common neural basis of seeing and feeling disgust.” Neuron 40: 655–664. Wilson, M. and Knoblich, G. 2005. “The case for motor involvement in perceiving conspecifics.” Psychological Bulletin 131: 460–473. Woodward, A.L. 1998. “Infants selectively encode the goal object of an actor’s reach.” Cognition 69: 1–34. Woodward, A.L. 2005. “The infant origins of intentional understanding.” Advances in Child Development and Behavior 33: 229–262. Zlatev, J. this volume. “The co-evolution of intersubjectivity and bodily mimesis.”
chapter 4
Engaging, sharing, knowing Some lessons from research in autism Peter Hobson and Jessica A. Hobson Our aim in this chapter is to consider how intersubjective co-ordination is integral to human forms of interpersonal engagement, sharing experiences with others, and acquiring knowledge about persons with minds. We dwell on three studies involving children and adolescents with autism, each concerned with different aspects of non-verbal communication in greetings and farewells, conversation, and imitation, respectively. Other researchers’ reactions to these studies illustrate how scientists tend to be sceptical of measures (however reliable) intended to capture the intersubjective dimension of personal relatedness. On a more theoretical note, we suggest that intersubjectivity acquires the structure that it does, and has the developmental implications that it does, in virtue of human beings’ propensity to identify with others’ attitudes.
1.
Introduction: Qualities of relatedness
Why do we need to bother ourselves with intersubjectivity? To many scientists, the concept has all the wrong kinds of qualities. It is vague; it seems to be trying to capture something that exists between or among individuals, a systemic property, rather than an identifiable feature or function of a given organism; and it is difficult to operationalize, to quantify or otherwise objectify. Moreover, it smacks of emotion – a matter that should not be so much of a problem, were the concept not framed in a manner so unaccommodating to cognitive/computational, information-processing models of the mind. Our aim in this chapter is to illustrate why the concept of intersubjectivity is indispensable for any account of the development of psychological functioning early in life, and pivotal for understanding the syndrome of autism. We begin by describing three studies of children and adolescents with autism, and indicate why the systemic and emotional qualities of intersubjectivity are so important for interpreting the findings. Each of the studies focuses on different aspects of
68 Peter Hobson and Jessica A. Hobson
non-verbal communication – participants’ greetings and farewells, their bodily expressions during conversation, and their sharing looks during tests of imitation – but they have in common a concern with what such expressions mean as expressions of affective relatedness and interpersonal co-ordination. A primary purpose of the studies is to pinpoint atypical qualities of relatedness among individuals with autism, and the results provide a basis for a discussion of what autism might reveal about the nature of intersubjectivity. We also consider the developmental implications of intersubjective engagement for the ability to share experiences with others and to arrive at knowledge of personswith-minds. Perhaps we should provide brief clarification of our use of four terms: social, interpersonal, intersubjective, and identification. We shall employ the word ‘social’ to refer to happenings between people, without prejudging the degree to which the participants in the exchanges experience each other as persons (rather than as, say, things). The word ‘interpersonal’ is intended to reflect a special form of relatedness to other embodied persons that includes the potential for intersubjective engagement (Trevarthen 1979), that is, connectedness and co-ordination between the subjective orientations of each person involved (see also Susswein and Racine this volume). We shall be suggesting that a process of ‘identifying with’ the attitudes of others is what structures intersubjectivity in human beings – but not, we believe, in other primates – and gives intersubjective transactions the power to shape the course of human cognitive as well as social development. Rather than attempting to define identification at this point – an especially difficult task, given that it is a process that operates on different levels at successive points in development – in what follows we shall illustrate its meaning through specific instances of its expression. Critically, to identify with someone else is to assume (and paradigmatically, be moved by) attitudes perceived in the other, in such a way that those attitudes become a part of one’s own experience (for example, when one shares experiences) and, potentially at least, part of one’s own emotional repertoire. The picture of scientific scepticism that we painted at the beginning of the chapter may seem an unfair caricature. So in describing our three empirical studies, we shall recount some stories of scientists’ reactions to our own research. In each case we shall convey the responses we received when we submitted the findings to mainstream academic journals.
Engaging, sharing, knowing
2.
Three empirical studies
2.1
Hello and goodbye
The first study we shall report appeared in a paper entitled: ‘Hello and goodbye: A study of social engagement in autism’ (Hobson and Lee 1998). In order to capture the spontaneous greetings and farewells of children and adolescents with and without autism relating to an unfamiliar person, a colleague Tony Lee videotaped participants as they entered and departed from a familiar but empty classroom in which there was a stranger (PH, the first author) to whom they were introduced, and from whom they later took their leave. In outline, the findings were as follows. Compared with participants without autism, there were about half as many of those with autism who gave spontaneous expressions of greeting in the ‘Hello’ episode, and a substantial proportion failed to respond even after prompting. All the young people without autism made eye contact, but a third of those with autism failed to do so; no fewer than 17 out of 24 of the former group smiled, but only six out of 24 of those with autism. In the ‘Goodbye’ episode, half the individuals without autism but only three of those with autism made eye contact and said a goodbye. And not only were there few participants with autism who waved in response to PH’s final prompt, but also their waves were strangely uncoordinated and limp. When we designed this study, we were aware that behavioural data would fail to do justice to the intersubjective phenomena we were attempting to capture, even though such conventional ratings might do the job of highlighting atypical forms of social exchange. We expected that we should have to resist others’ attempts to impose a conceptual framework in which the phenomena were reduced to the social transmission of non-verbal communicative cues. Therefore we also asked our judges to look at the greeting episode up to the time the child sat down at the table, and to rate the degree of personal engagement with PH. It turned out that different judges who made these ratings independently were in good agreement with each other. The results were that 14 participants without autism but only two with autism were judged to be in the most strongly engaged category, and only two without autism but 13 with autism in the least engaged category. What do these findings really mean? Here is how one female adolescent with autism negotiated the greeting and farewell. This person gave only the briefest
69
70 Peter Hobson and Jessica A. Hobson
glance towards PH as she entered, and then looked away. As Tony said ‘This is Peter’, she continued to look away for about a second, then looked towards PH without moving her rather set facial expression, and gave a brief and toneless ‘Unn’ in acknowledgement of PH’s presence. Then she looked away to one side, and maintained this lack of eye contact as she walked across the room with her hands linked together in front of her body. She sat down without looking at PH. Once seated, she did not look up at PH across the table. She fixed her gaze towards her lap. Throughout the sequence, she gave little sense of any emotional contact with either adult present. Then when she was told that our session was over, she stood up rather abruptly without making eye contact with PH, and only made any gesture towards PH when he said a first, rather insistent ‘Goodbye’ as she turned to leave. Even here, the gesture was to flap her left hand behind her vaguely in PH’s direction – a wave that hardly seemed like a wave, especially since she was still looking away – and her only remark was a rather nasal and flat ‘Bye’. PH’s final ‘Goodbye’ was met with the faintest of head-turns, another quiet (and hardly expressive) ‘Bye’, and what seemed like a stiff extension of her right wrist behind her body, which might have been a further wave. Although she had seemed aware of PH’s presence, he felt this involved little sense of himself as a person. When we reviewed the videotapes, something else struck us. This concerned PH’s own behaviour. Although he had been trying to relate to each participant in a consistent manner, it seemed that in being unable to sustain a fluency and spontaneity of exchange with the participants who had autism, his own behaviour and gestures became stiff and forced. We shall return to this observation in due course. When we first submitted our paper for publication, it carried the title: ‘Hello and goodbye: A study of interpersonal engagement in autism’. The journal editor who dealt with our manuscript favoured something more neutral about the greeting and farewell ‘behaviors’ of the children we studied. Thanks to our ratings of engagement, we were able to negotiate a compromise title, replacing the word ‘interpersonal’ with ‘social’. We had weathered what was to prove the first of a succession of encounters over our attempts to measure and describe patterns of intersubjective relatedness.
2.2 Head-nodding The second study (García-Pérez, Lee and Hobson 2007) arose out of a previous investigation in which Tony Lee had engaged adolescents with and without autism in conversation in the form of a semi-structured interview (Lee and Hobson 1998). We had videotaped the interviews, and now we wanted to test whether, as we supposed, the intersubjective impairments that characterize autism would be
Engaging, sharing, knowing
manifest in atypical patterns of interpersonal co-ordination in this conversational setting. Two previous studies of this issue by Capps, Kehres and Sigman (1998) and Tantam, Holmes and Cordess (1993) had yielded surprisingly few indications of such abnormalities. In keeping with these previous studies, we decided to apply behavioural measures such as the amount of smiling and head-shaking and nodding, but also predicted group differences when ratings of videotaped interactions were made of two relational characteristics: participants’ degree of affective engagement with the interviewer, and the flow of the dyadic exchange. Beyond this, on the basis of an hypothesis that individuals with autism are seldom ‘moved’ to adopt the bodily expressed psychological orientation of others (Hobson 1993a) – a phenomenon we consider to reflect a limited propensity to align one’s own subjective stance with that of someone else through the process of identification – we anticipated that participants with autism would show fewer episodes of nods and shakes of their heads and a smaller proportion of time looking to the interviewer’s face. We anticipated that these group differences might be more marked at those times when the interviewer was talking than in periods when they (the participants) were talking. The point here is that according to our hypothesis, there should be a specific difficulty when individuals with autism need to accommodate to and connect with someone else’s stance-in-talking, rather than simply needing to show non-verbal communicative expressions. The results were striking for the discrepancy between the very marked group differences that appeared on subjective (but objectively reliable) judgments of affective engagement and interactive flow between the conversational partners, and what seemed to be either absent, or subtle but modest, group differences on behavioural measures of amounts of looking, smiling, and head-nods/shakes. The participants with autism were rated as low in affective engagement and even more markedly discrepant from the control group in the smoothness of their exchanges (in keeping with clinical descriptions by Bosch 1970; Hobson 2002; Kanner 1943), yet they appeared more similar than different in the behavioural components of non-verbal communication. Perhaps there is something about the subtle yet powerful interplay between conversational partners that eludes capture by measures of behavioural events. And when we looked more closely at certain of the behavioural measures, there appeared to be tell-tale signs that all was not well in the interpersonal co-ordination of the exchanges. More specifically, we found that exactly as in the study by Capps et al. (1998), participants with autism often showed an absence of headshakes/nods when the partner was talking, even though the group difference was not significant when the participants themselves were talking. Therefore it seemed that the group difference was not reducible to a general disinclination to nod the head among participants with autism.
71
72
Peter Hobson and Jessica A. Hobson
What might this set of results signify? In our view, it probably signifies that children with autism are limited in the degree to which they identify with another person in conversation. We suggest that in the case of people who do not have autism, one individual nods in accordance with what he/she is saying when he/she is talking, and nods ‘in accordance with him/herself in identification with the other person’ when the other person is talking. In other words, it is because of the kind of engagement people have with the stance (and corresponding ideas) expressed by the other person’s speech and expressive behaviour – an engagement that leads one to adopt the other person’s cognitive-affective orientation in the act of comprehending the other – that the natural, unselfconscious kind of noddingin-communicating follows. Individuals with autism are specifically impaired in this kind of intersubjective linkage and attunement. This interpretation accords with other recent evidence that children with autism are limited in the degree to which they identify with the actions of others in imitative contexts (Hobson and Lee 1999; Hobson and Meyer 2005; Meyer and Hobson 2004). And it is in keeping with the fact that there were marked group differences in affective engagement and the flow of interpersonal exchanges during conversation. In fact, half-way through the study we made one additional prediction. We reasoned that if children with autism have a lowered propensity to identify with and feel moved by another person, then the other person might have a reciprocal difficulty in identifying with individuals who have autism. Therefore we predicted that the interviewer would also show less head-shaking/nodding specifically when the participant was talking, owing to the interviewer’s difficulty in identifying with the stance of the participant. This prediction was borne out – even though the interviewer did not look significantly less to the participants with autism when they were talking, nor was he lacking in smiles. Therefore the result was not simply a reflection of his looking less to the participants, nor showing less feeling. It was, we considered, a reflection of the fact that the intersubjective system of two people in relation to one another was awry. We were reminded of PH’s own stiffness towards participants with autism in the Hello-Goodbye study. In both cases, the workings (or not-workings) of intersubjectivity were reflected in each individual’s behaviour, and in all probability each individual’s experience, towards the other. When we submitted this paper for publication, the reviews were positive. However, one anonymous reviewer expressed concern that the interviewer had failed to interview all the participants in the same way, and recommended that we should control for this effect statistically. Otherwise, the reviewer explained, it is not possible to be sure that it was the participants rather than the interviewer who contributed to the outcome on measures of intersubjective engagement. Of course this raises an important point, for it is possible (though much in the results
Engaging, sharing, knowing
suggested was not actually the case) that the interviewer was systematically biased in his approach to the participants with autism. Yet here the dangers of trying to dissect the essentially interpersonal phenomenon of intersubjectivity into independent parts threatened to undermine the measurement of what was (necessarily) expressed by both components in an interlinked system.
2.3 Sharing looks and self/other-orientated imitation Our third study (Hobson and Hobson 2007) proved the most problematic of all when it came to arguing for its scientific respectability. This was because we invited independent raters to watch videotapes of participants interacting (one at a time) with an adult, and to judge the quality of each look they made to the adult’s face. We had not anticipated how strongly other researchers would contest the appropriateness or even the feasibility of judging what we defined as ‘sharing looks’ in distinction to ‘checking looks’ or ‘orientating looks’. The set-up (originally described in Meyer and Hobson 2004), involved a tester demonstrating actions that might or might not be imitated from the tester’s viewpoint, and then instructing the child: ‘Now you’. Imitation of self/other-orientation occurred when the child adopted the examiner’s demonstrated self/otheranchored orientation, thereby reversing the positioning of the object and directedness of the action. An example of this is when they saw the tester rolling a wheel far-from-herself and close-to-the-participant, and imitated by rolling the wheel far-from-themselves and close-to-the-tester. It turned out that participants with autism were less likely to respond in this way. What we now wanted to test, was our prediction that those participants who imitated the tester’s self/other-orientation would also be those most likely to manifest ‘sharing looks’ towards the tester in the imitation task itself. The rationale was that sharing looks, too, serve as indices of the quality of intersubjective engagement that implicates a degree of identification with the person related-to. The methodology was as follows. As a first step, two raters were found to agree in the amounts of time for which children directed their gaze to the tester. The next stage involved raters judging each look with respect to the quality and/or function of the look according to the following mutually exclusive and exhaustive scheme. ‘Sharing looks’ were defined as those looks directed to the tester that could be seen to express a participant sharing experience through interpersonal contact with the tester. They involved a deep gaze which conveyed personal involvement with reciprocity, depth and affective contact, in contrast to checking looks that involved superficial glances at the tester and were more superficial and lacking in mutuality. ‘Checking looks’ were defined as those looks towards the
73
74
Peter Hobson and Jessica A. Hobson
tester that were used in order to assess or check out either the situation, the tester’s response, or to determine what might happen next. ‘Orientating looks’ were those that appeared to occur in direct response to an action, sound, or movement on the part of the tester. It proved that mostly, such looks were easily distinguished: two independent judges agreed on 89% of the sample of looks they rated according to this three-way classification. The results were that each of the three forms of looking were less prevalent among participants with autism, and most of the participants in each group showed some ‘checking’ and ‘orientating’ looks. However, two-thirds of participants with autism never showed a ‘sharing’ look, whereas this was the case for one-third of the comparison group. Concerning our critical prediction that sharing looks, and only sharing looks, would relate to imitation of self/other-orientation, it turned out that indeed, participants in each group who showed sharing looks tended to be those who imitated the demonstrator’s self/other-orientation, whereas all those with the lowest scores of self/other-orientation also showed a complete absence of sharing looks. Our interpretation of the findings was that sharing involves a structure of interpersonal engagement (involving identification) that is easily overlooked until it becomes manifest through imitative self/ other-orientation. When the paper describing this study was sent out for review, all three of the anonymous reviewers made criticisms of the ratings and/or definition of ‘sharing looks’. The essence of their objections was that, despite the fact that we had established how the ratings could be made with high inter-rater reliability, there needed to be some better kind of behavioural operationalization that defined when a look could be considered a sharing look. We shall hold back from citing chapter and verse of the reviewers’ points, because this seems uncharitable when they are not here to argue their case. It should be evident from the two studies already described that we are not averse to measuring the kinds of behavioural components of communicative exchanges that these reviewers had in mind for our sharing looks. Yet it seems important to make two observations that, in our view, have a bearing on methodological approaches to measuring or evaluating intersubjectivity. Firstly, was it the case that we had failed to operationalize the concepts in terms of which our hypotheses and predictions were framed? Is it not standard science, for example in the domains of psychiatry and social psychology, to appraise all kinds of goings-on through human beings’ judgements of complex processes – with the only requirement being that independent judges arrive at acceptable levels of agreement in their ratings? Why would one suppose that alternative kinds of behavioural ratings would provide more reliable or more valid estimates of ‘sharing looks’? Of course this might be so, although as a matter of
Engaging, sharing, knowing
fact, our subsequent attempts to find some behavioural indices that corresponded with judgements of sharing looks failed miserably, so that (for example) they were often but often not accompanied by smiles, and participants sometimes showed smiles that were associated with other kinds of looks. The issue here is whether the kinds of subjective ratings applicable to intersubjective phenomena are going to be accorded the kind of status that appears to be justified – providing, of course, the usual scientific criteria of inter-rater reliabilities in judgement are satisfied. Secondly, and related to this, recall that our hypothesis was specifically concerned with looks that reflected sharing of experiences. Why? Because we considered that it was in virtue of the fact that sharing was happening that the looks reflected how a participant was so engaged with the tester through identification, that the imitation of self/other-orientation would be likely to occur. Now if this is the case, how are we to establish that any behavioural index or indices of sharing do indeed reflect sharing, without also making subjective judgements of sharing? Such behavioural components do not come ready-flagged, as it were. One reviewer pointed out that previous researchers had found ways of making behavioural ratings that (in his/her view) seemed to capture phenomena that were similar to those we were trying to measure. Yet if this judgement was not corroborated by ratings of sharing, then it seems the interpretation of the data as reflecting sharing must be questionable. At some point, measures of sharing need to enter the picture if claims are going to be made about the behavioural expressions and developmental implications of sharing.
2.4 Overview of the studies One theme has gained prominence through this research. In the Hello-Goodbye study, we saw that individuals with autism are less likely to orientate to and affectively engage with a stranger, or to depart with typical gestures of farewell. For example, all the participants with autism who waved did so with a strangely configured and ill-directed gesture that hardly seemed a wave at all. Why should that be? How do children without autism come to adopt and shape their waving, so that observed waves-from-others-to-self become waves-from-self-to-others? In the study of non-verbal communication during conversations involving a person with autism, again there was a lack of smooth and affectively co-ordinated exchanges, but also evidence of a subtle but deep failure of each conversational partner to link in with the subjective states that found bodily expression in headshakes and nods. Finally, in the tests of imitation and sharing looks, there appeared to be a relation between adopting the self/other-orientation of someone else, and making a connection with that person through sharing looks (see also
75
76
Peter Hobson and Jessica A. Hobson
Barresi and Moore this volume). Connectedness through identifying with another person’s psychological stance is what makes intersubjectivity a system of selves-in-relation-to-other. Studies such as these yield fresh insights into qualities of interpersonal engagement that are pivotal for human social and cognitive development. It is with clinical observations and research on childhood autism firmly in mind, that we turn to consider the nature and implications of intersubjectivity.
3.
The irreducibly social
The answer we give to the rhetorical question with which we began, is that it is only by bothering with intersubjectivity, that one can pinpoint what is essential to the social relations of human beings. If this were not enough, it is only if we accord intersubjectivity an appropriate place within our account of early human development that we shall be able to explain how social and cognitive development take the course that they do. Our purpose in the remainder of this chapter is to explicate what is involved in these claims, from two points of view; firstly, what intersubjectivity entails in terms of person-with-person engagement, and secondly, why it is so central an influence on the growth of the human mind, and so indispensable when we come to consider the kind of profound social impairment that occurs among children with autism. One important line of development that threads its way through an account of very young children’s increasingly differentiated and sophisticated social and cognitive lives concerns their ability to acquire concepts of other people that encompass people’s experiences and psychological orientations – their ‘minds’ – as well as bodies. We put the matter this way, with explicit reference to persons, for reasons that will soon become clear; in short, relative neglect of how the mind is a feature of embodied persons and selves has been damaging for much recent theorizing in developmental psychology. Having said this, such theorizing has also been hugely beneficial in re-focussing attention on the intimate connections among interpersonal understanding (as broadly conceived), communication, and thought. Our first claim, then, is that intersubjective relations between bodily expressive persons are at the core of what is irreducibly interpersonal. Perhaps the quintessential, but by no means only, example of intersubjectivity is to be found in human forms of sharing. To put it bluntly: you cannot share experiences with a stone, a tree, or a squirrel. It is only in a very limited sense that you can share experiences with a dog or chimpanzee – more limited, we believe, than with a two-month-old human infant, and infinitely more limited than with a 10-month-old. Already, then, we may distinguish between what is social, and
Engaging, sharing, knowing
what is interpersonal and more specifically, intersubjective in a uniquely human sense. One can have social relations with a dog or chimpanzee, just as dogs and chimpanzees can have social relations with conspecifics, and such relations are not completely devoid of intersubjectivity; on the other hand, one can relate to other people without this entailing much that is intersubjective. So it would seem worthwhile to think through what sharing might entail, so that we are in a better position to consider the origins and development of the ability to share in human ontogeny and phylogeny. In order to do so, it is necessary as a preliminary to distinguish between two kinds of sharing, what Trevarthen (1979; Trevarthen and Hubley 1978) called primary and secondary intersubjectivity. Primary intersubjectivity concerns the transactions that go on between two people, paradigmatically in face-to-face engagements, where the subjective states of each are closely co-ordinated one way or another, for example when they experience joy together, or one is angry and the other upset; secondary intersubjectivity concerns transactions that implicate shared experience of some real or imagined object or event external to the two or more people involved, for example when they share pleasure in watching cricket at Lord’s or argue about who pays for the tickets. Then of course one might wish to distinguish among many forms as well as degrees of intersubjectivity or (more specifically) sharing, so that sharing a joke is not the same as sharing a friendship. Our thesis is that human forms of sharing have qualities that are special to humankind, from very early in life. Let us focus on simple cases. Firstly, consider the following description of a typically developing two-month-old whom we videotaped during a still-face procedure as part of a study of mother-infant relations (previously described and discussed in Hobson 2005). When the mother assumed the still-face and unreactive posture as requested, the infant responded by becoming uneasy, restless, and jerky in her movements, and lost the infectious smiling and smooth tonguing movements that had been evident just moments before. Her bright, protracted gazes into her mother’s eyes were transformed into brief, checking glances. More important for the present purposes, after about 40 seconds her behaviour changed again, and she began to give longer looks to her mother accompanied by forced smiles. There was a strong impression that she was seeking to re-establish contact with her mother, trying to elicit a resumption of the joyful interpersonal exchange that was now missing. If all this is correct, then we see how the infant participates in experience with the other. Sharing experience with someone else is not merely like having one’s own experience of the world, and then adding something. It seems more like having one’s own subjective state and registering something of the other’s attitudes conjointly, in a qualitatively new form of experience (also Tronick et al. 1998).
77
78
Peter Hobson and Jessica A. Hobson
To explore this at a later point in development, we consider two further examples that come from videotaped interactions with typically developing children at the end of the first year of life, during semi-structured interactions with a tester. In the first example, a 13-month-old girl and a tester were seated across a table from one another. After playing together, the tester secured the child’s gaze to her face and then looked to her right while extending her right arm and finger into a point and exclaiming ‘look at that’. Initially, the child’s gaze lingered on the tester’s outstretched hand. Next she looked back to the tester’s face as if to ascertain what the tester might be trying to communicate. For a moment she seemed to dwell on her face and then suddenly she shifted her gaze to the target of the tester’s still outstretched finger. This case illustrates something about the means by which infants achieve shared reference by their psychological movement through the other. It is not simply that the other’s point or gaze or other gesture serves as a signpost to objects and events in the world. It is also that the infant is drawn into alignment with the other’s orientation toward a shared world. In a second videotaped interaction from a similar testing arrangement, a 12month-old girl swiftly followed the tester’s gaze and outstretched finger to locate a poster on the wall and then, giving herself barely enough time to take in the contents of the poster, quickly looked back to the tester’s face with an engaging smile. The tester (JH) spontaneously commented on this sharing, with a playful and affirming “Did you see Big Bird?” and the child immediately turned, still smiling, to look back at the poster again. Therefore sharing attention may involve more than being reorientated to see what another person sees, or even being moved to adopt a new attitude to the world; it can also mean that the people involved register and share that sharing is going on. Researchers have come a long way in measuring behavioural accompaniments to sharing experiences. Especially important and influential has been research into the forms of joint attention that occur at the end of the first year of life (Bruner 1983; Seibert, Hogan and Mundy 1982; Sugarman 1984; Walden and Ogan 1988). In joint attention, infants monitor and follow the gaze of others and point to, show, and/or alternate eye contact with reference to object and events in order to direct a person’s attention, share experiences, request things, or inform. In one form of joint attention (initiating joint attention, or IJA), infants use gaze and gesture to achieve sharing of experience with another (Kasari, Sigman, Mundy and Yirmiya 1990). Mundy, Kasari and Sigman (1992) provided evidence that directing the attention of others to objects, rather than to obtain objects, is especially associated with positive affect (also Bruner 1981; Hornik and Gunnar 1988; Rheingold, Hay and West 1976), and infants show more positive affect during joint attention involving both objects and caregivers than when playing with objects by themselves (Adamson
Engaging, sharing, knowing
and Bakeman 1985). Mundy et al. (1992) considered how such measures of positive affect in joint attention may contribute to operationalizing aspects of intersubjectivity, or the sharing of one’s inner subjective experiences with others. The important thing here is not to allow such concepts as joint attention, displays of positive affect, or even smiles to displace the meaning of intersubjective engagement, and in particular, to diminish the importance of joint attention as an expression of the human propensity to share the experience of sharing. In order to avoid this, it is essential to return to the meaning of what has been operationalized. By way of illustration, here are two examples of studies which seek behavioural evidence for what might be referred to as sharing looks (Venezia, Messinger, Thorp and Mundy 2004) and knowing looks (Rakoczy, Tomasello and Striano 2005). Both approaches involve the timing of smiles in relation to looks at a social partner during triadic experiences with objects, and both remain alert to the need to consider the intersubjective meaning toward which the evidence points. Venezia et al. (2004) provided evidence to suggest that although the frequency of smiling during joint attention remains stable between eight and 12 months, the timing changes so that the older infants begin to smile in anticipation of sharing the event with a social partner. These ‘anticipatory smiles’ involve smiling at an object and then gazing at the tester while smiling. The authors suggest that this developmental change may reveal increasing ability to communicate pre-existing positive affect about an object to another person, or be an index of social referencing in which the infants attempt to confirm their emotional response, but they also go further when they raise the possibility that anticipatory smiling “may index an intersubjective sense of the social partner as someone with whom experiences can be shared”(Venezia et al. 2004: 404). We would argue even further that such sharing looks might reflect the pleasure of psychological linkage with another person vis-à-vis a shared world – and after all, the infant is more likely to be smiling in the first place, if someone else is there sharing the experience. The smile toward the person with whom one is sharing might be expressive of sharing rather than communicating information. It might even be part of the communicative act: “We are enjoying this together.” Consider this vignette. A 15-month-old toddler seated across the table from a tester utters the word ‘Look’ as he extends his index finger into a point while gazing intently at the tester’s face. He ascertains that he has her attention, and then begins to point vaguely to one side while turning to a poster just behind him. He seems to pass this by, as he swings his body to look and point definitively to another poster on the wall to his right side. At the moment he extends his arm to a fully-fledged point, he looks back to the tester’s face with a broad smile. His gaze and smile seem to affirm that he has created an experience for them to share.
79
80 Peter Hobson and Jessica A. Hobson
He had decided to show something to his conversational partner, long before he knew what that something would be. Moments later, the tester activates a pop-up clown. When this emerges from hiding, the boy lets out a shriek of delight as he watches the clown dance, but his full exuberance (a way of characterizing the intense positive affect and social gregariousness that can be associated with initiating joint attention, Vaughan, van Hecke and Mundy personal communication September 2006) is only manifest when he lifts his gaze to connect with the tester with whom he is sharing the experience. At this moment, his face bursts into a smile: he is sharing the sharing. His heightened positive affect is an expression of the pleasure in sharing. Our second example comes from research conducted by Rakoczy et al. (2005) with 24-month-olds. These researchers gave an operationalized definition for smiles: “retracting both lip corners upwards and backwards”(Rakoczy et al. 2005: 61). Among the toddlers they studied, looks accompanied by smiles (let’s just call them smiles) were as likely to occur during the imitation of instrumental and pretence actions, but the smiles associated with pretence were more likely to coincide with looking at the social partner, while those associated with instrumental actions were more likely to have their onset while the child was still looking at the object. The investigators considered their data in terms of “some special interpersonal behavior by the child in pretending… and perhaps more ‘knowing smiles’ than in instrumental actions” (Rakoczy et al. 2005: 59). Here we find pointers to the interpersonal nature of symbolizing (also Hobson 1993), and again, reflections of intersubjective engagement.
4.
From engaging and sharing to knowing
The key idea here is that it takes emotional engagement to be moved by others, not least to be moved into sharing experiences. Movements in subjective orientation are essential for much of what is so special about human life – for transitions in thought as well as feeling – and their early beginnings are well illustrated by the phenomena of social referencing (e.g., Sorce et al. 1985). An especially important developmental implication of the ability to apprehend, respond to, and be moved by the subjective states of other people in relation to the world, is that such movements create mental space for negotiating attitudes and meanings. It is only once an infant has experience of shifting across person-anchored stances through assimilating another person’s attitudes – but at the same time, registering the source of those attitudes as ‘other’ – that it becomes possible for the infant to achieve understanding of what it means to have and to pass through alternative perspectives.
Engaging, sharing, knowing
Understanding what it means to have a perspective is critical if the infant is to develop the capacity knowingly to introduce new meanings/perspectives on to the materials of symbolic play. It is also part of achieving a conceptual grasp of the relation between ‘selves’ and the world, where the very idea of a self is that people are each a self with his or her own take on the world that may be shared or challenged, aligned-with or repudiated. So by the middle of the second year of life, pre-reflective and intuitive forms of role-taking are yielding both the means to conceptualise (symbols), and contents to conceptualize selves-with-attitudes in correspondence with a world that is the focus of attitudes and co-reference. It is here we find the beginnings of explicit forms of knowing, not least knowing about selves-with-minds. Humans think and know in virtue of having a form of life that is held in common with others, and in which agreements in judgement are possible (Wittgenstein 1958). In our view, all of this becomes possible through special forms of interpersonal engagement. Initially, there is dyadic engagement – as we described in the two-month-old infant above. Subsequently, there is engagement with others’ engagement with the world, through which one may be ‘moved’ in attitudes towards the world (as in social referencing). Out of this emerges the propensity to move within our own minds, from one psychological orientation to another as if from one person’s stance to another’s. It is in the propensity to dwell in the experience of the other, and to experience the world through the other – that is, to identify with the attitudes of another person from the stance of the other – that we believe the specialness of human forms of sharing are grounded.
5.
Four questions with brief replies
In the opening pages of this contribution, we summarized three studies in autism. Our aim was not only to illustrate the possible yield from such research for our understanding of typical as well as atypical development, nor merely to share our struggles to publish measures of intersubjectivity. We also wanted to introduce the notion that intersubjective engagement might be structured by the process of identifying with the bodily expressed attitudes and actions of other persons. We proceeded to develop the idea that intersubjective sharing is pivotal for development, and highlighted some of the motivational as well as cognitive implications of this fact. At this point, we shall not attempt to explicate the notion of identification in detail, nor try to unravel the details of development from interpersonal engagement to interpersonal understanding (and understanding of everything else, come to that) over the first two years of life. Instead, we offer some brief responses
81
82
Peter Hobson and Jessica A. Hobson
to four rhetorical questions that we consider important in thinking about intersubjectivity. We hope that the studies in autism cited earlier, and the vignettes of typically developing young children offered subsequently, will give personal colouring to this schematic account. Our four questions are as follows: Firstly, what is the structure to human interpersonal engagement that is so special, and that is unfolded (as it were) in the forms of joint attention and social referencing towards the end of the first year of life? Secondly, is this same mode of relatedness operative when a toddler achieves not simply interpersonal co-ordination, but also understanding of what it means to be a person or ‘self ’ who is able to respond to other persons and to move through a variety of person-anchored perspectives in his or her own mind? Thirdly, what has this to do with the origins of ‘theory of mind’? Finally, what does it mean that intersubjectivity works as a ‘system’ of self-in-relation-to-other, and what does autism tell us about this? In brief, our answers to these questions are as follows. We believe that in order for sharing to be possible by, say, the second month of life, there needs to be a capacity to register the attitudes expressed through another person’s body in such a way that they can be experienced in relation to one’s own state. This is what sharing entails, after all. We suppose this to be an early form of identifying with the attitudes of others. This natural propensity not only to respond to, but also to assume (in part) another persons’ feelings, gives motivational impetus to what we have referred to as being moved by others. Importantly, the boundary within the self-other mode of experience – and of course, this is prior to any concept of self and other – is also operative in settings of joint attention and social referencing. When an infant is moved to adopt the orientation of other people, it is vital that he or she has the pre-reflective capacity to register the source of the new orientation as ‘other’. This is important for the reason that the child has to reach the point of adopting such perspectives knowingly – and in part, by adopting an otherperson-anchored perspective on him or herself. This would not be possible without the differentiation of person-anchored stances within the child’s own mind. Now to ‘theory of mind’. This is a theoretically loaded expression that is – or should be – concerned with concepts of the mind such as those of thinking, believing, feeling, intending, and so on. In some quarters, the theory is framed in terms of a computational metaphor, with talk of representations, metarepresentations, and computations. Although there are strengths to this approach, among many limitations is the lack of a developmental theory (beyond innatism) to explain the acquisition of concepts of persons-with-minds. The present approach begins with interpersonal co-ordination of bodily-anchored expressive acts, and through a developmental pathway that opens such co-ordination to encompass
Engaging, sharing, knowing
co-orientation with others towards a shared world, leads to the acquisition of symbolic representations along with newfound abilities to relate to one’s own relations with that world. Of course there are developments between the second year of life, and again around the end of the third year when children seem to acquire a new grasp of the concept of belief. In our view, the critical acquisition here is the concept of ‘reality’, so that children can now grasp how people may ‘falsely hold as true of reality’ – and again, we take it that the notion of reality is that which transcends individual human viewpoints, insofar as it is ‘the way things are’. In other words, children who understand what it is to hold a belief grasp what Perner (1990) calls the representational relation. We think that it is through intersubjective engagement and communication with others that children come to see the force of this supra-personal characterization, this arbiter between my view and yours – reality as a given in relation to which humans should merely assent. Finally, the system of self-in-relation-to-other, although remarkably robust, is not immune to disruption from various sources. Autism is instructive for helping us see not only the implications of such disruption – impaired creative symbolic play, limited interpersonal (and ‘theory of mind’) understanding, and impaired self-monitoring probably among these (see also Barresi and Moore this volume) – but also how different forms of dysfunction may underlie the serious impairments in interpersonal engagement that characterize the syndrome. One of our special interests has been in congenital blindness (e.g. Hobson 2005), where the lack of visual (and probably to a lesser extent, affective) co-orientation with others towards a shared world is a serious risk factor for developing the syndrome of autism. Through the association between congenital blindness and autism, we may discern the importance of interpersonal engagement and sharing attitudes towards particular, visually perceived objects and events in the environment for coming to know about symbols, about selves, and more specifically about people’s minds.
6.
Methodology revisited
In this final section of the chapter, we return to issues of methodology. We have already indicated how one way to investigate intersubjectivity is to study human beings in whom the ability to engage with others is compromised, specifically those with childhood autism. Here we turn the argument around, and consider how in order to understand the nature of autism, and even to diagnose the condition, we need to be attuned to the significance of intersubjective engagement. Whatever the resistances to according intersubjective phenomena a central place in our
83
84
Peter Hobson and Jessica A. Hobson
a ccounts of autism and in our methodological approaches to its investigation, it is already the case that judgements about intersubjectivity pervade authoritative diagnostic approaches. Among individuals with autism, something is unusual or limited in the kind of interpersonal engagement we have been describing. In early clinical descriptions of the condition, Kanner (1943) highlighted the children’s “inborn autistic disturbances of affective contact” (Kanner 1943: 250). The phrase “affective contact” is notable for attempting to capture what it is like for a person to be in relation to another, an issue that is still marginalized in much theorizing about autism. Kanner captured something of this when he wrote that “people, so long as they left the child alone, figured in about the same manner as did the desk, the bookshelf, or the filing cabinet” (Kanner 1943: 246). It is instructive to consider how we measure whether or not a child has autism. One approach is to focus upon a relative lack of the presence of particular forms of social-communicative exchange described earlier. For example, Mundy (2003) describes the children’s fundamental social disturbance in terms of early and robust impairments in joint attention development. These are those specific forms of eye contact, affect and gestures used ‘for the singularly social purpose of sharing experiences with others’ (Mundy and Neal 2001). When one administers the Early Social Communication Scales (ESCS: Seibert et al. 1982), for example, young children with autism between the ages of three and seven years of age are likely to initiate one or two joint attention bids (Mundy et al. 1986), far below what one would expect in children without autism. This approach succeeds in characterizing a pivotal aspect of the children’s social-communication impairments, and such impairments have serious implications for the children’s language and cognitive development. Still one needs to ask in what sense such impairments in joint attention are the crux of the developmental psychopathology of autism and to what extent they represent prominent early manifestations of something deeper and yet earlier in the children’s atypical interpersonal engagement. But to return to the assessment of autism: one of the most widely-respected approaches to formal research diagnosis is the Autism Diagnostic Observation Schedule (ADOS-G: Lord et al. 2000), a semi-structured series of planned social presses to prompt requests, engagement in joint attention, and communication with the tester. It is very effective in eliciting certain kinds of interpersonal exchange that are often limited in autism. Not only this, but also the standardized way of administering and scoring the measure means one can interpret scores across research laboratories for diverse populations around the world. Interrater agreement standards are high. Items are typically scored on a three-point scale from 0 (no evidence of abnormality related to autism) ranging to 2 (definite evidence of
Engaging, sharing, knowing
abnormalities related to autism). Some of the items are as follows (and these are just a few examples): unusual eye contact, quality of social overtures, quality of social response, and overall quality of rapport. Note the degrees of appropriately subjective judgement in the ratings. Or again, here are a few examples of items from the Parent Interview for Autism (PIA: Stone and Hogan 1993). Parents are asked whether the child “enjoys interacting with familiar adults,” “enjoys playing with other children,” “looks through people as if they weren’t there,” “seems to be hard to reach or in his or her own world.” Our point here is that intersubjective judgements are a necessary and important part of conventional diagnostic procedures. So why is there such resistance to incorporating them into the broader domain of scientific investigation? Even though informally, eminent researchers will refer to the lack of “emotional sparkle” in imitative events (Rogers 2006) or the absence of “warm, joyful expressions” (Wetherby 2006) among children with autism – both examples from presentations at the recent International Meeting for Autism Research (IMFAR) in Montreal – we have a long way to go before the developmental significance of the phenomena that such expressions capture is fully appreciated. As a final flourish, here is one last study of our own (Hobson et al. 2006). In this study, we asked children and adolescents to pose for a picture. While one tester took the participant’s photograph with a Polaroid camera, a second tester filmed the encounter. All of the participants looked at the camera when their photograph was taken, and there were no group differences in the tendency to sustain or avert gaze during the episode. But when it came to the quality of these looks, there were marked and highly significant differences between the participants with and without autism. The participants without autism were judged to show self-consciousness in their looks away and their looks to the tester. Those with autism were judged to give blank looks away and to the tester. So, even when the numbers of looks were similar, the quality of the encounter – the feel of the exchange – contrasted sharply between the groups. Once again, here were looks that raters could judge reliably. Although one might now try to track down which behavioral/expressive characteristics corresponded with the ‘self-consciousness’ of participants’ looks, it was only through the ratings of intersubjectively attuned human beings that such looks could be identified as self-conscious in the first place. These issues matter when we think about treatment for children with autism. It is only when those involved in treatment cease to consider “emotion recognition” or “interpersonal skills” as abilities that can be taught or trained (even by computer), and instead seek ways to foster development in intersubjective engagement through efforts to draw individuals into appropriate kinds of socially co-ordinated experience by providing manageable and positively engaging exchanges
85
86 Peter Hobson and Jessica A. Hobson
(e.g., Gutstein, Burgess and Montfort 2007) do we come closer to finding ways to help children with autism shift to a more fruitful developmental trajectory. To conclude: there are indeed intersubjective foundations for engaging, sharing, and knowing. Intersubjectivity has self-other structure, in our view a structure inherent in the process of identifying with others. One way to explore this claim is to study children with autism. And one of the most promising startingpoints for understanding, diagnosing, and treating autism is to consider how intersubjective engagement shapes the development of social relations and creative symbolic thinking.
Acknowledgements This chapter was written while we were at the Center for Advanced Study in the Behavioural Sciences, Stanford, California. We are hugely grateful to the Center, and also to the Tavistock and Portman NHS Trust, for making our stay possible and so rewarding.
References Adamson, L. and Bakeman, R. 1985. “Affect and attention: Infants observed with mothers and peers.” Child Development 56: 582–593. Bosch, G. 1970. Infantile Autism. New York: Springer-Verlag. Bruner, J. 1981. “Learning how to do things with words.” In Human Growth and Development, J. Bruner and A. Garton (eds.), 62–83. London: Oxford University Press. Bruner, J. 1983. Child’s Talk: Learning to Use Language. New York: Norton. Capps, L., Kehres, J. and Sigman, M. 1998. “Conversational abilities among children with autism and children with developmental delays.” Autism 2: 325–344. García-Pérez, R.M., Lee, A. and Hobson, R.P. 2007. “On intersubjective engagement: A controlled study of nonverbal communication in autism.” Journal of Autism and Developmental Disorders 37: 1310–1322. Gutstein, S.E., Burgess, A.F. and Montfort, K. 2007. “Evaluation of the Relationship Development Intervention Program.” Autism 11: 397–411. Hobson, J.A. and Hobson, R.P. 2007. “Identification: The missing link between joint attention and imitation?” Development and Psychopathology, 19: 411–431. Hobson, R.P. 1993. Autism and the Development of Mind. Hove, Sussex: Erlbaum. Hobson, R.P. 2002/2004. The Cradle of Thought. London: Pan Macmillan & New York: Oxford University Press. Hobson, R.P. 2005. “Autism and emotion.” In Handbook of Autism and Pervasive Developmental Disorders (3rd ed), F.R. Volkmar, R. Paul, A. Klin and D. Cohen (eds.), 406–424. New Jersey: John Wiley & Sons, Inc.
Engaging, sharing, knowing
Hobson, R.P., Chidambi, G., Lee, A. and Meyer, J.A. 2006. “Foundations for self-awareness: An exploration through autism.” Monographs of the Society for Research in Child Development, Serial No. 284, 71. Hobson, R.P. and Lee, A. 1998. “Hello and goodbye: A study of social engagement in autism.” Journal of Autism and Developmental Disorders 28: 117–126. Hobson, R.P. and Lee, A. 1999. “Imitation and identification in autism.” Journal of Child Psychology and Psychiatry 40: 649–659. Hobson, R.P. and Meyer, J.A. 2005. “Interpersonal foundations for the self: The case of autism.” Developmental Science 8: 481–491. Hornick, R. and Gunnar, M. 1988. “A descriptive analysis of infant social referencing.” Child Development 59: 626–634. Kanner, L. 1943. “Autistic disturbances of affective contact.” Nervous Child 2: 217–250. Kasari, C., Sigman, M., Mundy, P., and Yirmiya, N. 1990. “Affect sharing in the context of joint attention interactions of normal, autistic, and mentally retarded children.” Journal of Autism and Developmental Disorders 20: 87–100. Landis J.R. and Koch G.G. 1977. “The measurement of observer agreement for categorical data.” Biometrics 33: 159–174. Lee, A. and Hobson, R.P. 1998. “On developing self-concepts: A controlled study of children and adolescents with autism.” Journal of Child Psychology and Psychiatry 39: 1131–1141. Lord, C., Risi, S., Lambrecht, L., Cook, E.H., Leventhal, B., DiLavore, P.C. and Rutter, M. 2000. “The Autism Diagnostic Observation Schedule – Generic: A standard measure of social and communication deficits associated with the spectrum of autism.” Journal of Autism and Developmental Disorders 30: 205–223. Meyer, J. A. and Hobson, R.P. 2004. “Orientation in relation to self and other: The case of autism.” Interaction Studies 5: 221–244. Mundy, P. 2003. “The neural basis of social impairments in autism: The role of the dorsal medial-frontal cortex and anterior cingulate system.” Journal of Child Psychology and Psychiatry 44: 793–809. Mundy, P., Kasari, C. and Sigman, M. 1992. “Nonverbal communication, affective sharing, and intersubjectivity.” Infant Behavior and Development 15: 377–381. Mundy, P. and Neal, R. 2001. “Neural plasticity, joint attention, and a transactional social-orienting model of autism.” International Review of Mental Retardation 23: 139–168. Mundy, P., Sigman, M.D., Ungerer, J. and Sherman, T. 1986. “Defining the social deficits of autism: The contribution of non- verbal communication measures.” Journal of Child Psychology and Psychiatry and Allied Disciplines 27: 657–669. Perner, J. 1990. Understanding the representational mind. Cambridge, MA: Cambridge University Press. Rakoczy, H., Tomasello, M., and Striano, T. 2005. “On tools and toys: How children learn to act on and pretend with ‘virgin objects’.” Developmental Science 8: 57–73. Rheingold, H., Hay, D. and West, M. 1976. “Sharing in the second year of life.” Child Development 83: 898–913. Rogers, S.J. 2006. “Imitation difficulties in autism.” Paper presented at the International Meeting for Autism Research, June 1 – 3, Montreal, Canada. Seibert, J.M., Hogan, A.E., and Mundy, P.C. 1982. “Assessing interactional competencies: The Early Social Communication Scales.” Infant Mental Health Journal 3: 244–245. Sorce, J.F., Emde, R.N., Campos, J. & Klinnert, M.D. (1985). “Maternal emotional signaling: Its effect on the visual cliff behavior of 1-year-olds.” Developmental Psychology 21: 195–200.
87
88
Peter Hobson and Jessica A. Hobson
Stone, W.L. and Hogan, K.L. 1993. “A structured parent interview for identifying young children with autism.” Journal of Autism and Developmental Disorders 23: 639–652. Sugarman, S. 1984. “The development of preverbal communication.” In The Acquisition of Communicative Competence, R.F. Schiefelbusch and J. Pickar (eds), 23–67. Baltimore: University Park Press. Tantam, D., Holmes, D. and Cordess, C. 1993. “Nonverbal expression in autism of Asperger type.” Journal of Autism and Developmental Disorders 23: 111–133. Trevarthen, C. 1979. “Communication and cooperation in early infancy. A description of primary intersubjectivity.” In Before Speech: The Beginning of Human Communication, M. Bullowa (ed.), 99–136. London: Cambridge University Press. Trevarthen, C. and Hubley, P. 1978. “Secondary inter subjectivity: Confidence, confiding, and acts of meaning in the first year.” In Action, Gesture and Symbol: The Emergence of Language, A. Lock (ed.), 183–229. London: Academic Press. Tronick E.Z., Bruschweiler-Stern N., Harrison A.M., Lyons-Ruth K., Morgan A.C., Nahum J.P., Sander L. and Stern D.N. 1998. “Dyadically expanded states of consciousness and the process of therapeutic change.” Infant Mental Health Journal 19: 290–299. Venezia, M., Messinger, D.S., Thorp, D. and Mundy, P. 2004. “The development of anticipatory smiling.” Infancy 6: 397–406. Walden, T.A. and Ogan, T.A. 1988. “The development of social referencing.” Child Development 59: 1230–1240. Wetherby, A.M. 2006. “Social communication profiles of children with autism spectrum disorders in the 2nd and 3rd years of life.” Paper presented at the International Meeting for Autism Research, June 1–3, Montreal, Canada. Wittgenstein, L. 1958. Philosophical Investigations. Oxford: Blackwell.
chapter 5
Coming to agreement Object use by infants and adults Cintia Rodríguez and Christiane Moro According to the “naturalistic view of the object” children give meaning to objects in a natural, direct, and spontaneous manner, without the need of others. The myth underlying the spontaneity of subject-object encounter is that, in contrast to the widely assumed opacity of “social” reality within modern psychological theory, there exists an alternative reality of “non-social physical” that is literal and transparent. We challenge this by adopting a pragmatic approach to objects. In everyday life, objects are situated in communicative contexts and used for doing things. During their first year of life, children achieve triadic interactions (baby-object-adult) involving very different degrees of agreement with adults concerning an object’s use and meaning by means of diverse semiotic systems in contexts of joint communicative action.
1.
Introduction
The ways in which the subject approaches the object constitutes one of the major themes in early cognitive development – in one way or in another the whole Geneva School was devoted to this issue. Despite this and the growing recent influence of Vygotsky, the “naturalistic view of the object” is predominant in studies of infant cognitive development. This naturalistic view implies that the child relates to the object in a natural, direct, and spontaneous way, often eliminating not only any action and use by the child upon the object, but also any adult-baby joint communicative action on the world (Rodríguez 2007). According to this viewpoint, the child is supposed to promote his early cognitive development through his encounters with an obvious and transparent reality which can be accessed without the need for any educational guidance through semiotic mediation, in order for him to share different degrees of meaning with the people surrounding him. The myth underlying this spontaneous encounter between the subject and the world is the existence of a literal, obvious, transparent, “non-social physical” reality, which declares itself nakedly, as opposed to the opacity of the “social” reality that
90 Cintia Rodríguez and Christiane Moro
exists only through communicative and conventional channels. This dichotomy between the opacity of the social and the immediacy of the meaning of objects has had a profound impact on research into early typically and atypically developing children. In this chapter we challenge this “naturalistic” view, and focus instead on a pragmatic and semiotic approach to objects. Objects are used for doing things in everyday life; these meanings and functions are socially established and are linked to the use we make of them. From the beginning of their lives, children are coopted by the people around them as co-protagonists of their activities when doing things with objects. Adults communicate with children through and about objects used in everyday contexts. They achieve different levels of agreement about the meaning of things being used in such contexts. As a consequence, children become involved. They are not in contact with any “syntactic” formal object, however, but with a pragmatic one, involving shared uses in communicative contexts where people do things with them as part of the social world. In the first section of this chapter we will challenge the widely established opposition in the theory of mind literature between the referential opacity of the mental world versus the self-evidence of so-called “physical reality”. Then we will question the solitary nature of the encounter between the child and the world and the “spontaneous” categorizations resulting therein. We will refer to the permanence of objects determined by their social function (before they have names). This permanence through use is already in place by the end of the first year of life and is presented by adults to children from the outset. Before objects are permanent to children they are regarded and used as permanent by adults. Then we will consider some voices arguing against the passivity in which babies are placed in laboratories and we will confront this with how the classical theories (Piaget, Wallon, Vygotsky) conceived subjects always as active, as always transforming their surroundings and hence as a key factor in development. In the next section we will consider the difference between what is meant in the literature by joint attention and our view of joint action, which implies a pragmatic and semiotic perspective where objects are an integral part of baby-adult communication. In the last section we present five observations of children (from 2- to 12-months-old) in contexts of triadic interaction with adults around the same object, and consider the different levels of agreement they reach within this 10 month interval between the first and last set of observations. As an essential tool of analysis, we have identified the different semiotic systems that are involved. Thanks to these, different levels of adult-child agreement about objects, events and situations in the world can be established about the meanings and the uses that can arise.
Coming to agreement
2.
The social nature of objects
2.1
Referential opacity of the social versus the evident physical reality
This dichotomy can be illustrated by an (extreme) excellent and well-known example that has contributed much to the persistence of this naturalistic and oversimplified view of the object. This example is provided during the 1980s and 1990s by traditional theories of mind that characterize mental states by their referential opacity as opposed to the evidence, transparency and literality of physical reality. The so-called “physical reality” is evident, transparent and literal (Leslie 1987). For instance, according to Baron-Cohen, “Belief, knowledge, desire, and pretence are all opaque mental states. That is, they suspend normal truth-conditions governing the propositions they prefix […] In contrast, ‘I saw a mouse’ is true only if I did indeed see one. Thus, perception is transparent, not opaque” (Baron-Cohen 1993: 65, emphasis added). As Costall and Dreier (2006: 4) point out: …the Theory of Mind approach, which has been dominant within developmental psychology for the last decade, has largely removed the issue of children’s use of objects from the research agenda, not only because of its emphasis upon ‘mind’, but also because of its explicit separation of children’s understanding of other people from their understanding of things.
The vast influence of the theory of mind perspective when trying to explain what is missing in children with autism is well known. Autistic children have difficulties in pretend play, in true and false belief and in communication, whereas such impairment “will not fundamentally affect [their] apprehension of physical artifacts, including representational artifacts such as photographs or maps” (Leslie and Roth 1993: 92, emphasis in original). According to this view, autistic children should have no difficulties in their dealings with objects. In other words, the “literal” reality of objects that is assumed to be directly evident is unproblematic. Since the 1980s, this position has been extremely influential regarding views about what should be looked at in early stages of development. Although we cannot develop this here, we would like to stress that this view of the “good relation” . We would like to refer here to the claim made by Jeremy Carpendale and Charlie Lewis (2004) in relation to children’s developing understanding of mind. They consider that “the development of children’s social understanding occurs within triadic interaction involving the child’s experience of the world as well as communicative interaction with others about their experience and beliefs” (2004: 79). There is maybe only a small, subtle difference in what we say: We consider that not only “social understanding”, but development (included what is called cognitive development) occurs within triadic interaction, following Vygotsky’s well known maxim: The path going from the child to the world, goes through another person.
91
92
Cintia Rodríguez and Christiane Moro
of autistic children with objects is contradicted by Williams, Kendell-Scott and Costall’s (2005) findings concerning the difficulty autistic children show with regard to using everyday objects in a conventional way (see also, Arango, Chávez, and Lasprilla 2003; Pardos and Rodríguez 2005; Moore 2004).
2.2 Early cognitive development: The kingdom of spontaneity The naturalistic view of the object in early cognitive development is not only the result of the impact of the classical theory of mind perspective, but is also characteristic of the enormous influence of the mainstream cognitive approaches that have typically ignored the world when trying to explain the mind. We encourage readers to have a careful look at the different sections included in recent infant development textbooks. Almost invariably one will find a section devoted to “social development”, clearly differentiated from another section dedicated to “cognitive development”. Now we invite the reader to concentrate on the “cognitive development” and see what it says about the relation between the child and the objects presented (or represented). In research, babies are frequently situated in a context (usually the laboratory) on their own, with no other people to interact with, thus implicitly eliminating social interaction as a cause of any cognitive development. In this solitary context their (active) actions on the world are neglected in favour of their (passive) reactions to the stimuli presented to them. When any characteristic of the object is included, most of the time it is merely shape or color (e.g., Bremner et al. 2005). In such contexts, the child becomes “a big solitary looking eye” who does not transform anything in the world but only reacts on his own towards what is being presented. A good example of this is Spelke’s nativist approach, according to which infants, from a very early age, have object permanence and representations of objects as coherent wholes without any social intervention, and without any action by the child. Her comment that “newborn infants also may have a functional system of object representation” (Spelke 1998: 185) seems an extremely vague and general statement. Mandler (2000, 2004b), for instance, adds something to this early representation of objects when she says that there is more than one kind of object categorization. She distinguishes two types: perceptual categorization and conceptual categorization. The latter is based not on what objects look like, but on what “objects do” (2000: 3), and “meaning accrues from what things do, not what they look like” (2004a: 168). Both perceptual and conceptual categories serve different functions: “Infants, just like adults, make their inductive generalizations on the basis of kind, not on the basis of perceptual similarity” (2004a: 199), and
Coming to agreement
when discussing the classification of animals by infants, she notes that “these data are at least partly culturally determined” (2004a: 204). Adding this pragmatic level seems to us essential. Nevertheless, it is extremely surprising to find Mandler go on to conclude with the further claim, especially in a chapter in a book entitled The Development of the Mediated Mind, dedicated to Katherine Nelson. In contrast to other chapters in this book, there was no mention here of the influence of sociocultural context. That is because infants are to some extent shielded from such influence by their lack of language…by and large, the early development of the ability to…categorize objects and to learn the important basics of language such as that it is used for communication, are all governed by universal factors common to infants in all cultures. It is when the foundations have been laid down and the naming practices of the culture begin to teach the infant which details are important that more cultural influence can be seen. (Mandler 2004b: 27–28, emphases added)
But how is it that the baby is “shielded” from the influence of other people precisely at a time when they are most vulnerable, and completely dependent on them for every aspect of their existence? Our own findings about how children start using everyday objects in a conventional way by the end of the first year stress the cultural and social aspects. If children start to understand objects as “sign of their [public] use”, it is because adults deploy an intense semiotic activity when using the objects with, or in the presence of, children. Thanks to this, children come to “categorize” them according to their social and public use (as adults do). The mediation through signs (not all signs are linguistic) with the object itself is the central tool of adult-child communication allowing the child to appropriate, over an extended process, these public meanings of use (Rodríguez and Moro 1998, 1999; Moro and Rodríguez 2005). It is possible that later concepts have their roots in this kind of canonical uses. The first conventional uses of objects could be regarded as “concepts in action” (Rodríguez 2006).
2.3 The permanence of the object: Only one? The permanence of the object has been investigated over many years. According to Piaget, the child is able during the second half of the first year to remove an obstacle in order to grasp the hidden object. In more recent research, in orders to test whether the child does or does not have object permanence earlier than Piaget suggested, images of autonomous moving objects on their own are projected into screens (Moore and Meltzoff 1999; Bremner et al. 2005). The visual reaction of the child is then recorded. There is no doubt that important findings are obtained with this method. Nevertheless, two things should be said. First, the
93
94 Cintia Rodríguez and Christiane Moro
infant’s knowledge of the object is an extremely complex thing that involves a long process of shared meanings through ontogenesis, at many levels, which cannot be simply reduced to its permanence as assessed with this paradigm. Given these preconceptions, it is hardly surprising to read that: “infant object permanence is still an enigma after four decades of research” (Moore and Meltzoff 1999: 623). According to Piaget (1937), there are degrees of permanence. In fact, he considers the object as a limit, in the mathematical sense: one is continually approaching objectivity but the object itself is never reached. However, there is another kind of permanence of objects – functional permanence – which is related to their everyday uses, social meanings and cultural functions whose acquisition by children takes place in educational contexts (Rodríguez and Moro 1998). This kind of permanence was largely ignored by Piaget and by mainstream cognitive psychology (Rodríguez 2006). Such functional permanence is different from the abstract conceptual categorization that is assumed in modern cognitive theory to be independent of education and culture, and it cannot be captured in the appearance/disappearance paradigm which simply ignores the function and cultural use of objects. (We will come later to this problem.) Third, in real life, as we previously pointed out, objects do not engage in their own spontaneous motion back and forth without the intervention of intentional agents. Finally, in real life, children are active subjects, not mere “re-acting” entities.
3.
Some voices against the passivity of the subject
Many voices from the ecological tradition are very active nowadays against the passivity imposed on subjects in experimental situations. As Alan Costall (2004: 76) has put it, “if we really did spend all our lives just waiting for things to happen to us (as the participants in psychology experiments are typically required to do), then whatever “activity” is involved in perceiving would necessarily be confined to internal processing.” Eleanor Gibson and Ann Pick (2000: 14), in opposition to common accounts of perception, refer with a great sense of humor to theories “that begin with a motionless creature haplessly bombarded by stimuli.” The stimuli presented belong to the evident, transparent and obvious reality to which the baby is supposed to be directly confronted. A similar objection to the absence of action is identified by some researchers in relation to the paradigm used, which tells only “part of the story” about the cognitive abilities of very young infants, as “the visual observation paradigm treats the infant as a couch potato who merely sits, watches, and offers an opinion (by showing varying degrees of interest)” (Willatts 1997: 132). If we check our not-so-remote past, this complaint about the neglect of actions in the world when
Coming to agreement
dealing with cognitive development is a recurrent theme. In their well known paper “If you want to get ahead, get a theory”, Karmiloff-Smith and Inhelder emphasized that “unlike previous Genevan research articles in which extensive quotations were given from what children said, this study’s protocols consist mainly of detailed descriptions of children’s actions” (1974: 200, emphasis added). To be fair to Piaget, we must say that their explicit disapproval is part of the Genevan movement on Microgenesis and problem solving, developed by Le groupe de stratégies (Inhelder et al. 1992) during the 70s and 80s while trying to restore the central place of actions upon the objects by subjects as the way of changing their representations, and thus provoking development. This criticism about Piaget’s neglect (“omission”) of action, which is very paradoxical given his epistemological emphasis upon the active subject, does not apply to his studies of sensorimotor development but rather to his work with children after 4 or 5 years old. In addition to this neglect of the active nature of the child, there has also been a disregard of communication between the baby and adults about objects in the world. In other words, we need to take seriously into account joint action in the world as a central scenario from where children build their meanings by acting in collaboration with others. Let us now take a brief look at the classic developmentalists and see which status was given by them to subjects’ actions upon objects and to others’ interventions when dealing with cognitive development.
4.
Action, communication and objects
4.1
Classical developmentalists on action, communication and objects
The situation in early developmental psychology has not always been the same. If we look back through the 20th century to the classics, the picture we find is very different – the American biologist Stephen Jay Gould, when complaining against the anti-intellectualist varnish of American culture, used to say that “the nuggets of the authentic discoveries” abound in the primary literature (1998: 14). Over the last 50 or 60 years when researchers are involved with cognitive development they usually look only at babies and objects. This was the classical Piagetian position.
. Even when researchers refer to development as “social-cognitive”, as soon as they consider objects or tasks involving objects, they treat them as the “non-social” part of the situation (e.g. Striano and Bertin 2005: 563)
95
96 Cintia Rodríguez and Christiane Moro
For him, the “social world” of conventions and communication had nothing to do with the origin of the intelligence, as there is no causal link between social facts and psychological development (Bronckart 1997). Nevertheless, Piaget was very much concerned with babies’ actions in the world – not just their responses to stimuli. The method employed by Piaget was observation of common situations from everyday life, which allowed him to look at microgenetic processes from a very qualitative standpoint; the complexity of the object was beyond doubt. According to Costall, both Piaget and Gibson insisted upon the primacy of “being in the world” (2004: 85) since “Contrary to the dominant approaches within cognitive psychology and artificial intelligence, they did not take our capacities for representation and symbolism for granted, but saw clearly that representational activities need to grounded in our interactions with our surroundings”. A classical objection had been addressed to Piaget in relation to the primacy of action: How active must a subject be to be considered active (see Moore 2004)? The recent discovery of mirror neurons in humans (see Barresi and Moore this volume; Zlatev this volume) – a number of areas of the brain are activated when people make a movement that involves reaching out and grasping an object, as well as when they watch grasping movements made by others (Corballis 2002: 47) – will provide probably answers to this important question. It is not by accident that Piaget was very much influenced by Köhler’s work on chimpanzees. If we look at the kind of situations Köhler considered as indicating that there was intelligence without language, we have to bear in mind that chimpanzees always had some practical problem to solve. They were active subjects inventing new solutions (see Leavens, Hopkins and Bard this volume, Pika this volume). Two other important figures in European Developmental Psychology are Vygotsky and the French psychologist Henri Wallon. Vygotsky is ambiguous about how many protagonists should be considered in early cognitive development. Sometimes it seems that there is room only for the child and the object. Whereas in his manuscript about the first year of life, Vygotsky says explicitly that the main way babies engage in activities at this early age is through other persons. He says something very important: “objects appear and disappear from the child’s visual field thanks to the other’s will” (1984/1996: 285). The subject considered by Vygotsky and by Wallon (1942/1970) was active, always involved in the transformation of the world thanks to the use of tools and signs. Wallon, for instance, used to say that biology is socially oriented. As stressed earlier, if we compare this situation with the subject being studied during the last 50 years by mainstream cognitive psychology, the picture is quite different:
Coming to agreement
1. No room is made for education and communication with other people (similar to Piaget). 2. The object under consideration is assumed to be “self-evident”, constitutes the literal reality, and is neither defined by its social use (absence of pragmatics of the object), nor by its roles in the communicative activities of the people about it, around it, and with it. 3. The subject no longer acts anymore, but only reacts following the stimuli presented by the experimenter (who is playing the active role). 4. There is an almost total lack of interest in the processes of construction, in microgenesis. There is also a disregard of qualitative analysis and longitudinal studies.
4.2 Acknowledging subjects as active means considering not only joint attention but also joint action Some researchers refer to the need to consider triadic interactions (adult-objectbaby), which focus on how the child shifts his interest from the adult towards the world and how he includes others and the world at the same time. This adultworld inclusion is known as joint attention. For instance, according to Legerstee et al. (1987) when children around 17 weeks of age start to be able to grasp a doll, they manifest less interest towards adults. This does not mean a halt in their communicative development, but rather this new phase of orientation towards an object prepares the way for a new kind of communication involving objects (1987: 228). By the end of the first year, communication changes dramatically and typically developing children start to refer to events taking place beyond the limits of interpersonal exchanges (Bakeman and Adamson 1986: 228). Other studies of joint attention have involved children with Down syndrome (Legerstee and Weintraub 1997). Others show how children as young as 6 months old use adults instrumentally; from 6 to 9 there was a rapid increase in their initiation of eye contact and stylized communication, they began to employ conventional symbolic means after 10 months, such as pointing, object offers, and nods (Mossier and Rogoff 1994: 71). Other studies suggest that a link exists between dyadic and triadic interactions in children aged 7 and 10 months, and the lack of age effects suggests a somewhat more gradual process of social cognitive development than that implied by a suddenly emerging ‘9-month-revolution’ as suggested by Tomasello (1999, see Striano and Rochat 1999; Striano and Bertin 2005). What distinguishes these views about joint attention from the main argument we develop here is that we need to know how the mutual understanding between adults and babies about objects in the world takes place. This is why we need to
97
98 Cintia Rodríguez and Christiane Moro
know how the long process of joint action on the world involving babies in interaction with adults works. In other words, we need to understand how the place of the adult as a mediator between the child and the world changes.
4.3 Triadic interactions through the first year of life: A pragmatic perspective If we agree that objects have functions (Nelson 1974), shared meanings, that are part of the social history, and which belong to systems of uses shared by the community (Sinha 2005), an important question then to ask as developmental psychologists is how, when, and through which processes, babies come to appropriate the pragmatics of objects situated in communicative contexts and used for doing things. How do children achieve, through diverse semiotic systems in contexts of triadic interactions very different degrees of agreement with adults about the meaning and uses of objects? These different agreements between adults and children are possible thanks to a long process of construction in which babies are actively involved with other people. This means putting back on the agenda of early development the study of processes with the help of observational and microgenetic methods. Our focus on objects according to their uses is far removed from the “culture-free” scenario of mainstream cognitive psychology. There are two major differences. Firstly, early cognitive development is not such a solitary business. We need to understand how the educative influences of other people on the baby operate in communicative contexts, which semiotic systems are at work in ontogenesis, and how they develop. This means substantially extending the Vygotskian hypothesis according to which communication through signs is taken to be a central cause of cognitive development. Secondly, there is an urgent need to restore to children their status as active agents, thereby rescuing them from the characteristic passive position where they can only re-act to stimuli. As a consequence, we reach the triadic interaction between baby-object-adult, in which both adults and children act together and communicate by different degrees of shared meanings about the uses they make of objects in the world. We come to an exceptional online scenario in which development takes place and new meanings grow from shared uses, and different levels of adult-child agreement are possible. Understanding these means seriously taking into account materiality as a central focus around which two different subjects communicate and establish a process of new, more powerful, shared meanings and conventions. The reality of objects, with their everyday social meanings of use, is a very complex issue. Our studies concerning the uses of objects made by children,
Coming to agreement
uring the second half of their first year and first half of the second, indicate that d many levels of representation, of meaning and use have to be carefully specified (Rodríguez and Moro 1998; Moro and Rodríguez 2005). When objects are considered from a social point of view, rather than reduced to a “non-social physical reality”, they may have multiple meanings, and can be used according to very different functions, both by the children and by the adults in the interaction. Distinguishing these different functions, putting them into a developmental perspective, and understanding the adult’s influence as a cause of cognitive development is an urgent task that early developmental psychology must address. As we pointed out when considering the permanence of the object from its functional perspective, adults in their interactions with babies use objects as permanent long before the babies themselves are able to consider them as permanent entities. This has important consequences for the child since children already live in a world that is considered and treated as permanent by those surrounding them. The same thing can be said about symbols. Before children are able to produce their first symbolic uses of objects the adults surrounding them produce symbols (see below, observation 1) when using objects with different functions. This has various didactic purposes, to show how things should be done, when producing gestures, or when referring to absent things, actions or events. The adult acts with the child as a “symbol maker”, thus introducing them into symbolic scenarios of uses of objects long before they are able to understand symbols as symbols, or, of course, to produce them. The same must be said in relation to other semiotic systems as ostensive signs involving objects. Before children are able to produce their first ostensive signs, giving or showing (Moro and Rodríguez 1991), or with a private self-reflexive function (Rodríguez and Palacios 2007), the adult segments the world by highlighting certain aspects of it in a “space of joint action”, thus provoking shared meanings whose complexity grows through development. First objects appear in children’s life when they are used by another person in a communicative context (see observations below). The same can be said about the conventional uses of objects. Adults use objects in the everyday life in a conventional way almost constantly, introducing directly the child into these practices, a long time before objects become for the child signs of their social and public use. Objects are first used (in shared contexts) before being understood by the child. After all, meaning arises from use. The same thing happens with indicative gestures. Before the child indicates with the help of a sophisticated pointing gesture, adults manage in very different ways to make clear which events or actions are indicated, provoking situations of joint action and attention. Of course the same thing happens with language.
99
100 Cintia Rodríguez and Christiane Moro
All this implies a long process and a variety of levels of adult-baby agreement in communicative contexts about the meaning of objects and events. This variety of levels of agreement involves different semiotic systems with different degrees of complexity. According to our view, infants do not begin to interact with objects only from 5 or 6 months of age as is usually claimed in the literature (Messer 1997). It depends how things are analysed, in which context, and how many partners are considered. When we look microgenetically at babies as young as 2 months in naturalistic interactions with adults and objects, we can see how adults bring babies into contact with objects in the world and how they do things with them. Of course, the main responsibility and the initiative come from the adult since the baby is not yet capable of finding them on their own. We will consider all this in the next section.
5.
Joint uses of objects by adults and babies from 2 to 12 months
We will present five observations – from 2- to 12-months-old – to illustrate the evolution of triadic interactions where the adult frequently acts as a guide in contexts involving the use of a very common object. Through them we will see very different levels of agreement between adults and children about the same object. The multiplication and differentiation of new meanings is possible in contexts of shared use. New meanings involving different semiotic systems grow little by little through shared contexts of use and joint action. The videotaping took place in the families’ homes in three suburban areas of Madrid. The same object, “Chico”, very commonly used in schools as well as in the family homes, was used with all three dyads. This consists of 6 hoops of different diameters that can be placed over a tapered pivot. The biggest goes at the bottom, the smallest at the top. Three transparent hoops are filled with small pieces of plastic, and when shaken sound as a rattle. The three other coloured hoops are empty, and so produce no “rattling effect”. Two main conventional uses can be realised with this object (although, as we can see in the observations, other symbolic uses are also evident): 1. The simplest consists in shaking the hoops as if they were rattles. 2. The most complex consist in inserting the hoops onto the support.
. Of course, these observations are only illustrations of what happens at each age.
Coming to agreement 101
Through these illustrative observations we will see how different levels of agreement between adults and infants can be reached about the same object. At the beginning the adult focuses on the “easiest” uses at the level at which the child can be included (ostensive signs are very important then), mainly shaking the hoops as rattles (see observations 1–3). Later the tendency is to use the object in more complex ways and place them onto the pivot (see observations 4 and 5).
5.1
Ostensive signs and immediate demonstrations allow the beginning of joint actions
Observation 1: Alejandro, 0;1,30. Duration: 52 sec. [Joint action (father-baby-hoops) where the father realises several rhythmic ostensive uses of a hoop as a rattle and two symbolic uses. Alejandro follows his father’s uses with great attention and allows himself to be introduced into them] [Alejandro is lying on the sofa; his father is leaning towards him] The father takes a hoop and produces an ostensive sign, showing, shaking it to Alejandro and saying: “this makes a sound.” Then, once again, he shakes the hoop close to Alejandro’s ear and stops, following a rhythm four times as follows: “Listen, do you like it?” approaching the hoop again towards the baby. Alejandro, who has been looking very attentively at his father all the time, now smiles at him, vocalises, and moves his body (arms and legs). “Yes, this one you like…it”, his father says. When for the fourth time his father shakes the hoop close to his ear, Alejandro stops looking at his father’s face and turns his head towards his father’s shaking action of the hoop. His father then leaves the hoop and takes a second hoop: “this one is bigger”, doing a new ostensive sign when shaking it. Alejandro watches with interest his father’s action. The father then transforms the hoop into a hat when he puts it on Alejandro’s head, “and we put it on top of your head.” Alejandro turns his head and the hoop slides from his head. His father takes it again, shaking it in front of the baby a couple of times. Alejandro attentively watches his father’s shaking action and at the same time directs his arm towards the hoop. […] “Look, let’s see if you can pull it off your hand” […] introducing the hoop onto Alejandro’s arm as if it were a bracelet, then shaking his hand. This shaking-baby’s-hand-with-the-hoop action once again provokes a rhythmic sound. Alejandro, who is looking towards his father all the time, co-participates thanks to the immediate demonstration of the object made by his father when he involves the child in the same use, in his shared space of rhythmic action, allowing some sharing of meaning. The child also thus becomes a protagonist in the double function of the object in its conventional use (as a rattle) and in its symbolic use (as a hat and as a bracelet).
102 Cintia Rodríguez and Christiane Moro
The child is not only confronted with an object, but an object intentionally used by his father presented to him in a certain way as a result of a particular point of view, with particular intentions, constituted by meanings of use by the adult. When an object is being used, this implies that a choice is being made, a perspective is being adopted; it is an object regarding which certain signs are applied. If we look carefully at the father’s action, we see that he is using different semiotic systems to communicate different things about and through the hoop to his son. Through its use, he transforms the object into something else, this particular thing that would not be accessible to this very young child in isolation. The hoop shown to the baby becomes something to look at, to listen to, and eventually to be used conjointly. The hoop becomes an object of shared attention and, in some degree, of shared action by both protagonists. This is only possible thanks to the father’s ostensive gestures when using the object, although this does not mean that the meaning given by each of them is identical. At the beginning of the observation, Alejandro looks at his father’s face with interest. Only after a while does he shift his attention and look at the ostensive actions of his father with the hoop. The father uses several semiotic systems and at a certain point the ostensive signs (hoop shown as a rattle) become a symbol for the father (when it is used as a hat or as a bracelet). Looking at this situation carefully allows us to understand that babies are exposed to complex uses related to different systems of signs (including symbols and language) from a very early age, long before they are able to understand them as such, as symbols or as words, and of producing them. Even with a two-month-old child, we are far from dealing with a re-acting subject, as a “big looking eye”. This is so because adults introduce the child constantly into their own ongoing meaningful activity. They intentionally introduce the child into different semiotic systems long before the child is able to understand and share the same meanings. The child himself is active at three levels: (1) As a subject who is very much interested in the other’s actions. (2) The adult introduces the child constantly into his own uses. (3) The child himself acts at his level with his own resources; at the very end he directs his arm towards the hoop used by his father, and this provokes in his father a symbolic new use of the object as bracelet.
5.2 Ostensive categorization uses by the father by sound and uses by the child Observation 2: Alejandro, 0;4,7. Duration: 37 sec. [Ostensive presentation / classification by the father of different hoops according to whether they make sounds or not]
Coming to agreement 103
Alejandro is sitting on his father’s lap. Both are looking towards the support with the hoops very close on the table. The father puts three hoops on his hands, two empty ones that are yellow and orange and one transparent with little pieces inside, and presents all of them to the child while exclaiming: “which one do you want? This one makes a sound” shaking the transparent as a rattle, then leaves it on the table. “Or do you want this one?” showing the orange one which has nothing inside (there is no sound). Alejandro, who is looking with great interest at what his father is doing, stretches out his hand and takes the orange one. He opens his mouth thus indicating he is going to use it for sucking, which is what he immediately does. Then, his father pulls out from the support another transparent hoop with little pieces inside and then creates new ostensive signs when showing them all again to Alejandro. Alejandro keeps the orange hoop in his hands and looks at the two others his father is showing him: “or do you prefer those?” […] Alejandro drops the orange ring and stretches his hand out towards the “rattle” hoop, trying to grasp it. But it is too far away to reach, and then he starts crying […].
As in observation 1, here the interaction between the two protagonists takes place through the use of the hoops, and once again the father does not perform the “complex” conventional use of introducing the hoops inside the support, as it would be too difficult to be shared with Alejandro at such an early age. Instead of this, he focuses on the easiest conventional use of some hoops as a rattle. He segments his presentation, realising a series of ostensive signs when showing them, according to a classification/categorization, from the point of view of the sounds that can be obtained versus no sounds when shaken. In his ostensive presentation the father is organizing the reality showed to the child according to this “rhythmic musical” criterion. This raises a very interesting question about the degree to which the early children’s categorisations are actually spontaneous. We see here how Alejandro’s father is active at providing a meaningful framework to the child according to the uses that can be made of the hoops. His intervention provides the child with a shared context of classification (of meaning subsequently). It comes to be a really important way of discretization, of putting order into the reality from a rhythmic musical and pragmatic point of view. As in observation 1, Alejandro is, at his level, very active. He gets readily involved in his father’s ostensive signs when he shows the hoops from a rhythmic point of view. The big difference with what happened at 2 months is that now he is much more involved with the uses of the hoops he is able to perform himself. He readily understands the meaning of the ostensive signs of his father as “take” them and “do something” – most of the time Alejandro just sucks them. A very important shift takes place now in relation to the place of the object in the interaction, as Alejandro participates always actively making a certain selection before
104 Cintia Rodríguez and Christiane Moro
the ostensive signs with the objects realised by his father. In this sense, it is a triadic interaction, with joint action. Both the father and child share important degrees of agreement about what to do with what is being shown. The father puts the object into a certain framework of meaning. The child accepts this “point of view” and does something with it. Whereas at 2 months Alejandro got involved most of the time in the uses made by his father, he is now able to use and choose the objects in a much more autonomous way.
5.3 The object as sign of its social use for the child related to its musicality Observation 3: Alejandro, 0;6. Duration: 86: sec. [Alejandro is able to obtain a variety of rhythmic sounds when he shakes the hoops. The father is providing the support, allowing the child an exploratory use of the object according to its musicality, sonority and rhythm] Alejandro is sitting on his own (while his father is anxious about whether he will lose his balance). In front of the child four of the six hoops are inserted into the support. He directs his hands towards the support too far from him. His father brings it toward the child. Alejandro then hits the hoops inserted into the support with both hands provoking a rhythmic sound. This action of rhythmically hitting the support gives way to trying to get one hoop, but without success as this is too difficult for him (even though he is looking attentively at the hoops). His father “reads” his intentions evaluating at the same time the difficulties he is having, then takes the last hoop from the support (the orange) dropping it closer to the child (between his knees and the support). In the meanwhile Alejandro is looking and trying to search out another hoop outside of the support too far from his reach. His father moves it towards Alejandro but he finally takes the (orange) hoop previously dropped by his father close to him. Alejandro then hits his hoop against the support or against the table obtaining a variety of sounds. His father holds the support close to the child (otherwise it would slip out of his reach).
If we compare this observation with the previous ones, we see a major difference. In observation 1, it is the father who, through different ostensive signs, presents the hoops and introduces the child directly into their use: the sounds they can produce when shaken. In the second observation he presents several hoops making an ostensive sign, proposing to the child to make a choice from the point of view of the musicality, something that Alejandro does at once. Whereas in this third observation the child is already able of doing it, obtaining with the hoops two types of musicality (when hitting the table or the support). The common agreement shared between the child and the father about the use of the object
Coming to agreement 105
is very easy to see. In this case the father does not give the hoops to the child, he only brings them closer in such a way that the child himself can realise and explore different uses related to their musicality. Both share the meaning of an object to be taken. The father reads his intentions and makes adjustments to facilitate the child to take the hoops. These adjustments are far less ostensive (bringing to attention) than the ostensive signs when presenting the hoops (observations 1 and 2) because, (1) the hoops are already the object of the child’s attention, (2) both protagonists share the definition of the (easiest) conventional use of the object (if hoop, then take and produce musicality with it). At 6 months of age the child is able to explore different kinds of rhythmic sounds on his own. The common feature of these three observations is that the uses selected, promoted and facilitated by the father are related to sonority and rhythm (uses more basic than those related to introduce the hoops into the support). What is different is the degree of involvement of each protagonist in the use. We will see that the situation is quite different in the next observations with Javier, a 9-month-old child and with Nerea, a 12-month-old girl, where as we will see theirs mothers promote the more complex conventional use of introducing the hoops into the support.
5.4 The child understands the ostensive signs but does not understand the pointing gestures related to the more complex conventional use of the object Observation 4: Javier, 0;9. Duration: 28 sec. [The mother tries to elicit the more complex conventional use of the objects (to introduce hoops into the support). Javier understands without problems his mother’s ostensive signs, but does not understand her pointing gestures as indicating a certain use of the object] Javier is sitting. His mother is sitting close to him. Javier is manipulating the support with both hands whereas the hoops are lying all around. Then his mother takes the biggest hoop and, while showing/offering it to the child, says: “this one first, it is bigger”, “insert it my son, insert it”. Javier looks at the hoop presented by his mother, and takes it with one hand while holding the support with the other, but as the support is not vertical the hoop touches it in a parallel position. The mother says “there, there” doing a pointing that is immediate (the finger is touching the support) and multiple (it happens more than once, in this case two times). “Insert it”, “here, here, here” putting the support, held by the child, in a quasi-vertical position and doing again an immediate and multiple pointing. “Put it here”. Javier manipulates the hoop very close to the top of the support looking at it very carefully but not following his mother’s
106 Cintia Rodríguez and Christiane Moro
indicative gesture as related to the conventional use of the object of inserting the hoop into the support. Then the mother says, “[the hoop] has pellets” interpreting maybe the interest of the child in looking at the hoop, “does it sound?” After attending to the hoop, Javier once again takes the support, brings the hoop towards it but without trying to insert it. Then he puts the support far away to the right, while keeping hold of the hoop in his other hand […].
Compared with the previous observations, this is the first time the mother uses highly elaborated conventional indexical signs (pointing gestures), as semiotic mediators, and, this time, not only as ostensive signs with the object, as was the case in the previous observations. This is first time also that the mother tries to get the child to recognise the more complex conventional use of the hoop by introducing it inside the support. In the previous observations the adult was much more concerned about the use of the hoops according to their sounds (the other possible conventional use of the object). It is true that the degree to which the child is introduced into this use is very different from what happened with Alejandro in observation 1 (where the musicality comes from the father’s use) to observation 3 (where the child explores different sonorities and rhythms). What seems to us evident is that at his level Alejandro from 2 to 6 months was always part of this universe of sound and rhythm. Now we would like to stress how the ostensive signs function here. The mother shows/offers the hoop and Javier readily understands his mother’s intentions about this particular use of the object: he takes the object being offered. Clearly, there is no problem with this level of agreement between mother and child. Both share this kind of meaning related to this object use. Nevertheless, with the indexical signs (the pointing gestures in relation to the conventional use expected) things are quite different. The mother points towards the support and shows her intentions about the way the child should use the object (i.e., the more complex conventional use of the hoops by introducing them into the support), but Javier does not follow her intentions in relation to this particular use. He does not interpret the signs according to the use she proposes. To interpret correctly the pointing gesture of the mother means to use this indexical sign as a tool to help doing the conventional use of the hoop by introducing it into the support. To make this inference: if pointing gesture, then introduce this hoop into this support, is very complex and non-transparent. This requires sharing several levels of signs (and several semiotic systems involving the meaning of the pointing, the social meaning of the object, as “sign of its use”, and the articulation between both, considering that the object becomes a sign of its use thanks to the ostensive and indexical signs applied to it). Objects themselves do not demonstrate how they should be used. We always have to do that on their behalf. This is why objects are opaque
Coming to agreement 107
entities. Under no circumstances is their meaning self-evident. For the child to understand his mother’s intentions necessarily means to be able to articulate them with the state of affairs of the world. In this case Javier has no problem understanding the ostensive gestures, but does not understand the conventional indexical one relating to a public use. Parent and child share important meanings related about what to do, but only to a certain degree. Only later in development do more complex uses (canonical, or symbolic, for instance) become systematic.
5.5 Agreements about more complex use: the child understands the ostensive and the pointing gestures related to the use of the object Observation 5: Nerea, 0;12 Duration: 25 sec. [Mother elicits the most complex conventional use of the objects (introduce hoops into support). Nerea readily understands ostensive signs, as well as the pointing gestures of her mother as related to the conventional use of the object] [At the beginning of the session, Nerea does not make any connection between the hoops and the support as the place where they should be inserted but sucks them instead. Her mother makes several distant demonstrations (ostensive signs of the whole conventional use) showing how to do it. Finally Nerea starts trying doing it herself and she introduces herself into the new meanings according to the practice proposed by her mother] Nerea and her mother are sitting on the floor. The support is standing between both of them. One hoop is already inserted into the support. The mother inserts another saying: “come on I’ll insert them…” and, while she is saying this, Nerea moves the red hoop in her hand towards the support in an attempt to comply with the conventional use of the object. But it is too far away, and so she fails. Her mother, changing her initial intention of introducing the hoops herself, says, “Yes, this one” pointing towards Nerea’s approaching hoop. As it is too far, Nerea withdraws the hoop keeping it on her hand. She looks with great attention at her mother’s action. “Right, the red, put it here, here” indicating with an immediate gesture – touching the support with several fingers and multiple as she does it two times, provoking a sound. She says again: “here, put it here”, realising an immediate multiple pointing towards the support. And taking the whole support and shaking it to indicate where to insert the object, the mother exclaims, “The red one”, while pointing towards the hoop held by Nerea. “Here”, pointing again towards the support with an immediate and multiple pointing. Nerea, who has been closely attending to her mother’s gestures, once again moves the hoop towards the support in an
108 Cintia Rodríguez and Christiane Moro
attempt to insert it, but it runs away and falls down. The mother takes it, and segments her action into two phases: First she brings the hoop towards the support, and says, “look”, and then she stops for a while with the hoop in the air above the pivot. In the second phase, she says “like this”, inserts the hoop and thus completes the action. “And now, the orange one”, she says, showing/giving this new hoop to Nerea. Nerea, who was reaching out towards her mother’s action, immediately takes it. The mother says “here” pointing to the support, and Nerea “reads” her mother’s intentions as she interprets correctly the meaning of the pointing as related to the use of the object directing the hoop towards the place indicated by her mother. Finally she inserts the to hoop around the support and her mother claps her hands, “very good”. “Now, the yellow one” says the mother, giving it to Nerea, who again without hesitation takes it and direct towards the support. However, as she brings the hoop towards the support at the wrong angle (i.e. parallel rather than vertically), the hoop hits the support and so she fails to insert it. Even though she did not succeed, her intention was to achieve the conventional use of the object.
If we compare what happened in observations 1 to 3 and in observations 4 and 5, we see how the mother’s focus with the oldest children Javier (observation 4) and Nerea (observation 5) in the more complex conventional use of the object of introducing the hoops into the support, a thing that was completely ignored by the adult with the youngest. In this latter case, adults were more focused on the conventional use of the hoop as a rattle. In other words, adults are extremely sensitive to the abilities of the children, towards what they can and cannot do. This is how they ignore some uses (and meanings) of the object (too complex for a 2 or a 4 months old child) and privilege those uses allowing “levels of agreement” with the child. If we then compare Javier and Nerea we see that in the case of Javier he understands the meaning of the ostensive gestures of his mother “to take the object offered”, whereas the pointing gestures related to the use was too complex for him and he does not direct the hoop towards the support being pointed to by his mother. With Nerea the situation was similar at the beginning of the session, but in the observation we have presented, she was already able to understand many of the semiotic mediators at work when her mother was trying to elicit from her the conventional use of the object. Consider for instance, her comprehension of the different modalities of the pointing gestures, of the distant demonstrations or the ostensive signs when the mother shows the hoops to Nerea. At this moment, the object on its own becomes for Nerea “the sign of the (complex) use that should be done” with it. New and much more complex levels of agreement between the child and adult now begin to emerge about how to use the object.
Coming to agreement 109
As an open question for further research, we would like to stress that on another study concerning the private ostensive and pointing gestures that Nerea produces with the hoops and the support at 18 months (Rodríguez and Palacios 2007) we see certain parallelism with the segmentations of the use made by her mother that we have seen in this last observation. Whether or not there is any connection between both activities needs further research.
6.
Conclusions
Twentieth century philosophy has centered its reflection on language (Eco 1997). “The linguistic turn” has strongly affected our vision of language putting the question of its meaning into a pragmatic perspective of everyday contexts of use (Wittgenstein 1953; see Itkonen, this volume). There is no need to remind how this pragmatic position has been influential in Psychology. Bruner’s work with babies is an excellent example (Tomasello 2001; Shotter 2001). However, this pragmatic perspective has hardly been extended to objects, the very things we do things with. The emphasis placed by Bruner (1975) on contexts-of-use in relation to language acquisition has, paradoxically, not been applied to objects. Everything happens as if language was used, but objects were not. Perhaps this happens because objects seem to be evident, no matter how complex they may be. The naturalistic view of objects, according to which they are obvious and show their meaning in a literal and direct way, is predominant nowadays in early developmental research (Rodríguez 2006, 2007). This position has been challenged in this chapter. In opposition to the oversimplification implied by the naturalistic view of objects, we have situated them from a semiotic and pragmatic position. That is to say, we have considered their use(s) in the everyday life. Objects cannot be reduced to any physical self-evident “non-social” reality (Costall 2004; Rodríguez and Moro 1999, 2002; Sinha 2005, Sinha and Rodríguez this volume). Many of the things we do with them are only possible within communicative contexts, and communication itself changes according to the object or use selected. From a developmental perspective, very different levels of adult-baby agreement can be reached when they interact around an object. As examples illustrating this oversimplification, we have focussed on a few topics extremely well known in the literature, such as the dichotomy in the theory of mind research between the opacity of the social world versus the evident physical reality. This dichotomy does not fit with our findings when looking at babyobject-adult interaction. The so-called “physical” reality is complex as objects are included into normative practices.
110 Cintia Rodríguez and Christiane Moro
Our position is not far from Racine’s (2004) who, following Wittgenstein, distances himself from a naturalistic logic of meaning: “… intentions, beliefs and desires are not in our head; they exist and are understood in language-games” (2004: 271). What Racine says is that the theories of mind we use should not be placed exclusively inside the head because they are actually born, and take place in language-games. Our focus here refers to what happens below the level of language games. This means that it exists, taking Wittgenstein’s words, in sign-games with objects (Rodríguez 2006) long before – and probably as a pre-condition for – the language games to appear later in development. Two subjects, such as an adult and a 2 months old baby, can reach important levels of agreement around the use of an object because adults make clear their intentions when using objects in a communicative context. Concerning the question of the subject when dealing with objects, in our view, there is an urgent need to place the active child back on the agenda of Psychology, in place of a merely reactive one. Often studies in early cognitive development belong to the “kingdom of spontaneity”, ignoring that children interact with others since the beginning of their life in contexts of joint action. The permanence of the object is another important but oversimplified topic. How many kinds of object permanence are there? According to our view, it is necessary to include something like a pragmatic permanence that takes into account its social use. After all, if objects have public names, they have also public functions. How does this affect the conventional use of the object (once the child has got it) to the object’s permanence? Is there any link? After all, objects are used according to certain rules shared by the community and these represent some kind of stability. The conventional use procures for the object a permanence of its social use, a sort of “label” of use (Rodríguez 2007). This means a detachment from the concrete and strict individuality of the case. A similar thing can be said in relation to early categorizations. Is there any relation between the public uses of objects and early conceptual categorizations? To use an object in a conventional way implies the application of a certain level of categorization. Maybe we should conclude that early categorisations made by children are not so culture free, not as spontaneous as it has been widely assured. When we put the object into this pragmatic position, and examine the uses of objects microgenetically within triadic interactions – baby-object-adult- we discover the multiple scenarios at work and how each new shared meaning emerges through use in such communicative contexts. Objects do not afford the same things in different moments in ontogenesis and cannot be excluded from normative practices. Adults seem to “know” this very well.
Coming to agreement
In the observations presented in this chapter, we have seen how children reach, through use, very different degrees of agreement with adults – involving diverse semiotic systems – around the meaning of objects. All observations involve situations of joint action between two protagonists and an object (not only joint attention where babies most of the time do not act). Different kinds and levels of conventional uses and different semiotic systems are at work. When objects are considered from this pragmatic position, the fiction that sees them as a literal, self-evident, non-social reality disappears. In fact many communicative processes take place through them. Meanings can be very diverse, as different semiotic systems with different levels of complexity are involved. The child reads the social meaning of the object and the adult’s intentions because signs and objects are articulated through use. Dealing with all that implies putting the child in an active position (as its common place among the classical theorists). Triadic interactions appear as a royal way of considering the processes at work long before the child is able to actively involve the adult and the world in a same communicative act. The “magic number three” as the unit of analysis – with the educational role of adults – may help to clarify how this growth of meaning takes place, starting with the simplest signs, with the simplest uses. The triadic interaction we are dealing with cannot be only joint attention, that is to say, a communality of attention around something (see Zlatev this volume). We need more; we need to know how what we call joint action works and evolves, how both protagonists transform the object by using it. A long time before the child is able to produce his first ostensive, conventional and symbolic uses of objects, once they become “signs of its [social] use”, his first pointing gestures towards something, or his first words, the adult acts with the child as a “symbol maker”, produces ostensive gestures with objects, points to them to make clear her intentions, uses the world in a canonical manner and talks to the child almost constantly. This implies a long process and an enormous variety of levels of adult-baby agreement about objects and situations during the first year of life. In one word, the meanings of objects in everyday life are a function of their use. If objects are not excluded from normative practices, then we need to know how this pragmatic position of the object and communication with others affects early cognitive development. After all, we do things with words, but we also do things with things (Costall and Dreier 2006), in agreement with others, and this is extremely important from the very beginning of life. Therefore, we have to understand how these sign games where objects are included work in development. Many levels of conventions refer not only to language but also to objects. Early cognitive development and linguistic theories cannot ignore all that any longer.
111
112 Cintia Rodríguez and Christiane Moro
Acknowledgements We would like to thank Alan Costall and the Editors of this book for their excellent comments and suggestions on this chapter.
References Arango, S., Chávez, L. and Lasprilla, A. 2003. Uso de un Objeto por Seis Niños Autistas. Unpublished manuscript, Universidad del Valle, Instituto de Psicología. Bakeman, R. and Adamson, L. 1986. “Infants’ conventionalized acts: Gestures and words with mothers and peers.” Infant Behavior and Development 9: 215–230. Baron-Cohen, S. 1993. “From attention-goal psychology to belief-desire psychology: The development of a theory of mind, and its dysfunction.” In Understanding Other Minds: Perspectives from Autism, S. Baron-Cohen, H. Tager-Flusberg and D. Cohen (eds.), 59–82. Oxford: Oxford Medical Publications Barresi, J. and Moore, C. this volume. “The neuroscience of social understanding.” Bremner, G., Slater, A., Foster, K., Johnson, S., Mason, U., Cheshire, A. and Spring, J. 2005. “Conditions for young infants’ perception of objects trajectories.” Child Development 76: 1029–1043. Bronckart, J.-P. 1997. “Semiotic interaction and cognitive construction.” Archives de Psychologie 65: 95–106. Bruner, J. 1975. “From communication to language: A psychological perspective.” Cognition 3: 255–287. Bruner, J. 1990. Acts of Meaning. Cambridge, MA: Harvard University Press. Carpendale, J.I.M., and Lewis, C. 2004. “Constructing an understanding of mind: The development of children’s social understanding within social interaction.” Behavioral and Brain Sciences 27: 79–151. Corballis, M. 2002. From Hand to Mouth. The Origins of Language. Princeton: Princeton University Press. Costall, A. 2004. “From direct perception to the primacy of action.” In Theories of Infant Development, G. Bremner and A. Slater (eds.), 70–89. Oxford: Blackwell. Costall, A. and Dreier, O. 2006. Doing Things with Things. The Design and Use of Everyday Objects. Hamphshire: Ashgate. Donald, M. 2001. A Mind So Rare. The Evolution of Human Consciousness. New York: W.W. Norton. Eco, U. 1999 [1997]. Kant y el Ornitorrinco. Barcelona: Lumen. Gibson, E.J. and Pick, A. 2000. An Ecological Approach to Perceptual Learning and Development. Oxford: Oxford University Press. Gould, S.J. 2001 [1998]. Les Coquillages de Léonard: Réflexions sur L’histoire Naturelle. París: Seuil. Inhelder, B., Céllerier, G., Ackermann, E., Blanchet, A., Boder, A., de Caprona, D., Ducret, J.-J. and Saada-Robert M. 1992. Le Cheminement des Découvertes chez l’enfant. Recherche sur les Microgenèses Cognitives. Neuchâtel-Paris: Delachaux et Niestlé. Itkonen, E. this volume. “The central role of normativity for language and linguistics.”
Coming to agreement 113
Karmiloff-Smith, A. and Inhelder, B. 1974. “If you want to get ahead get a Theory”. Cognition 3: 195–212 Leavens, D.A., Hoppkins, W.D. and Bard, K.A. this volume. “The heterochronic origins of explicit reference.” Legerstee, M., Pomerleau, A., Malcuit, G. and Feider, H. 1987. “The development of infants’ responses to people and a doll: Implications for research in communication.” Infant Behavior and Development 10: 81–95. Legerstee, M. and Weintraub, J. 1997. “The integration of person and object attention in infants with and without Down syndrome.” Infant Behavior and Development 20: 71–82. Leslie, A. 1987. “A ‘Language of Thought’ approach to early pretense.” In Symbolism and Knowledge/Symbolisme et connaissance, J. Montangero, A. Tryphon and S. Dionnet (eds.), 133–144. Genève: Cahiers de la Fondation des Archives Jean Piaget, 8. Leslie, A. and Roth, D. 1993. “What autism teaches us about metarepresentation.” In Understanding Other Minds. Perspectives from Autism, S. Baron-Cohen, H. Tager-Flusberg and D. Cohen (eds.), 83–111. Oxford: Oxford Medical Publications. Messer, D. 1997. “Referential communication: Making sense of the social and physical world.” In Infant Development: Recent Advances, G. Bremner, A. Slater and G. Butterworth (eds.), 291–309. Sussex: Psychology Press. Mandler, J. 2000. “Perceptual and conceptual processes in infancy.” Journal of Cognition and Development 1: 3–36. Mandler, J. 2004a. The Foundations of Mind. Origins of Conceptual Thought. Oxford: Oxford University Press. Mandler, J. 2004b. “Two kinds of knowledge acquisition.” In The Development of the Mediated Mind: Sociocultural Context and Cognitive Development, J. Lucariello, J. Hudson, R. Fivush and P. Bauer (eds.), 13–32. Mahwah, London: LEA. Moore, C. 2004. George and Sam. London: Penguin Books Moore, K. and Meltzoff, A.N. 1999. “New findings on object permanence: A developmental difference between two types of occlusion.” British Journal of Developmental Psychology 17: 563–584. Moro, C. and Rodríguez, C. 1991. “¿Por qué tiende el niño el objeto hacia el adulto? La construcción social de la significación de los objetos?” Infancia y Aprendizaje 53: 99–118. Moro, C. and Rodríguez, C. 2005. L’objet et la Construction de son Usage chez le Bébé: Une Approche Sémiotique du Développement Préverbal. Berne – New York: Peter Lang. Mossier, C. and Rogoff, B. 1994. “Infants’ instrumental use of their mothers to achieve their goals.” Child Development 65: 70–79. Nelson, K. 1974. “Concept, word and sentence: Interrelations in acquisition and development.” Psychological Review 81: 267–285 Pardos, A. and Rodríguez, C. 2005. “The importance of the use of objects in the early detection of autism.” Paper presented in the symposium Uses of objects and semiotic mediation in impaired children. First ISCAR Congress, Seville, 20–24 September. Piaget, J. 1977 [1937]. La Construction du Réel chez L’enfant. Neuchâtel-Paris: Delachaux et Niestlé. Pika, S. this volume. “What is the nature of the gestural communication of great apes?” Racine, T.P. 2004. “Wittgenstein’s internalistic logic and children’s theories of mind. ” In Social Interaction and the Development of Knowledge, J.I.M. Carpendale and U. Müller (eds.), 257–276. Mahwah, NJ: Erlbaum.
114 Cintia Rodríguez and Christiane Moro
Rodríguez, C. 2007. “Object use, communication and signs. The triadic basis of early cognitive development.” In The Cambridge Handbook of Socio-Cultural Psychology, J. Valsiner and A. Rosa (eds.), 257–276. New York: Cambridge University Press. Rodríguez, C. 2006. Del Ritmo al Símbolo: Los Signos en el Nacimiento de la Inteligencia. Barcelona: ICE-Horsori. Rodríguez, C. and Moro, C. 1998. “El uso convencional también hace permanentes a los objetos.” Infancia y Aprendizaje 84: 67–83. Rodríguez, C. and Moro, C. 1999. El Mágico Número Tres. Cuando los Niños aún no Hablan. Barcelona: Paidós. Rodríguez, C. and Moro C. 2002. “Objeto, comunicación y símbolo. Una mirada a los primeros usos simbólicos de los objetos.” Estudios de Psicología 23: 323–33. Rodríguez, C. and Palacios, P. 2007. Do private gestures have a self-regulatory function? A case study. Infant Behavior and Development 30: 180–194. Shotter, J. 2001. “Towards a third revolution in Psychology: From inner mental representations to dialogically-structured social practices.” In Jerome Bruner. Language, Culture and Self, D. Bakhurst and S.G. Shanker (eds.), 167–183. London: Sage Publications. Sinha, C. 2005. “Blending out of the background: Play, props and staging in the material world.” Journal of Pragmatics 37: 1537–1554. Sinha, C. and Rodríguez, C. this volume. “Language and the signifying object: From convention to imagination.” Spelke, E. 1998. “Nativism, empiricism, and the origins of knowledge.” Infant Behavior and Development 21: 181–200. Striano, T. and Bertin, E. 2005. “Social-cognitive skills between 5 and 10 months of age.” British Journal of Developmental Psychology 23: 559–568. Striano, T. and Rochat, P. 1999. “Developmental link between dyadic and triadic social competence in infancy.” British Journal of Developmental Psychology 17: 551–562. Thelen, E., and Smith, L. 1998 [1994]. A Dynamic Systems Approach to the Development of Cognition and Action. Cambridge: The MIT Press. Tomasello, M. 2001. “Bruner on language acquisition.” In Jerome Bruner. Language, Culture and Self, D. Bakhurst and S. G. Shanker (eds.), 31–49. London: Sage. Vygotsky, L. 1996 [1984]. “El primer año.” In Obras escogidas IV. Psicología infantil, L. Vygotsky, 275–318. Madrid: Visor. Wallon, H. 1970 [1942]. De L’acte à la Pensée. Paris: Flammarion. Williams, E., Kendell-Scott, L. and Costall, A. (2005). “Parents’ experiences of introducing everyday object use to their children with autism.” Autism 9: 521–540. Willatts, P. 1997. “Beyond the ‘couch potato’ infant: How infants use their knowledge to regulate action, solve problems and achieve goals.” In Infant Development: Recent Advances, G. Bremner, A. Slater and G. Butterworth (eds.), 109–135. Sussex: Psychology Press, Erlbaum. Wittgenstein, L. 1958 [1953]. Philosophical investigations. Englewood Cliffs: Prentice Hall. Zlatev, J. this volume. “The co-evolution of intersubjectivity and bodily mimesis.”
chapter 6
The role of intersubjectivity in the development of intentional communication Ingar Brinck The present account explains (i) which elements of nonverbal reference are intersubjective, (ii) what major effects intersubjectivity has on the general development of intentional communication and at what stages, and (iii) how intersubjectivity contributes to triggering the general capacity for nonverbal reference in the second year of life. First, intersubjectivity is analysed in terms of a sharing of experiences that is either mutual or individual, and either dyadic or triadic. Then it is shown that nonverbal reference presupposes intersubjectivity in communicative intent indicating and referential behaviour, and indirectly in modifications of previous behaviour in response to communication failure. It is argued that different forms of intersubjectivity entail different types of communicative skills. A comprehensive analysis of data on gaze-related intersubjective behaviour in young infants shows that interaffectivity and interattentionality enable referential skills early in development and together allow for complex behaviour. Early referential skills, it is proposed, arise by other mechanisms than in nonverbal reference. Reliable and consistent use of nonverbal reference occurs when interaffectivity and interattentionality coalesce with interintentionality, which affords general cognitive skills that together permit a decontextualisation of communicative behaviour.
1.
Introductory remarks on the approach and method
Few people would disagree with the statement that intersubjectivity plays a critical role for language acquisition and is central to nonverbal reference, yet not many would agree about its exact significance. Intersubjectivity is a complex relation that manifests itself in many types of context and in different ways, which makes it difficult to elucidate exactly how it contributes to these capacities. The aim of the present chapter is to account for the role that intersubjectivity plays for the
116 Ingar Brinck
evelopment of nonverbal, intentional communication in human infants. The ded velopmental psychologist’s definition of intersubjectivity in terms of a sharing of experiences will provide the starting-point for an approach in three steps. First, it will be determined which elements in the act of nonverbal reference are intersubjective, then the major effects of intersubjectivity on the developmental trajectory of intentional communication will be established, and finally it will be explained how intersubjectivity contributes to elicit the capacity for nonverbal reference. The approach is interdisciplinary and takes experimental research in developmental psychology and related areas as the foundation for generating new hypotheses and explanations from a general, cognitive perspective. In the present context this will mean to analyze and systematize existing data on intersubjective and referential skills in infants, evaluate current explanations of such skills, establish connections among different types of data that can be expected to have a general explanatory value, and develop the necessary theoretical and conceptual framework for explaining any observations that are made during the course of the investigation.
2.
Sharing experiences requires complementary capacities
In developmental psychology, intersubjectivity is frequently defined as a deliberate sharing of experiences about objects and events (cf. Trevarthen and Hubley 1978; Stern 1985). There is no reason to contest the central part of the definition, that the infant’s capacity for intersubjectivity concerns a ‘sharing of experiences’. We will return to it below. The first part of the definition that describes intersubjectivity as ‘deliberate’ is ambiguous. A weak interpretation in terms of goal-directedness or intentionality is presupposed by the concept of intersubjectivity, which implies interaction, and is therefore unproblematic. Any wider implications that stronger interpretations might have are of no particular concern to the present discussion, and will not be further considered. The last part of the definition according to which experiences are ‘about’ something expresses an obvious truth, yet is easily misunderstood. Intersubjectivity can . In line with current praxis in developmental psychology and research on language acquisition, the term ‘intentional communication’ will be used to refer to nonverbal, referential communication by gesture, gaze, vocalisation, etc., in preverbal human infants. . Experiences are ‘felt qualities’, i.e., the qualitative ways in which objects, states, and events present themselves to conscious awareness when perceived from a first-person perspective. They vary in intensity and vividness, and are sometimes value-laden.
Intersubjectivity and intentional communication 117
be triadic (relating two subjects relative to a third element) or dyadic (relating two subjects). In the former case, the subjects are said to exchange experiences around an object. In the latter case, the experiences concern the subjects themselves and the interaction between them, and therefore reflect the fundamental reflexivity of the relation of intersubjectivity. The definition of intersubjectivity as a sharing of experiences draws attention to the interactive aspect of the relation. Intersubjectivity first materializes in the form of early imitation between the newborn infant and an adult (usually its caretaker), and soon develops into mutual engagement. During these exchanges, infant and adult characteristically take a second-person perspective toward each other. Mutual engagement develops in the first month from the repetitive but active and rhythmic matching of facial expressions of emotion between adult and infant. In turn, proto-conversation is built around turn-taking, that is, the reciprocal co-ordination and sequencing of more or less spontaneous behaviours and actions in time (Trevarthen 1979; Trevarthen and Aitken 2001). By two months, the infant can produce differentiated responses to the adult’s attention and so take turns, and one month later, begins to actively call others’ attention to the self, as it seems, to initiate mutual engagement (Reddy 2005). The newborn infant’s primordial experience of similarity with the other is a precondition for developing intersubjectivity. Repeated episodes of mutual engagement during the first months in life further promote the implicit recognition of similarity between self and other. Meltzoff and Brooks (2001) argue that this recognition is based in a cross-modal mapping of felt and observed actions, which prepares the infant for adjusting his or her individual behaviour to the needs and demands of the other, as in later and more complex forms of interaction. It seems reasonable to think that the cross-modal mapping of actions causally depends on activity in the mirror neurons. Research on mirror neurons has revealed that manual actions are recognised by a mapping of the observed action onto a motor representation of it in the observer’s brain. When an observer is watching an agent perform an action, there is a concurrent activation of the motor circuits that would have been recruited had the observer performed the action herself (Gallese, Keysers and Rizzolatti 2004: 397). Furthermore, noticing another agent’s facial expressions of emotion will activate similar areas of the observer’s brain as of the agent whose face is observed, and gives rise to similar sensations and negative or positive experiences (Gallese et al. 2004; Rizzolatti et al. 2002; Wicker et al. 2003; Barresi and Moore, this volume). In line with this, the neural correlates . Other brain regions, e.g., the inferior prioretal cortex, make it possible for the brain (hence, the agent) to discriminate actions of the self from those of another (observed) agent. See Barresi and Moore (this volume) on the role of “mirror neurons” for intersubjectivity.
118 Ingar Brinck
of mutual engagement can be described in terms of spreading neural activation in the brain’s motor representations, which trains the perception-action system to recognise and produce instrumental actions and facial expression of emotion and to fine-tune its reactions (cf. Decety and Ingvar 1990; Jeannerod 1997). Meltzoff and Brooks (2001) maintain that the similarity between self and other presents itself directly to the senses of the infant in mutual engagement. Yet thinking of mutual engagement as based in an experience of mere similarity makes it hard to see how it might engender a dynamic interaction or exchange. Reddy (2003, 2005) resolves the dilemma by proposing that the experience of self-other equivalence presupposes both the infant’s original identification with the other and the experience of being different. Once made, the point is obvious: Similarity is only noticeable against the background of an experienced dissimilarity. It explains why mutual engagement is in fact dynamic and not purely contemplative although intersubjectivity presupposes sameness, or identity. Reddy’s remark implies that the experience that we have of the self and its relation to others is inherently double. In other words, self-awareness presupposes the awareness of another subject who will reflect the similarity by displaying points of difference. To follow up on what was said above about the neural basis of self-other equivalence, the present idea also can be expressed by reference to the perception-action system in the brain. Original identification is likely to be based in the activity of mirror neurons that automatically map observations of behaviour onto the corresponding areas in the brain of the observer. The experience of difference then will emerge from the need to calibrate the motor representations in the brain that are caused by the perception of other agents’ behaviour. The sharing of experiences requires complementary capacities. First, there is the capacity to recognise the experiences of another individual, and second, there is the capacity to make available one’s own experiences to somebody else. Even if the two correspond, they are distinct. The newborn infant needs to practise both, and learns how to harmonize them by engaging in early imitation soon after birth. Given that both the similarity and difference between self and other are perceivable, the infant will directly experience the bi-directional relation between self and other in mutual engagement, and eventually develop an intuitive understanding of intersubjectivity. Furthermore, since early intersubjectivity is grounded in concrete contexts of reciprocal interaction, the two complementary capacities automatically are brought together.
. The present view has affinities to the theory advanced by Goldman (2006), who argues that mind-reading is simulation, and that some forms of mind-reading, which do not depend on verbal capacities, are based in the mirror neuron system.
3.
Intersubjectivity and intentional communication 119
Sharing experiences individually or mutually
It might seem problematic that while the definition of intersubjectivity emphasizes a mutual sharing of experiences in the second person, just a fraction of the infant’s skills that are characterised as intersubjective in the literature is genuinely reciprocal. For instance, nonreciprocal gaze following in the direction of a target involves checking where the other agent is looking by attending to head and body orientation, visually searching for a target in common space, and then localising the target by gaze alternation between the agent and salient items in the shared context. In such cases, experiences are shared on an individual basis in the third person. Since behaviour that is built around individual sharing is not reciprocal in the sense of involving attention contact, it is a fair question whether there are other, independent reasons for calling it ‘intersubjective’, except for its being a sharing of experiences by definition. If not, it might seem that individual sharing is intersubjective in a merely derived or figurative sense, and we might choose not to call behaviour that relies on it ‘intersubjective’. However, it is clear that this behaviour requires a sharing of experiences, even if both parties are not aware of doing so. If an agent’s perceptual experiences could not be shared, or accessed, by observation, behaviour such as nonreciprocal gaze following would be impossible. That individual sharing is intersubjective is also evidenced by the fact that similar mechanisms in the brain underlie individual and mutual sharing of experiences. As described in the previous section, mirror neurons make it possible for another agent’s experiences to resonate in the observer. The underlying mapping mechanism will only function properly if it originates from the experience of selfother equivalence and receives training in episodes of mutual engagement. Besides, to decide to call only behaviour that involves attention contact and mutual engagement ‘intersubjective’ will prove impractical, given that in developmental and comparative psychology the term often is used for social behaviour that involves attending to another agent’s attention – whether or not the agents have attention contact. Thus, sharing is individual when a subject gains access to another subject’s experiences via observation (this still concerns attending to the other subject’s attention). In contrast to mutual sharing, it is one-way, distanced, and instrumental. Another example is gaze alternation as part of social referencing, when the infant seeks information from the adult by looking at his or her facial expression of emotion. Many states of mind can be manifest in overt behaviour and then are perceptually available to others in the third person (Bühler 1934). Thus attention, emotion, attitude, interest, and perceptual knowledge can be read from posture, movement, gesture, facial expression, gaze, head and body turn, and (nonverbal) vocalisation. Attention reading, the recognition of goal-directed intentions in
120 Ingar Brinck
others from observations of their behaviour relative to salient entities in the local context, is a generic form of individual sharing (Brinck 2004). A mutual sharing of experiences occurs when two agents gain access to each other’s experiences via either observation or attention contact. Mutual sharing is bi-directional. It does not necessarily entail attention contact, because the agents may be looking repeatedly towards each others’ faces but not simultaneously. It will still be a case of mutual sharing, since each agent is aware of the other’s experiences – even if they are not mutually aware of sharing experiences. Two examples of mutual sharing that do include attention contact are (triadic) joint attention and (triadic) ritualised behaviour in the form of structured play. Thus, consider the repeated giving and taking back of an object while exchanging emotions around it by vocalisation and facial expression, as illustrated by the toddler who repeatedly hands an old sock to her parents’ dinner guest, expecting to receive it in return from the adult.
4.
The act of nonverbal reference
To explain what is intersubjective in intentional communication and how it is so calls for an account of intentional communication that identifies its basic elements. Such an account is offered below, where the range of intentionally communicative behaviour that has been observed among human preverbal infants is categorised into four types, according to the standard uses the behaviour has been observed to have in acts of nonverbal reference. Intentional communication may be defined as the nonverbal, spontaneous and purposively produced social interaction between (typically) two agents relative to a distal object in common space. Its primary use is to establish joint attention to a third entity, typically for some further purpose, according to the sender’s needs and desires (Brinck 2001, 2003, 2004). The process that leads to joint attention is flexible, and can be adjusted to meet the behaviourally manifest idiosyncrasies and expectations of individual agents. The act of nonverbal reference, with the function of directing the observer’s attention to a distal target, is the paradigmatic example of intentional communication. Although its principal means is the
. The present analysis of intentional communication globally agrees with the analyses in Bard (1992), Bates et al. (1979), Leavens, Hopkins and Thomas (2004), and Leavens, Russell, and Hopkins (2005), and applies to human infants and Great apes alike.
Intersubjectivity and intentional communication 121
pointing gesture, it can be accomplished by other means, such as gaze or head and body turn in the direction of the target. Intentionally communicative acts are complex in the sense of being composed from a selection of behaviour. First, the varieties of behaviour that occur in intentional communication can be categorised into four distinct types according to the functions that such behaviour has been observed to have in contexts of use. The types consequently constitute the basic acts or units of the complex act of nonverbal reference. Which of these acts should be performed to achieve nonverbal reference on a given occasion is determined in the context of use. Below, the act of nonverbal reference is described from the sender’s perspective in terms of the four types of behaviour that are used to perform it (cf. Bard 1992; Bates et al. 1979; Leavens, Hopkins and Thomas 2004; and Leavens, Russell and Hopkins 2005). Taken together they are intended to subsume the multitude of communicative behaviour that recurs in nonverbal reference. Each type of behaviour will be characterised, first, operationally by its observable features, and then, in terms of its use, or function, in concrete situations. The present concept of a ‘function’ identifies behaviour by the observable use it has across contexts. It stands in contrast to the concept of a teleological or design function, which identifies behaviour in terms of its intended goal: the effect that is achieved by performing the behaviour successfully.
. A wide interpretation of the pointing gesture is used in the research on intentional communication, typically as the extension of the arm and hand towards a distal target, with or without the index finger outstretched. . Categorising behaviour with respect to its major detectable use in acts of nonverbal reference may be valuable for comparative psychology, which investigates cognitive differences between species, in simplifying the identification and explanation of behavioural similarities across species in spite of existing dissimilarities among surface properties. The present account of the basic units of referential acts provides the tools for further investigation of the varieties of behaviour in intentional communication. . It is very difficult to check whether a teleological interpretation of nonverbal behaviour is correct, which is why an interpretation in terms of use generally is preferable to one in terms of desired effect. Austin (1962) introduced a distinction between an utterance’s illocutionary force and perlocutionary effect that is similar to the distinction made here. It captures the difference between what is achieved in saying something and by saying something. The illocutionary force concerns changes that an utterance (if successful) will produce as a result of its regular use and meaning (informational content) in the context of utterance, but not any changes that occur in the wider environment. The perlocutionary effect concerns changes that an utterance can be used to achieve outside of the utterance context given that the non-linguistic contextual conditions are appropriate. Brinck (2003) argues against definitions of pointing in terms of perlocutionary effect, or what an act of pointing is meant to achieve, and claims instead that declarative
122 Ingar Brinck
i. Preparatory (attention getting) behaviour, e.g., gesturing, vocalising, and similarly conspicuous sounds and behaviours, drawing the observer’s attention to the sender. Given that the behaviour is successful, and the observer reacts to it by turning his or her attention to the sender, it will permit the sender to subsequently manipulate the observer’s behaviour via his or her attention. Thereby the behaviour prepares for the ensuing interaction between sender and observer. ii. Communicative-intent indicating behaviour, e.g., looks to the observer’s face and eyes, gesturing, gazing, vocalising, and touching, performed relative to the attentional status of the observer. These communicative-intent indicators signal the sender’s attempt to have attention contact and interact face-to-face with the observer. One might say that they make the sender’s intention to communicate mutually manifest to sender and observer. iii. Referential behaviour, e.g., pointing, gazing, visual orienting towards a (distal) target, head and body orienting in the direction of the target, gaze alternation between the observer’s face and the target, and looking back, performed relative to the attentional status of the observer. Referential behaviour displays the sender’s interest in, on the one hand, a target of attention in common space, and on the other, the observer’s attention. If successful, it will make the observer shift his or her attention to the sender’s target. In guiding the observer’s attention to the target, referential behaviour fixes the content of individual acts of nonverbal reference.
pointing is an illocutionary act with an indicating function, and that it can be used for different purposes, and thus can have different perlocutionary effects. . Developmental psychology defines the pointing gesture of human infants by what an act of pointing is meant to achieve, thus distinguishing between an imperative and a declarative form. Pointing can be used imperatively to request an object, or declaratively to achieve joint attention to an object for some further purpose, such as exchanging experiences of it with the observer, informing the observer about its location, initiating play that involves it, etc. (Bates 1976; Brinck 2003). Declarative pointing usually is characterised in referential terms, while imperative pointing is described as partly communicative, partly instrumental. Yet, given that imperative pointing is performed relative to the observer’s attentional state, it involves a referential element in directing the observer’s attention to the requested object (cf. Hopkins, Leavens, and Bard, this volume). Henceforth, the term ‘referential pointing’ will be used for any pointing gesture that satisfies this requirement.
Intersubjectivity and intentional communication 123
iv. Essentially intentional behaviour, i.e., persistent behaviour until reward, and elaboration of behaviour when repeated attempts to communicate fail. Essentially intentional behaviour shows that the sender understands that “different means may be directed toward the same end and that the same means may be used for different ends” (Tomasello and Call 1997: 361), and consequently is distinctive of intentional communication. Given that the four basic acts have counterparts in the real world, intentional communication has systematic properties. The acts have a similar use, or function (‘meaning’) across contexts, independently of individual agents, and in different combinations. Several distinct behaviours share the same function and so can realise the same act. This permits using them selectively on different occasions to meet contingent contextual demands. As a consequence of this flexibility in use, both as to the selection of basic acts and the behaviour that realises the selected act, nonverbal reference can take a great many forms. Essentially intentional behaviour has a fundamentally different function than the other basic acts, which gives it a unique status (Bates et al. 1979). It reinforces the sender’s previous act by introducing changes in the way the act was expressed, while preserving its quality, or content. In spite of the fact that over time essentially intentional behaviour has become the scientist’s litmus test for intentional capacities in the sender, it does not have a strictly communicative function, and is not in general necessary for the success of acts of nonverbal reference. To be more precise, essentially intentional behaviour constitutes a meta-operation that the sender uses to enhance previous behaviour that has failed to achieve the act of nonverbal reference. This makes it a resource for improving and repairing on-going communicative behaviour.
5.
Intersubjectivity in intentional communication
Having identified and described the building-blocks of acts of nonverbal reference, it is now possible to examine whether intentional communication is intersubjective. Remember that the present definition of intersubjectivity is in terms of sharing experiences, and that both mutual and individual ways of sharing have been specified. Of the four behaviour types, only preparatory (attention-getting) behaviour is not intersubjective. It neither depends on, nor directly issues in a sharing of sorts, but its role is exactly to prepare for the sharing of experiences. It functions as a signal that during normal circumstances will elicit a certain response in the observer, namely, a re-orientation of her attention towards the sender. This response
124 Ingar Brinck
may be explained in terms of reflexive behaviour as a reaction to the perception of a salient event in common space, say, the sender’s waving his arms, vocalising, or tapping on the ground. In contrast, communicative-intent indicators and referential behaviour are both intersubjective. This is demonstrated by the fact that unless they cause a sharing of experiences, they will misfire. Communicative-intent indicating behaviour is successful in case the observer notices the sender’s attempts to communicate, and in addition alerts the sender to this recognition, preferably by establishing attention contact with him or her. Referential behaviour is successful if it makes the observer attend to the sender’s object of attention in common space. It does not necessarily require that the agents are mutually aware of sharing experiences to be effective, although reciprocal intersubjectivity is typical for intentional communication. Whenever salient properties of the shared context can be expected to direct the observer’s attention towards the target, the sender will not have to invest herself in the other agent. As a result of the sender’s purposive behaviour, contextual scaffolding sometimes can replace mutual sharing relative to referential behaviour. Essentially intentional behaviour, finally, is directed at improving the conditions for communication by changing the way in which a particular act of nonverbal reference is expressed. It inherits its intersubjectivity in the context of use from the particular behaviour for whose enhancement it has been invoked. This means that essentially intentional behaviour presupposes the capacity for intersubjectivity in the sender. The present inquiry into the intersubjectivity of intentional communication suggests that although intersubjectivity often is ascribed to intentional communication on a general level, in fact, only a limited set of intentionally communicative behaviour is fundamentally intersubjective. Such behaviour has the function of either making manifest the sender’s intention to communicate by the attempts to engage in attention contact with the observer, or guiding the observer’s attention towards the sender’s target of attention in common space, by the sender’s referential behaviour performed relative to the observer’s attention. To conclude, intersubjectivity plays an integral role for intentional communication. Yet, it is still uncertain whether intersubjectivity is the triggering factor of intentional communication that elicits its onset. Data concerning the intersubjective and communicative skills in young infants that apparently speak against this hypothesis will be presented in the next section. The claim that intersubjective skills pave the way for intentional communication will be investigated relative to these data in Section 7.
6.
Intersubjectivity and intentional communication 125
Referential skills in early intersubjective behaviour
There is a wealth of experimental and observational data on intersubjective behaviour in human infants. According to the received view intersubjectivity and intentional communication are intimately related, and in contemporary research, intersubjectivity is one of the most commonly cited triggering factors of nonverbal reference. Despite this, recent data about mainly referential and gaze-related behaviour in young infants cast doubt on the received view by showing that quite a few intersubjective skills emerge several weeks, even months, before the capacity for nonverbal reference. First, point following: 7-month-olds have been reported to follow point when the gesture is initiated by eye contact and a subsequent head and gaze turn towards the object (Striano and Bertin 2005). This is puzzling, considering that normally the pointing gesture emerges around 10 months of age, and referential pointing is produced reliably some time between 12 and 15 months. Why does intentional communication occur so late, when apparently similar capacities emerge much earlier? The fact that point following pertains to the passive role of an observer does not in itself constitute a problem for the infant in this respect, because as a rule, infants first learn to understand communicative acts, and only shortly after that to perform them (Camaioni et al. 2004). Second, gaze following: A preference for eye contact that is similar to the one that occurs in contexts of intentional communication has been observed already in 2–5-days-old babies (Farroni et al. 2002). They attend to direct, but not averted gaze. Experiments on slightly older infants indicate that eye contact in newborns may facilitate later face processing and perception of gaze direction (Farroni, Johnson and Csibra 2004). Thus, 3–4-month-olds have been observed to follow gaze-shift (D’Entremont, Hains and Muir 1997), and perceived lateral motion will cue spatial location in 4-month-olds if preceded by eye contact (Farroni et al. 2003). The cueing effect is analogous to the one that results from referential gaze reading in older infants. These data suggest that communicative-intent indicating behaviour has an early precursor in the baby’s preference for direct gaze. They both signal the sender’s attempt to communicate to the observer. In contrast, by 6 months of age, infants will match direction of gaze as signalled by mere head turn within their own visual field (D’Entremont 2000; D’Entremont et al. 1997; Morales, Mundy and Rojas 1998). By this age an initial eye contact is less important for the matching of direction than it is for younger infants. Third, gaze alternation: Around 7 or 8 months, infants start to alternate gaze in contexts of social referencing (Campos and Steinberg 1981). In ambiguous or distressing situations, they look to the adult’s face and then back at the object, seeking
126 Ingar Brinck
emotional information from the adult by looking at his or her face, and, as it seems, using it to evaluate the situation. A similar behaviour occurs in unpredictable situations, and after infants’ intentional actions (Reddy 2005), when emotional cues appear to be sought for confirmation of the infant’s evaluation of the event. The significance of these data is not fully clear. However, granted that infants engage in attention contact, follow point and line of regard, and alternate gaze several months before the onset of intentional communication, the view that all the relevant intersubjective skills emerge at a precise age seems incorrect (cf. Tomasello 1995, 1999 on the 9-month revolution). The data seem to support the opposing view that intersubjectivity develops gradually (Reddy 2005; Striano and Bertin 2005; Striano and Rochat 1999). However, infant communication does show signs of a transition from one kind of behaviour to another around 12 months, when triadic interaction becomes more important and elaborate. This suggests that while intersubjectivity is crucial for intentional communication, the presence of intersubjective skills as such does not imply skills for intentional communication.
7.
Why does nonverbal reference not emerge before 12 months?
The question is why nonverbal reference does not emerge before 12 months, since the capacity for it appears to be in place much earlier. Three answers will be considered with regard to this issue. A first answer is based in the research on socio-emotional behaviour and autism. It takes its starting-point in the claim that human infants have a natural inclination to interact with their caregivers by exchanging facial expressions of emotion, irrespective of other goals than sharing experiences (Hobson 2002; Tomasello 1999; Trevarthen 1979). Infants who go on to be diagnosed with autism behave differently. For instance, they do not show a distinct preference for visual attention contact or for pointing merely to share attention. Observed differences in the communicative behaviour of autistic and non-autistic children are taken as evidence that certain intersubjective abilities, missing in autistic children, are necessary for intentional communication. Among these we find social reciprocity, emotional relatedness, the recognition of psychological self-other equivalence (by matching of mental states), and the capacity to identify with the other (Barresi and Moore, this volume; Hobson 2002; Hobson 2005; Meltzoff and Brooks 2001). Further evidence for the view that socio-emotional factors cause the emergence of intentional communication comes from experimental research on apes that test for typically human intersubjective capacities. The tests demonstrate behavioural differences between human infants and apes that in certain respects parallel those between non-autistic and autistic
Intersubjectivity and intentional communication 127
children (cf. Tomasello 1999; Tomasello et al. 2005). Because the observed differences are correlated with differences in the capacity for intentional communication, they are held to support the view that intentional communication requires specifically human, socio-cultural capacities. A second answer relies on experimental evidence about, on the one hand, perceptual skills for reading individual goal-directed action, and on the other, referential skills for understanding action, bodily manifest emotion, and gaze as linking agent and world. These skills manifest themselves in behaviour such as gaze following and social referencing and are held to be in place by 9 to 12 months, if not before (Brooks and Meltzoff 2005; Moses et al. 2001; Woodward 2005). Csibra (2003) holds that they represent two kinds of action understanding that operate independently of each other, but rely on a similar blind tracking of perceptual cues in respectively instrumental and communicative contexts. The tracking capacity functions without the attribution of mental, intentional, or representational states. Csibra (2003: 448) further conjectures that the two skills combine into a higher-order mentalistic understanding during the second year. In focusing on capacities for dyadic emotional engagement, the first answer emphasizes the importance of primary, face-to-face intersubjectivity for engaging with others, leaving the referential relation between agent and object aside (Trevarthen and Hubley 1978). The second answer revolves around capacities for attention reading and co-ordinated joint attention that do not involve social reciprocity. It stresses the role of secondary, “side-by-side” intersubjectivity (facing the shared target) for intentionality, but neglects the importance of explicit attentional engagement for triadic relations. Nevertheless, there is no reason to think of the two as mutually exclusive; on the contrary, they are complementary. As argued in Section 5, both kinds of intersubjectivity are required for intentional communication. Communicative-intent indicating behaviour involves face-toface intersubjectivity, or a mutual sharing of experiences by attention contact, whereas referential behaviour relies on “side-by-side” intersubjectivity. Both answers provide explanations of why the early presence of intersubjective capacities does not trigger intentional communication. The first one refers to an insufficient understanding of socio-cultural practices by dyadic infants as the reason why they develop intentional communication only at 12 months. Even so, it does not specify which those precise mechanisms are that socio-cultural practises can modulate, or in what respects they are conducive to intentional communication. The second answer explains early referential behaviour as the side effect of a blind tracking of motion cues, and is detailed enough to have a bearing on the data. Still, the hypothesis that action understanding is mechanistic and
128 Ingar Brinck
automatic before the age of 1 year, and that by 2 years the automatic skills combine to yield an interpretative system, does not account for the progress from one level to the other. Tomasello (1999) provides a third, reconciliatory answer in an attempt to combine the previous two approaches. He asserts that both dimensions of intersubjectivity – the motivation to share emotions and the understanding of intentional action – are necessary for human-like cognition. In the same vein, Tomasello and Rakoczy (2003: 125) declare that “the understanding of persons as intentional agents who have a perspective on the world that can be followed into, directed, and shared” constitutes a crucial social-cognitive skill. Tomasello et al. (2005) emphasise the importance of shared (“we”) intentionality and of collaborative (as against competitive) co-operation for achieving typically human ends.10 They maintain that “dialogic cognitive representations” emerge by 14 months that integrate the first- and third-person perspectives via internalisation and allow for specifically human, collaborative, cultural practices (2005: 683f., 689). However, it is not clear that the present form of integration of perspectives would permit such practices, or exactly how it would do so. The claim that dialogic representations permit specifically human forms of collaboration trades on an ambiguity in the concept of a third-person perspective. Although infants can integrate the first- and third-person perspectives relative to a shared (physical) space, this ability will not extend to regular kinds of future-directed collaborative co-operation. Such co-operation requires representations that do not contain indexical and demonstrative components (Brinck and Gärdenfors 2003), and are not available to the infant by 14 months. Furthermore, because of the all-inclusive nature of the theory, the significance of dialogic cognitive representations for the onset of intentional communication is uncertain. None of the answers fully explain why global capacities for intentional communication emerge by 12–15 months and not before. A satisfactory account would have to be explicit about the details. Nonetheless, recognising the continuous development of nonverbal reference is important for identifying changes in intersubjective behaviour that may be relevant. Next, we return to the data presented in Section 6 to search for a pattern that might indicate the direction in which to pursue the inquiry.
10. Brinck and Gärdenfors (2003) express a similar view. They describe collaborative co-operation as directed at future goals that are unrelated to the context of (inter)action. Sharing intentions about future goals requires the capacity for sharing representational content that does not depend on the actual context. In contrast, competitive co-operation concerns goals that exist as resources in the shared environment and sometimes appear in the context of (inter)action.
8.
Intersubjectivity and intentional communication 129
The effect of direct gaze and eye contact on gaze following
Direct gaze and eye contact have been seen to have a strong impact on the capacity for intersubjective behaviour. 4-month-olds follow gaze if it is preceded by eye contact, 7-month-olds alternate gaze in contexts of social referencing, and follow point if initiated by eye contact and a subsequent head turn towards the target. Gaze will be perceived differently and provoke other reactions if initially oriented towards the infant’s attention, than if directly oriented towards a target of action. Eye contact prompts a referential reading (in the direction of the target) of behaviour in the infant, first of averted gaze, and a few months later of the pointing gesture. Do these data support the hypothesis that infants have capacities for intentional communication already by this age? It will be argued that the answer is negative, and that the referential reading is an effect of contextual enhancement. The context in which referential, gaze-related behaviour is performed clearly plays a decisive role for how the young infant interprets the behaviour. Together, the physical layout and the sender’s behaviour simultaneously bolster and determine the infant’s response, which makes it likely that the mechanism that produces the response is a property of the local environment rather than of the infant. Consequently, the referential response to gaze requires an increased awareness of the layout of the physical context, but does not imply that the infant has acquired similar capacities to those demanded for nonverbal reference. On the contrary, to benefit from the positive effects of eye contact on the capacity for reading behaviour referentially, further cognitive capacities than those acquired during the first months in life are not necessary. Seen from an evolutionary point of view, this is not surprising, since recycling is nature’s way of increasing performance while minimizing the costs. Gaze plays a vital role for initiating and maintaining dyadic engagement in early infancy. Such engagement in turn underlies the understanding of attention, the experience of self-other equivalence, and intersubjectivity (see Section 2). Given that early infant-adult interaction is centred on attention contact, it comes as no surprise that the adult’s attempt to establish eye contact a few months later still will signal an invitation to interact to the infant, who as a consequence will expect the adult’s ensuing action to be of concern to the self. In summary, gaze has a particular social function for the infant in episodes of prolonged eye contact that most probably originates from dyadic engagements in early infancy. By 3 or 4 months, infants are capable of recognising the social function of gaze in contexts of joint attention. The information that the infant picks up from having had eye contact with the adult concerns the adult’s readiness to interact with the infant. Direct gaze has an imperative dimension that directs the infant to respond to the adult’s action and follow his or her gaze in the direction of a target.
130 Ingar Brinck
The fact that eye contact enhances the infant’s capacity for the referential reading of behaviour strengthens the view that intersubjectivity is a major driving force in the development of intentional communication. By 6 months, eye contact loses its importance for gaze following. That this behaviour changes its function indicates that at this stage of development direct gaze is primarily supportive of other behaviour, and cognitive rather than genuinely communicative. Thus, in Section 4, it was argued that the functions of the four basic acts of intentional communication are stable and do not vary between contexts or users. Because it is unnecessary to negotiate the meaning of these acts, more effort can be put into producing and responding to the message. Furthermore, the expression of an act can be adjusted to the context without a change in meaning. In contrast to the behaviour discussed above, it is uncertain that the intersubjective skills that occur in contexts of social referencing predict capacities for intentional communication. The nature of social referencing makes it unlikely that the infant by this age would use referential skills, although such skills are available. Moses et al. (2001) submit that there is evidence for referential understanding in the emotions domain by 12 months. Yet infants begin to seek information from adults about distal objects already by 7 months by alternating gaze between object and adult. This behaviour might be taken to demonstrate spontaneous referential behaviour already by this age. However, it is not very probable that gaze alternation is referential in social referencing. First, it would mean that the capacity for nonverbal reference would be functional in an isolated domain for a long period, several months, before it was generalised, which is atypical. Second, the significant information to the infant concerns the mother’s attitude to the object, not the object itself. The infant is seeking information about a target that has been individuated before social referencing is initiated and that moreover motivates initiating it. Therefore the infant is unlikely to attend to the referential relation between the adult’s facial expression and the target, but rather attends to how the target is affecting the adult emotionally. Third, on the assumption that this information is available by a from a cognitive point of view cheaper strategy than the referential one, the referential interpretation of social referencing is gratuitous. An alternative, more plausible interpretation of social referencing is that the behaviour ultimately relies on learning about the function of facial expressions of emotion during early dyadic engagement. Eventually this knowledge generalises to the adult’s responses to events in the surroundings. The infant knows that facial expression signals the observer’s reaction to a salient target. When perceiving a salient object of uncertain value, information about what to do can be found by looking to the adult’s face. In all likelihood, gaze alternation here indicates that the infant is monitoring the situation. Even though intersubjectivity can enable behaviour typical for intentional communication, evidently it does not do
Intersubjectivity and intentional communication 131
so automatically, but only in certain kinds of contexts. To young infants, reading behaviour referentially is demanding. Since there is no need for it in social referencing, the cost would hardly be motivated. To conclude this section, the examination of the data has established that some intersubjective behaviour constitutes precursors to intentional communication, but that intersubjective skills nevertheless do not entail skills for intentional communication. The fact that referential gaze will acquire different functions in different contexts of action depending on whether eye contact is involved makes it interesting to pursue the analysis of intersubjectivity in a new direction. Below the hypothesis will be investigated that the influence of intersubjectivity on the capacity for intentional communication is not uniform, but differs according to which kind of experience a given behaviour concerns.
9.
Interaffectivity, interattentionality, interintentionality
Stern’s (1985) distinction between interaffectivity, interattentionality, and interintentionality will constitute the basis for developing the proposal that not merely how experiences are shared (mutually or individually, in dyadic or triadic behaviour) matters for the developmental trajectory of intentional communication, but what kind of experiences also matters. It will be argued that the three forms of intersubjectivity enable distinct forms of behaviour; thus, referential behaviour depends on interattentionality, and communicative-intent indicators on interaffectivity. Additional data on gaze-related and referential behaviour will be discussed with the double purpose of illustrating and testing the adequacy of the hypothesis. The distinction between emotional, intentional, and attentional mental states corresponds to the traditional classification of mental acts in terms of affective, cognitive, or conative function. The analogy reveals the major ways in which a sharing of experiences may contribute to over-all behaviour. Affect concerns the experience and expression of emotion and is associated with positive or negative attitude and evaluation; cognition concerns the monitoring and control of intentional states from a general, sometimes a meta-, perspective; and conation concerns selective attention, interest, and action readiness with regard to a concrete context. Depending on which kind of experiences the agents are sharing, the interaction will be predominantly interaffective, interattentional, or interintentional. Stern (1985) describes the three forms of intersubjectivity as follows (italics added): Interaffectivity consists in the infant’s matching its own “feeling state as experienced within” with the feeling state “seen ‘on’ or ‘in’ another” (1985: 132), and still by 12 months, affective exchange is “the predominant mode and substance of communications” (ibid: 133). Interattentionality means that the infant has “some
132 Ingar Brinck
sense that” persons, including the infant itself, can have individual, different, attentional foci, which can be “brought into alignment and shared” (ibid: 130). Interintentionality “impl[ies] that the infant attributes an internal, mental state” to the adult, namely, “comprehension of the infant’s intention and the capacity to intend to satisfy that intention”. Intention is a shareable, but not necessarily selfaware, experience (ibid: 131). The above definitions involve assumptions about the qualitative nature of intersubjective states, which make them problematic for the present inquiry into the nature of intersubjectivity. Because the definitions focus on qualitative experiences instead of observable behaviour, they cannot be used to identify intersubjective behaviour except from the first-person perspective. This may reduce their value for some types of empirical and experimental research. To avoid these problems while preserving Stern’s intuitions, the concepts will be redefined as follows:11
Interaffectivity is the simultaneous matching of affects and emotions to the affects and emotions displayed by another agent in overt behaviour. It emerges in the guise of emotional contagion, and the infant soon develops skills for monitoring the emotions of self and other. Interaffectivity is dyadic, and underlies behaviour such as proto-conversation, social referencing, and attention contact. Interattentionality is the alignment of attention to the attention displayed by another agent in overt behaviour. It can equally well be a result of contagion as of purposive behaviour. Attentional states spring from arousal, and reflect the agent’s interest and action readiness with respect to a target of action. Therefore, interattentionality only makes sense relative to a context of action, although it is not always explicitly organised around a shared focus of attention or third entity. It supports behaviour such as gaze alternation, gaze following, attention reading, and forms of joint attention that do not require a mutual sharing of experiences. Interintentionality is the sharing of information with another agent about the intentions and beliefs of the self and others, first by ostensive, bodily-based means, such as gaze, gesture, and vocalisation, later in development by symbolic means, for instance, verbally. It is essentially triadic and permits taking an allocentric or de-centred perspective.
11. The descriptions of the three forms of intersubjectivity are not intended to reveal the whole truth about intersubjectivity, nor do they constitute an attempt to replace phenomenological descriptions.
Intersubjectivity and intentional communication 133
These three forms of intersubjectivity are not intrinsically related, and can develop independently of each other. During normal conditions, they interact in different ways at different stages to enable complex behaviour. The infant begins to explore interaffectivity soon after birth in mutual engagement, and later interattentionality, whereas the capacity for interintentionality emerges some time after 12 months and continues to develop for years to come. Irregularities in their respective developmental trajectories give rise to weaknesses or disturbances in an agent’s action repertoire relative to the range of behaviour that the failing form of intersubjectivity normally supports. Data about gaze reading will now be used to exemplify how different forms of intersubjectivity sustain behaviour, and illustrate how the interaction between interattentionality and interaffectivity causes developmental changes in intersubjective behaviour during the infant’s first 6 months in life. By four months, infants react differentially to perceived motion depending on the triggering conditions (Farroni et al. 2000, 2003). Cue-driven saccades are elicited when the model’s pupils shift from a central position to either side, and occur before the appearance of the target is recorded. The pupils do not have to occur in the context of an upright face to have this effect. Target-driven saccades, in the direction of the target, are elicited only after a period of eye contact with an upright face. The infant will then react equally to directed movement of either head or eye gaze (pupil shift). Both responses depend on the capacity for interattentionality, but occur by different processing. Similarly to the data presented in Section 8, the data on the effect of triggering conditions on response demonstrate that from quite early on infants intuitively read gaze referentially in case the gaze shift is preceded by eye contact. Eye contact has a special role for social interaction, appreciated already in the first month, which shows that mutual attention appeals to other ways of behaving than do contingent perceptual cues (Farroni et al. 2000; Farroni et al. 2002; cf. Emory 2000). In view of the previous discussion, the social function of gaze can be identified with reference to interaffectivity. Thus, the fact that eye contact produces a mutual sharing of affect would cause the infant to pursue the interaction. However, eye contact will also cancel cue-driven gaze shift, because only reflexive responses to unidentified stimuli are cue-driven.12 Consequently, eye contact might instead cause target-driven gaze shift indirectly, by eliminating the 12. Gaze is re-oriented automatically by salient, contingent behaviour, which trigger quick and effortless, cued gaze shifts (Driver et al. 1999; cf. Chawarska, Klin and Volkmar 2003). Because the shift occurs while the behaviour is processed in the ‘early vision’ or pre-attention system of the brain that is unavailable to conscious awareness, the subject cannot access the process to directly intervene and inhibit behaviour matching.
134 Ingar Brinck
cue-driven alternative. In spite of this, it is attractive to explain the enhancing role of eye contact for referential gaze reading as an effect of interaffectivity, because eye contact later returns in a similar role and again results in a referential reading, but of other actions, such as pointing. Therefore interaffectivity seems to be the cause of the referential response, rather than for this response to occur merely because the cue-driven gaze shift was cancelled. By 6 months, the infant begins to interpret gaze as goal-directed even in the absence of eye contact. This behaviour indicates a developmental generalisation of the response to gaze to target-driven saccades. The infant now can read any gaze referentially, and the referential reading will replace the earlier cue-driven reaction to gaze shift in most contexts, although cue-driven saccades remain operative in reflexive behaviour such as contagious attention (also in the adult). Hence, interaffectivity has stopped being a factor in gaze following, although direct gaze retains its social function. Slightly later, eye contact with the adult will prompt the infant to follow pointing. Another developmental generalisation has occurred, of the infant’s response to direct gaze to new forms of communicative behaviour. Eye contact can now be used to disambiguate novel types of interaction, and thus enhances the general capacity for intentional communication. To summarize, shortly after 6 months the infant can discriminate between the referential and social uses of gaze. This means that several months before the onset of intentional communication, two mechanisms for reading gaze have emerged – an instrumental one for intentions to act and a communicative one for intentions to interact. The instrumental mechanism attributes a referential function to the gaze of other agents, by which gaze indicates the sender’s target of attention and current interest. The communicative mechanism causes direct gaze to signal communicative intent to the observer, and enhances the observer’s attention to the other agent. The influence of direct gaze on social interaction suggests that intersubjectivity can be an instrument for meta-cognition. Interaffectivity enables an understanding of communicative intent-indicating behaviour by relating the agents in the mutual exchange of emotions from a second person-perspective. Interattentionality enables referential behaviour by causing a target-driven, still unattended and implicit, attention shift in the observer of the behaviour, which reflects the understanding of action as goal-directed. In mutual engagement, interaffectivity will introduce interattentionality into the monitoring of the interaction by drawing the agents’ attention to their mutual affective experiences, and then to the fact that they are mutually attending to each other. Consequently, interaffectivity and interattentionality together are able to dynamically increase the agents’ control over the on-going interaction, making the interaction mutually transparent.
Intersubjectivity and intentional communication 135
The chapter so far has investigated and expanded on the major elements of intersubjectivity in infants aged up to about 9 months. It remains to determine the triggering factors of the over-all capacity for intentional communication. In the following section, it is claimed that by 12 months of age, infants start to familiarize themselves with a few of the general skills that interintentionality leads to, and attempt to disengage from the previously mandatory egocentric and situated point of view. This cognitive change characterises the decontextualisation of communicative skills, which is necessary for the complete mastery of intentional communication.
10. The onset of intentional communication Some time between 12 and 15 months, infants begin to reliably produce and understand referential pointing. This behaviour is usually taken as a sign of the capacity for intentional communication, because it shows that the infant’s grasp of the gesture is consistent and bi-directional. It is hard to say which those single factors are that make it possible for the infant to access intentional communication at this very moment, because by this age, behaviour is quite complex and results from the interaction between several underlying capacities. Moreover it is enhanced by environmental and interactional scaffolding.13 It seems clear, though, that one factor is distinctive of the period: the decontextualisation of both communicative and cognitive skills (Bates et al. 1979). Context-independent cognitive skills start developing, and interintentionality will slowly mesh with the two existing forms of intersubjectivity. This will radically enhance the infant’s ability to control the environment. Two abilities are especially important for the development of nonverbal reference. First, there is the ability to distinguish instrumental, goal-directed intentionality that is ‘pulled’ by the context of action from communicative, intention-guided intentionality that the agent ‘pushes’ into the world. The infant must recognise the difference between goal-directed and intention-guided actions to fully appreciate the reference relation, in referential pointing as in verbal language. Second, there is the meta-cognitive ability to regulate behaviour with respect to a general principle or rule of action. By 15 months, most infants can both produce and understand acts of nonverbal reference, although they do not achieve this by a similar mechanism for reading speaker intention as to adults. They still exploit physical-functional action 13. See Brinck (2007) for the constitutive role of environmental scaffolding for the evolution of cognition.
136 Ingar Brinck
properties and contextual cues to understand gaze, point, and reach alike (Sodian and Thoermer 2001; Thoermer and Sodian 2004). The fact that gaze following now may be driven by any contextual features that happen to support it on a given occasion shows that infants by this age do not yet conceive of the sender-object relation as determined by the sender’s intention. Still, their use is reliable and appears consistent to an observer. Apparently they have acquired a general method for mastering referential acts, which in most contexts is good enough. They can use a range of referential behaviour in different combinations, in a variety of contexts, and during different conditions. The major requirement for having this competence is the ability to detach communicative behaviour from the original context in which it was learnt, and then to extend its use to an open-ended number of contexts. This means that understanding that the sender’s intention determines the reference relation is less important for mastering referential behaviour than is the ability to generalise. Decontextualisation is a gradual process that proceeds in what might seem like a random manner. In previous sections it has been argued that interaffectivity and interattentionality contribute to a general understanding of behaviour by increasing the infant’s awareness of social interaction and the process of communication, and also by providing the instruments for managing this process in particular contexts. Accordingly, decontextualisation concerns changes in on the one hand the infant’s global understanding of communication, and on the other, concrete contexts of interaction. It has several aspects. Behaviour can be extended to new kinds of contexts, while keeping the original function; it can be generalised to cover new situations, which will produce a change in the original function; it may be detached from any specific context of use by abstraction, losing exactitude while becoming inclusive; and it may be idealised, and changed into a general behaviour-guiding principle that controls certain contexts of (inter)action. This will transfer the behaviour to the metacognitive domain. Although robust skills for intersubjective behaviour occur already by 6 months, they are only intermittently available to the infant, because they are linked to particular contexts and patterns of (inter)action, usually the ones in which they were learned. Much of the behaviour shows traces of ritualisation, in contrast to intentionally communicative behaviour that characteristically is decontextualised (Brinck 2001; Brinck 2003). What from a general perspective appears to be identical, equally demanding contexts may still be handled in very different ways by the same infant, because behavioural competence is not a simple matter of development, but depends on the local affordances and constraints. To optimize performance, behaviour is fine-tuned to local properties – physical and
Intersubjectivity and intentional communication 137
functional as well as social ones. As a consequence, in one and the same infant one may find quite different behaviours in two related situations. In Section 9, it was argued that the sharing of experiences concerns affect, attention, or intention, and that each of these kinds of experience enables certain behaviour. This view suggests that communication depends on intersubjectivity in quite specific ways. Interaffectivity and interattentionality are operative soon after birth and support the development of communication during the first year, while skills for interintentionality that sustain complex and general forms of intentional communication develop later. Thus separate intersubjective behaviours first develop in parallel, but after a few months start to interact. Decontextualisation begins towards the end of the first year. Eventually the many manifestations of intersubjectivity organise themselves into the four functional units (the basic acts) that are constitutive of intentional communication (see Section 4). In contrast to views that describe the development of intentional communication as clear-cut, linear, and stage-like, the present one has similarities with both Woodward’s (2005: 124) and Reddy’s (2005). Woodward attests that infants “accrue knowledge about particular actions gradually during the first year of life”, while Reddy argues that the development of attention is continuous and proceeds from engagement. Woodward’s findings suggest that infants initially encode actions at a detailed level, which causes them to respond to similar actions in different ways. Woodward (2005: 125) further proposes that there are “varied developmental relations” between different levels of responses, a statement that is in agreement with the present view. Nevertheless, a systematic account of the development of intentional communication is within reach. Although intersubjectivity is complex and its development follows variable trajectories, its progress is continuous. Mapping it out not only provides for tying the emergence of various behaviours to specific periods in time and understanding how they interact, but also for elucidating the general principles behind the development of intersubjectivity as well as of intentional communication.
Acknowledgements Work on this chapter was gracefully supported by The Swedish Research Council and has profited from research conducted within the project Stages in the Evolution and Development of Sign Use (SEDSU), funded by the European Commission under the FP6 programme. I am grateful to Jordan Zlatev, Tim Racine, and Chris Sinha for comments on earlier versions of the chapter.
138 Ingar Brinck
References Austin, J.L. 1962. How to Do Things with Words. Oxford: Oxford University Press. Bard, K. 1992. “Intentional behaviour and intentional communication in young free-ranging Orangutans.” Child Development 63: 1186–1197. Barresi and Moore, this volume. “The neuroscience of social understanding.” Bates, E. (ed.). 1976. Language and Context. The Acquisition of Pragmatics. New York: Academic Press. Bates, E., Benigni, L., Bretherton, I., Camaioni, L. and Volterra, V. 1979. The Emergence of Symbols. New York: Academic Press. Brinck, I. 2001. “Attention and the evolution of intentional communication.” Pragmatics and Cognition 9 (2): 255–272. Brinck, I. 2003. “The pragmatics of imperative and declarative pointing.” Cognitive Science Quarterly 3 (4): 429–446. Brinck, I. 2004. “Joint attention, triangulation and radical interpretation: A problem and its solution.” Dialectica 58 (2): 179–205. Brinck, I. 2007. “Situated cognition, dynamic systems, and art. On artistic creativity and aesthetic experience.” JanusHead 9 (2): 407–431. Brinck, I. and Gärdenfors, P. 2003. “Co-operation and communication in apes and humans.” Mind and Language 18 (5): 484–501. Brooks, R. and Meltzoff, A.N. 2002. “The importance of eyes: How infants interpret adult looking behaviour.” Developmental Psychology 38 (6): 958–966. Brooks, R. and Meltzoff, A.N. 2005. “The development of gaze following and its relation to language.” Developmental Science. 8 (6): 535–543. Bühler, K. 1934. Rpt. Theory of Language. The Representational Function of Language. Amsterdam: John Benjamins. Camaioni, L., Perucchini, P., Bellagamba, F. and Colonnesi, C. 2004. “The role of declarative pointing in developing a theory of mind.” Infancy 5 (3): 291–308. Campos, J.J. and Steinberg, C.R. 1981. “Perception, appraisal, and emotion: The onset of social referencing.” In Infant Social Cognition: Empirical and Theoretical Considerations, M.E. Lamb and L.R. Sherrod (eds.), 273–314. Hillsdale, N.J.: Erlbaum. Chawarska, K., Klin, A. and Volkmar, F. 2003. “Automatic attention cueing through eye movement in 2-year-old children with autism.” Child Development 74 (4): 1108–1122. Csibra, G. 2003. “Teleological and referential understanding of action in infancy.” Philos. Trans. R Soc. B Biol. Sci. 29: 447–458. Decety, J. and Ingvar, D.H. 1990. “Brain structures participating in mental simulation of motor behavior: A neuropsychological interpretation.” Acta Psychologica 73: 13–24. D’Entremont, B. 2000. “A perceptual-attentional explanation of gaze following in 3- to 6months-olds.” Developmental Science 3: 302–311. D’Entremont, B., Hains, S.M.J., and Muir, D.W. 1997. “A demonstration of gaze following in 3- to 6-month-olds.” Infant Behavior and Development 20: 569–572. Driver J., Davis G., Ricciardelli P., Kidd P., Maxwell E. and Baron-Cohen S. 1999. “Gaze perception triggers reflexive visuospatial orienting.” Visual Cognition 6 (5): 509–54. Emory, N.J. 2000. “The eyes have it: The neuroethology, function and evolution of social gaze.” Neuroscience and Biobehavioral Reviews 24: 581–604.
Intersubjectivity and intentional communication 139
Farroni, T., Johnson, M.H., Brockbank, M. and Simion, F. 2000. “Infants’ use of gaze direction to cue attention: The importance of perceived motion.” Visual Cognition 7 (6): 705–718. Farroni, T., Csibra, G., Simion, F. and Johnson, M.H. 2002. “Eye contact detection in humans from birth.” PNAS 99 (14): 9602–9605. Farroni, T., Mansfield, E.M., Lai, C. and Johnson, M.H. 2003. “Infants perceiving and acting on the eyes: Tests of an evolutionary hypothesis.” Journal of Experimental Child Psychology 85: 199–212. Farroni, T., Johnson, M.H. and Csibra, G. 2004. “Mechanisms of eye gaze perception during infancy.” Journal of Cognitive Neuroscience 16 (8): 1320–1326. Gallese, V., Keysers, C., and Rizzolatti, G. 2004. “A unifying view of the basis of social cognition.” Trends in Cognitive Sciences 8 (9): 396–403. Goldman, A. 2006. Simulating Minds: The Philosophy, Psychology and Neuroscience of Mindreading, New York: Oxford University Press. Hobson, R.P. 2002. The Cradle of Thought. London: Pan Macmillan. Hobson, R.P. 2005. “What puts jointness into joint attention?” In Joint Attention, Communication, and Other Minds, N. Eilan, C. Hoerl, T. McCormack and J. Roessler (eds.), 185–204. Oxford: Oxford University Press. Jeannerod, M. 1997. The Cognitive Neuroscience of Action. Oxford: Blackwell Publ. Leavens, D.A., Hopkins, W.D., and Thomas, R.K. 2004. “Referential communication by chimpanzees (Pan troglodytes).” Journal of Comparative Psychology 118 (1): 48–57. Leavens, D.A., Russell, J.L., and Hopkins, W.D. 2005. “Intentionality as measured in the persistence and elaboration of communication by chimpanzees (Pan troglodytes).” Child Development 76: 291–30. Leavens, D.A., Hopkins, W.D. and Bard, K. this volume. “The heterochronic origins of explicit reference.” Meltzoff, A.N. and Brooks, R. 2001. ““Like me” as a building block for understanding other minds: Bodily acts, attention, and intention.” In Intentions and Intentionality. Foundations of Social Science, B. Malle, L.J. Moses and D.A. Baldwin (eds.), 171–191. Cambridge, Mass.: MIT Press. Morales, M., Mundy, P. and Rojas, J. 1998. “Gaze following and language development in sixmonth-olds.” Infant Behavior and Development 21: 373–377. Moses, L.J., Baldwin, D.A., Rosicky, J.G. and Tidball, G. 2001. “Evidence for referential understanding in the emotions domain at twelve and eighteen months.” Child Development 72 (3): 718 –735. Reddy, V. 2003. “On being the object of attention: Implications for self-other consciousness.” Trends in Cognitive Sciences 7 (9): 397–402. Reddy, V. 2005. “Before the third element.” In Joint Attention, Communication, and Other Minds, N. Eilan, C. Hoerl, T. McCormack and J. Roessler (eds.), 85–109. Oxford: Oxford University Press. Rizzolatti, G., Fadiga, L., Fogazzi, L. and Gallese, V. 2002. “From mirror neurons to imitation: facts and speculations.” In The Imitative Mind, A.N. Meltzoff and W. Prinz (eds.), 247–266. Cambridge: Cambridge University Press. Sodian, B. and Thoermer, C. 2004. “Infants’ understanding of looking, pointing, and reaching as cues to goal-directed action.” Journal of Cognition and Development. 5 (3), 289–316. Stern, D.N. 1985. The Interpersonal World of the Infant. New York, NY: Basic Books.
140 Ingar Brinck
Striano, T. and Bertin, E. 2005. “Social-cognitive skills between 5 and 10 months of age.” British Journal of Developmental Psychology 23: 1–11. Striano, T. and Rochat, P. 1999. “Developmental links between dyadic and triadic social competence in infancy.” British Journal of Developmental Psychology 17: 551–562. Thoermer, C. and Sodian, B. 2001. “Preverbal infants’ understanding of referential gestures.” First Language 21: 245–264. Tomasello, M. 1998. “Reference: Intending that others jointly attend.” Pragmatics and Cognition, 6 (1/2): 229–243. Tomasello, M. 1999. The Cultural Origins of Human Cognition. Cambridge, MA: Harvard University Press. Tomasello, M. and Call J. 1997. Primate Cognition. New York: Oxford University Press. Tomasello, M., Carpenter, M., Call, J., Behne, T., and Moll, H. 2005. “Understanding and sharing intentions: The origins of cultural cognition.” Behavioral and Brain Sciences 28: 675–735. Tomasello, M. and Rakoczy, H. 2003. “What makes human cognition unique? From individual to collective shared intentionality.” Mind and Language 18 (2): 121–147. Trevarthen, C. 1979. “Communication and cooperation in early infancy: A description of primary intersubjectivity.” In Before Speech, M. Bullowa (ed.), 321–347. Cambridge: Cambridge University Press. Trevarthen, C. and Aitken, K. 2001. “Infant intersubjectivity: Research, theory and clinical applications.” Journal of Child Psychology and Psychiatry 42 (1): 3–48. Trevarthen, C. and Hubley, P. 1978. “Secondary intersubjectivity: Confidence, confiding, and acts of meaning in the first year.” In Action, Gesture, and Symbol: The Emergence of Language, A. Lock (ed.), 183–229. New York: Academic Press. Wicker, B., Keysers, C., Plailly, J., Royet, J.-P., Gallese, V., and Rizzolatti, G. 2003. “Both of us disgusted in my insula: The common neural basis of seeing and feeling disgust.” Neuron 40: 655–664. Woodward, A.L. 2003. “Infants’ developing understanding of the link between looker and object.” Developmental Science 6: 297–311. Woodward, A.L. 2005. “Infants’ understanding of the actions involved in joint attention.” In Joint Attention, Communication, and Other Minds, N. Eilan, C. Hoerl, T. McCormack and J. Roessler (eds.), 85–109. Oxford: Oxford University Press.
chapter 7
Sharing mental states Causal and definitional issues in intersubjectivity Noah Susswein and Timothy P. Racine In this chapter we analyse ‘intersubjectivity’ and related psychological concepts. We focus on distinguishing between causal and definitional issues in early social development, between categorical explanations of what an organism is doing and causal explanations of how or why it is doing it. We argue that intersubjectivity is a taxonomic rather than a causal explanatory concept, a technical concept used to classify interactive behaviours and abilities rather than to denote vehicles or causes of those behaviours and abilities. We begin by examining the idea that intersubjective engagement involves the sharing of mental states and argue that the role of mental states and experience in intersubjective engagement is misconstrued. In the final sections we consider the meaning of declarative pointing.
1.
Introduction
Trevarthen’s (1977, 1979; Trevarthen and Hubley 1978) landmark research on early infant social development explicitly focussed the attention of developmental psychologists on the issue of intersubjectivity. In longitudinal observations of the first few months of life, Trevarthen showed that even 2-month-old infants have richly coordinated and complex interactions with their caregivers and respond differently to persons than objects. He described these early activities “in terms of mutual intentionality and sharing of mental state” (Trevarthen 1977: 228). Based on this elegant and careful research, Trevarthen proposed an innatist theory to explain this primary intersubjectivity in which he argued, “the infant is born with awareness specifically receptive to subjective states in other persons” (Trevarthen and Aitken 2001: 4). Intersubjectivity on this view is meant to denote a biologically specified psychological capacity that is a causal precondition for early human social interaction.
142 Noah Susswein and Timothy P. Racine
Infants’ behaviour towards animate and inanimate objects is strikingly different early in life and their interactions with others become increasingly coordinated and complex. This coordinated activity involves mutual sensitivity to one another’s emotional and attentional states, forms of activity that seem lacking in non-human primates (Tomasello and Carpenter 2005) and which are truncated in children with autism (Baron-Cohen 1995; Hobson and Hobson this volume). But the degree to which infants’ discriminatory and interactive abilities and preferences can be explained in terms of their experience is unclear, although infants’ motivation to interact may be partly explained by the fact that they experience pleasure while interacting. Thus, although there can be no doubt about the importance of Trevarthen’s findings, there can be doubt about his interpretation of those findings. Trevarthen’s is just one characterization of intersubjectivity by developmental psychologists. Reimers and Fogel (1992: 82), for example, conceive of intersubjectivity as “a shared understanding of what an interaction is about.” This argument turns in part on what exactly Reimers and Fogel mean by ‘understand’ (Racine and Carpendale 2007b, 2007c). This definition could be read as meaning that intersubjectivity is not a vehicle of interactive abilities, but that it is manifest in and inseparable from interaction. This emphasises the logical connections between understanding and doing, whereas Trevarthen’s use of ‘awareness’ seems to locate understanding in the experience of persons able to do such and such. Babies discriminate between persons and objects early in life, preferentially attend to faces, manifest pleasure in interaction, imitate, play, etc. In short, they do all of things that count as ‘intersubjective engagement’. However, conceiving of intersubjectivity as denoting a psychological vehicle of infants’ interactive abilities is problematic, conflating causal and definitional relations. And the claim of innate intersubjectivity goes beyond the claim that human biology is some sort of precondition for human social life. Specifying the unique structural properties of a life form is an important part of understanding how it is able to do what it does; e.g., the eye, optic nerve, visual cortex, etc. are vehicles of our ability to see. A causal explanation of visual perceptive abilities would involve specifying how processes in these vehicles interact with specified features of the environment in the process of seeing. It is not clear that ‘intersubjectivity’ similarly refers to vehicles of our interactive abilities, although it may appear to do so. In this chapter, we explore the distinction between causal and definitional issues in early social development. We distinguish between conceptual explanations of what an organism is a doing and causal explanations of how they are doing it (Dupré 1993). We argue that intersubjectivity is a refinement of our commonsense notion of ‘interaction’, specifying a form of interaction that is chronologically primary and developmentally significant in that its absence typically foreshadows
Sharing mental states 143
s erious disruption in a person’s social cognitive and emotional functioning. Rather than answering a how question by denoting a vehicle of interactive abilities, we argue that intersubjectivity is a taxonomic concept that speaks to questions of what an organism is capable of doing. In the first sections we examine the familiar idea that intersubjective engagement involves the sharing of mental states.
2.
Sharing mental states
Intersubjectivity appears relatively well understood in the sense that we know it when we see it. People typically agree on whether two (or more) agents are doing something together or are just near one another by happenstance. Echoing Trevarthen, intersubjectivity is “understood widely as involving matched or shared mental states between or among individuals” (Barresi and Moore this volume; see also Gallagher and Hutto this volume; Hobson and Hobson this volume). We think this definition is unclear and incomplete. What exactly does it mean to share a mental state with another? Does it mean something in addition to coordinating one’s actions in shared activities with others? Let us first take sharing a mental state to mean having the same or similar thoughts or feelings. We assume that no one would claim that 2-month-olds and their caregivers typically literally think and feel the same things during protoconversations. For example, a caregiver can experience feelings of aching pride and can see a resemblance between the infant and her father, but infants cannot. So we cannot define intersubjectivity as an interaction involving two agents having the same thoughts and feelings. However, dramatic differences in emotional states between infants and caregivers might be treated as clear criterion for a lack of intersubjective engagement. For example, when a caregiver tries to soothe a wailing infant, they might be said to be interacting without being intersubjectively engaged. So perhaps episodes of intersubjective engagement are partially characterized by a global similarity in mental state, by two parties experiencing some form of pleasure. However, two cheerful strangers standing side by side on an elevator would satisfy this condition, so this definition is obviously too inclusive. It is misleading to define intersubjective engagement in terms of shared mental states, if it is also necessary that two agents be coordinating their activities with one another. Perhaps both the interactional requirement and the emphasis on mental states can be met by stipulating that parents and infants must share a common attentional state for intersubjectivity to occur. However, as with the cheerful elevator passengers above, a distinction is often made in the research literature between passive coordinated attention where agents might be looking at the same state of affairs by happenstance versus both agents being aware that their attention is
144 Noah Susswein and Timothy P. Racine
coordinated (e.g., Bakeman and Adamson 1984). So even in clear cases of agents sharing an attentional state by attending to the same stimulus, it is necessary to stipulate that they be attending to the stimulus together if activity is to count as intersubjective engagement. From a third person perspective, infant and caregiver might be described as sharing the state of ‘attending to each other’ (i.e. A and B are intersubjectively engaged just in case A is attending to B while B is attending to A). This liberal definition of ‘shared attentional states’ focuses on coordinated activity rather than two actors having the same or similar experiences. That is, knowing that two actors are attending to one another may not entail knowing what it is like to be behind the eyes of either of the pair. So there are reasons to question the assumption that intersubjectivity involves ‘shared mental states’ even in the sense of having similar thoughts, feelings, or perceptual experiences, although it may require that interacting parties both manifest a degree of pleasure. There does seem to be a role for ‘shared mental states’ in considerations of infant intersubjectivity, but a more limited role than is widely assumed. Perhaps ‘manifesting pleasure or interest in coordinated interactions’ is a more accurate description of intersubjectivity.
2.1 Understanding and experience We must distinguish between intersubjectivity as “shared mental states” and intersubjectivity as “shared understanding”. The first notion of intersubjectivity seems to focus on experience whereas the second emphasises intersubjectivity as a kind of knowledge. Because it is widely assumed that understanding is a type of mental state, it may seem like these definitions of intersubjectivity are roughly equivalent. We argue that they are quite different, and that it is very misleading to conceive of understanding as a mental state. Mental states have genuine duration, beginnings and ends and vary in intensity (Bennett and Hacker 2003; Racine 2004; Wittgenstein 1958). They are states that a person is in for some period of time, unlike dispositions or traits, which might characterize a person’s thoughts or feelings for a lifetime. The category of mental states covers a lot of mental ground, including emotional states such as states of anxiety; appetitive states, such as states of hunger or thirst; perceptual states, such as seeing an apple or hearing a piano; and attentional states, such as attending to one’ hunger or to the apple one sees on the table. This list is meant to be illustrative, not exhaustive. The category of ‘mental states’ is diverse, populous, and not sharply bounded. Nor is it sharply divided, as illustrated by the overlapping of attentional states with perceptual and appetitive states in the list above. However, not everything mental should be construed as a mental
Sharing mental states 145
state. ‘Understanding’ is one example. First, understanding, unlike a mental state, is a potentiality, not an actuality (Baker and Hacker 1984; Kenny 1989). For example, to ‘understand attention’ is, among other things, to be able to gaze follow and direct others’ attention. But one’s understanding of attention does not hinge upon continuously exercising those abilities, that is, it does not hinge upon being in a mental state of attending to others attention. Furthermore, ‘understanding’ is typically general, a family of abilities (gaze following, social referencing, and declarative pointing all index infants understanding of attention), rather than a specific ability such as the ability to pass a particular perspective taking task. To ascribe an understanding of X to a person is to characterize what they are able to do, not what they currently or continuously experience (see also Leavens, Hopkins, and Bard this volume). If an infant follows others’ gazes, observers know that this infant has some rudimentary knowledge of attention. But observers do not know what it is like for the infant to follow others’ gazes. The criteria for ascribing an understanding of attention – what counts as understanding attention – are actions, not experiences of actions. Discussing the experience of infants always risks adultocentrism, so we will use a different example to further illustrate this point. Whether a person experiences algebra as dreadfully boring or deeply satisfyingly is irrelevant to considerations of her understanding of the subject. It is the action of correctly solving for x or y and not ones experience of doing so that determines whether one understands some aspect of algebra. That said, we might expect a person who greatly enjoys algebra to be or go on to become better at it than persons who enjoy it less or not at all, due the algebra-enjoyer’s spending more time practicing, etc. So it would be false to say that persons’ abilities to V are unrelated to their experience of Ving. However, a person’s enjoyment of an activity and their skill at it are distinct, as many hobbyists know firsthand. It might be tempting to think of understanding as an experience or mental state because we do have experiences of trying to understand things, coming to understand them, realizing that we do not or only partially understand something we thought we understood, remembering or relearning something we previously understood, etc. These experiences are related to the phenomena of understanding. But they are experiences of trying, learning, realizing, remembering, and relearning, not experiences of understanding per se. Furthermore, such experiences would seem to necessarily involve attentional states, and might also involve perceptual states, emotional states, and even appetitive states, for example, coming to understand that one always gets hungry soon after eating at a particular restaurant. So there are connections between mental states and the phenomena of understanding. However, there does not appear to be any reason to regard understanding itself as a type of mental state. We develop a more positive assessment of
146 Noah Susswein and Timothy P. Racine
what understanding is in the next section by considering the new forms of understanding that are often characterized as new forms of intersubjectivity.
2.2 Understanding, intersubjectivity, and abilities Many theorists argue that the new activities that appear towards the end of the first year, such as reliable gaze following, pointing and social referencing herald a shift from a basic to a more sophisticated form of intersubjective engagement. Trevarthen and Hubley (1978) characterize this new level of engagement as secondary intersubjectivity, which Bretherton and colleagues claim is suggestive of a burgeoning “implicit theory of mind” (Bretherton 1991; Bretherton, McNew and Beeghly-Smith 1981) and which Tomasello (1995, 1999) argues may show that infants are able to experience the intentions of others as similar to or different from their own. Developmentalists often refer to the appearance of these shared activities that require simultaneous visual attention on the part of both infant and caregiver as episodes of joint attention (see chapters by Brinck and Leavens et al. this volume and in Eilan, Hoerl, McCormack and Roessler 2005; Kita 2003; Moore and Dunham 1995; for a review see Racine and Carpendale 2007c). Many accounts of joint attention treat the emergence of these skills as behavioural effects of the infants’ having new kinds of experiences or of possessing some rudimentary understanding of psychological concepts. A possibly caricatured but illustrative paraphrase of such views is that infants discover or hypothesize that people possess mental things called ‘attention’ and ‘intention’, and to use this knowledge to navigate the social world. For example, gaze following or pointing skills are sometimes conceived of as varied behavioural consequences of a general conceptual insight, as the claim that infants come to ‘represent’ other persons as ‘beings with attentional and/or intentional capacities’ (Tomasello 1995, 1999) seems to suggest. We agree that joint attention behaviours are related to new experiences and new understandings, but not that these relations are of cause and effect. First we consider the notion of new experiences. There is a clear sense in which experiences can be causally related to new understandings. For example, the experience of listening to good lecture may cause a person to understand a difficult argument. And experiencing a serious illness may lead someone to understand the value of good health and supportive family. In these clichéd examples, the experiences
. Although Tomasello’s recent revision of his theory has tempered some of his earlier claims (Tomasello, Carpenter, Call, Behne and Moll 2005), his revised theory is still based on a mentalist metaphysics (Racine and Carpendale 2007a, 2007b, 2007c).
Sharing mental states 147
precede the new understandings, in keeping with the conventional requirement that operative causes precede their effects (Mill 1843/1875). An operative cause is that which brings about a change. However, causal explanations can also involve causal preconditions, the vehicles by which a change is brought about. For example, the operative cause of hearing an infant cry would be that infant’s crying while the vehicles of hearing her cry would include the eardrum, auditory cortex, etc. We think it is misleading to regard new experiences of others’ attention as either operative causes or as causal vehicles of joint attention behaviours. The statement ‘I understand the argument because of her lecture’ bears a superficial grammatical similarity to ‘infants begin react to others’ attention because they begin to experience others’ attention’. However, in this latter case, ‘because’ does not specify a contingent, causal relation, as in ‘A understands something because B explained it’. Rather, it is more like a semantic stipulation: the fact that infants react differentially is claimed to mean that others’ attention are objects of their experience. In one sense, the emergence of joint attention behaviours or secondary intersubjectivity necessarily involves new experiences; the emergence of gaze following entails a new experience of following an others’ gaze, and the emergence of declarative pointing entails a new experience of directing others’ attention. However, such ‘experiences of others’ attention’ and ‘reactions to others’ attention’ are not perfectly correlated but conceptually related, as are being an unmarried man’ and ‘being a bachelor.’ It is not as if infants’ experience of others’ attention can be detected independent of their reactions to others’ attention (Leavens et al. this volume). Thus, the ‘experience of others attention’ cannot be detected prior to ‘reactions to others’ attention.’ So it is seems misleading to regard new experiences as operative causes of new behaviours because operative causes precede their effects. But the bigger problem is logical, not chronological, nor methodological. There is simply no way to determine what any person, infant or adult, experiences independent of observing their behaviour, or at least, some consequence of their behaviour, e.g. reading an autobiography (Bennett and Hacker 2003). To be clear, this is not to say that experiencing is behaving. A comatose person may hear or see without behaving at all. But if it is impossible to assess what infants experience independent of what they do, then it cannot be an empirical claim that new experiences lead to new forms of interactive behaviours. It is also awkward to regard new understandings as the causes of such new behaviours. Rather than a more advanced understanding of mentality causally underlying these behaviours, we think it is clearer, albeit less familiar, to say that a more advanced understanding of mentality is manifest in these behaviours. Gaze following, pointing, social referencing, object-directed imitation and so forth are among the most rudimentary behaviours that count as ‘understanding attention’.
148 Noah Susswein and Timothy P. Racine
To ‘understand attention’ is, among other things, to be able to gaze follow and direct others attention. Being able to explain what ‘attention’ means or use ‘attention’ correctly in a sentence are more advanced criteria of ‘understanding attention’. What an agent understands determines what they can do. But ‘determine’ can mean both ‘cause’ and ‘define’. A man’s marital status determines whether or not he is a bachelor; this relation is logical. If this man has such objectionable body odor as to drive away all would-be spouses, this relationship is causal. It is mistaken to regard e.g. ‘understanding of attention’ as cause and gaze following, pointing, social referencing as effects, because these behaviours define rudimentary forms of understanding attention. We list these behaviours when explaining what it means to say that infants begin to understand attention at around one year of age. Ascribing an understanding of attention to infants specifies what they are capable of doing, not how or why they do it (Dupré 1993). That is, the relationship between joint attention behaviours and ‘understanding attention’ is logical rather than causal. Dyadic interactive behaviours – manifesting pleasure or interest while interacting with another – define primary intersubjectivity. Triadic interactive behaviours define secondary intersubjectivity and ‘an understanding of attention’. Invoking a state or stage secondary intersubjectivity should not be viewed as a causal explanation of these behaviours but a categorical explanation. Rather than cause and effect, the relations between primary intersubjectivity and dyadic interactions, and between secondary intersubjectivity and triadic interaction behaviours are relations between types and tokens.
3.
Intersubjectivity as a taxonomic concept
We have said a lot about we think intersubjectivity is not. Now we elaborate on what we think it is. “Intersubjectivity’ is a theoretically motivated taxonomic concept that helps researchers distinguish different forms of interaction that unfold in early development. A taxonomy is a scientific system of classification. A general question about how children develop social understanding must be broken down into smaller parts, into questions about more specific abilities, in order to be studied empirically. In distinguishing, for example, between primary and secondary intersubjectivity social developmental theorists are defining technical concepts for the purposes of dividing development into interesting stages or domains. ‘Intersubjectivity’ does not explain how it is that human infants are motivated to play and proto-converse with others. Reliable neurological differences between, for example, children with autistic spectrum disorders and non-autistic children may help answer how questions, specifying the causal preconditions necessary
Sharing mental states 149
for doing what we call ‘intersubjectivity’. Other operative causal questions related to intersubjectivite abilities may involve asking which if any environmental factors reliably predict the early emergence of or more skilful interactive abilities. However, for any such empirical questions to be asked about social cognitive development, conceptual questions regarding what counts as social cognition and development must be specified in advance. Not all genuine explanation is causal explanation. An explanation of social understanding involves specifying what counts as ‘social understanding’. And an explanation of infant development in terms of phases of intersubjectivity explains infant development in the same sense that a species/phylum/ family/genus diagram explains the animal kingdom. That is, to discuss the emergence of ‘secondary intersubjectivity’ is to pick out a class of actions – triadic interactions involving two persons attending one another as well as to some other feature of the environment – as theoretically interesting. Where a naïve observer may see nothing more than baby’s first point, not obviously more alluring than baby’s first tooth, the developmentalist appreciates the theoretical significance of the first pointing gesture, and teaches her students to do the same. But in doing so, she is teaching a technical, taxonomic vocabulary, not a lesson in discovered causes of different behaviours. We might also ask, in ascribing an ‘understanding of attention’ or a stage of secondary intersubjectivity to infants who gaze follow, point, socially reference and so on whether developmentalists have discovered or created order. We argue that this is a case of creating order, of specifying a theoretically interesting aspect of what infants do rather than discovering how they do it. This is important because phenomena must be conceptualized in order to be investigated. However, conceptualization is a creative act. Categories do not exist independent of language even though some categories (e.g., ‘human infants’) are more natural than others (e.g., ‘things over one pound’) in the sense that knowing that X is a human infant allows one to know much more about X than knowing that X is a thing that weighs over 1 pound (Dupré 1993: 62–64). Creating as well as discovering order is an essential aspect of empirical science. But mistaking creations for discoveries is, by definition, mistaken.
3.1
Ontological diversity of the mental
Does our denial that primary and secondary intersubjectivity are causes of dyadic, and triadic behaviours respectively entail a general scepticism about mind or mental causation? It does not. However, we do believe that causal explanations are only one type of explanation in which psychological concepts appear, and that the role of causal explanations of behaviour in social understanding is often
150 Noah Susswein and Timothy P. Racine
e xaggerated. In this section we wish to explain why we think our analysis is not at all anti-realist about mind. It is widely assumed that mental states are causes of behaviours (Racine and Carpendale 2007b, 2007c). And there are cases in which it is perfectly sensible to regard a mental state as the cause of a particular behaviour, for example, when a person shudders involuntarily at the memory of a disturbing image. In a different vein, our reflective thinking often provides us with reasons for choosing one course of action over another. Thus, it would be false to claim that mental phenomena are never causes of behaviour or, more generally, that mental phenomena are somehow irrelevant to considerations of behaviour. But it is grossly overgeneralized to think that mental states are causes of all behaviours. For one thing, this conflates actions with reactions. And if we erase the distinction between action and reaction, we vitiate the concept of responsibility, the practice of apologizing, the distinction between ‘on purpose’ and, ‘by accident’, and a host of other distinctions that seem to partially define human social life. An understanding of behaviour as, in a variety of ways, purposeful is an important aspect of the very commonsense view of the mind (‘folk psychology’) that developmentalists study in nascent forms. For example, to claim that A Ved intentionally is to claim that A is responsible for his Ving to a degree that A is not if A Ved unintentionally, out of ignorance or by accident. Responsibility and causality are related but distinct. A person can be held responsible for an outcome that she only indirectly caused, e.g. hiring another person to commit a crime. We also typically hold persons responsible to some degree for the mental states of jealousy or drunkenness that may cause them to act foolishly. And we readily forgive others for causing damage to our belongings if they didn’t do it on purpose. That these are truisms is precisely our point. A common-sense understanding of action involves more than causal analyses of behaviour. It seems that the psychology of the folk is subtler than most accounts of ‘folk psychology’ would have it. A thorny feature of psychological predicates is that they are used in a variety of relatively unrelated ways (Bennett and Hacker 2003; ter Hark 1990). When subsuming mental state concepts under the superordinate category of ‘the mind’, it is easy not to notice this feature. It might seem counterintuitive or anti-realist to suggest that the ‘mentalness’ of psychological predicates is a consequence of, rather than justification for, treating ‘mental’ as relatively uniform and sharply bounded category (e.g., as opposed to ‘physical’ or ‘behavioural’). But to claim that ‘the mental’ is a heterogeneous category is not to claim that it is a senseless category (as might be, for example ‘colourless green ideas’). The mental is an ontologically diverse category. And the diversity of the mental can be illustrated by examining the use of even a single term. For example, ‘think’ can be used to issue a threat (“I think you better leave now”), to make a request (“do you think
Sharing mental states 151
you can pass me the salt?”), to issue a command (“I think you’ve had enough ice cream”), to express uncertainly (“it’s raining, I think”), to delay action (“okay, I’m going to think about that), and far less frequently to actually express subjective experiences (“I am thinking of the time that we went to Barbados”). That is, a single mental state term can be manifestly used to perform very different social acts (Austin 1975). Now, some readers may object that we are too focused here on the word ‘think’ itself, and insist that it is the nature of the phenomena of thinking, and not the utterance of the word ‘think’ that is really at issue here. We agree that it is the phenomena of thinking that is at issue. But, given that the only way to determine which phenomena count as ‘thinking’ is to examine the application of the concept, we insist that considering the use of ‘think’ is of central importance to determination of what thinking is. However, for those who are not persuaded by this argument and hold that ontological and semantic questions are entirely separate, we offer another, roughly non-linguistic example of ontological diversity of a single mental concept. Any human being can ‘intend to grab a rock’. But only a human being in a very specific social context can ‘intend to write a check’ (Dupré 1993). One cannot intend to V unless there is such a thing as Ving (unlike ‘imagining Ving’ which may mean either that such a thing as Ving exists, and one imagines oneself doing what counts as Ving, or that that one is imagining that there is such a thing as Ving, the meaning of which must be explained if one is to describe what one has imagined). It was simply not possible for even the cleverest of our Stone Age ancestors to intend to write a check, whereas even the most Neanderthal of contemporary humans can do so (provided she has a checking account). Thus, many intentions are ontologically dependent upon highly specific social structures, while other intentions (e.g. to pick up a rock) denote capacities that seem ‘universal’ (but extend far beyond the species boundary – seagulls also intend to pick up rocks). Again, the ‘mentality’ common to both of these intentions does not appear to be well explained by shared or even similar ontological statuses. Furthermore, beyond pragmatic diversity in the application of ‘think’ and ontological diversity within the phenomena of intending, there are logical differences of a different order between “mental state” concepts that, as we noted earlier, have to do with issues of duration and which mental phenomena can be sensibly regarded as states. Mental state concepts do not necessarily share much in common other than seeming to ‘refer to the inner’, but the examples of ‘think’ and ‘intention’ above suggest this view of mental predicates is misleading. This seeming to refer to the inner is a consequence and not a cause of, conceptualizing activities in psychological terms. Action explanations that involve properties of agents rather than situational factors refer to ‘the inner’. Although we typically think of our thoughts and feelings
152 Noah Susswein and Timothy P. Racine
as ‘inside us’, talk of ‘the inner’ is metaphorical. Lakoff and Johnson (2003) argue that common sense is fundamentally metaphorical and describe one such foundational ‘container’ metaphor, a tendency to conceive of non-physical entities as contained by or within a physical structure (Slaney and Maraun 2005). Conceiving of experiences and abilities as being inside of the agent whose experiences and abilities they are is an example of this container metaphor. The fact that is sometimes possible to conceal what we think and feel should not tempt one to think of ‘the inner’ in mistakenly concrete terms (Bennett and Hacker 2003: 88–90). Now, to state baldly that ‘the inner’ is metaphorical seems uncontroversial. But the corollary that it is the use of psychological concepts and not their putative inner referents that determines their meaning is more difficult. We describe a usage-based account of ‘understanding attention’ in the next section. However, first we wish to address a potential misunderstanding. We are not arguing for a ‘psychological nominalist’ or ‘eliminativist’ position that psychological concepts are mere artefacts of language, just a way of talking, or that we don’t really have intentions or understandings (Churchland 1986). We do really have intentions and understandings, but this ‘having’ is not a relation of containing. Rather, this ‘having’ is akin to possessing. Containing involves a vessel while possessing involves an agent. What makes something an agent is that it acts or behaves. Not all agents but only agents can be characterized in psychological terms; only agents have psychological properties. It is the connections between psychological phenomena and activity, not descriptions of activity, which we wish to elucidate here. Psychological concepts are heterogeneous, but if we must speak in generalities, we ought to think of the verb and adverbial forms of psychological predicates (e.g., ‘attending’, ‘intentionally’) as primary and the noun forms (attention, intention) as usually harmless reifications.
4.
Social interaction and meaning
In our earlier discussion of psychological capacities, we attempted to draw attention to activity, and argued that it is what persons do that determines whether or not they are attending to X, intended to Y, that they understand some aspect of attending and intending, or that they are in a stage of primary or secondary intersubjectivity. If attributions of psychological capacities and understandings are based on behavioural criteria, how do we know when such ascriptions are justified? Pointing, and especially declarative pointing (see Brinck; Leavens et al. this volume), is thought to unambiguously reveal infant’s understanding of attention. However, other than Bates and colleagues’ (Bates, Camaioni and Volterra 1975)
Sharing mental states 153
and Bruner’s (1983) pioneering work, the issue of how pointing in general comes to be recruited and used as an interactional (and what is often, and derivatively, understood to be a referential) device is largely absent in empirical work. If one takes an inductive approach to joint attention, ethnomethodological, and derivatively, conversation analytic (CA) methods of social interaction are useful (e.g., Antaki 2004; Atkinson and Heritage 1984; Garfinkel 1967; Sharrock and Coulter 2004; Turnbull 2003; Wootton 1997). Conversation analysts study the orderly structure of conversation in order to examine how interaction is accomplished. CA studies of early infant development are relatively rare because with preverbal infants there is, by definition, no conversation to analyze. Wootton (1994) focuses on third position repair sequences (i.e., where an infant has acted, the parent has responded to that action in the second turn/position, and the infant responds to that response in the third turn/position). An infant is in a position to display her congruence or lack thereof with the parent’s previous turn, which is a contingency that Wootton reports 12-month-olds can manage. However, a prior concern for us is to first reconcile the logic of ethnomethodological investigation with the concerns discussed earlier. CA approaches to meaning assume that the meaning that participants attribute to their interactions is manifest in the details of that sequential and negotiated activity. Conversation analysis is a means of using interactional sequences to instantiate criteria in ongoing interaction. We now turn to sequences of interaction to illustrate the claim that it is what they do that determines whether two agents have “a shared understanding of what an interaction is about” (Reimers and Fogel 1992).
4.1 The attribution of intention To demonstrate this approach, to which we will return in more detail in the next section, we present two brief sequences of interaction involving the second author’s daughter, T, shortly after her second birthday. These examples show how criteria for the application of intention were warranted in the contingencies of this child’s interactions with her mother, C, and that T’s mother attributed intention to her daughter’s activity. That is, mother and child both acted in ways that manifest an understanding of intention. Readers and the participants, to the degree that they possess reasonably complex language, can see the criteria that are contained therein. However, we do not see a mental state in T that is her goal-directed behaviour. Rather the ascription of mental predicate to T presupposes her ability to act in goal-directed ways.
154 Noah Susswein and Timothy P. Racine
T and C were recorded while sharing a meal (cf. Canfield 1993). The examples were transcribed according to CA conventions (e.g. Atkinson and Heritage 1984). Square brackets represent overlap in speech. Numbers in round brackets represent pauses in seconds. Dots in round brackets indicate noticeable pauses of under 0.2 seconds in length. Words in brackets represent a description of an action. A colon represents a drawn out syllable. An upward arrow indicates rising intonation in a pitch contour.
Example 1 1 C: the [skin is really soft on those kind] 2 T: [too sour mommy] i han havva drink 3 (.5) 4 C: [you’re going to have a drink] 5 [(T drinks from cup on table)]
We can see T doing something analogous to C in line 2 of Example 2 below. C’s intention in this case is to provide T with a fork with which to eat her dinner, which T acknowledges.
Example 2 1 C: no that’s mine (.) i think i forgot your fork thi (.5) 2 T: you did 3 C: i did (.5) oh i did (.) it’s right here
How are we to make sense of this activity? Grasping the meaning of an action – that is, what it is that an agent is doing – involves understanding the practice in which it is embedded: e.g., too sour mommy i han havva drink followed by (T drinks from cup on table). To grasp the local meanings of actions is to see intention in this particular social situation and in so doing to attribute intentions to the actor who performed the act in question: i.e., you’re going to have a drink.
4.2 Gesture as a window on intersubjectivity In order to further discuss pointing, we need to address the issue of reference. From a conversational analytic point of view, episodes involving joint attention are meaningful because both agents, for example, understand a request is being made, not simply because both agents appreciate the correspondence between an action and a referent. From this point of view, reference presupposes, and is parasitic upon, social practice. More generally, “reference consists in a sign’s having a . A full transcription of this interaction is available upon request from the second author.
Sharing mental states 155
role in a language-game” (Proudfoot and Copeland 2002: 338). That is, although human reference requires particular mental and neurological capacities, reference is not a mental or neurological phenomenon, but a social one (see also Leavens et al. this volume). It is not a mental or a neurological event which makes a first finger extended towards X a reference to X. It is the fact that this extended finger is a technique or practice of orienting others towards X in a particular context. A person may very well be imagining what they plan to prepare for dinner while they point toward X, but the point of this distracted individual is no less referential. That is, to know that A’s point towards X is a reference to X is not to know anything about A’s state of mind. Now, it is of course possible to misinterpret A’s point toward X as a point toward Y. And doing so would constitute a misunderstanding of A’s intention. But this would be a case of misinterpreting what kind of action that point was, not an incorrect guess as to the underlying cause of A’s pointing behaviour. When thinking about gesture as a window on intersubjective engagement, the canonical activity of interest is pointing. Declarative pointing has attracted particular attention because it is thought to not exist in non-human primates, apart from those who are language-trained and humanly enculturated, and is has also been argued to ground language acquisition (e.g. Butterworth 2003; Brinck this volume; Leavens et al. this volume). Developmentalists tend to think of pointing as evidence of joint attention, a criterion for ‘secondary intersubjectivity’. In terms of CA, we would like to see evidence that a mother and baby treat these gestures with the same understanding. Racine (2005) charted the development of the interactional resources that one 9- to 12-month-old infant had at her disposal when interacting with her mother. In so doing the roles that infant pointing, pointing-like gestures and reaching might play in interaction were displayed. We now report on two of these observations that might be seen to involve declarative pointing. In these examples, M represents mother; B represents baby. Vocal (V) elements of the interaction, gaze (G) and positions of the right hand (RH) and left hand (LH) are identified on separate lines. The temporal dimension of the interaction is represented by movement from left to right along these lines and the horizontal relations between the various channels were kept as tight as possible. Manual actions are surrounded by ( ). Overlaps in actions (including speech) are surrounded by [ ]. Some less relevant aspects of the interaction are simply summarized as gloss transcriptions. Mother’s hand gestures and positions are noted only when they make relevant the infant’s behaviour. In cases of an empty line (typically when either M or B did not vocalize), the line is not listed. All talk is transcribed orthographically. Numbers in round brackets represent pauses in seconds, but timings have meaning only within the vocal line. Dots in
156 Noah Susswein and Timothy P. Racine
round brackets indicate noticeable pauses of under 0.2 seconds in length. Words in brackets represent a description of an action. A colon represents a drawn out syllable. Further conventions were adopted for this dataset. In cases where a vocalization was whispered it is surrounded by o o. Audible inbreathing or outbreathing accompanying utterances is represented by > and < respectively. In cases where a vocalization was inaudible the utterance is surrounded by ( ). Gaze was inferred from orientation of the head. In cases where the target of gaze was less clear it is surrounded by ( ). When targets are not stated in the gaze line, M represents gaze at parent and C represents gaze at infant. Dashes (---) represent sustained gaze at a target. With respect to hand positions (RH or LH), dashes are used to indicate static positioning of the hand and commas to represent movement of the hand from an initial position to another in the sequence. The following additional conventions are employed: Re = reach, and Po = point. In the first of two related examples, at the 11-month visit B does what would be typically coded as a declarative act with outstretched arm and index finger prominent but without all other fingers in palm. However, M seems to treat this as ‘asking for permission’ or to at least minimally signal intent. Thus, this is not just a point used to direct attention to enhance interaction (Liszkowski, Carpenter, Striano and Tomasello 2006; Moore and D’Entremont 2001). The function of M’s behaviour seems to be a confirmation of some sort given that it is accompanied by the utterance “I know”, but is also a prohibition because the infant ceases to advance towards the prohibited object. Example 1 1 B: (G) camera -- - - - - - - - - - - - - - - , , , M - - - - - - - - - - - - - - - - - - - - (LH) at side - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - (RH) open hand (shoulder level) -, Po (camera) - - , lowers arm - - - - (walks towards camera, ~7’ from where M is seated, stops 2’ from target) M: (V) no I don’t think we can (chuckle) (1.0) yeh (.) i know (.) thats a camera (G) B - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
At the 12-month visit, B gets up from M’s lap and again takes steps towards the camera that was forbidden at the previous month’s visit. While she gets up out of M’s lap, her RH pointing finger begins to extend. B takes 2 steps towards the camera with her RH changing into a grasping gesture. As in the previous visit, we argue that the pointing gesture needs to be made sense of in relation to the prohibitory social situation.
Sharing mental states 157
Example 2 1 B: (G) camera - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - (LH) getting up - - , at side - - - - - - - - - - - - - - - - - - - - - - - - - - - - (RH) getting up - - , Po - , , Re (not fully extended) - - , drops to side - (walks towards camera) (stops) (walks backwards towards M) M: (V) <> (0.5) Oget over hereO (G) B - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
4.3 The meaning of declarative pointing How are we to make sense of this activity? As noted earlier, a basic form of intersubjectivity is defined by interactions showing an ability to differentiate between people and objects, whereas another form of intersubjectivity manifests itself in interactions showing an ability to sequence simultaneous activity with people and objects. In the two examples we see clear evidence of satisfying the criteria for this secondary form of intersubjectivity. In Example 1, at 11 months of age B gets out of M’s lap while playing with her to approach the researchers’ video camera while the mother chuckles at this after stating no I don’t think we can. The infant then stops and points at the camera and shifts her gaze towards her mother. Her mother acknowledges her turn by stating I know and the infant lowers her pointing arm and proceeds no further. In Example 4, videotaped one month later, the point is briefer and turns into a reach and there is no eye contact towards the mother. However, B again does not proceed any further towards the camera when M whispers get over here. Thus, unlike some of the other interactions reported by in Racine (2005), these are rich tightly coupled interactions that clearly involve an understanding of what these interactions are about on the part of both mother and infant. To focus on the role of the pointing gestures themselves, M responds to the gesture in Example 1 with a verbal affirmation. But she is not responding to an extended index finger, but an entire sequence of interaction in a particular social situation. Setting aside for the moment what exactly the point means, the gesture and the mother’s words can mean what they mean in this circumstance of a camera being set up a few feet from a baby, that baby noticing the camera and approaching it, that baby noticing that her mother shows that she does want the baby to approach the camera, etc. (Canfield 1981; Racine 2004). The behavioural criteria apply in this particular social situation for humans and those who have . Bates et al. (1975) argued that infants of this age may begin to be more “confident in” their mothers attention and therefore do not need to visually check as often. However, Morissette, Ricard and Decaré (1995) report that visual checking increases from 12 to 18 months of age.
158 Noah Susswein and Timothy P. Racine
the capacity to act in ways similar to humans (Wittgenstein 1958). In Example 2, M exhales in a manner that seems to express mild frustration, which is supported by the fact that she then playfully instructs her infant to return to her lap and leave the camera alone. Thus, both points are incorporated into the interaction. And as one might expect interactions involving pointing gestures show intersubjectivity. The next issue with regard to the gestures is whether we would be justified in describing these as declarative points. A declarative point, as the name suggests, is meant to denote an index finger point that functions to comment on an object or some state of affairs rather than to request an object or action. But not only would calling these index finger extensions declarative radically reduce the complexity of these sequences of interaction, doing so would also not seem accurate. In the research literature, points are said to be declarative, imperative or informative (Bates et al. 1975; Liszkowski, Carpenter, Striano, and Tomasello 2006, but see Brinck this volume). There has also been the suggestion that infants might sometimes point to request object names (e.g., Carpendale and Lewis 2006). But the way these gestures are used in these examples (i.e., their meaning) does not seem to be to comment to the mother in a way that could be glossed “look, there’s a camera” (declarative), but rather “I want to touch that camera”, “I am going to touch that camera” or perhaps “can I touch that object?” or even “there’s an object that I am not supposed to touch.” The I want/I am going/Can I possibilities are all imperative in a sense in that they express desire for the object as opposed to wanting to share attention around it. The last possibility, of showing understanding of a prohibition, makes the most sense in this interactive context. This is because this interpretation is consistent with the infant’s prior turn of walking towards a forbidden camera and then stopping to point at it and with the mother’s subsequent turn of measured confirmation in Example 1 and overt ruling out in Example 2. But we are not arguing here for the creation of another category of pointing category; we are arguing for the importance of due care in the investigation of pointing.
5.
Summary and conclusion
Particular forms of social interaction define particular forms of intersubjectivity. Forms of intersubjectivity do not causally explain the existence of interactive abilities and behaviours. Very young typically developing infants discriminate between animate and inanimate objects, and their interactions with caregivers are richly coordinated and complex. Older infants can simultaneously attend to object and interlocutor. Intersubjectivity is a taxonomic, technical concept used to classify interactive behaviours and abilities rather than denoting the vehicles
Sharing mental states 159
or causes of those behaviours and abilities. Such taxonomic descriptions are easily mistaken for causal hypotheses. That is, ‘intersubjectivity’ and infants’ partial understanding of attention and intention appear to be misconstrued as the causes of their socially coordinated behaviour. But in truth, it is their socially coordinated behaviours which are logically primary and which justify ascribing intersubjectivity’ and ‘understanding intention’ and ‘attention’ to them. This chapter was written to remind developmental researchers that the relationship between causal and definitional issues in intersubjectivity is complicated, and not recognized as a problem to be addressed. Many interpretations of important empirical findings are predicated on problematic definitions of intersubjectivity and related psychological concepts. We hope that this chapter will stimulate further work on conceptual and empirical investigations of intersubjectivity.
Acknowledgments The empirical examples were taken from a study conducted by the second author, which was funded by the Human Early Learning Partnership and supported by doctoral fellowships from the Social Sciences and Humanities Research Council of Canada and the Michael Smith Foundation for Health Research. Preparation of this chapter was supported by grants from the Manitoba Health Research Council and the Social Sciences and Humanities Research Council of Canada to the second author. We thank Max Bibok, Jeremy Carpendale, Esa Itkonen, Bill Turnbull, Jordan Zlatev and Chris Sinha for helpful comments on earlier versions of this chapter.
References Antaki, C. 2004. “Reading minds or dealing with interactional implications.” Theory and Psychology 14: 667–683. Austin, J. 1975. How to do Things with Words (2nd ed.). New York: Oxford University Press. Atkinson, J.M. and Heritage, J. (eds.) 1984. Structures of Social Action: Studies in Conversation Analysis. Cambridge: Cambridge University Press. Bakeman, R. and Adamson, L. 1984. “Coordinating attention to people and objects in motherinfant and peer-infant interactions.” Child Development 55: 1278–1289. Baker, G.P. and Hacker, P.M.S. (1984). Skepticism, Rules and Language. Cambridge: Blackwell. Barresi, J. and Moore, C. this volume. “The neuroscience of social understanding.” Baron-Cohen, S. 1995. Mindblindness: An Essay on Autism and Theory of Mind. Cambridge, MA: MIT Press. Bates, E., Camaioni, L. and Volterra, V. 1975. “The acquisition of performatives prior to speech.” Merrill-Palmer Quarterly 21: 205–226.
160 Noah Susswein and Timothy P. Racine
Bennett, M.R. and Hacker, P.M.S. 2003. Philosophical Foundations of Neuroscience. Oxford: Blackwell. Bretherton, I. 1991. “Intentional communication and the development of an understanding of mind.” In Children’s Theories of Mind: Mental State and Social Understanding, D. Frye & C. Moore (eds.), 49–75. Hillsdale, NJ: Erlbaum. Bretherton, I., McNew, S. and Beeghly-Smith, M. 1981. “Early person knowledge as expressed in gestural and verbal communication: When do infants acquire a theory of mind?” In Infant Social Cognition: Empirical and Theoretical Considerations, M.E. Lamb & L.R. Sherrod (eds.), 333–373. Hillsdale, NJ: Erlbaum. Brinck, I. this volume, “The role of intersubjectivity in the development of intentional communication.” Bruner, J. 1983. Child’s Talk: Learning to Use Language. New York: Norton. Butterworth, G. 2003. “Pointing is the royal road to language for babies.” In Pointing: Where Language, Culture, and Cognition Meet, S. Kita (ed.), 9–33. Mahwah, NJ: Erlbaum. Canfield, J.V. 1981. Wittgenstein, Language and World. Amherst: University of Massachusetts Press. Canfield, J.V. 1993. “The living language: Wittgenstein and the empirical study of communication.” Language Sciences 15: 165–193. Carpendale, J.I.M. and Lewis, C. 2006. How Children Develop Social Understanding. Oxford: Blackwell. Churchland, P.S. 1986. Neurophilosophy. Cambridge, Mass: MIT Press. Dupré, J. 1993. The Disorder of Things: Metaphysical Foundations of the Disunity of Science. Harvard University Press. Eilan, N., Hoerl, C., McCormack, T. and Roessler, J. (eds.) 2005. Joint Attention: Communication and Other Minds. Oxford: Oxford University Press. Gallagher, S. and Hutto, D.D. this volume. “Understanding others through primary interaction and narrative practice.” Garfinkel, H. 1967. Studies in Ethnomethodology. Englewood Cliffs, NJ: Prentice-Hall. Hacker, P.M.S. 1996. Wittgenstein, Mind and Will. Cambridge: Blackwell. Hark, M. ter 1990. Beyond the Outer and the Inner: Wittgenstein’s Philosophy of Psychology. Dordrecht, The Netherlands: Kluwer. Hobson, R.P. and Hobson, J.A. this volume. “Engaging, sharing, knowing: Some lessons from research in autism.” Kenny, A. 1989. The Metaphysics of Mind. Oxford, England: Oxford University Press. Kita, S. 2003. Pointing: Where Language, Cognition and Culture Meet. Mahwah, NJ: Erlbaum. Lakoff, G. and Johnson, M. 2003. Metaphors We Live By (2nd ed.). Chicago: University of Chicago Press. Leavens, D.A., Hopkins, W.D. and Bard, K. this volume. “The heterochronic origins of explicit reference.” Liszkowski, U., Carpenter, M., Striano, T. and Tomasello, M. 2006. “Twelve- and 18-montholds point to provide information for others.” Journal of Cognition and Development 7: 173–187. Mill, J.S. 1875. System of Logic (8th ed). London: Longmans. (Original work published 1843). Moore C., and D’Entremont, B. “Developmental changes in pointing as a function of attentional focus.” Journal of Cognition and Development. 2: 109–129. Moore, C. and Dunham, P.J. (eds.) 1995. Joint Attention: Its Origins and Role in Development. Hillsdale, NJ: Erlbaum.
Sharing mental states 161
Morissette, P., Ricard, M. and Decaré, T.G. 1995. “Joint visual attention and pointing in infancy: A longitudinal study of comprehension.” British Journal of Developmental Psychology 13: 163–175. Proudfoot, D. and Copeland, B.J. 2002. “Wittgenstein’s deflationary account of reference.” Language and Communication 22: 331–351. Racine, T.P. 2004. “Wittgenstein’s internalistic logic and children’s theories of mind.” In Social Interaction and the Development of Knowledge, J.I.M. Carpendale and U. Müller (eds.), 257–276. Mahwah, NJ: Erlbaum. Racine, T.P. 2005. The Role of Shared Practice in the Origins of Joint Attention and Pointing. Unpublished doctoral thesis, Simon Fraser University, Burnaby, BC, Canada. Racine, T.P. and Carpendale, J.I.M. 2007a. “Shared practices, understanding, language and joint attention.” British Journal of Developmental Psychology 25: 45–54. Racine, T.P. and Carpendale, J.I.M. 2007b. “The embodiment of mental states.” In Body in Mind, Mind in Body: Developmental Perspectives on Embodiment and Consciousness, W.F. Overton, U. Müller and J. Newman (eds.), 159–190. Mahwah, NJ: Erlbaum. Racine, T.P. and Carpendale, J.I.M. 2007c. “The role of shared practice in joint attention.” British Journal of Developmental Psychology 25: 3–25. Reimers, M. and Fogel, A. 1992. “The evolutions of joint attention of objects between infants and mothers: Diversity and convergence.” Analise Psicologica 1: 81–89. Sharrock, W. and Coulter, J. 2004. “ToM: A critical commentary.” Theory & Psychology 14: 579–600. Slaney, K.L. and Maraun, M.D. 2005. “Analogy and metaphor running amok: An examination of the use of explanatory devices in neuroscience.” Journal of Theoretical and Philosophical Psychology 25: 153–172. Tomasello, M. 1995. “Joint attention as social cognition.” In Joint Attention: Its Origins and Role in Development, C. Moore and P. Dunham (eds.), 103–130. Hillsdale, NJ: Erlbaum. Tomasello, M. 1999. “Having intentions, understanding intentions, and understanding communicative intentions.” In Developing Theories of Intention: Social Understanding and SelfControl, P.D. Zelazo, J.W. Astington & D.R. Olson (eds.), 63–75. Mahwah, NJ: Erlbaum. Tomasello, M. and Carpenter, M. 2005. The emergence of social cognition in three young chimpanzees. Monographs of the Society for Research in Child Development, 70 (Serial No. 279). Tomasello, M., Carpenter, M., Call, J., Behne, T. and Moll, H. 2005. “Understanding and sharing intentions: The origins of cultural cognition.” Behavioral and Brain Sciences 28: 675–735. Trevarthen, C. 1977. “Descriptive analysis of infant communicative behavior.” In Studies in Mother-Infant Interaction, H.R. Schaffer (ed.), 227–270. New York: Academic Press. Trevarthen, C. 1979. “Communication and cooperation in early infancy. A description of primary intersubjectivity.” In Before Speech: The Beginning of Human Communication, M. Bullowa (ed.), 99–136. London: Cambridge University Press. Trevarthen, C. and Aitken, K.J. 2001. “Infant intersubjectivity: Research, theory, and clinical applications.” Journal of Child Psychology & Psychiatry 42: 3–48. Trevarthen, C. and Hubley, P. 1978. “Secondary subjectivity: Confidence, confiding, and acts of meaning in the first year.” In Action, Gesture and Symbol: The Emergence of Language, A. Lock (ed.), 183–229. London: Academic Press. Turnbull, W. 2003. Language in Action: Psychological Models of Talk. New York: Psychology Press.
162 Noah Susswein and Timothy P. Racine
Wittgenstein, L. 1958. Philosophical Investigations (3rd ed.). Englewood Cliffs, NJ: PrenticeHall. Wootton, A.J. 1994. “Object transfer, intersubjectivity and third position repair: Early developmental observations of one child.” Journal of Child Language 21: 543–564. Wootton, A.J. 1997. Interaction and the Development of Mind. Cambridge, MA: Cambridge University Press.
part ii
Evolution
chapter 8
What is the nature of the gestural communication of great apes? Simone Pika Human speech is frequently accompanied by movements of the arms and hands termed gestures. The majority of these gestures is invented spontaneously and is highly iconic but some gestures are used functionally in ways very similar to speech that is symbolically, referentially, based on intersubjectively learned and shared social conventions. Our closest living relatives, the great apes also use gestures in their natural communication in a variety of contexts such as play, grooming, sex and agonistic encounters. A deep understanding of apes’ gestural signalling might therefore be helpful to get insight into the evolutionary scenario of human communication and cognition. The present chapter investigates the nature of the gestural signalling of the four great apes, bonobos (Pan paniscus), chimpanzees (Pan troglodytes), gorillas (Gorilla gorilla) and orangutans (Pongo pygmaeus), with a special focus on the following three aspects: (1) the intentionality of gestures, (2) their referential use, and (3) similarities and differences to gestures in prelinguistic or just-linguistic human infants.
1.
Introduction
Human communication is unique in the animal kingdom in a variety of ways. Most importantly, human communication depends crucially on linguistic symbols, which are individually learned and intersubjectively shared social conventions used to direct the attentional and mental states of others to real or imaginary situations (Tomasello 1999). Human communication is also unique in the way it employs manual and other bodily gestures. For example, to our knowledge only human beings point to things deictically for conspecifics – the basic form of gestural reference – simply to share attention or to comment on events and objects (Tomasello et al. 2005). And only humans use gestures, ranging from conventionalized gestures such as ‘waving goodbye’ to iconic gestures such as drawing a circle in the air to depict the shape of the sun. Thus, although the majority of gestures are performed spontaneously and are highly iconic (McNeill 1992), some
166 Simone Pika
human gestures are used functionally in ways very similar to language, that is symbolically, referentially, based on intersubjectively learned and shared social conventions. Therefore the question arises: What is the nature of communicative signals in our closest living relatives, the non-human primates, and how do they relate to human gestures and language? The following chapter will address this question by focusing on the gestural signalling of the four great ape species, bonobos (Pan paniscus), chimpanzees (Pan troglodytes), gorillas (Gorilla gorilla) and orangutans (Pongo pygmaeus), with a special focus on the following three aspects: (1) the intentionality of gestures, (2) their referential use, and (3) similarities and differences to gestures in prelinguistic or just-linguistic human infants.
2.
State of the art: Communicative signals in primates
By looking for the evolutionary roots of human language, researchers quite naturally looked at the communication systems evolved in other animal species and especially in our closest living relatives, the non-human primates (hereafter primates). Until recently, the majority of studies focused on vocal communication (e.g., Marler 1980; Owings and Morton 1998; Seyfarth and Cheney 1997; Snowdon, Brown and Peterson 1982), which might be due to the analogy to human speech (Liebermann 1968; Seyfarth 1987; Snowdon 1988). This interest has been stimulated even further by evidence that primates and especially monkeys use vocalizations to communicate information about their social and physical environment, in addition to their emotional states (e.g., Cheney and Seyfarth 1990; Zuberbühler 2001). As used in the primate literature, animal vocalizations qualify as referential signals if they: (a) have a distinct acoustic structure, (b) are produced in response to a particular external object or event, and (c) elicit a similar response in nearby listeners as the external object or event normally does (Zuberbühler 2000b). The finding that vervet monkeys (Cercopithecus aethiops) use different alarm calls in association with different predators (leading to different escape responses in receivers, Seyfarth, Cheney and Marler 1980) raised the possibility that monkeys use vocalizations to make reference to outside entities (Cheney and Seyfarth 1990). Referential signals have been reported from various monkey species in their natural habitats (e.g., Zuberbühler 2000a; Zuberbühler 2003), suggesting that referential communication is a widespread and perhaps universal characteristic of primate communication. These findings have been . See Zlatev (this volume) on the difference between “triadic mimesis” which does not require conventionality, and language, which does.
Gestural communication of great apes 167
taken to suggest that primate referential abilities are the output of a cognitive ability that could be pivotal to language, namely the capacity to assign meaning to arbitrary sound utterances (Zuberbühler 2002). This conclusion remains controversial, however, because it has been shown since then that alarm calls of this type have arisen numerous times in evolution in species that also must organize different escape responses for different predators, including most prominently prairie dogs and domestic chickens (for an overview see, Owings and Morton 1998). In addition, related research on apes has provided mixed results. For example, Uhlenbroek (1996) has demonstrated that East African chimpanzees at Gombe (Pan troglodytes schweinfurthii) produce acoustic variants of ‘pant hoots’ in three different contexts: travel, food, and encounters with other community members. However, a comparable study on the same subspecies at the Mahale Mountains study site found no evidence of context specificity in ‘pant hoots’ (Marler and Hobbett 1975; Mitani and Brandt 1994; Mitani et al. 1992). In addition, Clark and Wrangham’s (1993; 1994) studies at Kanyawara have suggested that arrival ‘pant-hoots’ at fruiting trees provide information about the social context rather than about the food itself. However, recent research shows that some chimpanzee calls can differ in their fine acoustic structure depending on the eliciting context, a crucial prerequisite for calls to function referentially (Crockford and Boesch 2003; Slocombe and Zuberbühler 2005a). Furthermore, Slocombe and Zuberbühler (2005b) showed in an experimental setting that one chimpanzee was able to use the information conveyed by rough grunts given in two distinct contexts by his group members to guide his search for food. These results therefore suggest that the vocalizations of chimpanzees may also function referentially. This is consistent with the fact that great apes clearly have demonstrated referential abilities in captive conditions (see for laboratory trained apes, Rumbaugh 1977; Savage-Rumbaugh et al. 1993). Therefore, the absence of evidence might merely reflect a paucity of data, rather than a lack of referential abilities on behalf of the apes in the wild. However, one other possible explanation for the apparent lack of referential vocalizations in great apes might be that they are specialized in a different kind of referential skill – one based on the flexible use of manual gestural signals.
3.
Gestural signals
To date, studies on gestural communication in primates are very unevenly distributed among species (for an overview see, Tomasello and Call 2007). Almost no systematic studies exist focusing on the gestural signaling of monkeys. The most interesting observations concern hamadryas baboons (Papio hamadryas)
168 Simone Pika
which were observed to engage in notifying behaviour, before leaving the troop (Kummer 1968). The behaviour consists in approaching another animal and looking directly into their face, presumably to make sure that the recipient is attending before engaging in certain activities. In addition, Kummer and Kurt (1965) described a ground-slap behaviour that seems to serve as an attention getter and a kind of teasing behaviour during play. Maestripieri (1997; 1999) compared the gestural behaviour of three macaque species in captivity (Macaca arctoides, Macaca nemestrina, Macaca mulatta) and suggested that characteristics of a social structure, such as reduced influence of dominance and kinship may select along with group size for a wider gestural repertoire. In addition, he described a very interesting behaviour within mother-infant dyads: When pigtail macaque mothers want their infants to follow them and they do not, the mothers sometimes return and stare in the infant’s face (or even poke the infant) before leaving again (Maestripieri 1996). The gestural communication of apes has received much more research attention, but it has focused mainly on chimpanzees (Pan troglodytes) in captivity (Ladygina-Kohts 1935; Van Hooff 1973), and in the wild (Plooij 1979, 1987; Van Lawick-Goodall 1968). Goodall (1986) for instance observed that chimpanzees at the Gombe National Park use more than a dozen distinct gestures in a variety of contexts. In addition, signals such as the gesture leaf clipping (Nishida 1980) and the grooming hand clasp (McGrew and Tutin 1978) provided evidence for the existence of population-specific differences in chimpanzee communities in the wild. Concerning their closely related congener, the bonobo (Pan paniscus), SavageRumbaugh and colleagues (Savage-Rumbaugh, Wilkerson and Bakeman 1977; Savage-Rumbaugh and Wilkerson 1978) described the use of 20 gestures in a sexual context. In addition, de Waal (1988) provided a comparison of the gestural signaling of bonobos and chimpanzees and observed 15 distinct gestures for bonobos that are linked to particular situations. Tanner (1998) and Tanner and Byrne (1999) described the use of 30 gestures in a gorilla (Gorilla gorilla) group in captivity. For western lowland gorillas in the wild, Parnell and Buchanan-Smith (2001) reported a specific gesture called the splash display which is used to intimidate other silverbacks, and Fay (1989) observed hand-clapping behavior in females. Contrary to the African great ape species, the gestural communication of the Asian apes has received less research attention. MacKinnon (1974; however see also, Rijksen 1978) established a repertoire of tactile and visual gestures of wild Bornean (Pongo pygmaeus) and Sumatran orangutans (Pongo abelii). Overall, the above mentioned studies have provided evidence that great apes make frequent use of gestures in their everyday communication. However, they
Gestural communication of great apes 169
all use different definitions of the term ‘gesture’ or none at all, and did not focus on processes of social cognition such as the learning of the gestures or their intentional use (with the exception of Plooij 1979, 1987). The major aim of the present chapter is therefore to investigate whether communicative gestures in great apes are used as flexibly communicative strategies with individual decision making involving cognitive processes, which involve at least some degree of intersubjective understandings. This chapter will therefore provide an overview of the gestural signalling of all four great ape species and focus in detail on the following three aspects: (1) the intentional use of gestures, (2) their referential use, and (3) similarities and differences to gestures in prelinguistic or just-linguistic human infants. The presented quantitative data are based on recent papers on the gestural communication of subadult apes in captivity (Pongo pygmaeus: Liebal, Pika and Tomasello 2006; Gorilla gorilla: Pika, Liebal and Tomasello 2003; Pan paniscus: Pika, Liebal, and Tomasello 2005; Pan troglodytes: Tomasello et al. 1994, 1997, 1985, 1989).
4.
Gestures in the great apes: Empirical evidence
Gestures are a subset of communicative signals. They can be defined as expressive movements of the limbs or head and body postures that are directed toward a recipient, are mechanically ineffective and receive a voluntary response. The following behavioural criteria were used to infer their communicative intent: (1) gazing at the recipient, and/or (2) waiting after the signal had been produced, expecting a response. Thus, gestures that appear to have components of ritualised morphology (e.g., chest beat) are also included in this definition, if they meet these above mentioned criteria.
4.1 Overview: Gestural repertoires Based on auditory, tactile and visual components we formed three signal categories: (1) auditory gestures generate sound while performed, (2) tactile gestures include physical contact with the recipient, and (3) visual gestures generate a mainly visual component with no physical contact. In addition, the accompanying context was analyzed in own studies. The bonobos used 20 different distinct gestures (see Table 1), one auditory (5%), eight tactile (40%) and eleven visual gestures (55%), which were performed
170 Simone Pika
Table 1. Gestural repertoire of the great apes Pan paniscus
Pan troglodytes Gorilla gorilla
Pongo pygmaeus
Auditory
1
3
6
0
Visual Tactile Total average/ individual
11 8 20 11
18 9 30 9.5
16 11 33 20
14 12 26 16
mainly in the play context (55%), but also in the food (14%), travel (10%), nurse (5%), ride (5%), sex (5%), affiliative (3%), and agonistic contexts (3%). The chimpanzees used 30 different distinct gestures (see Table 1), three auditory (10%), nine tactile (30%), and 18 visual gestures (60%), in a variety of contexts including affiliation, agonistic, feeding and nursing, sexual, grooming, travel, and play. Play was the most important context accounting for between 47% and 70% of the gestures depending on the studies (Tomasello et al. 1994; 1997). Overall the gorillas performed 33 different distinct gestures, six auditory (18%), 11 tactile (33%) and 16 visual gestures (49%). These gestures occurred mainly in the play (40%) context, but also in the food (15%), ride (10%), nurse (10%), travel (10%) affiliative (10%) and agonistic (5%) context. The orangutans used 26 different distinct gestures, 12 tactile and 14 visual gestures. These gestures occurred mainly in the play (32%), feeding (26%), affiliative (17%), and agonistic context (7%) but also in the context of getting access to objects (2%), sex (2%), walking (2%), and nursing (2%). Overall, these data show that all four great ape species have multifaceted gestural repertoires of auditory, tactile and visual gestures, which are used in a variety of contexts.
4.2 Intentional action A crucial milestone in human ontogeny is the onset of intentional behaviour, which develops during the second half of the first year of life (Bates et al. 1979). Piaget (1952) defined an intention (in the psychological sense) as the differentiation of means and ends, but also emphasized the difficulty to create a valid definition. This view is supported by the history of psychology which attests to the difficulty of making a clear distinction between intentional and unintentional behaviour. However, work in the study of pre-linguistic communication in human infants has offered some relatively clear operational definitions that may be used
Gestural communication of great apes 171
to differentiate between these two types of behaviours. Elaborating on Piaget’s view, Bruner (1981) for instance noted that …an intention is present when an individual operates persistently toward achieving an end state, chooses among alternative means and/or routes to achieve that end state, persists in deploying means and corrects the deployment of means to get closer to the end state, and finally ceases the line of activity when specifiable features of the states are achieved.
It is worth mentioning that according to this definition much of intentional action takes place beyond the threshold of reportable awareness (Bruner 1981). Following Bratman (1989) an intention can be understood as a plan of action the organism chooses and commits itself to in pursuit of a goal. An intention thus includes both a means (action plan) and a goal. Contrary to the definition of Bruner (1981) in this definition the actor seems to be able to account for or be conscious of the nature of his intentions. In addition, distinctions have been made between perlocutionary and illocutionary acts (Bates et al. 1979), or communicative behaviour and intentionally communicative behavior (Golinkoff 1981). Perlocutionary acts are infant behaviours, in which communication occurs only because the receiver is adept at interpreting the behaviour of the child (von Glaserfeld 1974, 1976). For instance, an infant might cry because it can not reach a toy, which causes the mother to come and give her the toy. Although communicative behaviour occurred, the behaviour of the infant was not intentionally directed toward the mother. Illocutionary or intentionally communicative behaviours on the other hand are infant behaviours, “in which the sender is aware a priori of the effect that a signal will have on his listener, and he persists in that behavior until the effect is obtained or failure is clearly indicated” (Bates et al. 1979: 36). For example, the child turns its attention from the toy to the mother and whines at her. The whining becomes a socialcommunicatory act with the intention of obtaining the adult’s help. This distinction emphasizes the contributions of each partner to the communicative act and provides the behavioural tools which permit us to reliably identify intentional communication: a. alternations in eye contact between the goal and the intended communication partner, b. augmentations, additions and substitutions of signals until the goal has been obtained, and c. changes in the form of the signal toward the abbreviated and/or exaggerated patterns that are appropriate only for achieving a communicative goal.
172 Simone Pika
With respect to intersubjectivity, it can be argued that it is the illocutionary acts that are most relevant. To investigate whether the behaviour of apes qualifies as intentional acts, researchers have used the developmental Piagetian and the preverbal communication perspective. Using Speech Acts theory (see Austin 1962; Bates et al. 1975), Plooij (1978, 1979) for instance studied the development of communicative signals between mother-infant dyads in chimpanzees in the wild. He argued that with the onset of begging between the age of 9 and 12 ½ months, (also often called peering, Pika et al. 2005) followed by the use of gestures such as initiating tickling, grooming and approach, the chimpanzee infant understands his mother and conspecifics as social agents. This developmental stage therefore marks the onset of the use of imperative gestures (which are used to get another individual to help in attaining a goal) and the developmental shift from perlocutionary to illocutionary acts. Bard (1992) investigated the communicative abilities of orangutan infants in a food sharing context and focused on the transition from bifocal behaviour to behavioural sequences. Differentiation was made between ‘intentional behaviour’ as shown in bifocal behavioral sequences involving either objects or social agents (Case 1985) and ‘intentional communication’ (behavioral sequences involving coordination between social agents and objects, e.g., Bretherton, McNew and Beeghly-Smith 1981). Bard (1992) found that orangutans at the age from 1–6 months used intentional behaviours, whereas intentional communication was observed in older orangutans only, ranging from 2 ½ to 5 years of age. In intentional behaviour the action was either directed to the food, for instance, with a grasp coordinated with eating, or directed to the mother, for instance, by performing a pull on the mother’s body and subsequently eating. In intentional communication, on the other hand, the animal solicited food from the mother by using one open, cupped hand, palm up, held underneath but not necessarily touching the mother’s chin. To identify whether great apes use their gestures as intentional acts while communicating with group members, the presented studies focused on (1) Meansend dissociation of signalling behaviour and goal, and (2) Adjustment to social circumstances, such as adjustment to audience affects (for a detailed description of the methods and animals see, Liebal et al. 2006; Pika et al. 2003; Pika, Liebal, and Tomasello 2005; Tomasello et al. 1994; Tomasello et al. 1997). Means-ends dissociation is characterized by the flexible relation of signalling behaviour to the recipient and goal. For example, an individual uses a single gesture for several goals (touch for nursing and riding) or different gestures for the same goal (slap ground and body beat for play). Audience effects are characterized by differential use of gestures or other communicative signals as a function of the attentional states of the recipient.
Gestural communication of great apes 173
Figure 1. Means-ends dissociation. Average number of gestures used in a single context and average number of gestures used for several contexts. Error bars indicate the SD.
4.2.1 Means-ends dissociation Our results showed that bonobos used on average in every context approximately two (± 0.6) different gestures, the chimpanzees 3.2 (± 0.4), the gorillas 3.2 (± 1), and the orangutans 5.3 (±1.2) gestures (Figure 1). Concerning the use of gestures in different contexts, the bonobos utilized on average 2.7 (± 1.48) gestures in more than one context, the chimpanzees 1.3 (± 0.2), the gorillas 3.8 (± 2.6), and the orangutans 1.5 (± 0.9) gestures. 4.2.2 Adjustment to audience effects Focusing on audience effects we found a significant difference between the use of tactile and visual gestures among all species based on a variation in the degree of visual attention of the recipient (Wilcoxon-test: P < 0.05, for further details see, Liebal et al. 2006; Pika et al. 2003; Pika, Liebal, and Tomasello 2005; Tomasello et al. 1994). There was no significant difference between the uses of auditory versus visual gestures and auditory versus tactile gestures. On average, the bonobos performed 70% (± 10) of their visual gestures to an attending recipient, the chimpanzees 86% (± 2), the gorillas 89% (± 12), and the orangutans 98.8% (± 2). Tactile gestures were performed to an attending recipient in 51% (bonobos, ± 10), 48% (chimpanzees, ± 10), 66% (gorillas, ± 13), and 67% (orangutans, ± 10.3) of the cases only (see Figure 2).
174 Simone Pika
Figure 2. Audience effects. The y-axis indicates the percentage of gestures, the x-axis indicates the different species. The four different colours indicate the signal category and the attentional state of the recipient. Error bars indicate the SD.
In sum, these results reveal that great apes use different means, gestures in the same context interchangeably toward the same end, but also use the same means/gesture to achieve different ends/goals. Concerning audience effects, the findings show that great apes preferentially use visual gestures to an already attending recipient. Based on the key characteristics for intentional communication in human children, we can therefore conclude that great apes use gestures that classify as intentional acts. Other relevant studies on intentional communication focused mainly on audience effects. Tanner and Byrne (1993) for instance reported that a female gorilla repeatedly used her hands to hide her playface from a potential partner, indicating some flexible control of the otherwise involuntary facial expression – as well as a possible understanding of the role of visual attention in the process of gestural communication. Liebal et al. (2004) showed that chimpanzees tended to move into the attentional field of the recipient by walking in front of her and then performed visual gestures. Furthermore, in an experimental setting, Liebal et al. (2004) showed that all four great ape species take into account the attentional state of a human experimenter, by using visual gestures preferentially when they were facing the experimenter. In addition, anecdotal evidence for intentional communication is available for the language trained bonobo Kanzi. He was observed to hand a nut to a person
Gestural communication of great apes 175
who was supposed to crack it open. He then slapped the nut and placed a stone on top of it (Savage-Rumbaugh et al. 1986).
4.3 Referential gesturing Researchers working on communicative signals of pre-linguistic and just-linguistic children distinguish gestures in terms of direction and function. Concerning the direction of gestures, differentiation has been made between dyadic and triadic gestures. Dyadic gestures involve two individuals and are used to attract the attention of others to the self; triadic gestures are used to attract the attention of others to some entity, e.g. an event or an object. Although this “third entity” mainly denotes an outside entity (Bates 1976), it can also be a part of one’s body, e.g. referring to one’s own nose. Triadic gestures are therefore clearly referential and develop in human children at the age of 12 months (Bates et al. 1979). The use of these gestures has been linked with cognitive capacities such as mental state attribution (Camaioni 1993; Tomasello 1995), because the recipient must infer the signaller’s intended meaning. Concerning the function of gestures, differentiation has been made between protoimperative and protodeclarative gestures (Bates et al. 1975). Protoimperatives are defined as the child’s preverbal intentional use of the listener as an agent or tool in achieving some end (e.g. to request an object). Protodeclaratives are defined as the child’s preverbal effort to direct the adult’s attention to some event or object in the world. This approach suggests continuity between preverbal and later verbal communication and is useful when focusing on human children whose gestures precede speech (Bates et al. 1979). However, it is not coherent to use these terms for species who will never exhibit verbal communication (Leavens 2004; Leavens and Hopkins 1998). The term imperative will thus be used to refer to gestures being used to get another individual to help in attaining a physical goal, such as getting an object, playing, etc., and the term declarative will be used to characterize those gestures which are used to attain a non-physical goal, namely to draw another’s attention to an object or entity merely for the sake of sharing attention. Leavens et al. (this volume) and Baron-Cohen (1999) explicitly exclude protoimperative gestures from the category of intentional communication (in a similar vein see also, Povinelli et al. 2000; Povinelli et al. 2001), arguing that only protodeclarative gestures imply the signaller’s possession of a nascent theory of mind.
. See, however, Leavens (2004) for a critical view on defining these modes of communication by reference to underlying psychological processes or mental states.
176 Simone Pika
The majority of gestures used between great apes in their natural communication are dyadic (Pika et al. 2005). Exceptions are the gestures food begging (an animal holds out the hand, palm up to obtain food from another, see for orangutans, Bard 1992; see for chimpanzee Tomasello et al. 1994) food offer (an animal offers food placed on her arm to another one, Liebal et al. 2006) and pointing. These gestures are clearly triadic – a request to another for food or an offer of food to another is distal since the signaller is not touching the recipient. Food-begging and food-offer have been observed between conspecifics, but pointing has only been reported for captive chimpanzees interacting with their human experimenters (e.g., Leavens et al. 1996; Leavens et al. 2004) as well as human-raised or language trained apes (e.g., Gardner and Gardner 1969; Miles 1990; Patterson 1978a; Woodruff and Premack 1979). Although there is one anecdotal report about the declarative use of this gesture in a single bonobo in the wild (Vea and Sabater-Pi 1998), it is not clear yet whether these abilities represent natural communication abilities or are byproducts of living in a human encultured environment (Tomasello and Call 1997). It may be argued that apes have no ‘need’ for pointing because they rely on other behaviors that serve a similar function, such as detection of body orientation and eye gaze (Gomez 1991; Menzel 1974, 1973). While walking mainly quadrupedally their whole body is ‘pointing’ (Plooij 1987). Interestingly, Savage-Rumbaugh and colleagues (Savage and Bakeman 1978; Savage-Rumbaugh et al. 1977) and Tanner and Byrne (1996) described several gestures that they consider iconic uses of gestures. Iconic gestures are related to their referent by virtue of some actual physical resemblance between the two (Bates et al. 1979), such as a desired motion in space or the form of an action. Two individuals (one bonobo, one gorilla) seemed to signal with their hand, arm, or head to a playmate the direction in which they wanted her to move, the action they wanted her to perform, or the position they wanted her to take. Roth (1995) and Pika et al. (2003) however, who also focused on the occurrence of iconic gestures in three groups of bonobos and two groups of gorillas in captivity, did not observe any instances of the iconic use of gestures. It is possible that their analysis did not focus in sufficient detail on the receiver’s response to detect gestures of an iconic nature. Another explanation would be that gesturing of an iconic nature could be a developmental phenomenon, appearing only at adolescence and promoted by special social and physical conditions (Tanner and Byrne 1999). Concerning the function of gestures, the majority of studies have shown that apes mainly use imperative gesture in their natural communication. Focusing on human-raised and language-trained apes, Patterson (1978b) reported observations of ‘showing’ in one gorilla, and Savage-Rumbaugh (1988) in one bonobo. Furthermore, Savage-Rumbaugh et al. (1998) described how a bonobo female who had heard unusual sounds in the forest, directed the human caretaker’s attention
Gestural communication of great apes 177
Figure 3. The directed scratch gesture. ©Dorothee Classen
toward these sounds by looking and gesturing in that direction. It should be noted, however, that in all these cases interpretation is an issue. Interestingly, Pika and Mitani (2006) described the widespread use of a gesture, the so called directed scratch, by wild chimpanzees in the context of grooming (see Figure 3). The gesture involved one chimpanzee making a relatively loud and exaggerated scratching movement on a part of his body, which could be seen by his grooming partner. In the majority of the cases the indicated spot was groomed directly by the recipient. Pika and Mitani (2006) argue that (1) the gesture may be used communicatively to indicate a precise spot on the body, and (2) the recipient of the signal has an understanding of the intended meaning of the gesture. The authors suggest that directed scratches therefore may qualify as referential. In sum, the evidence shows that the majority of gestures of great apes used between conspecifics are imperative and dyadic (see also, Pika et al. 2005). In addition, the use of referential gestures supports earlier findings that certain important cognitive capacities pertaining to intersubjectivity are present in apes (for an overview see e.g., Zuberbühler, Tomasello & Call 1997).
178 Simone Pika
4.4 Similarities and differences of gestures in apes and human children Our and other researchers’ findings provide evidence that similar to prelinguistic or just-linguistic human children, apes use their gestures as intentional acts, by operating persistently toward achieving an end state, choosing among alternative means and adjusting their use of gestures to social circumstances. The majority of their gestures are used in dyadic interactions, whereas human children gesture from their very first attempts in addition to dyadic gestures triadically (Carpenter et al. 1998). However, it is worth noticing that quantitative comparisons are until now non-existent. Focusing on the type of gestures, human children perform conventionalizations, deictics, and symbolic gestures (Bates et al. 1979). Conventionalizations are gestures in which the signaller uses an effective behaviour for getting something done. For instance, many infants learn to request being picked up by raising their arms over their heads while approaching an adult. Great apes often do similar gestures, such as using a stylized arm-raise to initiate play, ritualized from actual acts of play hitting in the context of rough-and-tumble play. Many youngsters also conventionalize signals for asking their mother to lower her back so they can climb on. For example, a brief touch on the top of the rear end, ritualized from occasions on which they pushed her rear end down mechanically. The learning process involved in such cases is most likely an individual learning process called ‘ontogenetic ritualization,’ in which a communicatory signal is created by two individuals shaping each others’ behaviour in repeated instances of an interaction (Tomasello 1996). However, it does not involve understanding of communicative intentions or cultural (imitative) learning of any sort and therefore it does not create a shared communicative symbol. However, also note that the production of a variety of gestures of great apes, especially species-typical gestures (e.g. the chest beat in gorilla), seem to be due to genetic predisposition, while only the use and response has to be learned (Pika et al. 2003). The second type of gestures is deictics, which are designed to direct adult attention to outside entities. This does not automatically mean that the infant is gesturing in order to induce the adult to share attention with her on that third entity. Indeed, many infants use arm and index finger extension to orient their own attention to things and only understand later the function of the gesture (Carpenter et al. 1998; Franco and Butterworth 1996). Chimpanzees interacting with human caretakers point to request food and these gestures are clearly deictic. In addition, the gesture food offer seems to bear some relation with deictics, . Note that this is quite different from the notion of ‘conventions’, as used by e.g. Zlatev (this volume).
Gestural communication of great apes 179
ecause they are triadic, distal and direct the attention of the recipient to a speb cific event and object. The gesture directed scratch also resembles in some ways a deictic gesture. However, although it clearly involves some form of reference, this reference is self-directed. The third gesture type produced by human children, symbolic gestures, are communicative acts that are either associated with a referent metonymically (the gesture refers to an element or attribute of something to mean the thing itself) or iconically (Acredelo and Goodwyn 1988; Pizzuto and Volterra 2000). Examples include gestures such as: sniffing for a flower, panting for a dog, holding arms out for an airplane, raising arms for big things, and blowing for hot things. Empirically we do not know whether infants learn to produce symbolic gestures via ritualization or via imitation (Lock 1978), but it is much more likely that in most cases infants are learning these symbolic gestures via imitation. That is, they are learning by understanding an adult’s communicative intention in using the gesture first and then engaging in role reversal imitation to use the gesture themselves when they have ‘the same’ communicative intention. Although, group specific gestures such as leaf clipping (Nishida 1980), the grooming hand clasp (McGrew and Tutin 1978), somersault (Pika et al. 2005), and armshake (Pika et al. 2003) used by apes differ qualitatively from symbolic gestures in human children, their existence suggests that social learning based on intersubjectivity, in the form of some kind of group-specific cultural transmission, is at work. In addition, gestures such as directed scratches used by chimpanzees in the wild, provide evidence that (a) chimpanzees have an understanding of the intended meaning of the gestures, and (b) signallers and receivers shape a non-communicative signal into a communicative one with a distinct meaning. They thus might be useful tools to reconstruct the evolution and development of symbolic gestures. Thus, the crucial difference between the gestures of apes and those of prelinguistic or just-linguistic human children becomes obvious when focusing on the function of gestures: Apes mainly gesture for imperative purposes, while human children gesture for imperative purposes but also quite frequently for declarative purposes to direct the attention of others to an outside object or event, simply for the sake of sharing interest in it or commenting on it (Bates et al. 1975; Liszkowski et al. 2004). The propensity to communicate about outside entities and situations and comment on them seems to be unique for human beings and might have triggered the onset of symbolic communication, i.e. language. The question thus arises: why do humans comment on outside entities to share experiences? This behaviour is probably linked with an increased level of intersubjectivity that enables humans to understand other people as intentional agents with whom they may share experience (Tomasello et al. 2005). It therefore might have been
180 Simone Pika
erived from the need to create a new medium for social bonding in humans; a d medium to establish and service social relationship and to share experiences. In our closest living relatives, the bonobos and the chimpanzees, social grooming permeates virtually every aspect of social life. Grooming might therefore represent an ancient medium to evaluate and to invest into social relationships which was lost in our ancestors, who developed different means to perform this function through “vocal grooming” (Dunbar 1996) and/or gestural grooming, flowing in a later stage into linguistic communication.
5.
Conclusions
The ability to employ manual and bodily gestures provides a rich source of information about the nature of human and primate communication. Similar to the vocal modality, human beings use some of their gestures symbolically, based on intersubjectively learned and shared social conventions, to direct the attention of others referentially and for declarative purposes. This mode of communication clearly depends on the ability to be aware of others’ mentalities and an understanding of the intentional states of others, which might be unique to the human species (Tomasello et al. 2005). Primates use gestures mainly for imperative purposes and in dyadic interactions. However, many of their gestures – in contrast to their vocalizations – are clearly learned and are used intentionally, with adjustments for the attentional state of the recipient and means-ends dissociation. This shows that the capacity of great apes for intersubjectivity, while differing from that of humans, is not negligible. In addition, since apes in the wild seem to make use of group-specific and referential gestures, it seems plausible that the gestural modality of our closest living relatives was the modality within which symbolic communication first evolved (Arbib 2002; Condillac 1971; Corballis 2002; Hewes 1973; see also Hutto this volume). Future studies will hopefully shed light on potential evolutionary mechanisms by which the vocal and gestural signals of apes transformed into the linguistic and gestural symbols of human beings.
Acknowledgements I am grateful to Josep Call, Katja Liebal and Michael Tomasello, who shared their data with me. For comments on an earlier draft and lively discussions, I would like to thank Katja Liebal, Elena Nicoladis and Chris Sinha. I am indebted to Dorothee Classen, the artist of the drawing of directed scratch in Figure 3.
Gestural communication of great apes 181
References Acredelo, L.P. and Goodwyn, S.W. 1988. “Symbolic gesturing in normal infants.” Child Development 59: 450–466. Arbib, M.A. 2002. “The mirror system, imitation, and the evolution of language.” In Imitation in Animals and Artifacts. Complex Adaptive Systems, K. Dautenhahn and C.L. Nehaniv (eds.), 229–280. Cambridge, Masachusetts, USA: MIT Press. Austin, J. L. 1962. How to Do Things with Words. New York: Oxford University Press. Bard, K.A. 1992. “Intentional behaviour and intentional communication in young free-ranging orangutans.” Child Development 63: 1186–1197. Baron-Cohen, S. 1991. “Precursors to a theory of mind: Understanding attention in others.” In Natural Theories of Mind: Evolution, Development and Simulation of Everyday Mindreading, A. Whiten (ed.), 233–251. Oxford, UK: Blackwell. Bates, E. 1976. Language and Context: The Acquisition of Pragmatics. New York: Academic Press. Bates, E., Benigni, L., Bretherton, I., Camaioni, L. and Volterra, V. 1979. The Emergence of Symbols: Cognition and Communication in Infancy. New York: Academic Press. Bates, E., Camaioni, L. and Volterra, V. 1975. “The acquisition of performatives prior to speech.” Merrill-Palmer Quarterly 21(3): 205–226. Bratman, M.E. 1989. “Intention and personal policies.” In Philosophical perspectives, Philosophy of mind and action theory, J.E. Tomberlin (ed.), 443–469. Northridge: California State University. Bretherton, I., McNew, S. and Beeghly-Smith, M. 1981. “Early persons knowledge as expressed in gestural and verbal communication: When do infants require a ‘theory of mind’.” In Infant Social Cognition: Empirical and Theoretical Considerations, M. Lamb, E. and L.R. Sherrod (eds.), 333–373. Hillsdale, New York: Erlbaum. Bruner, J. 1981. “Intention in the structure of action and interaction.” In Advances in Infancy Research, L. Lipsitt (ed.), Vol. 1, 41–56. New Jersey: Ablex, Norwood. Camaioni, L. 1993. “The development of intentional communication: A re-analysis.” In New Perspectives in Early Communicative Development, J. Nadel and L. Camaioni (eds.), 82–96. London: Routledge. Carpenter, M., Nagell, K. and Tomasello, M. 1998. “Social cognition, joint attention, and communicative competence from 9 to 15 months of age.” Monographs of the Society for Research in Child Development 36(4): 176–179. Case, R. 1985. Intellectual Development: Birth to Adulthood. New York: Academic Press. Cheney, D.L. and Seyfarth, R.M. 1990. How Monkeys See the World. Chicago and London: University of Chicago Press. Clark, A.P. and Wrangham, R.W. 1993. “Acoustic analysis of wild chimpanzee pant hoots: Do Kibale forest chimpanzees have an acoustically distinct food arrival pant hoot?” American Journal of Primatology 31: 99–109. Clark, A.P. and Wrangham, R.W. 1994. “Chimpanzee arrival pant-hoots: Do they signify food or status?” International Journal of Primatology 15(2): 185–205. Condillac, E.B.D. 1971. An Essay on the Origin of Human Knowledge; Being a Supplement to Mr. Locke’s Essay on the Human Understanding. A Facism. Reproduction of the Translation of Thomas Nugent. Gainesville, Florida: Scholars’ facsimiles and reprints.
182 Simone Pika
Corballis, M.C. 2002. From Hand to Mouth, the Origins of Language. Princeton, New Jersey: Princeton University Press. Crockford, C. and Boesch, C. 2003. “Context-specific calls in wild chimpanzees, Pan troglodytes verus: Analysis of barks.” Animal Behaviour 66(1): 115–125. de Waal, F.B.M. 1988. “The communicative repertoire of captive bonobos (Pan paniscus) compared to that of chimpanzees.” Behaviour 106(3–4): 183–251. Dunbar, R. 1996. Grooming, Gossip and the Evolution of Language. London: Faber and Faber Ltd. Fay, J.M. 1989. “Hand-clapping in western lowland gorillas.” Mammalia 53(3): 457–458. Franco, F. and Butterworth, G. 1996. “Pointing and social awareness: Declaring and requesting in the second year.” Journal of Child Language 23: 307–336. Gardner, R.A. and Gardner, B. 1969. “Teaching sign language to a chimpanzee.” Science 165: 664–672. Golinkoff, R. 1981. “The influence of Piagetian theory on the study of the development of communication.” In New Directions in Piagetian Theory and Practice, I.E. Sigel and D.M. Brodzinsky and R. Golinkoff (eds.), 127–142. Hillsdale, New York: Erlbaum. Gomez, J.C. 1991. “Visual behaviour as a window for reading the mind of others in Primates.” In Natural Theories of Mind: Evolution, Development and Simulation of Everyday Mindreading, A. Whiten (ed.), 195–207. Oxford: Basil Blackwell. Goodall, J. 1986. The Chimpanzees of Gombe, Patterns of Behaviour. Cambridge, England: The Belknap Press of Harvard University Press. Hewes, G.W. 1973. “Primate communication and the gestural origin of language.” Current Anthropology. 12(1–2): 5–24. Hutto, D.D. this volume. “First communions: Mimetic sharing without theory of mind.” Kummer, H. 1968. Social Organization of Hamadryas Baboons. Chicago: University of Chicago Press. Kummer, H. and Kurt, F. 1965. “A comparison of social behaviour in captive and wild hamadryas baboons.” In The Baboon in Medical Research, H. Vagtborg (ed.), 1–16. Texas: University of Texas Press. Ladygina-Kohts, N.N. 1935. Infant Chimpanzee and Human Child. A Classic 1935 Comparative Study of Ape Emotions and Intelligence. New York: Oxford University Press. Leavens, D.A. 2004. “Manual deixis in apes and humans.” Interaction Studies 5: 387–408. Leavens, D.A. and Hopkins, W.D. 1998. “Intentional communication by chimpanzees: A crosssectional study of the use of referential gestures.” Developmental Psychology. 34: 813–822. Leavens, D.A., Hopkins, W.D. and Bard, K.A. 1996. “Indexical and referential pointing in chimpanzees (Pan troglodytes).” Journal of Comparative Psychology 110(4): 346–353. Leavens, D.A., Hopkins, W.D. and Thomas, R.K. 2004. “Referential communication by chimpanzees (Pan troglodytes).” Journal of Comparative Psychology 118(1): 48–57. Liebal, K., Call, J. and Tomasello, M. 2004. “Chimpanzee gesture sequences.” Primates 64: 377– 396. Liebal, K., Pika, S., Call, J. and Tomasello, M. 2004. “Great ape communicators move in front of recipients before producing visual gestures.” Interaction studies 5(2): 199–219. Liebal, K., Pika, S. and Tomasello, M. 2006. “Gestural communication of orangutans (Pongo pygmaeus).” Gesture 6(1): 1–38. Liebermann, P. 1968. “Primate vocalizations and human linguistic ability.” Journal of the Acoustic Society of America 44: 1574–1584.
Gestural communication of great apes 183
Liszkowski, U., Carpenter, M., Henning, A., Striano, T. and Tomasello, M. 2004. “Twelvemonths-olds point to share attention and interest.” Developmental Science 7(3): 297–307. Lock, A. 1978. Action, Gesture and Symbol: The Emergence of Language. New York: Academic Press. MacKinnon, J.R. 1974. “Behaviour and ecology of Orang Utans.” Animal Behaviour 22: 3–74. Maestripieri, D. 1996. “Maternal encouragement of infant locomotion in pigtail macaques (Macaca nemestrina).” Animal Behaviour 51: 603–610. Maestripieri, D. 1997. “Gestural communication in macaques.” Evolution of Communication 1(2): 193–222. Maestripieri, D. 1999. “Primate social organization, gestural repertoire size, and communication dynamics.” In The Origins of Language: What Nonhuman Primates Can Tell, B.J. King (ed.), 55–77. Santa Fe: School of American Research Press. Marler, P. 1980. Primate Vocalization: Affective or Symbolic? New York: Plenum Press. Marler, P. and Hobbett, L. 1975. “Individuality in a long-range vocalization of wild chimpanzees.” Zeitschrift für Tierpsychologie 38: 97–109. McGrew, W.C. and Tutin, C.E.G. 1978. “Evidence for a social custom in wild chimpanzees?” Man 13: 234–251. Menzel, E. 1974. “A group of young chimpanzees in a one-acre field.” In Behaviour of Nonhuman Primates, A. Schrier and F. Stollnitz (eds.), Vol. 5, 83–153. New York: Academic Press. Menzel, E.W. 1973. “Chimpanzee spatial memory organization.” Science 182(4115): 943–945. Miles, H.L. 1990. “The cognitive foundations for reference in a signing orangutan.” In Language and Intelligence in Monkeys and Apes, S.T. Parker and K.R. Gibson (eds.), 511–539. Cambridge: Cambridge University Press. Mitani, J.C. and Brandt, K.L. 1994. “Social factors influence the acoustic variability in the longdistance calls of male chimpanzees.” Ethology 96(3): 233–252. Mitani, J.C., Hasegawa, T., Gros-Louis, J., Marler, P. and Byrne, R.W. 1992. “Dialects in wild chimpanzees?” American Journal of Primatology 27(4): 233–243. Nishida, T. 1980. “The leaf-clipping display: A newly-discovered expressive gesture in wild chimpanzees.” Journal of Human Evolution 9: 117–128. Owings, D.H. and Morton, D.S. 1998. Animal Vocal Communication: A New Approach. Cambridge: Cambridge University Press. Parnell, R.J. and Buchanan-Smith, H.M. 2001. “Animal behaviour: An unusual social display by gorillas.” Nature 412: 294. Patterson, F. 1978a. “Conversations with a gorilla.” National Geographic 134(4): 438–465. Patterson, F. 1978b. “Linguistic capabilities of a lowland gorilla.” In Sign Language and Language Acquisition in Man and Ape, F.C.C. Peng (ed.), 161–201. Boulder, CO: Westview Press. Piaget, J. 1952. The Origins of Intelligence in Children. New York: Norton. Pika, S., Liebal, K., Call, J. and Tomasello, M. 2005. “The gestural communication of apes.” Gesture 5(1/2): 41–56. Pika, S., Liebal, K. and Tomasello, M. 2003. “Gestural communication in young gorillas (Gorilla gorilla): Gestural repertoire, learning and use.” American Journal of Primatology 60(3): 95–111. Pika, S., Liebal, K. and Tomasello, M. 2005. “Gestural communication in subadult bonobos (Pan paniscus): Gestural repertoire and use.” American Journal of Primatology 65(1): 39–51. Pika, S. and Mitani, J.C. 2006. “Referential gesturing in wild chimpanzees (Pan troglodytes).” Current Biology. 16(6): 191–192.
184 Simone Pika
Pizzuto, E. and Volterra, V. 2000. “Iconicity and transparency in sign languages: A cross-linguistic cross-cultural view.” In The Signs of Language Revisited: An anthology to honor Ursula Bellugi, K. Emmorey and H. Lane (eds.), 261–286. New York: Erlbaum. Plooij, F.X. 1978. “Some basic traits of language in wild chimpanzees?” In Action, Gesture and Symbol, A. Lock (ed.), 111–131. London: Academic Press. Plooij, F.X. 1979. “How wild chimpanzee babies trigger the onset of mother-infant play.” In Before Speech, M. Bullowa (ed.), 223–243. Cambridge: Cambridge University Press. Plooij, F.X. 1987. “Infant-ape behavioural development, the control of perception and, types of learning and symbolism.” In Symbolism and Knowledge, A. Tryphon and J. Montangero (eds.), Vol. 8, 29–58. Geneva: Jean Piaget Archives Foundation. Povinelli, D.J., Bering, J.M. and Giambrone, S. 2000. “Toward a science of other minds: Escaping the argument by analogy.” Cognitive Science 24: 509 – 541. Povinelli, D.J., Bering, J.M. and Giambrone, S. 2001. “Reasoning about beliefs: A human specialization?” Child Development 72: 691–695. Rijksen, H.D. 1978. A Field Study on Sumatran Orangutans. Wageningen: Mededelingen Landbouwhogeschool. Roth, R.R. 1995. “A study of gestural communication during sexual behavior in bonobo (Pan paniscus, Schwartz).” Unpublished PhD dissertation, University of Calgary, Calgary. Rumbaugh, D.M. 1977. Language Learning by a Chimpanzee. The Lana project. New York: Academic Press. Savage, S. and Bakeman, R. 1978. “Sexual morphology and behavior in Pan paniscus.” In Proceedings of the Sixth International Congress of Primatology, 613–616. New York: Academic Press. Savage-Rumbaugh, E.S. 1988. “A new look at ape language: Comprehension of vocal speech and syntax.” In Comparative Perspectives in Modern Psychology, Nebraska Symposium on Motivation, D.W. Leger (ed.), Vol. 35, 201–256. Lincoln, NB: University of Nebraska Press. Savage-Rumbaugh, E.S., McDonald, K., Sevcic, R.A., Hopkins, W.D. and Rupert, E. 1986. “Spontaneous symbol acquisition and communicative use by pygmy chimpanzees (Pan paniscus).” Journal of Experimental Psychology: General 115: 211–235. Savage-Rumbaugh, E.S., Murphy, J., Sevcic, R.A., Brakke, K.E., Williams, S.L. and Rumbaugh, D.M. 1993. “Language comprehension in ape and child.” Monographs of the Society for Research in Child Development 58(3–4): 1–256. Savage-Rumbaugh, E.S., Rumbaugh, D.M. and McDonald, K. 1985. “Language learning in two species of apes.” Neurosciences and Biobehavioral Review 9: 653–656. Savage-Rumbaugh, E.S., Shanker, S.G. and Taylor, T.J. 1998. Apes, Language, and the Human Mind. New York: Oxford University Press. Savage-Rumbaugh, E.S., Wilkerson, B.J. and Bakeman, R. 1977. “Spontaneous gestural communication among conspecifics in the pygmy chimpanzee (Pan paniscus).” In Progress in Ape Research, G.H. Bourne (ed.), 97–116. New York: Academic Press. Savage-Rumbaugh, S. and Wilkerson, B. 1978. “Socio-sexual behavior in Pan paniscus and Pan troglodytes: A comparative study.” Journal for Human Evolution 7: 327–344. Seyfarth, R.M. 1987. “Vocal communication and its relation to language.” In Primate societies, B. Smuts and D.L. Cheney and R. Seyfarth and R. Wrangham and T. Struhsaker (eds.), 440–451. Chicago: University of Chicago Press. Seyfarth, R.M. and Cheney, D.L. 1997. “Some general features of vocal development in nonhuman primates.” In Social Influences on Vocal Development, C. Snowdon and M. Hausberger (eds.), 249–273. Cambridge: Cambridge University Press.
Gestural communication of great apes 185
Seyfarth, R.M., Cheney, D.L. and Marler, P. 1980. “Vervet monkey alarm calls: Semantic communication in a free-ranging primate.” Animal Behaviour 28: 1070–1094. Slocombe, K.E. and Zuberbühler, K. 2005a. “Agonistic screams in wild chimpanzees vary as a function of social role.” Journal of Comparative Psychology 119(1): 67–77. Slocombe, K.E. and Zuberbühler, K. 2005b. “Functionally referential communication in a chimpanzee.” Current Biology 15: 1179–1784. Snowdon, C. 1988. “A comparative approach to vocal communication.” In Comparative Perspectives in Modern Psychology, Nebraska Symposium on Motivation, D.L. Leger (ed.), 145–199. Lincoln: University of Nebraska Press, Lincoln. Snowdon, C.T., Brown, C.H. and Petersen, M.R. 1982. Primate Communication. Cambridge: Cambridge University Press. Tanner, J.E. 1998. “Gestural communication in a group of zoo-living lowland gorillas.” Unpublished PhD, University of St. Andrews, St. Andrews. Tanner, J.E. and Byrne, R. 1996. “Representation of action through iconic gesture in a captive lowland gorilla.” Current Anthropology 37(1): 162–173. Tanner, J.E. and Byrne, R. 1999. “The development of spontaneous gestural communication in a group of zoo-living lowland gorillas.” In The Mentalities of Gorillas and Orangutans, Comparative Perspectives, S.T. Parker and R.W. Mitchell and H.L. Miles (eds.), 211–239. Cambridge: Cambridge University Press. Tanner, J.E. and Byrne, W.B. 1993. “Concealing facial evidence of mood: Perspective-taking in a captive Gorilla.” Primates 34(4): 451–457. Tomasello, M. 1995. “Joint attention as social cognition.” In Joint Attention: Its Origin and Role in Development, C. Moore and P.J. Dunham (eds.), 103–130. Hillsdale, New York: Erlbaum. Tomasello, M. 1996. “Do apes ape?” In Social Learning in Animals: The Roots of Culture, C.M.G. Heyes, B.G. Jr. (ed.), 319–346. San Diego: Academic Press, Inc. Tomasello, M. 1999. “Emulation learning and cultural learning.” Behavioral and Brain Sciences. 21: 703–704. Tomasello, M. 2003. Constructing a Language, a Usage-based Theory of Language Acquisition. Cambridge, Massachusetts, and London, England: Harvard University Press. Tomasello, M. and Call, J. 1997. Primate Cognition. New York: Oxford University Press. Tomasello, M. and Call, J. 2007. The Gestural Communication of Monkeys and Apes. Mahwah, New York: Lawrence Erlbaum Associates. Tomasello, M., Call, J., Nagell, K., Olguin, R. and Carpenter, M. 1994. “The learning and use of gestural signals by young chimpanzees: A trans-generational study.” Primates 35(2): 137–154. Tomasello, M., Call, J., Warren, J., Frost, T., Carpenter, M. and Nagell, K. 1997. “The ontogeny of chimpanzee gestural signals.” In Evolution of Communication, S. Wilcox, King, B. and Steels, L. (ed.), 224–259. Amsterdam/ Philadelphia: John Benjamins Publishing Company. Tomasello, M., Carpenter, M., Call, J., Behne, T. and Moll, H. 2005. “Understanding and sharing intentions: The origins of cultural cognition.” Behavioral and Brain Sciences 28: 1–17. Tomasello, M., George, B.L., Kruger, A.C., Farrar, M.J. and Evans, A. 1985. “The development of gestural communication in young chimpanzees.” Journal of Human Evolution 14: 175– 186. Tomasello, M., Gust, D. and Frost, G.T. 1989. “A longitudinal investigation of gestural communication in young chimpanzees.” Primates 30(1): 35–50. Uhlenbroek, C. 1996. “The structure and function of the long-distance calls given by male chimpanzees in Gombe National Park.” University of Bristol, Bristol.
186 Simone Pika
Van Hooff, J.A.R.A.M. 1973. “A structural analysis of the social behaviour of a semi-captive group of chimpanzees.” In Social Communication and Movement, Studies of Interaction and Expression in Man and Chimpanzee, M. von Cranach and I. Vine (eds.), 75–162. London & New York: Academic Press. Van Lawick-Goodall, J. 1968. “A preliminary report on expressive movements and communication in the Gombe stream chimpanzees.” In Primates. Studies in Adaptation and Variability, P.C. Jay (ed.), 313–374. New York: Holt, Rinehart, and Winston. Vea, J.J. and Sabater-Pi, J. 1998. “Spontaneous pointing behaviour in the wild pygmy chimpanzee (Pan paniscus).” Folia Primatologica 69(5): 289–290. von Glaserfeld, E. 1974. “Signs, communication, and language.” Journal of Human Evolution 3: 464–474. von Glaserfeld, E. 1976. “The development of language as purposive behavior.” Annals of the New York Academy of Sciences 280: 212–226. Woodruff, G. and Premack, D. 1979. “Intentional communication in the chimpanzee: The development of deception.” Cognition 7: 333–352. Zuberbühler, K. 2000a. “Interspecific semantic communication in two forest monkeys.” Proceedings of the Royal Society 267: 713–718. Zuberbühler, K. 2000b. “Referential labelling in Diana monkeys.” Animal Behaviour 59(5): 917–927. Zuberbühler, K. 2001. “Predator-specific alarm calls in Campbell’s monkeys, Cercopithecus campbelli.” Behavioral Ecology and Sociobiology 50(5): 414–422. Zuberbühler, K. 2002. “A syntactic rule in forest monkey communication.” Animal Behaviour 63: 293–299. Zlatev, J. this volume. “The co-evolution of intersubjectivity and bodily mimesis.”
chapter 9
The heterochronic origins of explicit reference David A. Leavens, William D. Hopkins and Kim A. Bard Explicit reference is the communicative capacity to intentionally pick out a specific object in the environment and make that object a manifest topic for shared attention. Pointing is the quintessential example of non-verbal, explicit reference. Chimpanzees, and other apes in captivity, spontaneously point without overt training. Because wild apes almost never point, and because both captive and wild apes are sampled from the same gene pool, this implies that, for apes, hominoid genes interact with certain environments to elicit pointing. We propose that changes in the patterns of hominid development interact with ape-like cognitive capacities to produce features of explicit reference in human infants, a capacity that emerges in our nearest living relatives when they experience similar circumstances.
1.
Introduction
Despite a number of claims to the contrary (e.g., Butterworth and Grover 1988; Petitto 1988; Povinelli, Bering and Giambrone 2003a) pointing is frequently displayed by captive apes (e.g., Leavens 2004; Leavens and Hopkins 1998, 1999). Captive apes usually point in apparent requests for delivery of food, but they will also point out the location of tools required to gain access to food (Call and Tomasello 1994; Russell et al. 2005; Whiten 2000). One chimpanzee, Clint, often pointed to experimenters’ shoes, which he subsequently manipulated with apparent satisfaction, upon presentation of said shoes (Leavens, Hopkins and Bard 1996). Pointing is, manifestly, a referential act, directing the attention, the movements or the actions of an observer to a specific locus. It has been argued that nonhumans both do not and cannot point because the psychological basis for their pointing-like behaviour differs from that of humans who point (e.g. Povinelli et al. 2003a; Tomasello 2006). We believe that the psychological aspects of pointing and other communicative acts are distributed between signaler and receiver, and that the psychological basis of pointing cannot be correctly attributed to an individual, but to that individual, and any and all
188 David A. Leavens et al.
observers, who form a communicative system. Thus, any appeal to unseen motivations or psychological bases of pointing individuals, specifically (and communicating entities, more generally), is as much an attribute of the psychology of the claimant as it is an attribute of the signaler (e.g. Johnson 2001). It is common contemporary practice to analyze patterns of communicative behaviour for evidence of hidden mental processes. For example, if a human child points to an object and alternates her gaze between that distant object and a social partner, this is interpreted to mean that the child is attempting to manipulate the “mind” of their communicative partner and considered to be evidence by many researchers that the child must, therefore, have a conception of others as mental beings (e.g. Baron-Cohen 1999; Tomasello 1995). Because, for many years, it was often erroneously stated that apes did not point to distant entities whilst alternating their gaze between these distant entities and their communicative partners (e.g. Butterworth and Grover 1988; Petitto 1988), this was taken as evidence that apes do not have conceptions of others as mental beings. In fact, as we will argue below, pointing to distant entities with gaze alternation between those entities and an observer is as ambiguous with respect to the signaler’s conceptions when that signaler is a human child as it is when an ape does the same thing. We will describe the human transition to intentional communication and describe a number of behavioural similarities between apes and young human children in their communicative signaling. Finally, we will outline an evolutionary scenario that might account for the near-ubiquity of pointing in human and captive ape populations, and the relative paucity of apparent pointing in wild apes. We turn, first to a brief consideration of how communication and cognition are instantiated in living systems.
2.
Communication is distributed, cognition is communicative
Communication is an interaction between transducing elements. Transduction is the codification of energy into information. Neither broadcast nor receipt of information constitutes a communicative act; communication is a distributed phenomenon, distributed across at least two transducing elements. Thus, cells of the same or different tissue composition within a body may communicate and organisms may communicate (transfer information) with organisms of the same or different species. If caught in an avalanche, the boulder that pins our leg may influence us, but it does not communicate with us, nor do we communicate with that boulder when we push it off our leg; communication requires at least two entities with transductive boundaries. As Gregory Bateson frequently noted (e.g. 1972a: 315), communication is about a “difference that makes a difference.”
Origins of explicit reference 189
Because communication is distributed across boundaries of transduction and because networks of transduction exist at all levels of living systems, from cells to organisms and societies, and because cognition (the discrimination and use of information) is hence an inherently communicative act, therefore it is a category error to interpret communicative behaviour as an index to unseen cognitive processes. Because cognition implies communication (between neurons, between aggregates of neurons, between individuals), therefore it is also a category error to attribute cognitive processes to an individual element, because cognition is a manifestation of communicative processes; i.e, information is distributed across at least two transductive boundaries. It might be argued that networks of communication between neurons comprise functional cognitive systems, or modules, that are properties of individual brains, but because of the distributed nature of cognitive processes, where no cognitive activity can develop or be manifested in a sensorimotor vacuum, all cognitive activity is co-constituted by organisms plus the physical consequences of action and sensation (e.g. Barrett and Henzi 2005; Bateson 1972b; Brinck 2007, this volume; Johnson 2001). We know, as an empirical fact, that organisms are not material objects with clear boundaries, or as William Bateson put it: “We commonly think of animals and plants as matter, but they are really systems through which matter is continually passing” (1906, in C. B. Bateson 1928: 209). The same general principle is true for communication and its special case: cognition. We are systems through which ideas (bits of information) are continually passing. The distributed nature of cognition across the transductive boundary of an individual can be masked by the sometimes deferred nature of environmental input and effects of action, due to memory processes influencing ongoing activity. But the fact that organisms can re-represent (as it were) environmental events only means that experience (or learning history) is important in understanding cognition; it does not mean that cognition can be isolated inside the skull. Gregory Bateson (1972c) expressed it this way: A priori it can be argued that all perception and all response, all behavior and all classes of behavior, all learning and all genetics, all neurophysiology and endocrinology, all organization and all evolution – one entire subject matter – must be regarded as communicational in nature, and therefore subject to the great generalizations or “laws” which apply to communicative phenomena. (1972c: 282–283)
In short, living systems are open systems at all levels of analysis. There are substantive implications for psychology, generally, and comparative psychology, in particular, of the multiscalar dependence of the systems we study, of which we wish to briefly note, here, the widespread dualistic assumption that individual brains constitute the loci for computations, the products of which then cause
190 David A. Leavens et al.
overt behaviour. We note that the communicative phenomena we discuss, here (pointing and its accompaniments), are interactive, distributed phenomena that are usually not manifest except in particular social and physical contexts (e.g. Leavens et al. 1996; Leavens, Hopkins & Bard 2005a), which we describe below. As Johnson (2001) argued, social cognition is manifest in communicative interaction; this suggests that media (contexts), modes, and both individual and relationship histories are all vital components of particular communicative episodes. Communication (construed as manifest cognition, rather than an index to hidden mental processes) is simultaneously a preface and a denouement. Because cognition has been historically defined as the mental processes of an individual, there is a widespread contemporary misconception that cognition is a property of individuals. It is not and cannot be, and therefore a revolution in our traditional approaches to the acquisition, storage, retrieval and use of information is overdue (see e.g. Barrett and Henzi 2005; Bateson 1972c; King 2004; Shanker and King 2002). In what follows, we will occasionally write of individuals making discriminations and displaying evidence for having certain concepts; this is shorthand for describing what organisms do in particular social, cultural, historical, and experimental contexts.
3.
Intentional communication and intersubjectivity
When we speak of intentional communication, we are specifying a sub-class of communication that is manifest in its flexible accommodation to the behavioural state of a social partner (e.g. Bard 1992; Bates, Camaioni and Volterra 1975; Leavens, Russell and Hopkins 2005b; Pika et al. 2005a; Sinha 2004; Sugarman 1984; Tomasello et al. 1994). Implicit within this definition is the idea that an organism who is intentionally communicating can perceive and respond to the independent agency of others and therefore intentional communication is manifest at higher than sub-organismal levels (e.g. Trevarthen 1998). There is a contemporary intellectual fashion towards re-defining intentional communication so as to limit it only to organisms who have concepts of others as mental agents (as contrasted with concepts of others as behavioural agents; e.g. Baron-Cohen 1999; cf. Povinelli, Bering and Giambrone 2000), but because we can perceive or attribute mentality only through manifest behaviour (see e.g. Brinck 2001; Mitchell 2000; Racine 2005), therefore all putative instances of mental state attribution reduce to either (a) behavioural analysis or (b) mental illness. Suppose we tell you that there is . Are we really trying to assert that only crazy people attribute mental states? No. What we are asserting is that there is no essential difference between appeals to mental states as causes of
Origins of explicit reference 191
an organism and ask you, “What does this organism desire/intend/believe, at this moment?” Is any statement about that organism’s intentional or epistemic status rational in the absence of any more information than that the organism exists? Obviously not. We interpret this inescapable opacity of mental states to indicate simply that whatever publicly available information an observer uses to make attributions of mental states must partially constitute (and therefore define) those mental states. The fact that people, the world over, attribute complex motives and beliefs to their pets, other animals, other people, and mythical entities does not constitute evidence for the independent existence of these complex motives and beliefs, which are not available to the senses. Thus, no organism actually discriminates or attributes mentality to other organisms on a purely empirical or inductive basis; people learn, for example, to characterize behaviour of others in symbolic terms that are, by their very nature, distributed within a language-using community. This is not to say that organisms do not perceive regularities in the behaviour of others, nor is it to say that many behavioural regularities are not publicly available; to interpret behaviour in terms of hypothetical constructs such as “belief ” or the notion that mental states are distinct from and, somehow, “cause” behaviour is to make a commitment to a currently fashionable, dualistic model of mental functioning that is historically situated in Western philosophy and its narrative structures (e.g., Gallagher and Hutto this volume; Susswein and Racine this volume). Commonsense models like these are acquired from our cultures, not from inductive observation (cf. Mitchell 2000). behaviour and appeals to, for example, demons as causes of behaviour (e.g. “The Devil made me do it”; see also Mitchell 2000; Susswein and Racine this volume; Thompson 1994). We cannot see, hear, smell, taste, feel, or take photographs, spectrograms, temperatures, weights, volumes or any other measure of either demons or mental states. To attribute behavioural phenomena to either demonic influences or mental state influences is, in both cases, a culturally situated manner of speaking. It is no more or less crazy to appeal to mental states than it is to appeal to demonic possession in describing what organisms do, depending upon the cultural precepts of the individual attempting to account for the behaviour of others. What definitely is irrational is to make these appeals to unseen entities in the complete absence of any behavioural information whatsoever. Thus, if cognitive scientists wish to use concepts like epistemic states in models of psychological processes, then is it incumbent upon them to supply definitions of these hypothetical constructs in measurable terms. . Are we advocating methodological behaviourism? Yes. The following quotation from the recent obituary of Gregory A. Kimble describes his position, with which we are in strong agreement: . . . Kimble believed that the so-called “cognitive revolution” had not in an Oedipal frenzy slain behaviorism, as was proclaimed by some cognitively oriented psychologists late in the 20th century. Instead, Kimble argued, cognitive psychology had not
192 David A. Leavens et al.
The concept of intersubjectivity was predicated on Trevarthen’s (e.g. 1977) observation that even very young babies act differently towards inert objects and animate agents, implying that the discrimination of independent agency occurs within two months of birth. According to Trevarthen, then, babies arrive more-or-less equipped to engage their worlds with two different motivations: a praxic mode for interaction with objects and a communicative mode for interaction with agents. In primary intersubjectivity (roughly 2–5 months of age), babies share emotional attitudes with their social partners, whereas in secondary intersubjectivity (after about 9 months of age) babies share emotional attitudes about events, objects and circumstances external to the dyad. In concrete terms, then, intentional communication is defined by the display of publicly observable behaviour which accommodates to the publicly observable behavioural correlates of intersubjective propensity (expressed emotion, orientation of gaze, etc.). If intentional communication requires the concept of independent agency and if even very young babies manifest this concept as young as 2 months, then why do so many researchers speak of the human developmental transition to intentional communication much later, at 9 months of age? That young babies clearly do discriminate states of engagement in their social partners is demonstrated by the still-face procedure (e.g., Adamson and Frick 2003, for review); babies react to sudden lapses in engagement by their mothers in dyadic contexts and they make bids to re-engage with one or both parents in triadic contexts, long before the traditional transition to intentional communication (see also Fivaz-Depeursinge and Corboz-Warnery 1999; Reddy 2003).
at all avoided the behaviorist requirement that intervening concepts and dependent variables be anchored to observables, in other words, to responses of some kind – be they overt muscular movements, verbal responses, or electrophysiological readings. Cognitive psychology, thus, could not escape its behavioristic roots. (Boneau and Wertheimer 2006: 632) Although none of us would describe ourselves as philosophical behaviourists, we are unanimous in believing that all essential theoretical concepts in cognitive science must rest on patterns of publicly observable behaviour, as broadly defined in this quotation. From a methodological standpoint, cognitive scientists have not escaped the same rigorous requirements for grounding their hypothetical processes in observable behaviour under which behaviourists operate (cf. MacCorquodale and Meehl 1948: esp. 105–106). In short, we believe that scientifically useful concepts can be operationalized, at least in principle (see e.g. Brinck in press). Finally, we note that philosophers, theologians and other scholars are not necessarily subject to the same narrow requirements for public availability of core theoretical concepts to which scientists are required to adhere; we would not like to be construed as implying that only scientific endeavour is worthwhile.
Origins of explicit reference 193
In contrast, most researchers evoke certain novel behavioural capacities that typically emerge at about 9 months of age as definitive of the transition to intentional communication, including pointing and use of other manual gestures to manipulate others to act on the world (e.g., Bates, Camaioni and Volterra 1975; Butterworth 2001, 2003) and, crucially, certain concomitants of manual gestures, such as visual monitoring of the social partner in explicitly triadic contexts (e.g., Bates et al. 1977; Franco and Butterworth 1996; Tomasello 1995). Another behavioural capacity that emerges near the end of the first year of life is the ability to follow another’s gaze or pointing gestures to increasingly specific loci in the environment (e.g. Butterworth and Grover 1988; Lock 2001). Thus, the currently mainstream view of the transition to intentional communication can be characterized as a focus on the dawning of “attentionality”; the capacity to monitor, capture, and redirect the attention of a social partner. Because the discrimination or attribution of intentional behaviour and attentional behavior are both predicated on (a) the specific interactive histories of the organisms involved (the level of trust, the frequency of interaction, the amount of joy, etc.), (b) the ongoing motivational states of the interactants, (c) spatial relations obtaining between the interactants, and (d) specific manifestations of contextual markers (i.e., proxemic, behavioural, and physical correlates of routines; cf. Savage-Rumbaugh 1991), then the central difference between early intentional communication (in the first year of life) and late “attentional communication” (near the end of the first year of life and continuing into the second year) is the development of the capacity to integrate actions on objects with communicative acts directed toward people; this is the advent of Piaget’s sensorimotor stage IV, or coordinated secondary circular reactions (cf. Sugarman 1984), or secondary intersubjectivity (Trevarthen and Hubley 1978). Whether or not one cares to argue that intentionality characterizes babies’ communication throughout the first year of life or that intentionality dawns later with the advent of triadic use and responses to deictic gestures, virtually all researchers agree that there is a developmental elaboration of communicative behaviours in humans, near the end of the first year of life. Despite much debate over whether this pattern is better characterized as a primary discontinuity in cognitive development (e.g. Baron-Cohen 1995; Lock 2001) or the product of continuous processes manifest in a developing organism of maturing motoric capabilities (cf. Moore and Corkum 1994; Reddy 2001, 2003), it is empirically true that, in many cultures, at the end of the 20th Century, babies begin to point to distant events, agents, and objects with gaze alternation between these elements and their social partners by about one year of age (Bates et al. 1975, 1977; Blake, O’Rourke and Borzellino 1994; Franco and Butterworth 1996; Leung and Rheingold 1981; see Butterworth 2001 and Lock 2001, for reviews). In these populations, during
194 David A. Leavens et al.
the second year of life, additional changes occur in how babies deploy their manual gestures and visual orienting behaviour: they become sensitive to whether their social partners are attending to themselves or to distant loci; in other words, there is a well-documented behavioural transition characterized by sensitivity to the behavioural correlates of visual attention in others, or “attentionality” (e.g. Bakeman and Adamson 1986; Franco and Gagliano 2001; O’Neill 1996). By 18 months of age, human babies in the cultures studied, to date, exhibit a robust capability to monitor, capture, and direct the attention of their social partners, through pointing; this is the capacity for explicit reference. In accordance with the introductory remarks, our position is that reporting on the capacity to monitor, capture, and direct attention is to describe typical behavioural development in particular cultural contexts; moreover, organisms can exhibit these capabilities in the absence of any explicit theory of mental functioning (see e.g. Brinck 2003; Doherty 2006; Mitchell 2000).
4.
The phylogeny of explicit reference
4.1
Pointing
Having briefly summarized the ontogeny, or development of explicit reference, we turn now to the phylogeny, or evolutionary history of this capacity. Humankind’s nearest living relatives are the African great apes: gorillas (Gorilla gorilla gorilla, chimpanzees (Pan troglodytes), and bonobos (Pan paniscus). We shared a common ancestor approximately seven million years ago (e.g. Hacia 2001). Orangutans (Pongo pygmaeus) are Asian apes with which we shared a common ancestor approximately 15 million years ago (Hacia 2001). Apes and humans are a group of close relatives that are relatively distantly related to monkeys: monkeys and apes shared a common ancestor approximately 30 million years ago (Steiper, Young and Sukarna 2004). Thus, humans and chimpanzees have approximately 23 million years of shared evolutionary history between the time at which we shared a common ancestor with monkeys and the time at which humans and chimpanzees diverged. To put this another way, the lineage of the last common ancestor of chimpanzees and humans existed for more than 75% of the length of the modern human lineage since the split with monkeys. In general, any strong claim that a species-specific cognitive or behavioural capacity (of which speech is the most salient example) evolved de novo in humans must therefore also claim (a) that the selective contexts in which these traits appeared are strictly limited to recent times (from slightly before the Miocene/Pliocene boundary to the present), (b) that the selective contexts pre-dating the Miocene/Pliocene boundary
Origins of explicit reference 195
are irrelevant to understanding both the evolution and the development of those traits, and therefore (c) the study of our nearest living relatives, the Asian and African great apes, will not produce data relevant to understanding the development in humans of the traits in question. To date, there is only one published report of unambiguous pointing by any wild ape, a bonobo (Veà and Sabater-Pi 1998). In this episode, a male bonobo pointed repeatedly (with outstretched arm and “ring” and “index” fingers; presumably the 2nd and 4th rays) towards the location of several human observers who were partially hidden behind some shrubbery, whilst looking back-andforth between these observers and the rest of his troop. Wild apes do extend their hands towards each other, in various contexts (e.g. van Lawick-Goodall 1968), but it is the triadic use of the outstretched arm and hand that seems to be exceedingly rare. Recently, Pika and Mitani (2006) reported that wild chimpanzees use a characteristic, directed scratching behaviour to elicit grooming at that part of the body from their grooming partners. Thus, although manual pointing by wild apes appears to be extremely rare, the report by Pika and Mitani (2006) suggests that the capacity for explicit reference may be expressed more commonly through different kinds of behaviours. In strong contrast to the rarity of pointing by wild apes, captive apes commonly and spontaneously point in the complete absence of overt training (e.g. Call and Tomasello 1994; Krause and Fouts 1997; Leavens and Hopkins 1998; Leavens et al. 2005a, b; Leavens, Hopkins, and Thomas 2004a; Menzel 1999; Miles 1990; de Waal 1982; Whiten 2000; reviewed by Leavens 2004; Leavens and Hopkins 1999; de Waal 2001). By far the most common context in which captive apes use pointing gestures is one in which they point to desirable, but unreachable food, in the presence of a human observer. A brief digression is warranted about what constitutes pointing behaviour. When human infants point with their whole hands (outstretched arms and most or all fingers extended), researchers have long termed this “reaching” (Blake et al. 1994; Leung and Rheingold 1981; Murphy and Messer 1977; see Leavens and Hopkins 1999, for review). The term “reach” has two primary meanings: (a) to attempt to grasp something and (b) to extend the arm and hand. Neither meaning captures the communicative significance of these gestures, which has been noted by many infancy researchers (e.g., Murphy and Messer 1977; Leung and Rheingold 1981). Franco and Butterworth (e.g. 1996) have employed the term “indicate” or “indicative” for these whole-handed gestures, but this implies the same function as the term “pointing.” For these reasons, we refer to these as “whole-hand points”. (Recently, at a public science lecture, a member of the audience asked, sardonically, if we stayed up nights developing this terminology). In this usage, we join, for example, Kendon and Versante (2003), Haviland (2003),
196 David A. Leavens et al.
and Wilkins (2003) in recognizing that adult humans in diverse cultures do, indeed, point with their whole hands in naturalistic contexts (see photographs and drawings in these sources). Pointing has an attention-directing function, and people can and do point with their eyes, with their lips (Enfield 2001; Wilkins 2003), with their whole hands (Wilkins 2003), and with their index fingers. Some researchers argue that pointing with the index finger has a special status as a human species-specific biological adaptation for definite reference (see, esp., Butterworth 2003), implying that the gesture is derived from our species-specific adaptations for language and speech, but because (a) apes point with their index fingers (see below) and (b) some humans do not point with their index fingers (Wilkins 2003), we believe that both the claim for the species-specificity of pointing and the alleged adaptive derivation of the gesture from adaptations for speech are challenged, at our present state of knowledge. Most captive apes who point, point with their whole hands (Call and Tomasello 1994; Leavens and Hopkins 1998, 1999; Leavens et al. 2004a; de Waal 1982). That these are communicative signals and not attempts to reach for obviously unreachable food is demonstrated by the necessity of an audience for the display of these gestures (Call and Tomasello 1994; Hostetter, Cantero and Hopkins 2001; Leavens et al. 1996; Leavens et al. 2004a; see Table 1). Thus, apes in captivity do not point to obviously unreachable items, with either their index fingers or with all fingers extended, in the absence of a human observer; these are communicative signals, not abbreviated reaches for obviously unreachable food.
Table 1. Apes require an audience to display points and other manual gestures in the presence of unreachable food: Summary of experimental studies. Study
Species
N (subjects)
% Presencea
Call and Tomasello (1994) Leavens, Hopkins, and Bard (1996) Hostetter, Cantero, and Hopkins (2001) Leavens, Hopkins, and Thomas (2004a) – Visible Banana Condition – Hidden Banana Condition – Experiment 2
Orangutans Chimpanzees Chimpanzees Chimpanzees
2 3 49
95b 99 97
101 101 35
98 98 100
Notes. aThis is the percent of trials in which subjects pointed (Call and Tomasello 1994), percent of gestures, some of which were points (Leavens et al. 1996; Hostetter et al. 2001), or percent of subjects who gestured, including those who pointed (Leavens et al. 2004a) in the presence, as compared to the absence of human observers. bIn 48 of the total of 96 experimental trials in this study, the human observer, although present, either had his eyes closed or was facing away from the subjects.
Origins of explicit reference 197
However, there is considerable diversity in the preferred form of pointing across captive groups of apes: language-trained apes point overwhelmingly with their index fingers (Figure 1). As Figure 1 shows, language-trained chimpanzees point preferentially with their index fingers. Similar observations were reported for two orangutans, Puti (not language-trained) and Chantek (language-trained) by Call and Tomasello (1994). There are numerous observations of language-trained apes that support the generalization that they point primarily, or at least very frequently, with their index fingers (Bodamer and Gardner 2002; Call and Tomasello 1994; Krause and Fouts 1997; Menzel 1999; Miles 1990; Savage-Rumbaugh 1986; Whiten 2000). It is not clear which aspects of language-training result in this group difference between different populations of captive apes, but it is clear that apes who have more prolonged and direct interactions with humans point more frequently with their index fingers than do apes who are raised with less human contact. Thus, for captive apes, the preferred form of pointing is attributable to differential environmental influences on communicative development; the form of pointing is attributable to epigenetic (i.e., other than exclusively genetic) processes (see Leavens 2004 and Leavens et al. 2005b, for elaboration of this specific point, and Sinha 2004, for more general considerations of epigenetic effects on human communicative development).
Figure 1. Indexicality index of pointing in captive chimpanzees: Differences in percentages of index-finger to whole-hand extensions to objects distal to both chimpanzees and their human observers. Negative numbers reflect a majority of whole-hand extensions, whereas positive numbers reflext a preponderance of index-finger extensions. Sources: Language-naive chimpanzees – Leavens et al. 1996, and a re-analysis of data reported in Leavens and Hopkins 1998; Language-trained chimpanzees – data from Kause and Fouts 1997, Experiments 1 and 2 combined. Figure and caption adapted from Leavens and Hopkins (1999, their Figure 1).
198 David A. Leavens et al.
4.2 Gaze alternation In humans, alternation of gaze while gesturing in triadic contexts is a defining behavioural criterion for the development of mature intentional communication in the second year of life (e.g. Bates et al. 1975, 1977; Franco and Butterworth 1996; Leung and Rheingold 1981). That human infants do alternate their gaze between distant entities and their human caregivers while pointing is well-established, but what it might signify for the infants, themselves, is really quite unclear. A rich interpretation of this behaviour was offered by Tomasello (1995) who argued that gaze alternation while pointing signifies “that the child understands that the adult is a separate person who has intentions and attention that may differ from its own” (p. 109), largely because he claims that babies direct affectively laden facial expressions toward adults, but not the objects to which they point. Empirically, this turns out not to be the case: babies frequently smile while pointing to distant entities before they turn to look at their social partners (e.g. Jones and Hong 2001; Leavens and Todd unpublished raw data). In an earlier paper (Leavens et al. 1996), we also suggested that this gaze alternation implied an awareness by the signaler that the recipient of the gesture had a distinct visual perspective, in accordance with Tomasello’s (1995) claim. Since that time, we have rejected this interpretation (e.g. Leavens 2004). A more cautious interpretation of this visual orienting behaviour is that babies are monitoring the effect of their gestures, implying that they have some expectations that their communicative bids will have social or instrumental consequences. In our studies of captive apes, housed in a research facility, we find that between 85% and 100% of individuals who gesture in the context of unreachable food also display gaze alternation between the food and the social partner (Table 2). As Table 2 illustrates, it takes human infants two years to reach the same levels of accompanying gaze alternation displayed by chimpanzees; unfortunately, there are no relevant data, to our knowledge, on the early development of gaze alternation in chimpanzees. Nevertheless, it is clear that chimpanzees raised in what may be considered to be impoverished conditions, nevertheless acquire this pattern of visual orientation in the absence of any explicit training to do so, just as people do. Also like people, chimpanzees in captivity monitor facial expressions of their human caregivers in social referencing contexts (Russell, Bard and Adamson 1997), thus alternating their gaze between social partners and distant entities in both information-providing and information-seeking contexts.
Origins of explicit reference 199
Table 2. Comparison between chimpanzees and human infants of the percent of subjects displaying gestures accompanied by gaze alternation. Species
Studya
Chimpanzees Leavens and Hopkins (1998) Leavens et al. (2004a) – Visible Banana Condition – Hidden Banana Condition – Experiment 2 Leavens et al. (2005b) – Predelivery: Banana – Predelivery: Half-Banana – Predelivery: Chow Humans Bates et al. (1977) – 9.5 Months – 10.5 Months – 11.5 Months – 12.5 Months Lempers (1979) – 9.0 Months – 12.0 Months – 14.0 Months Desrochers, Morissette, and Ricard (1995) – 9.0 Months – 12.0 Months – 15.0 Months – 18.0 Months – 24.0 Months
N (Subjects) % Subjects w/GAb 78
87
76 73 11
86 85 91
22 20 24
91 100 92
25 25 25 25
0 28 36 56
36 36 36
8 8 64
25 25 25 25 25
0 13 54 79 100
Notes. aAll chimpanzees sampled from the same population at the Yerkes National Primate Research Center, Atlanta, Georgia, U.S.A., and aged between 3 and 56 years. All human gestures are index-finger points. Methodological differences in the assessment of visual orienting behaviour between studies of apes and humans render these comparisons more qualitative than quantitative. Specifically, gaze alternation in apes was defined as looks to an experimenter during an observational interval that varied substantially between subjects and studies (typically, these observation intervals ranged between 1s and 60s), whereas human visual orienting was typically defined as looks to caregivers from 1s before point onset to 1s after point termination. b“Subjects w/GA” means “percent of subjects gesturing with gaze alternation.”
200 David A. Leavens et al.
4.3 Sensitivity to attentional status of an observer There is a widespread misconception that captive apes are relatively insensitive to the attentional status of others (e.g. Povinelli et al. 2000, 2003a), despite numerous demonstrations to the contrary. Table 3 lists a representative sampling of experimental and observational demonstrations of the sensitivity of chimpanzees to the visual attention of both conspecific and human observers. These studies include captive apes from both ends of an “enrichment” spectrum (i.e. the studies in Table 3 used apes in biomedical research centers and zoos, as well as much more socially enriched language-trained apes). In short, as the studies listed in Table 3 make clear, apes discriminate direct gaze, they follow gaze, and they selectively deploy their communicative signals in accordance with the attentional status of an observer. Apes without any special training display a procedural awareness of the “attentionality” that characterizes human infant communication between approximately 9 and 18 months of age. A recent spate of claims to the effect that in their pre-verbal signaling behaviour human infants evince evidence for some additional representational capacity not available to non-human primates is subject to criticism by appeal to the empirical data which clearly and almost unanimously demonstrate that apes also discriminate visual attention in their social partners. Table 3. Apes are sensitive to the visual attention of an observer: Representative studies. These studies variously demonstrated that apes discriminated direct gaze, followed the gaze of others, or displayed visual signals selectively when social partners were looking at them. Type
Study
Species
Observational Observational Observational Observational Observational Observational Observational Observational Observational Experimental Experimental Experimental Experimental Experimental Experimental
Tanner and Byrne (1993) Tomasello et al. (1994) Tanner and Byrne (1996) Tomasello et al. (1997) Pika, Liebal, and Tomasello (2003) Liebal, Call, and Tomasello (2004a) Liebal, Pika, and Tomasello (2004b) Pika, Liebal, and Tomasello (2005) Liebal, Pika, and Tomasello (2006) Call and Tomasello (1994) Itakura (1996) Povinelli and Eddy (1996a) Povinelli and Eddy (1996b) Krause and Fouts (1997) Povinelli and Eddy (1997)
Gorillaa Chimpanzees Gorilla Chimpanzees Gorillas Chimpanzees Siamangs Bonobos Orangutans Orangutans Orangutan Chimpanzeesb Chimpanzees Chimpanzees Chimpanzees
Origins of explicit reference 201
Table 3 (continued) Type
Study
Experimental Experimental Experimental Experimental Experimental Experimental Experimental Experimental Experimental Experimental Experimental Experimental Experimental Experimental Experimental
Tomasello, Call, and Hare (1998) Itakura and Tanaka (1998) Itakura, Agnetta, Hare and Tomasello (1999) Peignot and Anderson (1999) Povinelli, Bierschwale, and Cech (1999) Tomasello, Hare, and Agnetta (1999) Hare, Agnetta, Call, and Tomasello (2000) Hare, Call, and Tomasello (2001) Hostetter, Cantero, and Hopkins (2001) Bodamer and Gardner (2002) Okamoto et al. (2002) Povinelli, Theall, Reaux, and Dunphy-Lelii (2003) Liebal, Pika, Call, and Tomasello (2004c) Leavens, Hostetter, Wesley, and Hopkins (2004b) Braüer, Call, and Tomasello (2005)
Experimental Experimental Experimental Experimental
Species
Chimpanzees Chimpanzees, Orangutan Chimpanzees Gorillasc Chimpanzeesd Chimpanzees Chimpanzees Chimpanzees Chimpanzees Chimpanzees Chimpanzee Chimpanzees Chimpanzees Chimpanzees Bonobos, Chimpanzees, Gorillas, Orangutans Melis, Call, and Tomasello (2006) Chimpanzees Poss, Kuhar, Stoinski, and Hopkins (2006) Gorillas, Orangutans Hostetter, Russell, Freeman, and Hopkins (2007) Chimpanzees Hopkins, Russell, and Leavens (In press) Chimpanzees
Notes: aA gorilla covered her facial expression with her hand, implying an awareness of the social consequences of her expressions in the visual domain. bThis study is often cited as evidence against discrimination of human gaze by chimpanzees but, in fact, the chimpanzees readily discriminated human gaze in almost all experimental contexts either spontaneously or with a modicum of training. cThe gorillas in this study readily discriminated head orientation, but not eyes only. dThe authors interpreted their data to indicate that chimpanzees lacked a “high-level” model of mental functioning, but the chimpanzees in this study outperformed the human children in an object-choice task, using aspects of an experimenter’s attentional cues and, furthermore, reliably followed the experimenter’s gaze to a point behind them.
There are two empirical grounds on which some researchers base claims for a uniquely human cognitive adaptation in the domain of non-verbal communication: (a) an occasional absence of discrimination by apes of the focus of the eyes, specifically, and (b) the phenomenon of pointing to distant events or objects by human babies in the apparent absence of any attempt by the baby to manipulate the social partner to act on that distal element (so-called “protodeclarative” communication, Bates et al. 1975; see also Brinck this volume; Susswein and Racine this volume). With respect to discrimination of eye direction, we have argued that “whether discriminations are based on eye direction, head orientation, or other
202 David A. Leavens et al.
postural cues is irrelevant to the cognitive implications, which are simply that humans and great apes discriminate different states of visual attention in others” (Leavens et al. 2005a: 294). In other words, if an organism can use the behavioural correlates of visual attention in social agents to exercise choice over their modality of signaling, to find food, to effectively manipulate others to retrieve food, etc. – all of which have been well-demonstrated in apes – then the concept of visual attention is manifest in the interplay of that organism with its social environment, and this is true irrespective of the specific behavioural cues that organism might use (see Table 3). Every extant, published, alleged species difference between apes and humans in the capacity to discriminate and use visual attention in others is predicated on an experimental confound between early rearing history and species classification: apes are typically orphans raised in cages without primary, stable adult attachment figures and humans are raised by their biological parents in rich environments filled with laughter, joy, and frequent face-to-face interaction with their primary caregivers. Consider the following thought experiment: raise human boys from birth in the same relatively impoverished circumstances in which captive apes are typically raised. Let the comparison group be human girls raised by their biological parents in their homes, who are cherished, and unreservedly and reliably loved by their caregivers. Years later, assess the sensitivity of the boys and the girls to subtle cues of visual attention in human adults. Suppose the girls, unsurprisingly, perform better than the boys – would any researcher in their right mind attribute the difference to a gender difference between boys and girls? Of course not, rearing history is clearly confounded with the gender of the subjects. Yet substitute apes for boys and humans for girls in this research design and how often have researchers trumpeted a “species difference” between apes and humans in various aspects of sensitivity to visual attention (e.g. Povinelli and Eddy 1996a; Theall and Povinelli 1999)? If the practice of almost completely ignoring the effects of pre-experimental experience (or what used to be called the “preparation” of the organisms under scrutiny) were not so widespread in contemporary comparative psychology, this would be laughable (for a notable exception to this general methodological failing, see Carpenter, Tomasello and Savage-Rumbaugh 1995). The reports listed in Table 3 adequately demonstrate that captive apes discriminate different states of visual attention in others, despite the impoverishment of their early rearing histories. With respect to so-called protodeclarative pointing, there are very few reports of pointing by apes with the apparent goal of merely directing the attention of their social partners to distant goals; this scarcity has been noted by, among others, Baron-Cohen (1999), Butterworth (2001), Povinelli et al. (2003b), and Tomasello (1999). These authors suggest that the absence or scarcity of protodeclarative pointing in apes and in humans with autism is diagnostic of an inability
Origins of explicit reference 203
to represent the perspectives or mental states of others. However, in the original formulation by Bates et al. (1975), protoimperatives were described as pre-verbal attempts by human babies to elicit action from a social partner, and protodeclaratives were described as attempts to elicit positive emotional engagement from a social partner. Thus, in the original formulation, both protoimperative and protodeclarative pointing were presented as instrumental, imperative gestures, differing only in the apparent goals of the signaler (retrieval of objects and social responses, respectively). Hence, the term “protodeclarative” has undergone an equivocation, changing from a label for an instrumental act that signifies the same cognitive processes as protoimperative pointing, to a label for a communicative act that indexes a nascent theory of mind. This has occurred in the absence of any significant new empirical findings on typical human communicative development; indeed, recent research confirms the necessity of positive emotional engagement to satisfy babies who point protodecaratively (Liszkowski et al. 2004). If, as Bates and her colleagues originally claimed, babies sometimes point to elicit positive affective responses from their caregivers, then any apparent absence of such pointing in certain psychopathological human populations, or in nonhuman populations, can only signify a difference in motivation, not a difference in representational capacities, as the latter are not implicated in protodeclarative pointing. If human babies typically receive positive emotional consequences to their signaling behaviour, it is not implausible to suggest that they may increasingly act to bring about such consequences as they mature. Therefore, we suggest that if these kinds of consequences are necessary for the development of protodeclarative pointing in human children, then it is reasonable to suggest that humans learn to point protodeclaratively (cf. Moore and Corkum 1994). In other words, humans may develop protodeclarative pointing in human species-typical caregiving contexts that may not require human species-unique cognitive capacities. Moreover, empirically, there are several reports of apparent declarative pointing by apes (pointing in the absence of any evidence that apes are attempting to instrumentally manipulate a social partner to act on the indicated element): the single report of pointing by a wild ape was an apparent declarative (Veà and Sabater-Pi 1998), and there are numerous reports of apparently declarative pointing by language-trained apes (e.g. Miles 1990; Savage-Rumbaugh et al. 1998). These language-trained apes are notable in particular for having experienced unusually close emotional bonds with human caregivers; the putative species difference in propensity to point declaratively may, therefore, be attributable to differences in the degree of exposure to particular kinds of caregiving practices. To put this another way, humans may inculcate the motivation to share attention to distant events by making such shared attention reinforcing through extravagant displays of contingent joy. Organisms with histories of joyful consequences to shared attention might
204 David A. Leavens et al.
reasonably be expected to instigate such episodes in the future. The much-touted “species difference” in propensity to point declaratively may simply reflect different degrees of exposure to some, particularly Western, caregiving practices.
5.
Heterochrony and the Referential Problem Space
Both the propensity to point and the form of pointing in captive apes are influenced by rearing history (as may be also the motivation to engage in declarative behaviour). The propensity to point with outstretched arms and fingers can be characterized as a cultural difference between wild and captive apes, in the same way that differential propensities to point with the lips is a human cultural difference. Apes are, thus, malleable in their gestural repertoires (e.g. Bard 1998; Leavens et al. 2005b; Pika et al. 2005a; Tomasello et al. 1994). Given this flexibility, what is it about captive environments, which are so impoverished in many respects, that fosters the development of pointing in captive apes, in the absence of explicit training? We believe a plausible answer lies in consideration of the circumstances in which human infants begin to point. Human infants have both endogenous and exogenous barriers to free movement. With considerable inter-individual variability, humans do not achieve bipedal locomotion until approximately a year of age and mastery of this mode of locomotion takes several years (e.g. Cheron et al. 2001). Chimpanzees, on the other hand, are capable of independent quadrupedal locomotion (technically known as “knucklewalking”) by about five months of age (van Lawick-Goodall 1968). The significant delay in locomotor development in our species, relative to other primates, is largely attributable to maturational factors affecting the stability of the trunk in a vertical mode – humans are biomechanically unstable in this posture for much of the first year of life (e.g., Adolph and Berger 2005). Direct evidence for bipedal locomotion in hominids (Australopithecus afarensis) dates to over 3.5 million years ago with the footprints in the Laetoli lava beds (Leakey and Hay 1979) and more controversial claims exist for bipedal locomotion in much older hominids dated to nearly twice that old (Orrorin tugenensis, Senut, et al. 2001). It is an open question whether the relatively protracted, peripatetic ontogeny of locomotor development that characterizes modern humans was also characteristic of the earliest bipedal hominids or whether this is a more recent feature of human development related to the very oversized heads of infant representatives of the genus Homo, dating from about 2.5 million years ago. The difference between humans and apes in the attainment of independent locomotion constitutes a change in our lineage in the relative timing of this motoric competency, or heterochrony. In addition to these endogenous limitations that uniquely affect human babies,
Origins of explicit reference 205
secondary consequences of bipedal locomotion in adult caregivers give rise to widespread exogenous barriers to free movement in babies: babies are carried in restraining devices or left physically restrained in a variety of settings, such as cribs or feeding chairs, typically for their own safety (e.g. Super 1990). At the age at which modern human babies begin to point, near the end of the first year of life, they exhibit novel capacities for means-ends reasoning, or tool use (Bates, Thal and Marchman 1991; Leavens 2004; Sugarman 1984). Thus, at the same age at which babies begin to point and to otherwise use communication instrumentally, there is a concomitant advent of the use of indirect means to achieve goals, characteristic of late Piagetian sensorimotor sub-stage IV (coordinated secondary circular reactions) and early sub-stage V (tertiary circular reactions) cognitive development. These capacities are also well-demonstrated in the great apes (e.g. Bard 1990; Gibson 1996; Parker 1999; Potì and Spinozzi 1994). However, in the wild, because apes develop independent locomotor competence many months prior to the advent of means-ends reasoning, they are not dependent upon others to act on the world for them. In contrast, human infants experience multitudinous barriers to the direct attainment of distant objects and are reliant upon others to retrieve those objects for them: this is the Referential Problem Space, a related series of circumstances in which babies are dependent upon the successful capture and re-direction of the attention of their caregiver to specific loci for instrumental ends (Figure 2). In order to obtain distant items, babies manipulate their caregivers to deliver them, and this requires a means to capture and re-direct the attention of others. Because infants have long histories in which their caregivers have retrieved distant items for their manipulation, the caregivers become established means to an end. The innovation of pointing is that it combines an established means (caregiver) with a novel means (pointing) to numerous established and novel ends. When apes are raised in captivity, we put them directly into the Referential Problem Space. Captive apes have no access to food without the direct provisioning by human caregivers, so histories of dependencies upon caregivers are wellestablished in these populations. Pointing to request food or other items develops in this problem space (e.g. Call and Tomasello 1994; Krause and Fouts 1997; Leavens and Hopkins 1998; Leavens et al. 1996, 2004a, 2005a, b; Figure 2). Because wild apes develop early locomotor competence, they circumvent the Referential Problem Space: they are never reliant upon others to retrieve distant objects for them. They go on to manifest their problem-solving capacities in well-documented foraging contexts: fishing for termites, using stones to crack nuts, etc., but pointing or otherwise manipulating others to act vicariously on the world does not develop. In contrast, both human infants and captive chimpanzees face the Referential Problem Space: they cannot retrieve distant objects except through
206 David A. Leavens et al.
Figure 2. The Referential Problem Space. Wild chimpanzess, because of their ability to independently travel to virtually any object of interest, do not need to apply their problem-solving skills to referential contexts (downward-pointing arrow at 4–5 months indicates the onset of independent locomotion). Apes go on to display means-ends problem-solving capacities in foraging domains, but do not develop pointing as an instrumental tool because it is not required in those contexts (but see Pika and Mitani 2006). In contrast, humans and captive chimpanzees experience both barriers to direct attainment of desirable objects and long histories in which caregivers deliver desirable objects to them. By virtue of humans’ long-delayed development of bipedal locomotion (upward-pointing arrow), they are restricted in movement and dependent upon others to act on their world at a time in development in which they have sophisticated problemsolving capacities. Pointing emerges, then, in this problem space.
manipulation of others and they have the advanced sensorimotor problem-solving capacities to use existing means (human caregivers) to act on the world for them (Bard 1990; Leavens 2004). One implication of this hypothesis is that there is no human-specific adaptation for definite reference through non-verbal means. This suggestion runs counter to an existing body of theory that either explicitly or implicitly construes non-verbal reference as being derived from human species-specific cognitive adaptations for symbolic reference (e.g. Butterworth 2003). Because of the heterochronic changes
Origins of explicit reference 207
in locomotor independence in our lineage, possibly long before the development of speech our ancestors experienced novel ontogenetic conundrums for which the development of referential behaviour had tangible payoffs. Because apes in similar circumstances also spontaneously point, this implies that a trajectory into referential communication does not require adaptations for speech or adaptations for bipedal locomotion, as apes have evolved neither capacity. According to this view, pointing derives from an interaction between a particular set of environmental circumstances (the Referential Problem Space) and cognitive capacities for means-ends reasoning that are shared by humans and the great apes. Babies and some apes who point and then experience very positive emotional responses from their caregivers to this pointing, may come to point “declaratively,” yet nonetheless instrumentally to elicit these states of positive mutual engagement; i.e., they may generalize their pointing to request action on distant objects to contexts in which pointing leads to positive emotional, rather than physical reinforcement, when the social environment provides those affective contingencies (Leavens 2004; Moore and Corkum 1994). Thus, we do not believe that pointing by captive apes implies any cognitive capacity not also manifest in wild apes, in problem-solving contexts. Indeed, we believe one of the reasons captive apes so frequently point to request food from human observers may be because, in many captive contexts, humans are so oblivious to subtle indicators of gaze by chimpanzees that these apes are forced to deploy extraordinarily explicit means of capturing and re-directing human attention. Through pointing, apes are able to scaffold humans into more responsive interactions (cf. de Waal 2001).
Acknowledgements We thank the editors of this volume, Jordan Zlatev, Tim Racine, Chris Sinha, and Esa Itkonen for inviting this chapter. Special thanks to Tim Racine for his helpful feedback on several earlier versions of this manuscript. We gratefully acknowledge inspirational conversations about pointing and related matters with Robin Banerjee, Irwin Bernstein, Joanna Blake, Ingar Brinck, the late George Butterworth, Josep Call, Robert Corruccini, Deborah Custance, Susan Ford, Dorothy Fragaszy, Fabia Franco, Janet Frick, Juan-Carlos Gomez, R. Peter Hobson, Autumn Hostetter, Jim Hurford, Mark Krause, Maria Legerstee, Katja Liebal, Ulf Liszkowski, Chris Moore, B. E. Mulligan, Simone Pika, Vasu Reddy, Connie Russell, Jamie Russell, Chris Sinha, Roger Thomas, Mike Tomasello, Colwyn Trevarthen, Katherine Whitcome, Nicola Yuill, and many others.
208 David A. Leavens et al.
References Adamson, L.B. and Frick, J.E. 2003. “The Still-face: A history of a shared experimental paradigm.” Infancy 4: 451–473. Adolph, K.E. and Berger, S.E. 2005. “Physical and motor development.” In Developmental Science: An Advanced Textbook, 5th Ed., M.H. Bornstein and M.E. Lamb (eds.), 223–281. Mahwah, NJ: Lawrence Erlbaum Associates. Bakeman, R. and Adamson, L.B. 1986. “Infants’ conventionalized acts: Gestures and words with mothers and peers.” Infant Behavior and Development 9: 215–230. Bard, K.A. 1990. “’Social tool use’ by free-ranging orangutans: A Piagetian and developmental perspective on the manipulation of an animate object.” In “Language” and Intelligence in Monkeys and Apes: Comparative Developmental Perspectives, S.T. Parker and K.R. Gibson (eds.), 356–378. Cambridge: Cambridge University Press. Bard, K.A. 1992. “Intentional behavior and intentional communication in young free-ranging orangutans.” Child Development 62: 1186–1197. Bard, K.A. 1998. Social-experiential contributions to imitation and emotion in chimpanzees. In Intersubjective Communication and Emotion in Early Ontogeny, S. Bråten (ed.), 208–227. Cambridge: Cambridge University Press. Baron-Cohen, S. 1995. Mindblindness: An Essay on Autism and Theory of Mind. Cambridge, Mass.: MIT Press. Baron-Cohen, S. 1999. “The evolution of a theory of mind.” In The Descent of Mind: Psychological Perspectives on Hominid Evolution, M.C. Corballis and S.E.G. Lea (eds.), 261–277. Oxford, UK: Oxford University Press. Barrett, L. and Henzi, P. 2005. “The social nature of primate cognition.” Proceedings of the Royal Society B 272:1865–1875. Bates, E., Benigni, L., Bretherton, I., Camaioni, L. and Volterra, V. 1977. “From gesture to the first word: On the nature of cognitive and social prerequisites.” In Interaction, Conversation and the Development of Language, M. Lewis and L. Rosenblum (eds.), 247–307. New York: John Wiley and Sons. Bates, E., Camaioni, L. and Volterra, V. 1975. “Performatives prior to speech.” Merrill-Palmer Quarterly 21: 205–226. Bates, E., Thal, D. and Marchman, V. 1991. “Symbols and syntax: A Darwinian approach to language development.” In Biological and Behavioral Determinants of Language Development, N.A. Krasnegor, D.M. Rumbaugh, R.L. Schiefulbusch, and M. Studdert-Kennedy (eds.), 29–65. Hillsdale, NJ: Erlbaum. Bateson, C.B. 1928. William Bateson, F.R.S.: His Essays and Addresses, together with a Short Account of his Life. Cambridge: Cambridge University Press. Bateson, G. 1972a. “The Cybernetics of ‘Self ’: A Theory of Alcoholism.” In Steps to an Ecology of Mind, G. Bateson (ed.), 309–337. New York: Ballantine Books. [Original article published 1971 in Psychiatry 34: 1–18] Bateson, G. 1972b. “Form, Substance, and Difference.” In Steps to an Ecology of Mind, G. Bateson (ed.), 448–465. New York: Ballantine Books. [Original article published 1970 in General Semantics Bulletin, no. 37, Institute of General Semantics] Bateson, G. 1972c. “The Logical Categories of Learning and Communication.” In Steps to an Ecology of Mind, G. Bateson (ed.), 279–308. New York: Ballantine Books.
Origins of explicit reference 209
Blake, J., O’Rourke, P. and Borzellino, G. 1994. “Form and function in the development of pointing and reaching gestures.” Infant Behavior and Development 17: 195–203. Bodamer, M.D. and Gardner, R A. 2002. “How cross-fostered chimpanzees (Pan troglodytes) initiate and maintain conversations.” Journal of Comparative Psychology 116: 12–26. Boneau, C.A. and Wertheimer, M. 2006. “Gregory A. Kimble (1917–2006)” [Obituary]. American Psychologist 61: 632–633. Braüer, J., Call, J. and Tomasello, M. 2005. “All primates species follow gaze to distant locations and around barriers.” Journal of Comparative Psychology 119: 145–154. Brinck, I. 2001. “Attention and the evolution of intentional communication.” Pragmatics and Cognition 9: 255–272. Brinck, I. 2003. “The pragmatics of imperative and declarative pointing.” Cognitive Science Quarterly 3: 429–446. Brinck, I. 2007. “Situated cognition, dynamic systems, and art.” Janus Head, 9, 407–431. Brinck, I. In press. From similarity to uniqueness: Method and theory in comparative psychology. In Learning from animals? Examining the Nature of Human Uniqueness. L.S. RöskaHardy and E.M. Neumann-Held (eds.), London: Psychology Press. Butterworth, G. 2001. “Joint visual attention in infancy.” In The Blackwell Handbook of Infant Development, J.G. Bremner and A. Fogel (eds.), 213–240. Hove, U.K.: Blackwell. Butterworth, G. 2003. “Pointing is the royal road to language for babies.” In Pointing: Where Language, Culture, and Cognition Meet, S. Kita (ed.), 9–33. Mahwah, NJ: Erlbaum. Butterworth, G. and Grover, L. 1988. “The origins of referential communication in human infancy.” In Thought without Language, L. Weiskrantz (ed.), 5–24. Clarendon Press: Oxford. Call, J. and Tomasello, M. 1994. “Production and comprehension of referential pointing by orangutans (Pongo pygmaeus).” Journal of Comparative Psychology 108: 307–317. Carpenter, M., Tomasello, M. and Savage-Rumbaugh, S. 1995. “Joint attention and imitative learning in children, chimpanzees, and enculturated chimpanzees.” Social Development 4: 217–237. Cheron, G., Bouillot, E., Dan, B., Bengoetxea, A., Draye, J.-P. and Lacquaniti, F. 2001. “Development of a kinematic coordination pattern in toddler locomotion: Planar covariation.” Experimental Brain Research, 137: 455–466. Doherty, M. 2006. “The development of mentalistic gaze understanding.” Infant and Child Development 15: 179–186. Enfield, N. J. 2001. “‘Lip-pointing’: A discussion of form and function with reference to data from Laos.” Gesture 1: 185–212. Fivaz-Depeusinge, E. and Corboz-Warnery, A. 1999. The Primary Triangle: A Developmental Systems View of Mothers, Fathers, and Infants. New York: Basic Books. Franco, F. and Butterworth, G. 1996. “Pointing and social awareness: Declaring and requesting in the second year.” Journal of Child Language 23: 307–336. Franco, F. and Gagliano, A. 2001. “Toddlers’ pointing when joint attention is obstructed.” First Language 21: 289–321. Gibson, K. 1996. “The ontogeny and evolution of the brain, cognition, and language.” In Handbook of Human Symbolic Evolution, A. Lock and C.R. Peters (eds.), 407–431. Hove, U.K.: Blackwell. Hacia, J.G. 2001. “Genome of the apes.” Trends in Genetics 17: 637–645. Hare, B., Call, J., Agnetta, B. and Tomasello M. 2000. “Chimpanzees know what conspecifics do and do not see.” Animal Behaviour 59: 771–785.
210 David A. Leavens et al.
Hare, B., Call, J. and Tomasello, M. 2001. “Do chimpanzees know what conspecifics know?” Animal Behaviour 61: 139–151. Haviland, J.B. 2003. “How to point in Zincatán.” In Pointing: Where Language, Culture, and Cognition Meet, S. Kita (ed.), 139–169. Hillsdale, NJ: Erlbaum. Hopkins, W.D., Russell, J.L. and Leavens, D.A. In press. “Multi-modal communication and its social contextual use in captive chimpanzees (Pan troglodytes).” Animal Cognition. Hostetter, A.B., Cantero, M. and Hopkins, W.D. 2001. “Differential use of vocal and gestural communication in response to the attentional status of a human.” Journal of Comparative Psychology 115: 337–343. Hostetter, A.B., Russell, J.L., Freeman, H. and Hopkins, W.D. 2007. “Now you see me, now you don’t: Evidence that chimpanzees understand the role of the eyes in attention.” Animal Cognition, 10, 55–62. Itakura, S. 1996. “An exploratory study of gaze monitoring in nonhuman primates.” Japanese Psychological Research 38: 174–180. Itakura, S. and Tanaka, M. 1998. “Use of experimenter-given cues during object-choice tasks by chimpanzees (Pan troglodytes), an orangutan (Pongo pygmaeus), and human infants (Homo sapiens).” Journal of Comparative Psychology 112: 119–126. Itakura, S., Agnetta, B., Hare, B. and Tomasello, M. 1999. “Chimpanzee use of human and conspecific social cues to locate hidden food.” Developmental Science 2: 448–456. Johnson, C.M. 2001. “Distributed primate cognition: A review.” Animal Cognition 4: 167–183. Jones, S.S. and Hong, H.-W. 2001. “Onset of voluntary communication: Smiling looks to mother.” Infancy 2: 353–370. Kendon, A. and Versante, L. 2003. “Pointing by hand in ‘Neapolitan.’” In Pointing: Where Language, Culture, and Cognition meet, S. Kita (ed.), 109–137. Hillsdale, NJ: Erlbaum. King, B.J. 2004. The Dynamic Dance: Nonvocal Communication in African Great Apes. Cambridge, Mass.: Harvard University Press. Krause, M.A. and Fouts, R.S. 1997. “Chimpanzee (Pan troglodytes) pointing: Hand shapes, accuracy, and the role of eye gaze.” Journal of Comparative Psychology 111: 330–336. Lawick-Goodall, J. van 1968. Behaviour of free-living chimpanzees in the Gombe Stream area. Animal Behaviour Monographs 1: 163–311. Leakey, M.D. and Hay, R.L. 1979. “Pliocene footprints in the Laetolil Beds, at Laetoli, northern Tanzania.” Nature 278: 317–328. Leavens, D.A. 2004. “Manual deixis in apes and humans.” Interaction Studies 5: 387–408. Leavens, D.A. and Hopkins, W.D. 1998. “Intentional communication by chimpanzees: A crosssectional study of the use of referential gestures.” Developmental Psychology 34: 813–822. Leavens, D.A. and Hopkins, W.D. 1999. “The whole hand point: The structure and function of pointing from a comparative perspective.” Journal of Comparative Psychology 113: 417– 425. Leavens, D.A., Hopkins, W.D. and Bard, K.A. 1996. “Indexical and referential pointing in chimpanzees (Pan troglodytes).” Journal of Comparative Psychology 110: 346–353. Leavens, D.A., Hopkins, W.D. and Bard, K.A. 2005a. “Understanding the point of chimpanzee pointing: Epigenesis and ecological validity.” Current Directions in Psychological Science 14: 185–189. Leavens, D.A., Hopkins, W.D. and Thomas, R.K. 2004a. “Referential communication by chimpanzees (Pan troglodytes).” Journal of Comparative Psychology 118: 48–57.
Origins of explicit reference 211
Leavens, D.A., Hostetter, A.B., Wesley, M.J. and Hopkins, W.D. 2004b. “Tactical use of unimodal and bimodal communication by chimpanzees, Pan troglodytes.” Animal Behaviour 67: 467–476. Leavens, D.A., Russell, J.L. and Hopkins, W.D. 2005b. “Intentionality as measured in the persistence and elaboration of communication by chimpanzees (Pan troglodytes).” Child Development 76: 291–306. Leung, E.H.L. and Rheingold, H.L. 1981. “Development of pointing as a social gesture.” Developmental Psychology 17: 215–220. Liebal, K., Call, J. and Tomasello, M. 2004a. “Use of gesture sequences in chimpanzees.” American Journal of Primatology 64: 377–396. Liebal, K., Pika, S., Call, J. and Tomasello, M. 2004c. “To move or not to move: How apes adjust to the attentional state of others.” Interaction Studies 5: 199–219. Liebal, K., Pika, S. and Tomasello, M. 2004b. “Social communication in siamangs (Symphalangus syndactylus): Use of gestures and facial expressions.” Primates 45: 41–57. Liebal, K., Pika, S. and Tomasello, M. 2006. “Gestural communication of orangutans (Pongo pygmaeus).” Gesture 6: 1–38. Lillard, A.S. 1998. “Ethnopsychologies: Cultural variation in theories of mind.” Psychological Bulletin 123: 3–32. Liszkowski, U., Carpenter, M., Henning, A., Striano, T. and Tomasello, M. 2004. “TwelveMonth-Olds Point to Share Attention and Interest.” Developmental Science 7: 297–307. Lock, A. 2001. “Preverbal communication.” In The Blackwell Handbook of Infant Development, J.G. Bremner and A. Fogel (eds.), 379–403. Hove, U.K.: Blackwell. MacCorquodale, K. and Meehl, P.E. 1948. “On a distinction between hypothetical constructs and intervening variables.” Psychological Review 55: 95–107. Melis, A., Call, J. and Tomasello, M. 2006. “Chimpanzees (Pan troglodytes) conceal visual and auditory information from others.” Journal of Comparative Psychology 120: 154–162. Menzel, C.R. 1999. “Unprompted recall and reporting of hidden objects by a chimpanzee (Pan troglodytes) after extended delays.” Journal of Comparative Psychology 113: 426–434. Miles, H.L. 1990. “The cognitive foundations for reference in a signing orangutan.” In “Language” and Intelligence in Monkeys and Apes: Comparative Developmental Perspectives, S.T. Parker and K.R. Gibson (eds), 511–539. Cambridge: Cambridge University Press. Mitchell, R.W. 2000. “A proposal for the development of a mental vocabulary: With special reference to pretense and false belief.” In Children’s Reasoning and the Mind, P. Mitchell and K.J. Riggs (eds.), 37–65. Hove: U.K.: Psychology Press. Moore, C. and Corkum, V. 1994. “Social understanding at the end of the first year of life.” Developmental Review 14: 349–372. Murphy, C.M. and Messer, D.J. 1977. “Mothers, infants and pointing: A study of a gesture.” In Studies in Mother-infant Interaction, H.R. Schaffer (ed.), 325–354. London: Academic Press. O’Neill, D.K. 1996. “Two-year-old children’s sensitivity to a parent’s knowledge state when making requests.” Child Development 67: 659–677. Okamoto, S., Tomonaga, M., Ishii, K., Kawai, N., Tanaka, M. and Matsuzawa, T. 2002. “An infant chimpanzee (Pan troglodytes) follows human gaze.” Animal Cognition 5: 107–114. Parker, S.T. 1999. “The development of social roles in the play of an infant gorilla and its relationship to sensorimotor intellectual development.” In The Mentalities of Gorillas and Orangutans: Comparative Perspectives, S.T. Parker, R.W. Mitchell, and H. Lyn Miles (eds.), 367–393. Cambridge, U.K.: Cambridge University Press.
212 David A. Leavens et al.
Peignot, P. and Anderson, J.R. 1999. “Use of experimenter-given manual and facial cues by gorillas (Gorilla gorilla) in an object-choice task.” Journal of Comparative Psychology 113: 253–260. Petitto, L. 1988. “’Language’ in the prelinguistic child.” In Development of Language and Language Researchers, F. Kessel (ed.), 187–222. Hillsdale, NJ: Lawrence Erlbaum Associates. Pika, S., Liebal, K. and Tomasello, M. 2003. “Gestural communication in young gorillas (Gorilla gorilla): Gestural repertoire, learning, and use.” American Journal of Primatology 60: 95–111. Pika, S., Liebal, K., Call, J. and Tomasello, M. 2005a. “The gestural communication of apes.” Gesture 5: 41–56. Pika, S., Liebal, K. and Tomasello, M. 2005b. “The gestural repertoire of bonobos (Pan paniscus): Flexibility and use.” American Journal of Primatology 65: 39–61. Pika, S. and Mitani, J. 2006. “Referential gestural communication in wild chimpanzees (Pan troglodytes).” Current Biology 16: R191–R192. Poss, S.R., Kuhar, C., Stoinski, T.S. and Hopkins, W.D. 2006. “Differential use of attentional and visual communicative signaling by orangutans (Pongo pygmaeus) and gorillas (Gorilla gorilla) in response to the attentional status of a human.” American Journal of Primatology, 68, 978–992. Potì, P. and Spinozzi, G. 1994. “Early sensorimotor development in chimpanzees (Pan troglodytes).” Journal of Comparative Psychology 108: 93–103. Povinelli, D.J., Bering, J.M. and Giambrone, S. 2000. “Toward a science of other minds: Escaping the argument by analogy.” Cognitive Science 24: 509–541. Povinelli, D. J., Bering, J.M. and Giambrone, S. 2003. “Chimpanzee ‘pointing’: Another error of the argument by analogy?” In Pointing: Where Language, Culture, and Cognition Meet, S. Kita (ed.), 35–68. Hillsdale, NJ: Erlbaum. Povinelli, D.J., Bierschwale, D.T. and Cech, C.G. 1999. “Comprehension of seeing as a referential act in young children, but not juvenile chimpanzees.” British Journal of Developmental Psychology 17: 37–60. Povinelli, D.J. and Eddy, T.J. 1996a. “What young chimpanzees know about seeing.” Monographs of the Society for Research in Child Development 61(3, Serial No. 247). Povinelli, D.J. and Eddy, T.J. 1996b. “Chimpanzees: Joint visual attention.” Psychological Science 7: 129–135. Povinelli, D.J. and Eddy, T.J. 1997. “Specificity of gaze-following in young chimpanzees.” British Journal of Developmental Psychology 15: 213–222. Povinelli, D.J., Bering, J.M. and Giambrone, S. 2000. “Toward a science of other minds: Escaping the argument by analogy.” Cognitive Science 24: 509–541. Povinelli, D.J., Bering, J.M. and Giambrone, S. 2003a. “Chimpanzee ‘pointing’: Another error of the argument by analogy?” In Pointing: Where Language, Culture, and Cognition Meet, S. Kita (ed.), 35–68. Hillsdale, NJ: Erlbaum. Povinelli, D.J., Bierschwale, D.T. and Cech, C.G. 1999. “Comprehension of seeing as a referential act in young children, but not juvenile chimpanzees.” British Journal of Developmental Psychology 17: 37–60. Povinelli, D.J., Theall, L.A., Reaux, J.E. and Dunphy-Lelii, S. 2003b. “Chimpanzees spontaneously alter the location of their gestures to match the attentional orientation of others.” Animal Behaviour 66: 71–79. Racine, T.P. 2005. The Role of Shared Practice in the Origins of Joint Attention and Pointing. Unpublished doctoral thesis: Simon Fraser University, Burnaby, Canada.
Origins of explicit reference 213
Reddy, V. 2001. “Mind knowledge in the first year: Understanding attention and intention.” In Blackwell handbook of infant development, J.G. Bremner and A. Fogel (eds.), 241–264. Hove, U.K.: Blackwell. Reddy, V. 2003. “On being the object of attention: Implication for self-other consciousness.” Trends in Cognitive Science 7: 397–402. Russell, C.L., Bard, K.A. and Adamson, L.B. 1997. “Social referencing by young chimpanzees (Pan troglodytes).” Journal of Comparative Psychology 111: 185–193. Russell, J.L., Braccini, S., Buehler, N., Kachin, M.J., Schapiro, S.J. and Hopkins, W.D. 2005. “Chimpanzee (Pan troglodytes) intentional communication is not contingent upon food.” Animal Cognition 8: 263–274. Savage-Rumbaugh, E.S. 1986. Ape Language: From Conditioned Response to Symbol. New York: Columbia University Press. Savage-Rumbaugh, E.S. 1991. “Language learning in the bonobo How and why they learn.” In Biological and Behavioral Determinants of Language Development, N.A. Krasnegor, D.M. Rumbaugh, R.L. Scheifelbusch, and M. Studdert-Kennedy (eds), 209–233. Hillsdale, N.J.: Lawrence Erlbaum Associates. Savage-Rumbaugh, E.S., Shanker, S.G. and Taylor, T.J. 1998. Apes, Language, and the Human Mind. Oxford: Oxford University Press. Senut, B., Pickford, M., Gommery, D., Mein, P., Cheboi, K. and Coppens, Y. 2001. First hominid from the Miocene (Lukeino Formation, Kenya). Comptes Rendus de l’Académie de Sciences 332: 137–144. Shanker, S.G. and King, B.J. 2002. “The emergence of a new paradigm in ape language research.” Behavioral and Brain Sciences 25: 605–656. Sinha, C. 2004. “The evolution of language: From signals to symbols to system.” In Evolution of Communication Systems: A Comparative Approach, D. Kimbrough Oller and Ulrike Griebel (eds), 217–235. Vienna Series in Theoretical Biology. Cambridge, MA: MIT Press. Steiper, M.E., Young, N.M. and Sukarna, T.Y. 2004. “Genomic data support the hominoid slowdown and an Early Oligocene estimate for the hominoid-cercopithecoid divergence.” Proceedings of the National Academy of Sciences 101: 17021–17026. Sugarman, S. 1984. “The development of preverbal communication: Its contribution and limits in promoting the development of language.” In The Acquisition of Communicative Competence, R.L. Scheifelbush and J. Pickar (eds.), 23–67. Baltimore: University Park Press. Super, C.W. 1990. “The cultural regulation of infant and child activities.” In Activity, Energy Expenditure and Energy Requirements of Infants and Children, B. Schurch and N.S. Scrimshaw (eds.), 321–333. Lausanne: Nestle. Tanner, J.E. and Byrne, R.W. 1993. “Concealing evidence of mood: Evidence for perspectivetaking?” Primates 34: 451–457. Tanner, J.E. and Byrne, R.W. 1996. “Representation of action through iconic gesture in a captive lowland gorilla.” Current Anthropology 37: 162–173. Theall, L.A. and Povinelli, D.J. 1999. “Do chimpanzees tailor their gestural signals to fit the attentional states of others?” Animal Cognition 2: 207–214. Thompson, N.S. 1994. “The many perils of ejective anthropomorphism.” Behavior and Philosophy 22: 59–70. Tomasello, M. 1995. “Joint attention as social cognition.” In Joint Attention: Its Origins and Role in Development, C. Moore and P.J. Dunham (eds.), 103–130. Hillsdale, NJ: Lawrence Erlbaum Associates.
214 David A. Leavens et al.
Tomasello, M. 1999. The Cultural Origins of Human Cognition. Cambridge, MA: Harvard University Press. Tomasello, M. 2006. “Why don’t apes point?” In Roots of Human Sociality: Culture, Cognition and Interaction, N. Enfield and S.C. Levinson (eds.), 506–524. Oxford: Berg. Tomasello, M., Call, J. and Hare, B. 1998. “Five primate species follow the gaze of conspecifics.” Animal Behaviour 55: 1063–1069. Tomasello, M., Call, J., Nagell, K., Olguin, K. and Carpenter, M. 1994. “The learning and use of gestural signals by young chimpanzees: A trans-generational study.” Primates 35: 137– 154. Tomasello, M., Call, J., Warren, J., Frost, T., Carpenter, M. and Nagell, K. 1997. “The ontogeny of chimpanzee gestural signals: A comparison across groups and generations.” Evolution of Communication 1: 223–253. Trevarthen, C. 1977. “Descriptive analyses of infant communicative behavior.” In Studies in Mother-infant Interaction, H. R. Schaffer (ed.), 227–270. London: Academic Press. Trevarthen, C. 1998. “The concept and foundations of infant intersubjectivity.” In Intersubjective Communication and Emotion in Early Ontogeny, S. Bråten (ed.), 15–46. Cambridge: Cambridge University Press. Trevarthen, C. and Hubley, P. 1978. “Secondary intersubjectivity: Confidence, confiding and acts of meaning in the first year.” In Action, Gesture and Symbol, A. Lock (ed.), 183–229. New York: Academic Press. Veà, J.J. and Sabater-Pi, J. 1998. “Spontaneous pointing behaviour in the wild pygmy chimpanzee (Pan paniscus).” Folia Primatologica 69: 289–290. de Waal, F.B.M. 1982. Chimpanzee Politics: Power and Sex among Apes. New York: Harper and Row. de Waal, F.B.M. 2001, January 19. “Pointing primates: Sharing knowledge . . . without language.” Chronicle of Higher Education B7–B9. Whiten, A. 2000. “Chimpanzee cognition and the question of mental re-representation.” In Metarepresentation: A Multidisciplinary Perspective, D. Sperber (ed.), 139–167. Oxford: Oxford University Press. Wilkins, D. 2003. “Why pointing with the index finger is not a universal (in sociocultural and semiotic terms).” In Pointing: Where Language, Culture, and Cognition Meet, S. Kita (ed.), 171–215. Hillsdale, NJ: Erlbaum.
chapter 10
The co-evolution of intersubjectivity and bodily mimesis Jordan Zlatev This chapter presents an evolutionary and developmental model, according to which intersubjectivity is intimately tied to bodily mimesis – the use of the body for communicative and representational purposes – to an extent that intersubjectivity can be said to co-evolve with it. I review some relevant evidence concerning non-human primates which shows that feral and captive apes are capable of the first two levels (involving e.g. empathy, shared attention and imitation), but not of the third level which involves an understanding of communicative signs, i.e. triadic mimesis. In contrast, enculturated language-trained apes show some aspects of triadic mimesis, suggesting how our predecessors could have bootstrapped themselves to this level without language (and without a “theory of mind”). The emergence of language, on the other hand, opens the way to the highest two levels of intersubjectivity, bringing forth the understanding of “beliefs” and the use of folk psychology.
1.
Introduction
Many if not most would agree that there is a close relationship between intersubjectivity and language. But what more precisely is this relation? The first impediment to answering this question is definitional. As several of the contributions to this volume show, there are rather different understandings of the concept of intersubjectivity. For the purpose of this chapter, intersubjectivity will be taken to be the sharing of affective, perceptual and reflective experiences between two or more subjects. Such “sharing” can take different forms, some more immediate, while others more mediated by higher cognitive processes, e.g. what Barresi and Moore (this volume) call “understanding” as opposed to simply “sharing”. The phenomenon of joint attention (e.g. Moore and Dunham 1995), for example, would qualify as a paradigmatic form of perceptual intersubjectivity (Zlatev, Brinck and Andrén 2008).
216 Jordan Zlatev
If it is difficult to reach a consensus concerning the notion of intersubjectivity, it is even more so when it comes to language. Therefore I will not here argue for but simply assume that language is a conventional (normative) semiotic system for communication and thought. The signs constituting language are predominantly symbolic, i.e. conventional pairings of expression and content. The expressions can be spoken (oral), signed (manual-brachial) or written and the denoted concepts (or “uses” in a more action-oriented Wittgensteinian approach) are commonly known to those who are fluent in the language (see Itkonen 1978, this volume; Zlatev 2007a, 2007b). Given these provisional definitions, we can reformulate our question as follows: is intersubjectivity a prerequisite for the learning and use of language or is language a prerequisite for understanding others’ (and perhaps even one’s own) mind? Thus phrased, the question is a classic dilemma in the literature on “theory of mind”, with arguments in favor for each side of the dependence relation. On the one hand, Bloom (2000: 2) argues persuasivly that “… it is impossible to explain how children learn the meanings of a word without understanding of certain non-linguistic mental capacities, including how children think about the minds of others”. There is, indeed, strong evidence that children understand much about adults’ (visual) attention and communicative intentions prior to 18 months (Baldwin 1991, 1993; Tomasello 1999, 2003), and that such understanding appears pivotal for word learning. An often reported result is the following: a child in his second year of life is given a novel toy A to play with, while another toy B is placed out of view. As he is playing with toy A, the experimenter looks at toy B and says: “It’s an X”. The child looks at the experimenter, follows his gaze and discovers toy B for the first time. Importantly, the child assumes that X is the name of toy B, not toy A that he was playing with when he heard X for the first time. Such findings are problematic for pure associationist models of word learning (e.g. Plunkett 1998). On the other hand, there is accumulating evidence that language acquisition itself is a determining factor for the development of certain forms of intersubjectivity, especially those which have been linked to the understanding of (false) beliefs. For example, deaf children who are not exposed to signed language at an early age understand others’ (false) beliefs significantly later than those with signing parents, or hearing children (Peterson and Siegal 1995). Longitudinal co-relational studies indicate that language development predicts performance in tasks of “theory of mind”, but not vice versa (de Villiers and Pyers 1997; Astington and Jenkins 1999). Furthermore, training in sentential complement constructions (with or without . For an interesting discussion of these issues, though from a predominantly “theory of mind” perspective, cf. www.interdisciplines.org/coevolution.
Intersubjectivity and bodily mimesis 217
mental predicates) significantly improves performance in false belief tasks (de Villiers and Pyers 1997; Hale and Tager-Flusberg 2003; Lohmann and Tomasello 2003) and exposure to discourse involving different perspectives independently enhances false belief understanding (Lohmann and Tomasello 2003). So is it the chicken (language) or the egg (intersubjectivity) that comes first? In order to resolve this dilemma, we need first to clear up our conceptual dusty corners a bit more. Most importantly, it should be emphasized: Intersubjectivity ≠ “theory of mind”! Elsewhere (Zlatev 2007b) I address the different perspectives to social cognition these two concepts apply, but for present purposes it is important to point out the following three characteristics to the approach to intersubjectivity that is here adopted: 1. Intersubjectivity is not a unitary capacity: it involves understanding not only beliefs and other proposition-like entities, but other less explicit forms of consciousness: emotions, attentional foci and intentions (cf. Tomasello, Carpenter; Call, Behne and Moll 2005). 2. Intersubjectivity develops in a stage-like manner (in ontogeny and phylogeny) with “lower” stages serving as prerequisites for “higher” ones (e.g. the relationship between empathy and cognitive empathy, Preston and de Waal 2002). 3. Intersubjectivty is bodily-based: understanding others involves identifying with them on a direct bodily level (Merleau-Ponty 1962; Gallagher 2005, 2007; Gallagher and Hutto this volume; Hobson and Hobson this volume), with recent progress in understanding the neural underpinnings of this capacity (Gallese, Keyners and Rizzolatti 2004; Arbib 2005; Barresi and Moore this volume). These three characteristics allow linking intersubjectivity quite naturally to bodily mimesis (Donald 1991, 2001; Zlatev 2005, 2007a, 2007b), in a way first suggested by Zlatev, Persson and Gärdenfors (2005a). In this chapter, I will elaborate on this linkage, showing how five levels of the mimesis hierarchy – defined in the following section – correspond to different levels of intersubjectivity. By looking at recent evidence from primatology, and to a lesser degree child development, I will argue that the first two of these levels/stages of intersubjectivity are (to a considerable extent) common to human beings and great apes, and are therefore quite clearly pre/non-linguistic. The last two are specific for us as a species, but they are also clearly linked to language, and thus “post-mimetic”. The most theoretically
. See also Sinha and Rodruiguéz (this volume) and Hutto (this volume) for critiques of overly “mentalist” interpretations of intersubjectivity.
218 Jordan Zlatev
i nteresting level is the one in between, which appears to be both pre-linguistic and specific for human beings (in its full form). This is our capacity for triadic mimesis, as evidenced in e.g. declarative pointing and iconic gesturing (cf. Zlatev, Persson and Gärdenfors 2005b). I will argue that this social-cognitive capacity is the cradle for another cognitive capacity which appears to be uniquely human: third-order mentality (e.g. to see that you see that I see), which on its side is central for our ability to share (semantic) knowledge (cf. Itkonen, this volume). Thus, part of the story to be presented is an “egg-based” solution to the dilemma: intersubjectivity grounds language, which then propels the rocket to higher levels. However, if we inquire about the evolutionary origins of triadic mimesis, it appears likely that it is gestural communication itself that provided the evolutionary niche for its selection. This brings back focus to the precursors of language as a causal factor in the development of intersubjectivity. Thus, the story here told is one of co-evolution.
2.
Bodily mimesis and the mimesis hierarchy
In his influential theory of human evolution, Donald (1991) proposed that a form of cognition crucially based on mimesis, and a corresponding culture based on mimetic skills such as tool use, imitation, ritual dance and gestural communication mediated between the “episodic” cognition of the common ape-human ancestor and the emergence of language as a dominant mode of human communication (see also Hutto, this volume). Mimetic representations are according to Donald “conscious, self-initiated, representational acts that are intentional but not linguistic” (Donald 1991: 168). This rather broad definition includes a number of different skills such as imitation, the re-enactment of actions in imagination (and hence planning and rehearsal), and the use of iconic and deictic gestures for intentional communication. Others have suggested a similar “mimetic stage” in ontogeny, but have proposed quite different interpretations of its scope (Nelson 1996; Zlatev 2002, 2003), making it clear that the concept of mimesis requires a more precise definition. Building on Donald’s work, but taking into account some more recent evidence in social neuroscience (see Barresi and Moore, this volume; Zlatev 2007b, in press) and evidence on the mimetic capacities of non-human primates (Zlatev et al. 2005a), I have in a number of recent publication (Zlatev 2005, 2007a, 2007b) proposed the concept of bodily mimesis, which can be defined as follows:
Intersubjectivity and bodily mimesis 219
Def: A particular bodily act of cognition or communication is an act of bodily mimesis if and only if: a. It involves a cross-modal mapping between exteroception (i.e. perception of the environment, normally dominated by vision) and proprioception (perception of one’s own body, normally through kinesthetic sense); b. It is under conscious control and corresponds to – either iconically or indexically – to some action, object or event, while at the same time being differentiated from it by the subject; c. The subject intends the act to stand for some action, object or event for an addressee (and for the addressee to recognize this intention); d. Without the act being conventional-normative, and e. Without the act dividing (semi)compositionally into meaningful sub-acts that systematically relate to each other and other similar acts.
This definition allows us to clarify the relationship between bodily mimesis and a number of related phenomena, on the basis of an evolutionary and developmental model referred to as the mimesis hierarchy. The model is unashamedly “progressivist”, and defines each successive stage through the attainment of a new semiotic capacity: (b), (c) and the positive versions of (d) and (e), i.e. with conventionality-normativity (d-poss) and with compositionality (e-poss). At the same time, it is not a classical stage model in the spirit of Piaget, where each consecutive stage brings with it total reorganization, but a “layered model” (Stern 1985) where earlier capacities continue to co-exist with newer ones, which may subsume but not abolish them. The model (in its current version) distinguishes between 5 stages/levels, determined on the basis of the definition of bodily mimesis given above: 1. Proto-mimesis: only (a), e.g. neonatal mirroring, contagion, mutual gaze (cf. “primary intersubjectivity”, Trevarthen 1979) 2. Dyadic mimesis: only (a) and (b), e.g. action imitation, shared attention, mirror self-recognition 3. Triadic mimesis: only (a), (b) and (c): e.g. declarative pointing, iconic gestures, full joint attention (cf. “pantomime”, Arbib 2005) 4. Post-mimesis1: (a), (b), (c) and (d-poss), (cf. “protolanguage” Bickerton 2003; “protosign” Arbib 2005) 5. Post-mimesis2: (a), (b), (c), (d-poss) and (e-poss): e.g. spoken/signed language (cf. “symbolic reference”, Deacon 1997)
220 Jordan Zlatev
The first stage, proto-mimesis, is made possible by a cross-modal mapping between one’s felt bodily actions and the observed actions of others. This serves as a basis for bodily mimesis and intersubjectivity – but only a basis, since it is neither under (full) conscious control, nor representational. It can nevertheless account for certain forms of social cognition such as emotional and behavioral contagion and neonatal mirroring. It is in many ways similar of the notion of “primary intersubjectivity” and “interaffectivity” in the developmental literature (Trevarthen 1979; Stern 1985). For level 2, dyadic mimesis, a mimetic act needs to be volitional and representational, as in Donald’s original characterization of mimesis, given earlier. As condition (b) of the definition states, the notion of representation is understood in line with Piaget’s (1945) criterion of differentiation between “signifier” and “signified” from the subject’s point of view (cf. Sonesson 2007), adding the requirement that the signifier is a bodily act. Piaget’s example of an infant opening and closing her mouth to model the opening and closing of a matchbox would be an example of an iconic correspondence. Children’s acts of pointing for themselves in order to help guide their attention (Bates, Camaioni and Volterra 1975) would qualify as indexical mimetic acts. Level 3 is brought about by adding condition (c), which introduces the necessary triadic element in order to make bodily mimesis communicative: the representation or sign is intended to be recognized as such by an addressee, along with the communicative intention itself. This introduces a Gricean element of intentional communication (Grice 1957), involving intentional attitudes but not propositional ones (cf. Hutto, this volume). An example of an iconic sign that fulfills all three conditions is the miming of eating by pretending to move a spoon to one’s mouth (made behind a glass door) in order to communicate to a colleague a desire to go for lunch. An indexical mimetic sign would be, for example, a paradigmatic form of declarative pointing (Brinck 2003). Condition (d-poss) distinguishes triadic mimesis from post-mimesis, in which the communicative representations are conventional (i.e. commonly known) and normative (i.e. their application is governed by criteria of correctness, of which the users are at least to some degree aware). This qualifies them as being symbolic, though not in the sense of Deacon (1997) who insists above all on the property mentioned in (e), the presence of systematic semantic and grammatical relations . See Barresi and Moore (this volume) for a detailed neuroscientific model of how first-person and third-person information appears to be “matched” in the posterior parietal cortex. . The iconic, but not the indexical aspects of triadic mimesis are similar to the notion of pantomime, constituting Stage 5a in the somewhat similar evolutionary model proposed by Arbib (2003, 2005).
Intersubjectivity and bodily mimesis 221
between the symbols as a definitional criterion for “symbolic reference”. In previous work (Zlatev 2003, 2005) I combined these criteria and did not distinguish between the corresponding two levels of post-mimesis. While it is possible that conventionality and systematicity always come in tandem (in both evolution and ontogeny) I now believe (in agreement with e.g. with Sonesson 2007) that we should not make this an inherent characteristic of the definition of symbolicity. It is at least possible that the “one-word stage” in childhood, and the linguistic skills of language taught apes such as Kanzi (Savage-Rumbaugh and Lewin 1994) qualify as a “protolanguage”, as suggested by Bickerton (2003), i.e. as an inventory of symbols, but with very little knowledge of their interrelations. The more “holophrastic” version of this hypothesis – according to which the first expressions are/were more like formulaic sentences such as How are you? than words (Zlatev 1997; Wray 2000) – has been referred to as “protosign” (Arbib 2003, 2005). The two possibilities are not exclusive and it seems that some early expressions are more “wordlike” while others are more formulaic, and it is well-known in the child language literature that some children vary in their preference for the two strategies (Nelson 1985). It is the pervasive grammatical and semantic systematicity of all human languages – that has traditionally attracted most of the attention of philosophers and linguists – that brings about the final level: Post-mimesis2. However, as the model implies, this level would be impossible if it did not, at least in part, rest on the earlier levels. Nelson and Shaw’s (2002) poignant definition of language as a “socially shared symbolic system” implies this as well: language may indeed be called a “system”, but if it were not for socially shared symbolism, it would not be language.
3.
The mimesis hierarchy and levels of intersubjectivity
What gives us ground to apply the mimesis hierarchy to the development and evolution of intersubjectivity? As suggested in the introduction: above all the approach arising from the combination of phenomenology and neuroscience that links social cognition and embodiment, perhaps best summarized by Gallagher (2005, 2007). It was originally Husserl, who over one hundred years ago first criticized the intellectualist perspective on the “understanding of other minds” as a matter of inference or analogy from the knowledge of one’s own mind. As . One case in which language acquisition relies (much) less on bodily mimesis is autism. Predictably, however, even the language of high-functioning persons with autism displays semantic and pragmatic abnormalities (Menyuk and Quill 1985; Zlatev forthcoming).
222 Jordan Zlatev
summarized by Gallagher (2007: 286): “For Husserl, understanding another person is not a matter of intellectual inference, but a matter of sensory activations that are unified in or by the animate organism or lived body that is perceiving another animate organism.” This perspective on intersubjectivity was further elaborated by Merleau-Ponty (1962), in particular through the notion of the “corporeal schema” which serves as “a normal means of knowing other bodies” (Merleau-Ponty 2003: 218). This was originally stated long before neuroscience corroborated the role of the body in understanding others through various mirror neuron systems (Rizzolatti et al. 1996; Gallese et al. 2004; Arbib 2005; Zlatev in press), which link the perception of another person and the subject’s own proprioception and action. Gallagher’s (2005, 2007) major contribution to this discussion has been to elaborate the distinction between body schema and body image where the first is pre-conscious and serves as a precondition and backdrop for intentionality, while the latter is “a (sometimes conscious) system of perceptions, attitudes, beliefs and dispositions pertaining to one’s own body” (Gallagher 2005: 37). While one may still question this distinction, and require a finer division within the latter – the perceptions of the body are quite different from the beliefs and attitudes towards it – I will in general accept it, and reply on it in the discussion of the first two levels of intersubjectivity. Within this “embodied” and action-oriented perspective on the understanding of others as well as the self, bodily mimesis becomes clearly relevant. A central question in relating the mimesis hierarchy and intersubjectivity is whether the respective level serves as a precondition and a causal factor for the development of corresponding skills of intersubjectivity. Or is it rather that independently reached insights into the minds of others makes increasingly complex forms of bodily mimesis possible? The problem for the latter scenario is that one would need to account for the emergence of “theory of mind” skills (or “modules”) independently, which remains a problem for both evolutionary and child psychology (cf. Hutto 2008, this volume). Furthermore, from a Vygotskyan perspective stating the priority of the “inter-personal” to the “intra-personal” in cultural development (see Zlatev et al. this volume), one would expect bodily mimesis to bootstrap (in ontogeny) and provide selection pressures for (in phylogeny) the development of more refined skills in “mind reading”. At the same time, as with the chicken-and-egg relationship between intersubjetctivity and language, we will see that the causality is not unidirectional, and the relationship between mimesis and intersubjectivity may be more pertinently described in terms of co-evolution.
3.1
Intersubjectivity and bodily mimesis 223
Proto-mimesis
According to the definition of bodily mimesis we can regard some of the most basic forms of intersubjectivity as proto-mimetic to the extent that they consist of interpersonal interactions that involve cross-modal mapping between proprioception and the (visual) perception of others, but lack the characteristics volition and differentiation. In terms of the distinction between body schema and body image (Gallagher 2005) mentioned above, proto-mimesis can be said to involve (above all) the body schema, which is largely innate (in the sense of being present at birth) and pre-conscious, rather than the body image, which is gradually constructed with experience and accessible to consciousness. Proto-mimetic forms of intersubjectivity do not require a conceptual differentiation between self and other, which is necessary for establishing a correspondence relation between them. This is not to say that the young infant lives in a completely undifferentiated world in which there is no awareness of self whatsoever, as pointed out by Stern (1985). Nevertheless, even a modern developmental psychologist who emphasizes the role of the awareness of others’ attention and the presence of affective self-consciousness in the first months of life points out that “older infants reveal a greater focus on the self and the younger ones reveal a more immersed, less detached focus on the other” (Reddy 2003: 401). This “more immersed, less detached” quality of the earliest forms of intersubjectivity motivates the classification of them as proto-mimetic. Can this analysis be extended to the (early) interpersonal relations among apes? Neonatal mirroring has also been observed, and appears to be common in chimpanzees (Myowa-Yamakoshi 2001; Myowa-Yamakoshi et al. 2004). Since this is typically attributed to a form of identification with the person imitated when children are concerned (Meltzoff and Gopnik 1993; Gallagher 2005), it can be viewed as evidence that at least chimpanzees, and possibly other apes and nonhuman primates possess the capacity for basic proto-mimetic intersubjectivity. The function of such “mirroring” can be related to what is possibly the most basic form of intersubjectivity, both ontogenetically and phylogenetically: the ability to share emotions, or empathy (Einfühlung). As a proto-mimetic, non-representational capacity, this is testified in early infancy and sometimes referred to as interaffectivity (Stern 1985). The well-known experiments described by Trevarthen (1992) show that parent-infant interactions in the first few months take the form of a reciprocal rhythmic “dance”, and that frustration follows if this “attunement” is disrupted (notice the musical metaphors). The suggestion is that emotions such as joy and suffering are perceived directly, possibly involving mirror-neuron
224 Jordan Zlatev
s tructures similar to those involved in action recognition and imitation, rather than inferences to underlying states (Gallese et al. 2004). Preston and de Waal (2002) have argued persuasivly that as a basic biological mechanism involving the linkage of perception and action, empathy is available to most if not all mammal species. Defining empathy as “any process where the attended perception of the object’s state generates a state in the subject that is more applicable to the object’s state or situation than to the subject’s own prior state or situation” (ibid: 4), they see a clear evolutionary motivation for its emergence in the ability to recognize and understand the behavior of con-specifics. It is an open question how much of such matching between the visually perceived body of the other and the proprioceptively perceived body of oneself is domain-general – and thus can be expected to be general across species – and how much is specialized in the form of species-specific communicative signals such as facial expressions. It is characteristic that signals such as the famous “play-face” expression of great apes (an evolutionary precursor to the human smile) typically carry emotional rather than referential meaning. A second mode of intersubjectivity that appears to be of a proto-mimetic nature, at least in human children, is attention. Reddy (2000, 2003, 2005) has argued that prior to awareness of the other’s attention to an external object and much prior to joint attention appearing around 12 months (see below), children “show an awareness of others as attending beings, as well as an awareness of self as an object of others’ attention” (Reddy 2003: 357), displayed in phenomena such as eye-contact, intense smiling, coyness, “calling” vocalizations, showing-off etc. Since awareness of self (at this stage) is largely proprioceptive, while awareness of the attention of others (in seeing children) is mostly based on vision, this satisfies the first criterion for bodily mimesis. Reddy’s claim is that such dyadic (though not dyadic-mimetic) interactions underpin later developments in intersubjectivity. Evidence for this is the observation that autistic children show difficulties even with such simple interpersonal engagements, “suggesting that whatever is going on in dyadic attentional engagements may indeed be critical, not just as a source of information and experience about attentional behavior, or as a scaffold for the subsequent development of awareness of attention, but also as evidence of awareness of attention” (Reddy 2005: 95). Until recently it has not been clear whether such awareness of another’s attention exists in the interaction between infant apes and their mothers, but in a recent study, Bard et al. (2005) report that the rates of mutual gaze between infants and their mothers are virtually the same in 3-month old human children and 3-month old chimpanzees; 18–20 and 17 times per hour, respectively (though humans tend to engage in much longer bouts of mutual gaze). Furthermore, the authors noticed a “cultural” difference between the apes at Primate Research Institute, Japan and
Intersubjectivity and bodily mimesis 225
those at Yerkes National Primate Research Center, USA, with the ones in Japan engaging in mutual gaze at much higher rates (22 vs. 12 times per hour), while the ape mothers in the USA cradled their infants more often (71% vs. 40% of the total time). Intriguingly, a similar inverse correlation between visual and tactile contact has been observed in human societies, with traditional cultures favoring touch and Western ones gaze: “With reduced physical contact found in Western societies, mutual engagement shifts to the visual system, arguably a more evolutionarily derived pattern” (Bard et al. 2005: 621). Since apes do not seem to differ from humans in the capacity to perform this shift, and possibly even transmit it culturally to their descendents, this supports the conclusions from the studies on neonatal imitation that the difference between the species on the proto-mimetic level is not of a qualitative nature.
3.2 Dyadic mimesis In the previous discussion, I proposed that proto-mimetic intersubjectivity, without a clear differentiation between self and other is based on the (mostly) unattended body schema and similarly unattended mechanisms for “body copying”. On the other hand, dyadic mimetic intersubjectivity is based on the conscious control of the movements of one’s body and attention to their correspondence to the body of another, whereby one can imagine what the other experiences on the basis of one’s own experiences in similar circumstances. In the terminology of Gallagher (2005), I propose that the role that bodily mimesis proper plays for the development of intersubjectivity implies not the body schema but the body image. While the body schema and the body image normally interact, Gallagher (2005) shows how in certain pathologies they can be disassociated. Below, I will describe how understanding others’ emotions, attention and intentions can be seen as intimately related to dyadic mimesis. Whereas (simple) empathy is proto-mimetic, what Preston and de Waal (2002) call cognitive empathy requires a differentiation between subject and object where “the subject is thought to use perspective-taking processes to imagine or . In a recently completed study, however, we found remarkable differences in both rates and durations of mutual gaze between ape and human mother-infant dyads, with five dyads per group, and somewhat older ages for the infants than those in the study of Bard et al. (2005): 5–8 months. The rates for the apes (3 chimpanzee, one bonobo and one gorilla) was on average 2/ hour, while those for the 5 human infants (observed in Lund, Sweden) was as many as 35/hour on average (Zlatev 2008). There are many methodological issues that need to be addressed, but these findings suggest the need for precaution before concluding that there are few differences between ape and human proto-mimesis (and thus, primary intersubjectivity).
226 Jordan Zlatev
project into the place of the object” (ibid: 18). Evidence that this is not an isolated phenomenon, but shows a more advanced level of intersubjectivity is the fact that cognitive empathy “appears to emerge developmentally and phylogenetically with other ‘markers of mind’ … including perspective taking …, mirror self-recognition …, deception, and tool-use.” (ibid: 18). Research concerning cognitive empathy in apes has focused on their consolation behavior, which is well-attested in at least chimpanzees, but has not been found in monkey species (de Waal and Aureli 1996) or any other mammalian species. Consolation is cognitively more complex than simple empathy since the consoling individual not only feels that somebody else experiences a particular negative emotion, but also intends to help relieve this, implying an ability to imagine the more positive emotional state. This supports the interpretation that cognitive empathy involves a more sophisticated representational capacity than what is necessary for simple empathy. Since dyadic mimesis involves both the ability to identify with the other, and at the same time to differentiate between self and other, a natural hypothesis is that it is dyadic mimesis, implicated in e.g. imitation (Zlatev et al. 2005a) that scaffolds the development of such representational capacity (cf. Hutto this volume, for a similar proposal). Since dyadic mimesis allows to “place oneself in the shoes of others”, it also gives the opportunity to understand what someone else is attending to. Such “second-order attention” is well testified among great apes (Hare et al. 2000). When two individuals become aware that both are attending to the same object, what results is shared attention. This comes a good deal towards the construction of a “consensual reality”, but does not quite reach it. To make a given object X fully intersubjective between you and me, I would need not only to see that you see X, (second-order attention, see Figure 1a), but also to see that you see that I see X (third-order attention, see Figure 1b) and vice versa – which is one interpretation of what it means to engage in joint attention. Full joint attention thus involves third-order mentality and possibly because of this (see below) appears to be beyond the cognitive capacities of apes (Tomasello 1999). It also goes beyond dyadic mimesis, so I will leave this capacity for the time being, but return to it in the following subsection. Concerning the understanding of another’s intentions, it was the received view until the end of the last century that apes cannot do this (e.g. Tomasello 1999). Recently, however, there has been mounting evidence that (at least) “[c]himpanzees . The terms “shared attention” and “joint attention” have, unfortunately, not been standardized in the literature and are often used interchangebaly. One exception is Emery (2000), who however uses the terms in nearly the reverse sense as that adopted here, with “shared attention” being the more high-order phenomenon.
Intersubjectivity and bodily mimesis 227
Figure 1a. Shared attention: Second-order attention: “I see that you see X” (and vice versa, though only the second-order arrow for one of the participants is shown)
Figure 1b. Joint attention: Third-order attention: “I see that you see that I see X” (and vice versa, though only the third-order arrow for one of the participants is shown)
understand psychological states – the question is which ones and to what extent”, which is the title of Tomasello, Call and Hare (2003). A wealth of experiments supports this claim. For example, a subordinate and a dominant chimpanzee compete for food placed on the subordinate’s side of two barriers, so that in some cases only the subordinate, but not the dominant can see the food and monitor the visual access of her competitor. The results showed that the subordinates preferentially retrieved the food that dominants could not see (and had not seen in the past), implying that chimpanzees are aware of the perceptual states of con-specifics. Together with awareness of the competitor’s goal (i.e. to obtain the food) this allows the “prediction” of the other’s actions and acting accordingly. A dyadic-mimetic interpretation of these facts is that such an understanding can be obtained through the “projection” of one’s own perceptual and motivational state
228 Jordan Zlatev
onto the other (“I would get the food if I were in her place!”) at the same time as distinguishing between the self and the other, and does not require explicit (propositional) reasoning or inference. There is mounting evidence that even non-enculturated and language-naïve apes are capable of such mimetic enactments. At the same time, it has not been shown that (non-enculturated) apes are capable of understanding another’s mental states about their own mental states, which would involve, as with joint attention, third-order mentality. In natural settings, some cases of deception may be interpreted in a way to involve thirdorder mentality, but do not require this. For example de Waal (1982) describes the chimpanzee Yeroen who has had a fight with Nikkie, and continues to fake a limp only when in the presence of Nikkie, apparently in an act of wishing to provoke Nikkie’s empathy: “I wish to make you see that I hurt”. The more parsimonious explanation, however, is that Yeroen has learned from previous experience that he is not bothered by Nikkie when he is hurt: “He may have learned from incidents in the past in which he is seriously wounded that his rival was less hard on him during periods when he was (of necessity) limping” (ibid: 35–36), so he mimes the appropriate behavior. Here we have a clear correspondence between dyadic mimesis and second-order intersubjectivity. Notice that Yeroen’s limping was not a form of intentional communication – obviously he did not wish Nikkie to understand that he was faking a limp – if he did, that would be a case of triadic mimesis. The conclusion that can be drawn from these various examples is that wild apes as well as those who are exposed to different degrees of human contact (captive, nursery-raised and laboratory-trained apes), but are not raised in a “something like a human cultural environment” and thus enculturated (Call and Tomasello 1996: 372) can indeed understand second-order mentality. However, such apes do not seem able to master third-order mentality (in the domains of emotion, attention nor intention) in which their own mental state needs to be either intentionally communicated – in collaboration, or hidden – in competition. This corresponds well with the capacity of apes for dyadic mimesis, but their relative difficulty with triadic mimesis, as argued below.
. While it has still not been conclusively shown that apes are capable of such mental “projection”, and it is conceivable how the evidence can be explained in a more behaviouristic manner involving learning generalizations over other individuals’ behaviour in relation to food, the mental explanation is (a) ultimately more parsimonious (cf. Tomasello and Call 2006) and (b) consistent with the performance of human children in roughly comparable stages of development.
Intersubjectivity and bodily mimesis 229
3.3 Triadic mimesis In the case of triadic mimesis, there is not only an understanding of the representational relation between one’s bodily motion and the object, action or event it corresponds to, but an understanding that such a representational relation can be used communicatively. This requires some understanding that the representation (sign) has the same meaning for the addressee as for the sender. This involves at least second-order mentality, which was shown above to correlate with dyadic mimesis. But “having the same meaning” is a reflexive notion and implies at least some degree of third-order mentality (see Itkonen this volume). Consider the simple example of what knowing the meaning of the word cat implies: (1) I know that cat means “a small furry animal that meows”. (2) I expect you know that cat means “a small furry animal that meows”. (3) I expect that you know that I know that cat means “a small furry animal that meows”.
While it is possible for intentional communication to begin without a full realization of (3), it is practically inevitable that discursive experience (including failures in communication) will promote the development of third-order mentality. Therefore it is possible that triadic mimesis was one of the major driving forces behind the development of intersubjectivity in hominid evolution. Unlike competing hypotheses related to “Machiavellian intelligence” (Byrne and Whiten 1988), this puts the focus on cooperation rather than competition (see also Brinck and Gärdenfors 2003; Tomasello et al. 2005). A prediction from this hypothesis is that enculturated apes – and these have all been taught at least some degree of sign use – will develop higher-level skills of intersubjectivity. There is support for this prediction. In summarizing some 200 studies of the role of human influence, Call and Tomasello (1996) conclude that “[t]he sociocognitive domains in which humans seem to have the highest effect on apes are intentional communication and social learning” (ibid: 391). As stated out earlier, wild apes do not seem to be capable of engaging in full, third-order joint attention. Furthermore, as Tomasello (1999: 21) points out, wild apes do not (a) point to objects; (b) hold up objects to show them to others; (c) take someone along to a place to show them something; (d) actively offer something to someone; and (e) intentionally teach other individuals new behaviors. Tomasello’s original account of these absences was based on the claim that apes are unable to understand another’s intentions. Given the more recent evidence, this explanation is no longer tenable, and indeed Tomasello et al. (2005) suggest instead that the crucial difference between apes and humans involves the motivation to
230 Jordan Zlatev
participate in joint collaborative engagements, and the lack of this motivation prevents apes from constructing “dialogical cognitive representations”. The explanation I propose is similar but more specific: non-enculturated apes fail to master triadic mimesis, and related to that, the ability to engage in third-order mentality. The motivational difference between apes and humans appealed to by Tomasello et al. (2005) cannot be the full explanation since enculturated apes such as Koko, Kanzi and Chantek manage at least (a) and apparently communicative skills (b–e) listed above as well (Miles 1999), even if in restricted forms. This seems to imply that the human cultural environments of the enculturated apes have taught them the basics of intentional, sign-mediated communication, and thereby (the roots of) third-order mentality. How this could occur can be seen again with respect to joint attention, which can be seen as emerging from second-order attention combined with the recognition of another’s intention concerning my attention: “I see that you see X” (second-order attention, Figure 1a), and furthermore “I realize that you want me to look at X” (Brinck 2001). In other words, joint attention can be brought forth by understanding a simple form of communicative intention, combined with already existing second-order attention. Thus, communicating the intention to jointly attend may be said to involve the simplest kind of triadic mimesis: whatever kind of sign that is used to convey that intention (see the example in the next paragraph) can be said to stand for that intention for both sender and interpreter. Without enculturation, experiments indicate that apes do not understand communicative intentions. A rather typical example is an experiment by Tomasello, Call and Gluckman (1997), where the authors in different ways indicated for both chimpanzees and two- to three-year-old children which out of three containers contained a reward: by pointing to the correct container; by placing a marker on top of the correct container; and holding up a replica of the correct container. Tomasello (1999: 102) summarizes the results of the experiment as follows: Children already knew about pointing, but they did not know about using markers and replicas as communicative signs. They nevertheless used these novel signs very effectively to find the reward. In contrast, no ape was able to do this for any of the communicative signs that they did not know before the experiment. One explanation of these results is that the apes were not able to understand that the human beings had intentions toward their own attentional states. The apes thus treated the communicative attempts of the human as discriminative cues on par . Consisting of what Wittgenstein (1953) called the “forms of life”, which provide the necessary context for the emergence and functioning of intentional communication and language.
Intersubjectivity and bodily mimesis 231
with all other types of discriminative cues that have to be laboriously learned over repeated experiences. The children, in contrast, treated each communicative attempt as an expression of the adult’s intention to direct their attention in ways relevant to the current situation. [my emphasis]
In other words, while the children clearly understood the communicative intentions of the experimenter, the apes did not. This interpretation is supported by a similar experiment designed to test “false beliefs” (Call and Tomasello 1999), in which the enculturated and language-taught orangutan Chantek clearly performed differently from all the other apes in understanding a human communicator’s signals. Even though this was not the goal of the experiment, and Chantek did not score better than the other apes in the false beliefs task, his much better performance could be explained by considering that he understood the signals as communicative signs (in this case indexes), rather than as “discriminative cues”. Finally, we can consider the case of captive apes living in a zoo and thus involved with at least some degree of interaction with human culture. In their study of spontaneous gestural communication in a group of gorillas in the San Francisco zoo, Tanner and Byrne (1996, 1999), found a wealth of gestures used by several members of the group, in particular by the adult male Kubie, some of which seemed to be used in a communicative way so that: [w]hether the receiving partner was a human or another ape, the signaling ape made sure that visual contact was established (except for tactile close gestures), and seemed to understand both the other’s potential actions and what the partner might, in turn, understand from his (the signaler’s) performance of gestures. (Tanner and Byrne 1999: 231, [my emphasis])
We can conclude that triadic mimetic intersubjectivity, involving understanding not only of con-specifics’ intentions, but their communicative intentions, and consequently a degree of third-order mentality, appears to be difficult but not completely beyond the cognitive potential of apes. To realize this potential, however, apes need an environment that is rich in opportunities for developing triadic mimesis, i.e. a particular form of enculturation. Thus triadic mimesis may be said to be within apes’ “Zone of Proximal Development” (ZPD), albeit in its periphery.10 If “enculturation” provides the ZPD for present-day apes, it is reasonable to suppose that it did the same for some particularly social group of hominids through a form of “self-domestication” giving rise to a bootstrapping spiral of sign use and intersubjectivity. In the terms of Donald (2001), triadic mimesis must have been within
10. ZPD is the notion introduced by Vygotsky (1978) to refer to skills that children could acquire with the help of adults, but not alone.
232 Jordan Zlatev
the common ancestor’s “zone of proximal evolution”, and is therefore a likely candidate for constituting “the missing link” in human cognitive evolution.
3.4 Post-mimesis What differentiates post-mimetic, or symbolic, cognition from mimesis is the use of fully conventional signs, interrelated within a system (Deacon 1997; Zlatev 2003). The most obvious example of post-mimesis, involving all the previous features but also symbolicity is a conventional, institutionalized signed language such as ASL (Stokoe 1960) or Swedish Sign Language (Ahlgren 2003). What is the relation between acquiring such a system and intersubjectivity? Prior to addressing this question, let us make the distinction between Post-mimesis1 (protolanguage) and Post-mimesis2 (language), pointed out in Section 2, where only the latter has (extensive) systematicity. A case can be made for apes acquiring the first but hardly the second.
3.4.1 Post-mimesis1: Protolanguage Evidence from four of the most successful projects involving the teaching of language to great apes – the chimpanzee (Pan troglodytes) Washoe (Fouts 1972, 1973), the gorillas Koko and Michael (Patterson 1978, 1980) and the bonobos (Pan paniscus) Kanzi and Panbanisha (Savage-Rumbaugh and Lewin 1994; Savage-Rumbaugh et al. 1998) and the orangutan Chantek (Miles 1990) – has shown that some of the characteristics of language are within the grasp of our nearest non-human relatives. As with children a precondition for the success of these projects has been a cultural environment rich with intersubjectivity and a variety of activities to stimulate communication (Miles 1990). The “ape language” literature contains rather convincing evidence that apes can: – comprehend the referential (representational) function of spoken words, ASL signs and visual lexigrams, and combinations of these; – use the sign-tokens in the absence of their referents, i.e. “displacement” (Hockett 1960); – acquire a considerable vocabulary of words/signs, according to some measurements extending 600 signs, but even according to the most conservative criteria no less than 140 signs – understand novel combinations of spoken or signed words; – produce novel combinations of signs. The following have also been reported, but are considerably less well documented:
Intersubjectivity and bodily mimesis 233
– apes can regard the acquired signs as conventional-normative (consensual), to the point of correcting their teachers if the latter do not use these appro priately; – apes can use language for a number of different functions (speech acts), including labeling, answering, expressing emotion, arguing and insulting; – apes can use language not only for communication, but for thinking (private speech). It is therefore possible to agree with Miles (1999: 204), that all great apes “have the intelligence for a rudimentary, referential, generalizable, imitative, displaceable, symbol system” – but with an emphasis on rudimentary. It has, for example, not yet been clearly demonstrated whether the spoken or signed utterances of apes conform to consistent principles of grammatical organization. Greenfield and Savage-Rumbaugh (1990, 1991) describe two ordering “rules” in the two symbol combinations of Kanzi, but the data show at best a weak statistical correlation between preferred order and semantic (communicative) function. The most plausible conclusion to the prolonged “ape language” debate therefore seems to be a tie between the extreme proponents and opponents: apes such as Koko and Kanzi can be said to have acquired a form of protolanguage, which is different from both mimesis due to conventionality, and full language – due to a lack of systematicity which, on its part, is necessary for the production of narratives. But can it even be truthfully said that Koko and Kanzi have acquired semantic conventions? A convention (Lewis 1969; Clark 1996) or a norm (Itkonen 1978) exists as a form of common knowledge among the members of the group that share the convention. A common explication of common knowledge is that it consists of third-order knowledge: “I know that you know that I know X” (Itkonen 1978, this volume). If this “knowledge” must be in explicit propositional form, then it is unlikely that we can attribute it to the language-taught apes, making it dubious whether we could even call their communicative acquisitions protolanguage. However, is it even warranted to make this requirement when it comes to children? While I earlier argued that triadic mimesis is connected to third-order mentality, it is not necessary that the understanding on all three orders is explicit enough to be a matter of belief, i.e. a propositional representation that is actively held to be true. Consider again the three orders of knowing the conventional meaning of the word cat given as (3) above and repeated for convenience: (3) I expect that you know that I know that cat means “a small furry animal that meows”.
234 Jordan Zlatev
The highest order thought, my expectation that you know that I know, is not a belief for the 4 year old child, since it is taken for granted, without pondering on whether it is true or not. For younger children, even the second order thought is unlikely to be propositional, as evidenced by their inability to understand beliefs proper – false or otherwise. Perhaps it best to call the most basic form of shared cultural knowledge a sharing of expectations: we both expect each of us to behave in a certain way given certain conditions (e.g. a red light) and are, however dimly, both aware that this expectation is mutual, and thereby binding. Given this, (3) can be reformulated in a Wittgensteinian manner into (4): (4) I expect that we are both using cat to mean “a small furry animal that meows”.
If we are prepared to attribute a degree of semantic knowledge to 2–3 year old children in terms such as (4), then I doubt whether we have good, non-anthropocentric reasons not to do likewise with Chantek and Kanzi. Furthermore, as mentioned earlier, Chantek’s experiences with protolanguage obviously bootstrapped his understanding of communicative intentions and third-order mentality, even if he, as all apes so far, was not able to pass a false-belief task (Call and Tomasello 1999).
3.4.2 Post-mimesis2: Language Irrespective of modality – spoken, signed or written – language is characterized by a form of combinatorialness that is unprecedented in animal communication. Recent studies of the spontaneous emergence of Nicaraguan Sign Language (NSL) during the past 25 years show that signed languages have their origin in (triadic) mimesis, but quickly acquire the properties of conventionality and systematicity. Senghas, Kita and Özyürek (2004: 1791) compared the co-speech gestures of Nicaraguan speakers of Spanish, with the signing of three “cohorts”, or generations, of learners of NSL and could document some aspects of this transition in detail: The movements of the hands and body in the sign language are clearly derived from a gestural source. Nonetheless, the analyses reveal a qualitative difference between gesturing and signing. In gesture, manner and path were integrated by expressing them simultaneously and holistically, the way they occur in the motion [event] itself. Despite this analogue, holistic nature of the gesturing that surrounded them, the first cohort of children, who started building NSL in the late 1970s, evidently introduced the possibility of dissecting out manner and path and assembling them into a sequence of elemental units. As second and third cohorts learned the language in the mid 1980s and 1990s, they rapidly made this
Intersubjectivity and bodily mimesis 235
segmented, sequenced construction the preferred means of expressing motion events. NSL thus quickly acquired the discrete, combinatorial nature that is hallmark of language. [my emphasis]
Given their mimetic-gestural origin signed languages have a much greater degree of iconicity than spoken languages and it has been proposed that this plays a role in their faster acquisition by (deaf) children (Brown 1977). Recent studies have questioned this, however, since only a minority of the signs of signed language have transparent iconic meanings, and in a study of 22 children acquiring ASL it was shown that “of the 44 different signs produced by the children before the age of 13 months, 36% were classified as iconic, 30% as metonymic, and 34% as arbitrary” (Bonvillan and Patterson 1999: 253).11 In this study, the authors compared the rate and pattern of acquisition of ASL by children and that of two gorillas, Koko and Michael. Despite certain differences – the children’s acquisition was (unsurprisingly) faster – it was shown that “that the similarities in early development across the species outweigh the differences” (ibid: 260). Thus, it can be concluded that gorillas, and by inference other great apes, not only can acquire the basics of a post-mimetic symbolic system such as ASL, but that they do this in a similar way. Interestingly, however, the gorillas seemed to rely somewhat more on the iconicity of the signs in comparison with the children, so that the proportion of their first 46 signs was somewhat different to the one reported above: 42% iconic, 32% metonymic and 26% arbitrary, while for the first 10 signs this difference was even clearer: 60% iconic, 20% metonymic and 20% arbitrary. This suggests that the apes relied to a greater degree on triadic mimesis than the children in their acquisition of the sign language, which also would explain why the children quickly progressed beyond the initial level of vocabulary acquisition to learn the systematic character, i.e. the grammar, of the language, while the apes “stagnated” on a simple, protolanguage level. With the rapid development of grammar and vocabulary around the age of 4, most children also become capable of understanding that others have or lack knowledge or have “false beliefs” (e.g., Perner 1991; Mitchell 1997), implying a metarepresentational capacity. It appears that these two developments are closely connected, and that acquiring a language, spoken or signed, is a major causal factor for developing a fully-fledged “theory of mind”. At least four different sides to language (use) combine to promote metarepresentational capacity:
11. “Metonymic” signs are such that involve some degree of iconicity between the sign and the referent, but ”the tie between the sign and its meaning is not readily apparent – one would be unlikely to guess the meaning of a metonymic sign simply seeing it produced” (Bonvillan and Patterson 1999: 252).
236 Jordan Zlatev
– Language is a conventional symbolic system, and as such its mastery implies third-order mentality, which would carry with it training in the understanding of others’ beliefs. – Two specific (universal) features of human languages are (a) mental predicates such as “think”, “believe” and “know” and (b) sentential complement constructions such as “say that X”. If one can meaningfully formulate sentences such as “I know that you think that X”, then one should be able to think the corresponding thought. – Not just the semantic/grammatical structure of language, but its use in discourse would promote the understanding of others as “mental agents”: There are at least “three kinds of discourse, each of which requires [children] to take the perspective of another person in a way that goes beyond the perspective-taking inherent in comprehending individual linguistic symbols and constructions.” (Tomasello 1999: 173): disagreements, repairs/explanations and meta-discourse. – Closely related to the above is the “narrative practice hypothesis” that with linguistic proficiency (usually) comes first apprenticeship and then various degrees of mastery in understanding and producing narratives, through which children become familiar with both the core structure of folk psychology and its norm-governed possibilities for using it in practice (Hutto 2008; Gallager and Hutto, this volume). As pointed out in the introduction, there is accumulating evidence for a strong connection from language to the understanding of beliefs and folk psychology in the case of children (e.g. Peterson and Siegal 1995; de Villiers and Pyers 1997; Astington and Jenkins 1999; Hale and Tager-Flusberg 2003; Lohmann and Tomasello 2003). There is also negative evidence for apes, enculturated or not: as mentioned, Call and Tomasello (1999) used a non-verbal false belief task with chimpanzees and orangutans as well as with human children. While the children’s performance on verbal and nonverbal false belief tasks was highly correlated, supporting the hypothesis of a possible causal connection, none of the apes including Chantek, could pass the nonverbal false belief task even though they succeeded in all of the control trials indicating mastery of the general task demands. A prediction from the present analysis would be that if Chantek, or any of the other “language apes” that have been the subject of so much controversy, could progress in their language development from protolanguage to (systematic and narrative) language, they would also be able to pass false belief tasks. It is indicative that no such evidence has so far been offered.
Intersubjectivity and bodily mimesis 237
Table 1. The mimesis hierarchy, intersubjectivity skills and “types of mentality” Level 1. Proto-mimesis
2. Dyadic mimesis
3. Triadic mimesis
4. Post-mimesis1: protolanguage 5. Post-mimesis2: language
4.
Intersubjectivity skills
Type of mentality
– – – – – – –
1st order: lack of complete differentiation between self and other 2nd order: understanding the other through “projection” (identification, but differentiation)
neonatal imitation (simple) empathy mutual attention cognitive empathy shared attention understanding other’s intentions (in competitive contexts) – joint attention – having and understanding communicative intentions – semantic conventions
3rd order (expectations)
– (false) belief understanding
3rd order and higher (beliefs)
3rd order (attention and intentions)
Summary and conclusions
In this chapter I have argued that there is a close connection between the 5 levels of the evolutionary and developmental model referred to as the mimesis hierarchy and corresponding skills in intersubjectivity. There is furthermore a connection between the five levels and what we can call the “type of mentality” involved – reminding that mentality refers to various kinds of states and processes of consciousness, and not only to “propositional attitudes”. These correlations are summarized in Table 1. Proto-mimesis is crucially implicated in mutual attention and the awareness of others’ feelings, through a species-general capacity for empathy that has possibly been further developed in the “ultra-social” species Homo sapiens. Dyadic mimesis leads to the ability to map between one’s own body and that of others in a more detached, differentiated manner, and in this way understand others’ emotions, i.e. cognitive empathy, shared attention and even intentions through a (conscious) process of “projection”: what would I see/feel/wish if I were you. Unlike earlier claims to the contrary, newer evidence and analyses show that apes do not have much difficulties with this level and that they have the capacity for second-order mentality. One of the main claims of this chapter is that the crucial step in the evolution and development of human intersubjectivity involves triadic mimesis, implying having and understanding others’ communicative intentions. I have argued that
238 Jordan Zlatev
this requires third-order mentality: “I want you to do X (e.g. share attention on an object) by recognizing my intention that you do this” from the sender’s perspective and “I understand that you want me to do X” from the recipient’s, but also pointed out that this need not (and does not at this stage) involve beliefs and propositions. Triadic mimesis is clearly difficult for apes to attain, especially in natural conditions. However, through enculturation and especially through extensive sign use, some understanding of communicative intentions seems to be within the reach of apes’ “Zone of Proximal Development”. Evidence for this is the relative mastery of joint attention by enculturants, and as argued this can be seen to originate in (dyadic mimetic) second-order attention combined with the understanding of the other’s intention that I attend. Post-mimesis1 or “protolanguage” implies some understanding of semantic conventions, which I suggested can emerge as shared expectations of common usage, with little if any explicit third-order (propositional) knowledge. The long-term studies of language-taught enculturated apes suggest that this level as well, with much persistence, is at least in part accessible to our nearest animal cousins. Post-mimesis2, which is identical to language as we know it, has on top of everything else the command of a conventional/normative system for communication and thought. Arguably it is first with this level that the real payoff of using the same system for both meta-functions comes into play, giving us the cognitive benefits of (logical) reasoning, inference, long-term planning etc. that we take pride in as a species. It gives us, but no other creature on our planet, a metarepresentational capacity, allowing (at least) second-order beliefs, e.g. “I think that you know (or don’t)”. In summary, bodily mimesis – in its proto, dyadic and triadic forms – can be argued to be a (and possibly the) major factor in the evolution of intersubjectivity, with higher mimetic levels bringing along with them more advanced forms of intersubjectivity such as joint attention and third-order mentality. Sign use itself was suggested to be a driving force in the development of an understanding of third-order mentality, and the performance of enculturants shows that this achievement is within the reach of apes, albeit in special conditions. Therefore one can conclude that (mimetic) sign use was possibly within the “zone of proximal evolution” (Donald 2001) of the common ape-human ancestor, considerably more so than language, characterized by full conventionality and systematicity. To return to the chicken-and-egg question at the beginning of this chapter, the intersubjective skills in the first three levels of the mimesis hierarchy (see Table 1) are indeed “pre-linguistic” according to the analysis offered in this chapter, and serve as a ground for language – in both evolution and ontogenetic development. However, since they are not “theory of mind modules”, but social skills arising through face-to-face and body-to-body interactions, it would be incorrect
Intersubjectivity and bodily mimesis 239
to say that intersubjectivity per se is a prerequisite for language. It is rather the mimetic “first communions” (Hutto this volume), in which the various skills of intersubjectivity are a natural part that prepare the way for language. The emergence of the latter marks a major transition, or even two such transitions: to conventionality and systematicity, and it is possible that the two are necessarily linked (Deacon 2003). The understanding of (false) beliefs and folk psychological reasoning are therefore post-mimetic forms of intersubjectivity, since they are based on language, either spoken or signed. This analysis does not contradict the claims of those who like Bloom (2000) argue that the acquisition of language presupposes “theory of mind” skills such as joint attention and communicative intentions. What it does contradict is classing such skills as “theory of mind” modules or competencies, since they are triadic mimetic phenomena that are far from being theoretical. Finally, I should point out that dividing intersubjectivity along different evolutionary/developmental levels as here suggested could help resolve even other controversies, such as those voiced by Sinha and Rodriguéz (this volume) on the relationship between intersubjectivity and common knowledge. Instead of setting the two in opposition it is quite possible to analyze common knowledge as an advanced, post-mimetic and language-dependent form of intersubjectivty, which I take to be the intention of Itkonen (1978, this volume).
Acknowledgments This chapter originates from joint work with Tomas Persson and Peter Gärdenfors, which gave rise to the analysis presented by Zlatev, Persson and Gärdenfors (2005a, 2005b). Göran Sonesson, Ingar Brinck, Chris Sinha, Esa Itkonen and Mats Andrén have always been very helpful with feedback related to this and related topics. The Lund University project Language, Gesture and Pictures in Semiotic Development and the EU-project Stages on the Evolution and Development of Sign Use (SEDSU) provided the funding and the interdisciplinary framework for the type of cognitive semiotic investigations that this chapter has dealt with.
References Ahlgren, I. 2003. “Teckenspråk.” In Sveriges officiella minoritetsspråk. Finska, meänkieli, samiska, romani, jiddisch och teckenspråk. En kort presentation. Stockholm: Nordstedt.
240 Jordan Zlatev
Arbib, M. 2003. “The evolving mirror system: A neural basis for language readiness.” In Language Evolution, M. Christiansen and S. Kirby (eds.), 182–200, Oxford: Oxford University Press. Arbib, M. 2005. “From monkey-like action recognition to human language: An evolutionary framework for neurolinguistics.” Behavioral and Brain Sciences 28/2: 105–124. Astington, J.W. and Jenkins, J.M. 1999. “A longitudinal study of the relation between language and theory-of-mind development.” Developmental Psychology 35/5: 1311–1320. Baldwin, D. 1991. “Infants’ contributions to the achievement of joint reference.” Child Development 62: 875–890. Baldwin, D. 1993. “Infants’ ability to consult the speaker for clues to word reference.” Journal of Child Language 20: 395–418. Bard K.A., Myowa-Yamakoshi, M., Tomonaga, M., Tanaka, M., Costall, A. and Matsuzawa, T. 2005. “Group differences in the mutual gaze of chimpanzees.” Developmental Psychology 41: 616–624. Barresi, J. and Moore, C. this volume. “The neuroscience of social understanding.” Bates, E., Camaioni, L. and Volterra, V. 1975. “Performatives prior to speech.” Merrill-Palmer Quarterly 21: 205–226. Bickerton, D. 2003. “Symbol and structure: A comprehensive framework for language evolution.” In Language Evolution, M. Christiansen and S. Kirby (eds), 77–93, Oxford: Oxford University Press. Bloom, P. 2000. How Children Learn the Meaning of Words. Cambridge, MA: MIT Press. Bonvillan, J.D. and Patterson, F.G. 1999. “Early sign-language acquisition: Comparison between children and gorillas.” In The Mentalities of Gorillas and Orangutans: Comparative Perspectives, S.T. Parker, R.W. Mitchell and H.L. Miles (eds.), 240–264. Cambridge: Cambridge University Press. Brinck, I. 2001. “Attention and the evolution of intentional communication.” Pragmatics and Cognition 9/2: 255–272. Brinck, I. 2003. “The pragmatics of imperative and declarative pointing.” Cognitive Science Quarterly 3 (4): 429–446. Brinck, I. and Gärdenfors, P. 2003. “Co-operation and communication in apes and humans.” Mind and Language 18(5): 484–501. Brown, R. 1977. “Why are signed languages easier to learn than spoken languages?” In Proceedings of the National Symposium on Sign Language Research and Training, W. Stokoe (ed.), 9–24. Silver Spring, MD: National Association of the Deaf. Byrne, R. and Whiten, A. 1988. Michiavelian Intelligence. Social Expetise an the Evolution of Intellect in Monkeys, Apes and Humans. Oxford: Oxford University Press. Call, J. and Tomasello, M. 1996. “The effect of humans in the cognitive development of apes.” In Reaching into Thought: The Minds of Great Apes. A. Russon, K. Bard and S. Parker (eds.), 371–403. Cambridge: Cambridge University Press. Call, J. and Tomasello, M. 1999. “A nonverbal false belief task: The performance of children and great apes.” Child Development 70/2: 381–395. Clark, H. 1996. Using Language. Cambridge: Cambridge University Press. de Villiers, J. and Pyers, J. 1997. “Complementing cognition: The relationship between language and theory of mind.” In Proceedings of the 21st Annual Boston University Conference on Language Development. Somerville, MA: Cascadillia Press. de Waal, F.B.M 1982. Chimpanzee Politics. London: Cape.
Intersubjectivity and bodily mimesis 241
de Waal, F.B.M. and Aureli, F. 1996. Consolation, Reconciliation and a Possible Cognitive Difference Between Macaques and Chimpanzees. Cambridge: Cambridge University Press. Deacon, T. 1997. The Symbolic Species: The Co-Evolution of Language and the Brain. New York: Norton. Deacon, T. 2003. “Universal grammar and semiotic constraints.” In Language Evolution, M. Christiansen and S. Kirby (eds.), 111–139. Oxford: Oxford University Press. Donald, M. 1991. Origins of the Modern Mind. Three Stages in the Evolution of Culture and Cognition. Harvard: Harvard University Press. Donald, M. 1998. “Mimesis and the Executive Suite: Missing links in language evolution.” In Approaches to the Evolution of Language, J. Hurford, M. Studdert-Kennedy and C. Knight (eds.), 44–67. Cambridge: Cambridge University Press. Donald, M. 2001. A Mind So Rare. The Evolution of Human Consciousness. New York: Norton. Emery, N.J. 2000. “The eyes have it: the neuroethology, function and evolution of social gaze.” Neuroscience and Biobehavioural Reviews 24: 581–604. Fouts, R.S. 1972. “Use of guidance in teaching sign language to a chimpanzee (Pan troglodytes).” Journal of Comparative and Physiological Psychology 80/3: 515–22. Fouts, R.S. 1973. “Acquisition and testing of gestural signs in four young chimpanzees.” Science 180: 978–980. Gallagher, S. 2005. How the Body Shapes the Mind. Oxford: Oxford University Press. Gallagher, S. 2007. “Phenomenological and experimental contributions to understanding embodied experience.” In Body, Language and Mind. Vol 1. Embodiment, T. Ziemke, J. Zlatev and R. Frank (eds.), 271–293. Berlin: Mouton. Gallagher, S. and Hutto, D.D. this volume. “Understanding others through primary interaction and narrative practice.” Gallese, V., Keyners, C. and Rizzolatti, G. 2004. “A unifying view of the basis of social cognition.” Trends in Cognitive Sciences 8/9: 396–403. Greenfield, P.M. and Savage-Rumbaugh. E.S. 1990. “Grammatical combination in Pan paniscus: Processes of learning and invention in the evolution and development of language.” In “Language” and Intelligence in Monkeys and Apes, S.T. Parker and K. Gibson (eds.), 540–578. Cambridge: Cambridge University Press. Greenfield, P.M. and Savage-Rumbaugh, E.S. 1991. “Imitation, grammatical development, and the invention of protogrammar.” In Biological and Behaviorial Determinants of Language Development, N. Krasnegor, D. Rumbaugh, M. Studdert-Kennedy and R. Schiefelbusch (eds.), 235–258. Hillsdale, NJ: Erlbaum. Grice, P. (1957). “Meaning.” Philosophical Review 66: 377–88. Hale, C.M. and Tager-Flusberg, H. 2003. “The influence of language on theory of mind: A training study.” Developmental Science 6/3: 346–359. Hare, B., Call, J. and Tomasello, M. 2001. “Do chimpanzees know what conspecifics know?” Animal Behaviour 61: 139–51. Hare, B., Call, J., Agnetta, B. and Tomasello, M. 2000. “Chimpanzees know what conspecifics do and do not see.” Animal Behaviour 59: 771–85. Hockett, C.F. 1960. “The origin of speech.” Scientific American, 23: 89–96. Hobson, R.P. and Hobson, J.A. this volume. “Engaging, sharing, knowing: Some lessons from research in autism.” Hutto, D.D. 2008. Folk Psychological Narratives: The Socio-cultural Basis of Understanding Reasons. Cambridge MA: MIT Press. Hutto, D.D. this volume. “First communions: Mimetic sharing without theory of mind.”
242 Jordan Zlatev
Itkonen, E. 1978. Grammatical Theory and Metascience. Amsterdam: Benjamins. Itkonen, E. this volume. “The central role of normativity for language and linguistics.” Lewis, D.K. 1969. Convention: A Philosophical Study. Cambridge MA: Harvard University Press. Lohmann, H. and Tomasello, M. 2003. “The role of language in the development of false belief understanding: A training study.” Child Development 74/4: 1130–1144. Meltzoff, A. and Gopnik, A. 1993. “The role of imitation in understanding other persons and developing a theory of mind.” In Understanding other Minds: Perspectives from Autism, S. Baron-Cohen, H. Tager-Flusberg and D.J. Cohen (eds.) 335–366. Oxford: Oxford University Press. Menyuk, P. and Quill, K. 1985. “Semantic problems in autistic children.” In Communication Problems in Autism E. Schoper and G. Mesibov (eds.), 127–145. New York: Plenum Press. Merleau-Ponty, M. 1962 [1945]. Phenomenology of Perception. Routledge and Kegan Paul. Merleau-Ponty, M. 2003. La Nature: Notes, Cours du College de France. Trans. Robert Vallier, Northwestern University Press. Miles, H.L. 1990. “The cognitive foundations for reference in a signing orangutan.” In “Language” and Intelligence in Monkeys and Apes, S.T. Parker and K.R. Gibson (eds.), 511–539. Cambridge: Cambridge University Press. Miles, H.L. 1999. “Symbolic communication with and by great apes.” In The Mentalities of Gorillas and Orangutans, S. Taylor Parker, R.W. Mitchell and H.L. Miles (eds.), 197–210. Cambridge: Cambridge University Press. Mitchell, P. 1997. Introduction to Theory of Mind: Children, Autism and Apes. London: Arnold. Moore, C. and Dunham, P.J. (eds). 1995. Joint Attention: Its Origins and Role in Development. Hilldale, NJ: Lawrence Erlbaum. Myowa-Yamakoshi, M. 2001. “Evolutionary foundation and development of imitation.” In Primate Origins of Human Cognition and Behavior, T. Matsuzawa (ed.), 349–367. Dordrecht: Springer. Myowa-Yamakoshi, M., Tomonaga, M., Tanaka, M. and Matsuzawa, T. 2004. “Imitation in neonatal chimpanzees (Pan troglodytes).” Developmental Science 7/4: 437–42. Nelson, K. 1985. Making Sense. The Acquisition of Shared Meaning. London: Academic Press. Nelson, K. 1996. Language in Cognitive Development. The Emergence of the Mediated Mind. Cambrdige: Cambridge University Press. Nelson, K. and Shaw, L.K. 2002. “Developing a socially shared symbolic system.” In Language, Literacy and Cognitive Development, J. Byrnes and E. Amseli (eds), 27–57. Hillsdale, NJ: Lawrence Erlbaum. Patterson, F. 1978. “The gestures of a gorilla: Language acquisition in another pongid.” Brain and Language 5: 72–97. Patterson, F. 1980. “Innovative use of language in a gorilla: A case study.” In Children’s Language Vol 2, K. Nelson (ed.), 497–561. New York: Garnder Press. Perner, J. 1991. Understanding the Representational Mind. Cambridge, MA: MIT Press. Perner, J., Leekam, S. and Wimmer, H. 1987. “Three-year-old’s difficulty with false belief: The case for a conceptual deficit.” British Journal of Developmental Psychology 5: 125–137. Peterson, C.C. and Siegal, M. 1995. “Deafness, conversation and the theory of mind.” Journal of Child Psychology and Psychiatry and Allied Disciplines 36: 459–474. Piaget, J. 1945. La Formation du Symbole Chez l’enfant. Delachaux et Niestlé. Plunkett, K. 1998. “Language and connectionism.” Language and Cognitive Processes 13: 105– 127.
Intersubjectivity and bodily mimesis 243
Preston, S. and de Waal, F.B.M. 2002. “Empathy: Its ultimate and proximal bases.” Behavioral and Brain Sciences, 25: 1–72. Reddy, V. 2000. “Coyness in early infancy.” Developmental Science 3/2: 186–192. Reddy, V. 2003. “On being an object of attention: Implications of self-other-consciousness.” Trends in Cognitive Science 7/9: 397–402. Reddy, V. 2005. “Before the “third element”: Understanding attention to self.” In Joint Attention, N. Klein (ed.), 85–109. Oxford: Oxford University Press. Rizzolatti, G. and Arbib, M. 1998. “Language within our grasp.” Trends in Neurosciences 21: 188–194. Rizzolatti, G., Fadiga, L., Gallese, V. and Fogassi, L. 1996. “Premotor cortex and the recognition of motor actions.” Cognitive Brain Research 3: 131–141. Saussure, F. de 1916. Cours de Linguistique Générale [Course in General Linguistics]. Paris: Payot. Savage-Rumbaugh, S., Shanker, S., and Taylor, T. 1998. Apes, Language and the Human Mind. Oxford: Oxford University Press. Savage-Rumbaugh, S. and Lewin. R. 1994. Kanzi: The Ape at the Brink of the Human Mind. New York: John Wiley. Searle, J. 1992. The Rediscovery of the Mind. Cambridge, Mass.: MIT Press. Senghas, A., Kita, S. and Özyürek, A. 2004. “Children creating core properties of language: Evidence from an emerging sign language in Nicaragua.” Science 305: 1779–1782. Sinha, C. and Rodríguez, C. this volume. “Language and the signifying object: From convention to imagination.” Sonesson, G. 2007. “From the meaning of embodiment to the embodiment of meaning: A study in phenomenological semiotics.” In Body, Language and Mind. Vol 1. Embodiment, T. Ziemke, J. Zlatev and R. Frank (eds.), 85–127. Berlin: Mouton de Gruyter. Stern, D. 1985. The Interpersonal World of the Infant: A View from Psychoanalysis and Developmental Psychology. New York: Basic Books. Stokoe, W. 1960. “Sign language structure: An outline of the visual communication system of the American deaf.” Studies in Linguistics, Occasional Paper 8, University of Buffalo, New York. Tanner, J.E. and Byrne, R.W. 1996. “Representation of action through iconic gesture in a captive lowland gorilla.” Current Anthropology 37/1: 162–73. Tanner, J.E. and Byrne, R.W. 1999. “The development of spontaneous gestural communication in a group of zoo-living lowland gorillas.” In The Mentalities of Gorillas and Orangutans – Comparative perspectives, S.T. Parker, R.W. Mitchell and H.L. Miles (eds.), 211–239. Cambridge: Cambridge University Press. Terrace, H.S., Petitto, L.A., Sanders, R.J. and Bever, T.G. 1981. Science 211/4477: 87–88. Tomasello, M. 1999. The Cultural Origins of Human Cognition. Cambridge, MA: Harvard University Press. Tomasello, M. 2003. Constructing a Language: A Usage-based Theory of Language Acquisition. Cambridge, MA: Harvard University Press. Tomasello, M., Call, J. and Gluckman, A. 1997. “Comprehension of novel communicative signs by apes and human children.” Child Development 68: 1067–1080. Tomasello, M., Call, J. and Hare, B. 2003. “Chimpanzees understand psychological states – the question is which ones and to what extent.” Trends in Cognitive Sciences 7/4: 153–156. Tomasello, M., Carpenter, M., Call, J., Behne, T. and Moll, H. 2005. “Understanding and sharing intentions: The origins of cultural cognition.” Behavioral and Brain Sciences: 28: 675–735.
244 Jordan Zlatev
Tomasello, M. and Call, J. 2006. “Do chimpazees know what others see – or only what they are looking at?” In Rational Animals, S.L. Hurley and M. Nudds (eds), 371–384. Oxford: Oxford University Press. Trevarthen, C. 1979. “Communication and cooperation in early infancy: A description of primary intersubjectivity.” In Before Speech, M. Bullowa (ed.), 321–347. Cambridge: Cambridge University Press Trevarthen, C. 1992. “An infant’s motives for speaking and thinking in the culture.” In The Dialogical Alternative, H. Wold (ed.), 99–137. Oslo: Scandinavian University Press. Vygotsky, L. S. 1962. Thought and Language. Cambridge MA: MIT Press. Vygotsky, L. S. 1978. Mind in Society. The Development of Higher Psychological Processes. Cambridge, Mass: Harvard University Press. Wittgenstein, L. 1953. Philosophical Investigations. London: Basil Blackwell. Woll, B. and Kyle, J. 2004. “Sign language.” Encyclopedia of Language and Linguistics. Oxford: Elsevier. Wray, A. 2000. “Holistic utterances in protolanguage: the link from primates to humans.” In The Evolutionary Emergence of Language: Social Function and the Origins of Linguistic Form, C. Knight, M. Studdert-Kennedy and J. Hurford (eds.), 285–302. Cambridge: Cambridge University Press. Zlatev, J. 1997. Situated Embodiment. Studies in the Emergence of Spatial Meaning. Stockholm: Gotab. Zlatev, J. 2002. “Mimesis: The “missing link” between signals and symbols in phylogeny and ontogeny.” In Mimesis, Sign and the Evolution of Language, A. Pajunen (ed.), 93–122. University of Turku Press. Zlatev, J. 2003. “Meaning = Life (+ Culture). An outline of a unified biocultural theory of meaning.” Evolution of Communication 4/2: 253–296. Zlatev, J. 2005. “What’s in a schema? Bodily mimesis and the grounding of language.” In From Perception to Meaning: Image Schemas in Cognitive Linguistics, B. Hampe (ed.), 323–342. Berlin: Mouton de Gruyter. Zlatev, J. 2007a. “Language, embodiment and mimesis.” In Body, Language and Mind. Vol 1. Embodiment, T. Ziemke, J. Zlatev and R. Frank (eds.), 297–337. Berlin: Mouton de Gruyter. Zlatev, J. 2007b. “Intersubjectivity, mimetic schemas and the emergence of language.” Intellectica. 2007/2–3 (46–47): 123–152. Zlatev, J. 2008. SEDSU research related to intersubjectivity and conventions. Deliverable 19. EU-FP6 Project: N° 012984, Stages in the Evolution and Development of Sign Use, Nest2003-Path-3: “What it means to be human?” Zlatev, J. in press. “From proto-mimesis to language: Evidence from primatology and social neuroscience.” Journal of Physiology, Paris. Zlatev, J. forthcoming. “Autsim as an impairment in bodily mimesis” Zlatev, J. Brinck, I. and Andrén, M. 2008. “Stages in the development of perceptual intersubjectivity.” In Enacting Intersubjectivity, F. Morganti, A Carassa and G. Riva (eds), 117–132. Amsterdam: IOS Press. Zlatev, J., Persson, T. and Gärdenfors, P. 2005a. “Bodily mimesis as “the missing link” in human cognitive evolution.” Lund University Cognitive Studies 121. Zlatev, J., Persson, T. and Gärdenfors, P. 2005b. “Triadic bodily mimesis is the difference.” Commentary on Tomasello et al. (2005) Behavioral and Brain Sciences: 28: 675–735. Zlatev, J., Racine, T., Sinha, C., and Itkonen, E. this volume. “Intersubjectivity: What makes us human?”
chapter 11
First communions Mimetic sharing without theory of mind Daniel D. Hutto It is widely held that the gradual development of metarepresentational Theory of Mind (ToM) abilities constituted at least one important hominid upgrade. Are such abilities really needed to explain hominid (i) tool-making, (ii) social cohesion, or even (iii) basic interpretative and language formation/learning capabilities? I propose an alternative explanation of what underlies these sophisticated capacities – the Mimetic Ability Hypothesis (MAH). MAH claims that a vastly increased capacity for recreative imagination best explains the kinds of sophisticated intersubjective engagements of which hominids would have been capable – and that these constituted an important basis for the development of complex language. This proposal puts the idea of the evolution of ToM devices under considerable strain. How did humans bridge the tremendous gap between symbolic thought and the nonsymbolic forms of intelligence that still dominate the rest of the animal kingMerlin Donald, Origins of the Modern Mind dom?
1.
The missing cognitive link
Many today claim that full-fledged theory of mind (ToM) abilities – specifically those requiring mastery of the concept of belief – must have emerged at a point in our pre-history sometime after the human line broke from that of the chimpanzees (Call and Tomasello 1999, 2005; Papineau 2003; Povinelli and Vonk 2003; Tomasello, Call and Hare 2003a; Sterelny 2003). This is thought to have occurred during “the million or so years preceeding modern recorded history, the late Stone Age or Pleistocene. This period is referred to by evolutionary psychologists as the Environment of Evolutionary Adaptiveness” (Dupré 2001: 21). The big question is just how much later did full-fledged folk psychological abilities arrive on the scene and why. I claim that they are very late-developing, socio-culturally based and not necessarily universal to human thinking (see
246 Daniel D. Hutto
allagher and Hutto this volume; Hutto 2004, 2007a and b, 2008). In contrast, G those who defend the existence of ToM modules (ToMMs) typically argue that very many important activities of our hominid forerunners necessarily depended on the having of mature ToM abilities. Hence, they posit the formation of dedicated mechanisms which gifted our ancestors with such capacities and, being builtin by biological evolution, these would have been inherited by our species. How might we decide between these possibilities? Cognitive archaeology – the attempt to understand ancestral minds by drawing on insights of psychology applied to remnants of pre-history – is a highly speculative business. Not only is the archaeological record gappy, with only fragmentary material evidence having been preserved (and much of it yet to be discovered), there are no live subjects to test. This means we do not know, with any certainty, which kind of activities our ancestors engaged in and what abilities they may have had. And, as shown by the debates over the cognitive capacities of non-human primates, even when we have live subjects to examine there is scope for competing interpretations about how precisely to characterise and explain the basis of social-psychological abilities (Call and Tomasello 2005; Povinelli and Vonk 2004). By comparision, the task of deciding between rival conjectures about prehistoric cognition is trickier still. Matters are helped somewhat by an emerging consensus about the level, if not the nature, of the cognitive capacities of chimpanzees. Placed alongside what we know about the abilities of modern humans – both infants and adults – this yields at least a very rough sense of the cognitive distance that hominids must have covered: i.e. in thinking about the common ancestor we shared with chimpanzees we know roughly where our forerunners must have started from and where they ended up; even if details of the precise route taken remain obscure. Despite its sketchiness, the archaeological record gives us a general picture, though to be sure a changeable and contestable one, of the large features of the terrain they covered and the likely timing of their specific movements. However imperfect, this is the evidence against which we must test the plausibility of proposals about our ancestors’ likely cognitive powers and what may have driven them. It is beyond doubt that there was a cognitive change of considerable magnitude (or a series of such) that took place over the period spanning from the emergence of a common ape-human ancestor approximately 6 million years ago to the appearance of Homo sapiens (see Figure 1). If we assume a neo-Darwinian perspective it is likely that such changes will have happened gradually, presumably under a variety of selection pressures. Recognising these limitations, in the remainder of this chapter I review the existing evidence and challenge the assumptions that lend prima facie support to the familiar and widely held view that the gradual development of sophisticated mindreading devices must have constituted at least one important hominid upgrade.
First communions 247
Figure 1. A Simplified Evolutionary Tree of the Hominids (reprinted from How Homo Became Sapiens: On the Evolution of Thinking. Gärdenfors, P. p. 7, Oxford University Press, Copyright © 2003. Used with permission from P. Gärdenfors)
2.
Imitation and social learning
Going back to the very first hominids, there is no reason to think that their lifestyles offered challenges of a significantly different kind from that faced by the apes – certainly, no drastic change occurred in this respect until long after the passing of Australopithecus afarensis and even after the arrival of Homo habilis. With the latter’s appearance we see the first major spurt in brain size (which increased roughly 1.5 times in cranial capacity compared to that of the australopithecines). Indeed, the fossil record tells a story of just two such instances of encephalisation in human pre-history. The first, and most remarkable, roughly coincides with the emergence of simple Oldowan tool manufacture – a craft supervised by the first members of the Homo line (see Figure 1). Although the tools made during this period would have been extremely basic in many respects by today’s standards, their fashioning would have constituted an impressive achievement – a genuine innovation – when compared with what had come before (or rather what hadn’t). The stone knapping techniques involved in the fashioning of such tools would have required a good sensitivity to fractural dynamics and strong hand-eye coordination, capacities that far outstrip those required for making simple repetitive actions (Mithen 1996, 2000a; Wynn 2000). An off-line imaginative ability to
248 Daniel D. Hutto
re-enact and practice complex routines would also have been needed, not only for the technical proficiency of fashioning the final products but also for the collection and preparation of materials. The manipulation of images – perceptual re-enactments – looks likely to have funded these aspects of the early tool-maker’s craft. Clearly, not even weak ToM abilities would have been needed for the individual acts of making such tools, but they might have played a crucial part in enabling the social learning upon which the tool industries themselves were founded. For more than an indivudal’s potential to fashion such tools would have been needed to keep the practice alive; such crafts had to be maintained over time. Could it be that ToM abilities might have played a critical pedagogical role in ensuring this? In assessing this idea, let us simply accept the consensus view that “the Paleolithic record suggests very strong social learning of such skills” (Mithen 2000b: 496). Very well, but what is the exact connection with ToM abilities? Mithen suggests that: … it seems most probable that these technical skills and traditions were intentionally taught from generation to generation, or acquired by passive watching and active imitation. A modern-like theory of mind appears essential to either task. Instructed learning requires that both the teacher and the novice take account of what is in each other’s mind. (Mithen 2000b: 496, emphasis added, 2002, see also Baron-Cohen 1999: 263)
Despite this rather bold statement, the idea that passive watching and active imitation implies ToM abilities of any sort is surely false. Demonstrating this is of particular importance because there is a strong independent reason to suppose that hominids must have had impressive imitative abilities. Humans are natural mimics and our basic abilities in this regard seem to be inherited, though they may be elaborated and extended by epigenesis (see Sinha 2004). Human neonates, as is well known, engage in facial imitation even at the tender age of thirty-two hours old (indeed it has been claimed that they are capable of this when they are less than an hour old) (Meltzoff and Moore 1977; Meltzoff and Moore 1994; Gopnik and Meltzoff 1997: 131). Young chimpanzees have shown similar abilities (see Myowa-Yamakoshi et al 2004). Yet the claim is not that humans alone are natural-born mimics, it is that we are also mimics of the first rank. Recent work on social responding carefully distinguishes a number of different forms, including: stimulus enhancement, goal emulation, response priming and imitation proper (Billard and Arbib 2002: 344; . Donald puts the point beautifully: “Innovative tool use could have occurred countless thousands of times without resulting in an established toolmaking industry, unless the individual who ‘invented’ the tool could remember and re-enact or reproduce the operations involved and then communicate them to others” (Donald 1991: 179).
First communions 249
Rizzolatti, Graighero and Fadiga 2002: 52; Hurley 2005). Stimulus enhancement and response priming involve the triggering of an action that is already within the repertoire of the individual, while goal emulation can lead to a new action leading to a desired goal but without copying the means of the model. True or complex imitation, by way of contrast, stands out in that it requires the capacity to copy both the novel ends and means of another’s complex action. When compared with humans, apes are much less good at complex forms of imitation (see Donald 2005: 285–6; Jones 2005). The fact is that it remains a matter of some dispute whether they are capable of true imitation at all. Monkeys are able to copy simple actions, such as a movement sequence tied to a particular goal but they cannot reproduce complex ones involving hierarchically structured repertoires. They can copy bodily movements but without adoption of specific goal structures with multiple sub-steps. Chimpanzees too may only be capable of limited forms of imitation involving uncomplicated movements – their performances only becoming reliable after considerable training. Exactly what level of ability chimpanzees really have in this regard remains a somewhat open question (Zlatev, Persson and Gärdenfors 2005; Whiten, Horner and Marshal-Pescini 2005). But what is not in doubt is that their abilities stand in stark contrast to those of young human infants who are able to masterfully copy complex and novel movements and actions – even those segmented into discrete ordered parts – with no training and little effort (Meltzoff and Moore 1977). Putting these thoughts together, it seems hard to deny that distinctively human imitative abilities are inherited from the hominids, but also considerably extend them. I take this as established common ground. But, pace Mithen, there is simply no reason to think that imitative abilities in any way entail ToM abilities. To see this, it helps to have a clear picture of just what imitation involves. Bermúdez neatly characterises the problem that infants must solve in order to imitate faces: Facial imitation involves matching a seen gesture with an unseen gesture, since in normal circumstances one is aware of one’s own face only haptically and proprioceptively. If successful facial imitation is to take place, a visual awareness of someone else’s face must be apprehended so it can be reproduced on one’s own (Bermúdez 1998: 125) face.
If we model what is going on in the infant on the way a suitably well informed adult might attempt to solve this sort of problem, using a set of explicit propositional instructions, then their ability to converge on precisely the right gesture to be imitated will be regarded as, “the product of inference-like processes [that] are not merely reflexive” (Gopnik and Meltzoff 1997: 130). And it would seem to follow that, “Very young infants represent a variety of aspects of human action, they can make inferences on the basis of these representations, they think of themselves
250 Daniel D. Hutto
and others as fundamentally sharing the same psychological states” (Gopnik and Meltzoff 1997: 133). If so, we must presuppose that infants are aware of at least some sort of substantive self/other contrast from birth – and this might imply some sort of ToM capacity (Gopnik 2004: 22). But this line of thought raises a number of difficult questions. What is the character of this neonatal understanding of the self-other contrast? What is the precise content of the representations that these infants are allegedly using, and what is their origin? And, crucially, what account can be given of the sub-personal mechanisms that make use of these representations in order to effect the appropriate manipulation of infants’ faces and bodies? In offering a straight choice between inferentialist accounts and merely reflexive ones, theory theorists, such as Gopnik and Meltzoff, have missed a trick. Research into Mirror Neuron Systems (MNSs) holds out the promise of a better way of understanding the mechanics of simian and human neonatal abilities to imitate buccal and facial expressions and other gestures (Gallese 2003; Rizzolatti 2005). Even though research in this area is still at the early stage, it is widely agreed that “a mechanism with the characteristics of the mirror system appears to have the potentiality to give a neurophysiological, mechanistic explanation of imitation” (see Rizzolatti et al. 2002: 55). This is an instance in which we have some reason to believe that the promissory note for future explanations might actually get cashed. The fact that MNSs of apes are not as sophisticated as the human variety could explain their limited capacity for complex forms of imitation; the differences are empirically well-established, having been demonstrated by various brain imaging studies (Knoblich and Jordan 2002: 115; Heiser et al. 2003; Arbib 2005). Also, more importantly, in the human case the potential fit with the basic type of inherited mimetic abilities to be explained is plausibly of the right level. It is easy enough to imagine that our early hominid ancestors might have started their mimetic careers with abilities akin to those of “young children [who] spontaneously imitate adults in a mirror-like fashion” (Wohlschläger and Bekkering 2002: 102, emphasis added). Thus, just as children only begin to make the appropriate adjustments for cross-lateral differences when imitating others as they get older, so too hominid . For these reasons the theory-theory approach compares negatively with the account that Gallagher advances, which makes appeal to body schemas. He claims that “the imitating subject depends on a complex background of embodied processes, a body-schema system involving visual, proprioceptive, and vestibular information… This intermodal intra-corporeal communication then, is the basis for an inter-corporeal communication” (Gallagher 2005: 76). Although Gallagher suggests that infants may be capable of experiencing a difference between self and other, he argues that “the concept of the self starts out closer to an embodied sense than to a cognitive or psychological understanding” (Gallagher 2005: 79).
First communions 251
mimetic capacities may have become similarly enriched over evolutionary time. Looking to non-representational, enactive accounts for an explanation of imitative abilities is surely a better bet than positing modules with the relevant declarative knowledge to do such work (see Hutto 2008). Sterelny (2003) identifies and deftly defuses the assumption that can make it look as if a ToM might be needed for these basic kinds of imitative task: The link between imitation and theory of mind depends on the supposition that imitation involves a translation between points of view: the mimic represents something like the model’s motor pattern as seen by an onlooker, and turns it into a representation of a motor pattern as seen by the agent himself. But that is not the only possibility… If the mimic represents the model’s behaviour functionally – pick up the rock in the grasping hand, hold the nut facing away, place it on smooth hard surface – there is no need to transform between points of view. (Sterelny 2003: 64)
There are good reasons to opt for the simpler functional interpretation of this process, appealing to mirror neuron ystems or resonance pattern research in order to understand its mechanics. Although I have said that the mirror systems neuron of humans are much more sophisticated than those of other primates, we must not suppose that the basic acts of imitation which they sponsor have any of the standard features of mentalistically-based simulations per se. Certainly, they do not implicate the kinds of simulation procedures that involve the manipulation and attribution of propositional attitudes (see Goldman 2005: 82; Gallagher 2007). Here it pays to remember that Gallese and Goldman, who were the first to claim that the discovery of mirror neurons might lend respectable empirical backing to simulation theory (ST), only ever took the evidence to show that there were primitive precursors that might be related to explicit, mentalistic forms of
. Interestingly, those who suffer from autism have difficulty in replicating the manner and style of another’s response (see Hobson 2007; Hobson and Hobson this volume). . To have a content-involving thought it is not enough that an organism is merely intentionally directed at a possible situation or state of affairs, for the latter might be understood in purely extensional terms. I therefore distinguish intentional attitudes from propositional attitudes, reserving the latter title for those attitudes that are content-involving. This is not just a terminological stipulation, though I recognize that some would claim that attitudes of both sorts are directed at propositions, albeit in different ways. In my view intentional attitudes are not directed at propositions (only possible situations), yet they exhibit intentional directedness all the same. The reader is free to assume that my use of terms is decided by fiat. The main point is that, within the class of intentionally-directed attitudes we can distinguish between those of the extensional and intensional varieties. For further detail about and justification of this distinction see Hutto 2008, Chs. 3–5.
252 Daniel D. Hutto
simulation (Gallese and Goldman 1998: 498). As such, it is clear that the kind of responsiveness associated with these precursor abilities cannot be identified with or understood in terms of ToM capacities of either the “theory theory” or the “simulation theory” variety. The bottom line is that there is simply no need or warrant to postulate any inferential or theoretical activity on the part of these imitators. And it is mimetic abilities of this sort that suffice to explain how technical skills are acquired and developed by attending either to one’s own or another’s action routines (whether selectively or otherwise) so that these can be recalled and re-enacted. In this context, it is important to take note of something that Donald has taken pains to underscore: “The process that generates these action-patterns relies on a principle of perceptual resemblance; accordingly I have labeled the skill ‘mimesis’ or ‘mimetic skill’” (Donald 1999: 145). Worse still, the truth is that Mithen’s claim that our ancestors would have needed ToM abilities for the social transmission of technical skills does not hold up even if we imagine that basic acts of imitation do involve representing the contents of others’ minds. Thus, even if we suppose that the minds of ancient tool-makers were filled with sets of instructions about how to fashion their artefacts, representing these would have been of little use to the trainee. A clutch of rules – here imagined as a series of conditional statements – is the wrong sort of medium for technical training and instruction of the kind needed to learn basic tool manufacture. Quite the opposite; it is often the case when learning a practical craft that mastering a set of explicit rules is a positive hindrance. This fact will be salient to anyone who has tried to build a do-it-yourself product using only a pictureless set of instructions, with no blueprint for guidance. Call this the “Ikea constraint”. Technical training is not about passing on declarative knowledge that, but of engendering a kind of know-how. Novices typically get the knack by direct hands-on practice. And where this is not possible, they must attend to what the other does and attempt to re-enact the appropriate steps using their visuo-motoric imagination in a cross-modal way. Consider what would have been needed for acquiring the skill of fashioning the Levallois flake, focusing on the diagram provided in Figure 2.
. The fashioning of the Levallois flake was a highly sophisticated prehistoric tool-making technique that originally dates back to the Lower Palaeolithic and which was retained into the Upper Palaeolithic and beyond. Its sophistication makes it a useful test case for deciding if the transmission of tool-making skills must have required the exercise of theory-of-mind capacities. My proposal is that such activities look better suited for explanation in terms of hominid capacities for imaginative re-enactment. And there are yet more deflationary proposals about the origins of such flakes afoot. Should those turn out to be correct they too would undermine
First communions 253
Figure 2. Technique for fashioning the Levallois flake (reproduced from (http://anthro. palomar.edu/homo2/archaic_culture.htm, Used with permission from Dennis O’Neil)
The need for such first-hand and hands-on training is felt even today, as when surgeons are given the chance to ‘see one done’ before performing operations in theatre. It is therefore hardly surprising that verbal instruction, even when it is readily available as an accompaniment, appears to play a minor role in craft apprenticeship, even amongst modern humans (Donald 1991: 213; Wynn 1991). Mimesis of the kind discussed is perfectly suited to enable the non-genetic copying required for the passing down of technical skills through the generations by imitative means. Social training of this kind takes the form of showing not saying (of imagining not propositional thinking). This is consistent with the fact that the mimetic skills in question develop over time; this happened in phylogeny and is likely to be recapitulated in ontogeny too, in an epigenetic manner: imitative learners might become more selective and discriminating in the sorts of routines that they choose to mimic (see Harris and Want 2005; Sinha 2004).
3.
The mimetic ability hypothesis
If ToMMs are neither implicated in the maintenance of tool industries nor the basis of imitative capacities, perhaps they were needed to fund the sophisticated the claim that theory-of-mind capacities must have sponsored the social learning that made tool-making industries possible.
254 Daniel D. Hutto
social skills of the hominids. This seems most unlikely. H. habilis looks to have been living in groups only marginally larger in size than that of australopithecines, and both would have kept within relatively tight geographical boundaries (Mithen 1996). There is no compelling reason to think that the social circumstances of the very early hominids would have changed enough to require more powerful tools for engaging in or monitoring social dynamics of a qualitatively different sort than those afforded to apes. Very much in line with current thinking about the abilities of our simian relatives in this regard, Mithen – who has done more than most in thinking about the likely stages of ToM development – surmises that the Oldowan tool makers would have had “an equivalent theory of mind ability to that found within chimpanzees today. As an alternative they would have been extremely clever behaviourists” (Mithen 2000b: 500). Things changed decisively with the coming of Homo ergaster/erectus in the Lower Paleolithic period. Their arrival was accompanied by a remarkable new way of life, which some regard as constituting the first Hominid Revolution. Unlike its predecessors, the extent of H. ergaster/erectus movements – coming out of Africa and spreading across Europe and Asia – were unprecedented. During this period the quality of tools improved quite dramatically, so much so that “archaeologists require months of training and practice to become good at creating Acheulian tools” (Donald 1991: 179). This could be explained by an increase in mimetic ability, which would have also conferred advantages with respect to other technical crafts such as shelter construction. But such skills would have also permitted new and more complex forms of social coordination. Many animals, but particularly those who form cohesive social units, faithfully produce and respond to characteristic expressive behaviours of others, normally conspecifics (Allen and Saidel 1998). They can signal shifts in emotional temperament or mood and otherwise indicate their readiness to engage in characteristic kinds of action. For example the barring of teeth, the arching of backs and the lowering of heads can be early warnings that the other is preparing to fight, or mate, or retreat, and so on. Being able to faithfully produce and respond to such recognisable behaviours makes basic social coordination possible. The mimetic ability with which hominids appear to have been gifted would have been qualitatively unlike these other animal signal systems, not just in its special
. There is current debate about whether the early African H. ergaster (meaning workman) or the later Asian H. erectus was the direct ancestor of modern humans. Indeed, there is debate over whether or not the former is merely a sub-species of the later. Either way, they will have established the Acheulian stone industry before their offspring will have left Africa to become H. heidelbergensis, our last common ancestor shared with the Neanderthals.
First communions 255
mirroring character but in being securely under voluntary control. Although it is likely that our ancestors would have used the full range of facial, vocal and postural gestures, at least to some limited extent, it is likely that manual gestures would have dominated (Donald 1991; Arbib 2005). It is likely that the tree-based living of their simian forefathers would have prepared the early hominids with prodigious dexterous freedom and control – a freedom which the shift to bipedalism would have allowed them to capitalise upon (Corballis 2003; Lieberman 2000: 151–153). Also, being self-cueing and self-regulating, the hominid ability to imitate would have constituted a major step beyond the more stereotypically circumscribed patterns of interaction that characterise the intersubjective engagements of other social creatures; those that depend mainly on inherited routines alone to structure the basic form of their engagements. A flexibility conferred by mimetic skills in conjunction with recreative imaginative abilities for practice and rehearsal would have had vastly increased the developmental possibilities for social expression and engagement (see Hutto 2007d; Sinha 2004; Zlatev this volume). This openness would have introduced new challenges; coordinating intersubjective interactions in stable and effective ways would require taming or regulating these newfound capacities for freedom of expression, at least within local communities. It is very plausible that this was achieved by the development of a kind of mimetic culture. Donald has convincingly argued that the establishment of such would have funded the emergence of games, rites, and well-defined norms of a kind unlike anything found in simian ‘societies’. A kind of mutual miming – which Donald calls ‘reciprocal mimesis’ – is the plausible basis of nonlinguistic conventions (Donald 1991: 6). Such interactions could have acted as powerful social glue. At the very least, the development of mimetic abilities which sponsored the emergence of a unique hominid culture is a credible explanation of changed living patterns and augmented technical and social abilities of H. ergaster/erectus, without the need to postulate that these hominids were in command of anything like a modern language. Certainly, a capacity for reciprocal miming and
. It seems safe to assume that such abilities were under the voluntary control of the hominids. Whether some other animals have some degree of control over their vocalisations is still an open question (Allen and Sidel 1998; Hauser and Marler 1993a, 1993b; Marler and Evans 1995; see also Corballis 2003: 202). . The development of domain-general mimetic abilities obviates the need to postulate the “emergence of an evolved mechanism for identifying, memorizing and reasoning about social norms, together with a powerful motivation to comply with such norms. And with norms and norm-based motivation added to the human phenotype, the stage would be set for much that is distinctive in human cultures” (Carruthers 2003: 75).
256 Daniel D. Hutto
the establishment of a mimetic culture could have played a central part in an impressive list of important activities and practices such as childrearing, coordinated hunting and gathering, food sharing, defining community recognised social ranks and statuses. Mimetic abilities look well-suited to explain norm governed social interactions of the sort needed for the remarkable and wide-ranging achievements of H. ergaster/erectus. Yet all of this would have been available “in the absence of language” (Donald 1991: 174). For, as Donald makes quite clear “Language is not necessary for the development of complex social roles and rules, but mimesis is essential” (Donald 1991: 175). A steady increase along this cognitive trajectory might explain why, although the second period of significant encephalisation that occurred with the advent of Homo sapiens was still a long way off, hominid brain sizes were increasing slightly all the while. And this needs explaining since having larger brains came at a heavy price. They are costly to run and feeding them on small stomachs requires a highquality diet – one that is not easy to acquire. Big brains will have made other demands too; especially for bipeds. Not only do they require more energy, they make birthing difficult (Lovejoy 1980). This has other consequences. Having bigger brained offspring is problematic for those who walk on two legs, for it meant that babies had to be pushed through rather narrow birth canals. The solution to this problem, having immature offspring, saddled hominids with all the burdens of dealing with prolonged periods of ‘childhood’ (Locke and Bogin 2006). This is not seen in other primates but is pronounced in humans. H. ergaster/erectus would have required dedicated practices of pedagogy and childcare. Thus, even though it is not possible to draw direct conclusions about specific modes of cognition based on brain size, we can conclude that there must have been some major trade-off (or trade-offs) for all of these changes. It seems plausible that expansion of domain-general (not general purpose) mimetic abilities, which may have reached a first plateau with H. habilis, may have been at least partly responsible for the growth in their neural volume. The further enhancement of these abilities, could have been the source, not only of the more advanced technical skills exhibited by H. ergaster/erectus but also the basis for their dramatically different kinds of social engagements. The technical and social advantages conferred by imaginative-imitative abilities would have had a ratcheting-up effect, independently spurring on and reinforcing their selection. This might explain why, despite the seemingly great achievements of hominids during the reign of H. ergaster/erectus, there is an extended period of steady but unremarkable brain growth – which lasted until the arrival of the early humans.
. For a discussion of the running costs of brains see Aiello and Wheeler (1995).
First communions 257
This is surely consistent with the hypothesis that with H. ergaster/erectus mimetic capacities had reached an apex, coming into full swing for the first time. To give this proposal a name, let us call it the Mimetic Ability Hypothesis (or MAH). As presented here the MAH has considerable scope for development. It makes no strong commitments as to the exact level of hominid mimetic abilities at the different stages nor does it say anything very precise about the timings of their emergence. The core claim of MAH does not require that we decide such issues. It only claims, weakly, that mimetic abilities (and not ToMMs) can potentially account for the most important technical and social feats of our immediate ancestors. To make this case in full would require going into too much detail for a short chapter. A slightly more developed version can be found in Hutto 2008, Ch. 11 and 12. But in support of the weak claim, in what follows, I concentrate on critically assessing and rejecting some of the more prominent reasons that have been offered for thinking that theory-of-mind abilities must have been necessary for the hominids.
4.
Why else ToMMs?
Once the MAH is articulated it puts great pressure on claims that our ancient ancestors must have used ToMMs in order to get by in their daily routines. In this regard, it is worth noting that the claim that mindreading devices would have been necessary for hominids gets much of its credibility from equivocation about the level of abilities we are seeking to understand. Thus it has been argued that ToM abilities would have been needed in order to share a plan or a goal, as required to develop and implement sophisticated hunting tactics or erecting various constructions (Baron-Cohen 1999: 264). Yet this thought must be weighed up against the fact that even wild chimpanzees and other group animals are quite capable of coordinating their hunting efforts, despite a manifest lack of mature ToM abilities (Boesch and Boesch-Achermann 2000; Brinck and Gärdenfors 2003). At the very least, such facts encourage taking extreme caution when drawing inferences about the degree of mentalising capacity that might have been needed by our ancestors. Some have exercised it. For example, Mithen speculates that – at most – only a desire-based psychology would have been needed in order to account for the kinds of behaviours that would have been witnessed from the time of H. ergaster/erectus to the rise of archaic humans. The trouble for those who postulate innate mindreading devices is that it is easy enough to understand a purely ‘desirebased psychology’ in terms of an appropriate capacity for unprincipled embodied
258 Daniel D. Hutto
e ngagements – the having of a kind of intentional attitude psychology (as discussed in Hutto 2006a, 2008). In this light, Dunbar’s claim that more would have been needed at this intermediate phase of hominid evolution becomes particularly important. He has offered special reasons for thinking that nothing short of full metarepresentational ToM abilities would have been required. For, he argues, it is only by having such abilities that those hominids could have managed to have lived in the large groups to which they were accustomed. Dunbar has independently established that there is a direct correlation between neocortical volume in primates and the maximal size of their workable social groups (Dunbar 1992, 1993). Extrapolating from this data, and estimating the brain size of H. ergaster/erectus, he suggests that these hominids may have been operating in social groups with as many as 150 members. There are good reasons to believe this might have been so. Groups of this size would have afforded greater protection and improved capacities to defend and access resources and new opportunities for predation – but at a price. For, upon reaching a critical ceiling they would have become hard to manage. The maintenance of social cohesion certainly would not have been possible using the same methods as their predecessors. A crucial factor in keeping a handle on this more complex social matrix would have been to keep tabs on personal relationships – both one’s own, as established with specific individuals, as well as monitoring the third-party alliances and interactions of others. A quantitative increase in domain-general capacities for re-identifying particulars and stronger working memory, not unlike that of apes, would be sufficient to explain this achievement. But keeping track of a social space and one’s place in it is one thing, securing and maintaining one’s relationships within it is quite another. Monkeys and apes manage the latter through one-to-one physical grooming, but this requires direct contact and interaction with a restricted number of others. While effective, this method is a thief of time and – in any case – once groups get too large and physically disperse such intensive interactions would have been impossible. Using cranial comparisons with apes and indexing these to their social habits as a base-line, Dunbar calculates that the budget of time required for physical grooming would have reached pressure point in groups of the size that H. ergaster/erectus would have been operating in (see also Mithen 1996: 111). If so, these hominids would have been forced to find fresh methods to substitute for personal grooming. Dunbar’s proposal is that they switched to linguistic as opposed to tactile means for achieving this – our early ancestors would have had to learn to gossip (Dunbar 2003, 2004). If we assume that these linguistic exchanges involved the conveyance of, and conversations about, propositional attitudes then mature ToM abilities are
First communions 259
e ntailed. In arguing that we should accept this, Dunbar assumes the truth of an intention-based semantics and a particular version of the communicative view of language. Crudely, on this view, the primary role of language is to clothe the pre-existing private thoughts of conversationalists – it provides the conventional forms that serve as the public medium for the sharing of these. Comprehension and production are understood in translational terms, involving the appropriate coding of outputs from and the decoding of inputs to the language of thought of each participant. This has been aptly named the ‘inner process model’ (Rowlands 2003: 76–77). It casts natural language signs in the role of mere facilitators, lacking any intrinsic representative power of their own. If one accepts this picture of the function of language, then metarepresentational devices would have been necessary pre-adaptations for such public exchanges. They would be crucially implicated in the translational processes since a hearer’s grasp of a speaker’s meaning would depend on deciphering the speaker’s communicative intention – understood as their sincere assertion; i.e. they would be giving expression to what they believe (Grice 1989). If public communication is, at root, an attempt by hearers to grasp what individual speakers have in mind – i.e. the content of what they intend to assert by their utterances – then anyone engaged in such activity must be presumed to have intact metarepresentational ToM abilities (and the relevant mechanisms to support these). In this way, folk psychological abilities would have played a vital role “in ensuring temporal cohesion of large dispersed groups” (Dunbar 2000: 250). Dunbar is explicit about the relevant implications: Theory of mind is probably essential for language, not so much because it is involved in the production of speech per se but because it provides a mechanism that both enables speakers to ensure that their message has got through and allows hearers to figure out what the speaker’s message actually is (subtext and all). (Dunbar 2003: 224, emphasis added)
Although in some quarters the communicative view of language still has the status of being the received view, it deserves to be treated with scepticism (Gauker 2003). In its strong form it rests on the assumption that pre-verbal individuals would have been capable of propositional thought prior to developing a public medium appropriate for the expression of such thoughts. Yet if my arguments against the very idea of content hold good, this idea is thrown into serious doubt and the communicative conception of language along with it (see Hutto 2006b, 2007c).10 10. The argument developed here only targets those versions of the communicative view of language that presuppose the existence of a meaning-conferring Language of Thought. It may be possible to accept a Gricean analysis of mature language, while rejecting the claim that beliefs
260 Daniel D. Hutto
These considerations alone suffice to make Dunbar’s conclusions precarious, to say the least. But his proposal is implausible in any case, if we accept widely held views about the arrival dates of a symbolic language with the sort of complexity that would have been needed for the reliable formation, encoding and public expression of content-involving propositional attitudes. Since they lacked vocal tracts of an equivalent kind to that of modern humans, “the anatomy necessary to produce the full range of human speech was absent in H.[ ergaster/]erectus and certain, if not all, Neanderthals” (Lieberman 2000: 136). It might be thought that this was only a barrier to the expression of content-involving propositional attitudes. But, as I argue elsewhere, if there could be no such attitudes prior to the establishment of public languages with stable meanings, then there would have been no thought contents for these early hominid speakers to express in any case (see Hutto 2007c). In all if we assume, in line with standard thinking, that the kinds of linguistic practices needed to support discursive conversational exchange were still a long way off, it is much more likely that grooming-at-distance was achieved by the making of “pleasant but meaningless noises” (Bickerton 2003: 79). This may have involved the exchanging of familiar idiosyncratic calls and would have required a reasonable vocal control – of the sort we find in the ‘duetting’ used by chimpanzees, gorillas and baboons; this is used by these animals to keep in touch with each other when they are out of one another’s sight. Signature calls used in this way may have served as a vocal analogue of the repeated actions normally involved in manual grooming. These performances would be, in key respects, like infantdirected speech. This is a more plausible hypothesis than Dunbar’s if we suppose, as seems likely, that, at this stage, hominids would only have had the capacity to engage in non-conversational exchanges which had “none of the properties of conversation” (Corballis 2003: 203). Building directly on Dunbar’s work, Dautenhahn (2002) has proposed that the capacity to produce and comprehend narratives, taking the form of conversational gossip, may have evolved in order to resolve the dilemma of living in larger social groups mentioned earlier: “the evolutionary origin of communicating in stories coevolved with the increasing social dynamics among our human ancestors, in particular the necessity to communicate about third-party relationships” (Dautenhahn 2002: 103–104, 2001: 252). Stated in this way, Dautenhahn’s proposal seems to would have been a prerequisite for it. In other words, linguistic utterances have representational content, the conventional meanings of which are mutually known, but in both ontogeny and phylogeny, language use does not get its life from belief-based understandings; rather it enables the establishment of such understandings, in time (see Zlatev 2007, this volume; cf. Eilan 2005: 13–14).
First communions 261
i nherit all of the problems that haunt Dunbar’s account. But, in fact, these evaporate if we distinguish between two importantly different kinds of narrative, those of a purely dramatic re-enactive sort and those which are linguistically based.11 A mimetic culture could have plausibly sponsored the former calling on established canonical forms, roles and figures in doing so. These would have been the obvious precursors to oral myth and story, but they cannot be identified with such. There is a more modest rendering of Dautenhahn’s scenario, according to which the dramatic re-enactments of social happenings would have taken the form of stories that were literally played out. These would have had a recognisable pre-narrative format and structure of the sort that would make these embodied stories ripe for verbal rendering (Dautenhahn 2001). Thus, as Donald suggests: … if hominids could comprehend and remember a complex event, such as the killing of an animal or the manufacture of a tool, they should have been capable of re-enacting such events, individually or in groups, once mimetic capacity was (Donald 1999: 146) established.
Regular re-enactments of events of special significance may have eventually become deeply ingrained in the social fabric, thus supporting the establishment of common customs and habits. Established dramatic re-enactments and nonlinguistic conventions would have been a powerful substitute means of ensuring social cohesion, supplanting or at least supplementing the physical grooming of individuals. Mimetic interactions of this kind would have helped to solidify within-group identities, obviating the need for more direct and physically taxing forms of one-to-one social maintenance.12 Either way, if only non-linguistic grooming methods were used then there is no need to suppose that the early hominids would have needed (mature) ToMs. Similar considerations defeat the claim that “Pretending necessarily requires a theory of mind” (Baron-Cohen 1999: 265, emphasis added). For if the MAH is even possibly true then a capacity for quite sophisticated forms of pretence – those powerful enough to have fuelled a robust ‘public theatre’ – surely did not rest on
11. We can thus re-name Dautenhahn’s Narrative Intelligence Hypothesis, which is a direct descendant of Dunbar’s, the Pre-Narrative Intelligence Hypothesis (PIH) to avoid confusion. This recognises that there are “two aspects of children’s narrative activity which are too often treated in mutual isolation: the discursive exposition of narratives in storytelling and their enactments in pretend play” (see Richner and Nicolopoulou 2001: 408). 12. This may explain why, being in the same form as the dramatic re-enactments of our ancient ancestors “Children’s first narrative productions occur in action, in episodes of symbolic play by groups of peers, accompanied by – rather than solely through – language. Play is an important developmental source of narrative” (Nelson 2003: 28).
262 Daniel D. Hutto
having a ToM.13 Here it helps to recall the recent evidence that has been effectively marshaled against Leslie’s claim that this sort of pretence involves metarepresentation (cf. Leslie 1987, 1994, Berguno and Bowler 2004). Moreover, there is a clear connection between the kinds of imaginative re-enactments that a mimetic culture would have sponsored and the basis for the kind of narrative competency young children first exhibit (see Hutto 2006h). Social dramas are, of course, the very stuff of many narratives – such engagements set the broad parameters for deciding which events are interesting; these provide the subject matter for much narration. Without question, stories could have been told about the actions of our ancient ancestors. Certainly, their lives would have been dramatic enough. Nevertheless, they – lacking the appropriate medium and established practice – were in no position to tell such stories. We must be clear about the order of appearance: narratives could not have been related orally or conversationally in the early stages of our pre-history on the assumption that “there can be no narrative without narration, a point sometimes overlooked by those who see human life in terms of narratives untold or waiting to be told” (Lamarque 2004: 394).14 And assuming that H. ergaster/erectus lacked a (sufficiently complex) language, the resources for conducting discursive conversation and telling stories would have been missing (Dunn 1991). In place of such discourse, a blossoming capacity for mimesis may have taken up the slack.
5.
First communions
Even if they lacked language it is widely supposed that the hominids as far back as H. ergaster/erectus would have used a kind of proto-language – one which had some but not all the features of a full fledged language. Specifically, if we follow Bickerton, such a ‘language’ would have consisted of grammarless utterances, 13. Carruthers puts a very late date on the development of a capacity for imaginative pretence, suggesting it only emerged at some point between 30–60,000 years ago, thus restricting it to H. sapiens sapiens. He supposes that its onset, along with the amassing of cross-domain knowledge which would have already been made possible through the medium of public language, resulted in an avalanche of creative thinking, cultural artefacts and novel modus vivendi for our ancient ancestors (Carruthers 1998: 115). He finds support for this hypothesis in the putative fact that if earlier hominids, such as H. ergaster/erectus, had exercised their imaginations then we would have seen a much greater impact of this in the patterns of their life. But Carruthers may be looking for the wrong sort of evidence, for instead of appearing in the form of symbolic art, the recreative imagination may have left its mark visibly on tool-making industries and invisibly on a mimetic culture. 14. Or more succinctly: “a story must be told, it is not found” (Lamarque 2004: 394).
First communions 263
rather like the one and two-word offerings of two-year olds; being comparable to pidgins. Thus it might be thought that metarepresentational abilities must be implicated in the formation of such a basic proto-language and its subsequent acquisition. The forging and learning of a public lexicon is based on a capacity for co-reference, and on the received view “children require some type of theory of mind. They need to have an idea of what the parent is thinking and how his/her mind differs from their own to be so very good at associating the correct utterance with the correct reference” (Mithen 2000b: 497). If so, perhaps metarepresentational mechanisms played an instrumental part in the development of the very first public ‘languages’. The assumption here is that the formation and learning of a public lexicon rests on the most sophisticated kind of pre-verbal intersubjective engagement – joint attention. And the idea is that joint attention looks hard to account for without invoking ToM abilities of some kind or other. Certainly, the capacity of human infants of around 1 year to attend jointly with others is the most sophisticated form of non-verbal intersubjective engagement. It involves participants not only attending to the same objects at the same time; it also requires that they mutually recognise that their attending has a common focus – that they both attend to one another’s attending to one another and some worldly point of interest. Intersubjective interactions of this kind are thus of a quite different order from more mirror-like forms of reciprocal imitation. The capacity builds upon the uniquely human capacities for declarative pointing, social referencing and, as will become clear momentarily, imaginative perspective shifting.15 Typically the scene is set for such engagements when one or other of the participants draws attention to some object or feature in the local surround, using the sort of strategies familiar to children once they have learnt to point declaratively, at around 9–12 months old (Tomasello 2003). Gaze monitoring, ‘checking back’ and other social referencing techniques are then used to ensure that the communicative triangle has been established and that it is maintained. In this way, joint attention involves attending not just to the object of the other’s interest but also to “the intention behind the behaviour (yet without entertaining higher order thoughts about it)” (Brinck 2004: 196). If we focus on this aspect of the phenomenon it might be assumed that it requires making some kind of simulative leap. Simulation is often presented as a kind of imaginative attempt to adopt another’s perspective on events, a coming 15. Joint attention of this kind has never been shown to be mastered by apes, not even adult apes. Indeed, the gulf between their abilities and ours in this domain is so wide that it suggests that to the extent that they can attend jointly at all it is “an evolutionarily different version” (Gómez 2005: 80; see also Leekam 2005: 225).
264 Daniel D. Hutto
to see how the world looks from the other’s perspective. It is as if I had climbed in behind your eyes, while at the same time recognising that I am not you. But there are problems even for those defenders of simulation theory, such as Gordon (1995, 1996), who only invokes a first-personal ‘transformation’ as opposed to a ‘projective’ version of the theory. For, unless such an act of simulation is assumed to take place within a context in which the simulators are somehow already independently aware that the other in question has a different point of view, the heuristic would be an utterly hopeless means of getting to grips with the perspective of others; at best it would be a means of becoming the other while losing one’s sense of self (see also Gallagher 2007). Whereas in joint attention what is required is both identification and recognition of difference (Hobson 2007). It involves having the experience of acting in and attending to a shared world, alongside others – in response to public objects. This is quite different from the experience of identifying with others or of merely acting on the world in response to objects. In such non-verbal engagements I ‘see’ what the other is attending to, I see that they are attending to it, and I see that they are attending to my attending to both the object and to my attending (see Zlatev this volume). Only in this way is the object recognised as a common focal point. The mechanisms that underpin the mutual connectedness that enable identification and common focus are likely to have their ancient roots in mirror neuron systems. Interestingly, key elements of the human mirror neuron system for grasping are found in Broca’s area – an area of the brain that relates to language production and comprehension. This has inspired researchers to speculate that there may be interesting, possibly quite tight, connections between this kind of intentional attunement and basic lexicon forming abilities (Rizolatti and Arbib 1998; Billard and Arbib 2002; Arbib 2005). For example, Arbib (2005) suggests that this part of the brain ‘evolved atop’ of those implicated in more basic modes of intersubjective engagement, precisely because it once subserved a manual, gesture-based form of communication which he calls proto-sign. It is plausible that resonance systems of this kind may have played a pivotal role in enabling humans to enjoy a shared world and develop a common language for describing it, ultimately the intentional attunements they subserve may be the basis for the ‘parity’ between speakers and hearers that enable first communions (Arbib 2003, 2005; Heiser, Iacoboni, Maeda, Marcus and Mazziotta 2003).16 These lines of research are especially attractive given the important link between joint attention and word learning. 16. It seems likely therefore that the first non-syntactic proto-language would have been manually-based. If so, it would follow that language evolution took a circuitous route – i.e. speechdominated language would not have emerged directly from primate call systems. Some claim that the latter is the more straightforward explanation (Dunbar 2003). That is certainly so, but
First communions 265
Yet, as noted above, to jointly attend requires more than mere identification. It also requires being able to see the other as other. This fact can make it look as if such triangulation must be in debt to the services of some kind of inferencebased ToM abilities, of the TT or ST sort. To attend to the other’s perceptual take in the appropriate way can certainly seem to require a capacity for conceptually distinguishing between self and other, and indeed of making ‘inferences from me to you’ based on assumptions of similarity by some means. Characterised in this way, joint attention might be thought to implicate full fledged mindreading abilities involving propositional attitudes. But it ain’t necessarily so. As always, there are richer and leaner interpretations, even when it comes to making sense of what is perhaps the most sophisticated form of non-verbal intersubjectivity (Carpendale and Lewis 2004). There is little doubt that joint attention depends on having a multi-tasking ability to shift one’s perspective across distinct axes – focusing, at different moments, on the common focal point, the other’s attention to the same, and also on the other’s attending to one’s own attending. It is quite plausible that such interactions therefore call on recreative imaginative perspective shifting capacities. But these should not be confused with mindreading activity of the folk psychological sort. As those who have done the most to explicate their nature make clear, they are non-propositional and non-simulative (Currie and Ravenscroft 2003: 96). Seeing another’s seeing (whether directed at the world or at my own seeing) does not involve representing the other’s cognitive take on things, it only involves imagining their perceptual ones (see Gallagher and Hutto, this volume). If this is right, even the most sophisticated form of non-verbal intersubjective engagement does not involve the manipulation or attribution of propositional attitudes. We need appeal to nothing more than capacities for intentional attunement and recreative imagination to explain pre-linguistic joint attention. Most importantly, such acts do not require their participants to make full fledged propositional attitude ascriptions. Infantile forms of attending to the attending of others cannot be explicated in terms of metarepresentational understanding, if that understanding is only operative at a later stage, after the concept of belief has been mastered (Moses 2001; Wellman and Phillips 2001; Woodward, Sommerville and Guajardo 2001). We have little choice but to conclude that non-verbal acts of joint attention are only based in responsiveness to intentional attitudes – not propositional ones. Infantile capacities to identify and respond to intentions-in-acting are quite distinct from the understanding of intentions that rest on having mastered the concepts of desire and belief (and how these interrelate to form reasons). there is simply no a priori reason to suppose that straightforward explanations are always the best ones.
266 Daniel D. Hutto
Baron-Cohen claims that “understanding that words refer presumes the concept of intention or goal” (Baron-Cohen 1999: 267). Yet, as just argued, there is every reason to think that even if this is true in some sense it does not imply a capacity for metarepresentation. Put otherwise, even if one were to insist that some ‘concept of intention’ is needed it could not be one that is equivalent to the mature folk psychological variety. It is much more plausible that even in the normal case, where basic language learning is supported by children and adults mutually engaging in pre-linguistic acts of joint attention; these only involve a mutual responsiveness to one another’s intentional attitudes. Not only that, but the argument suffers in any case since having joint attentional abilities is not even necessary for learning the basics of a language. Consider that: Children with autism show us just how useless a language capacity is without a theory of mind. Strip out a theory of mind from language use and you have an individual who might have some syntax, the ability to build a vocabulary and a semantic system but what would be missing from their language use and comprehension is pragmatics… Language without a theory of mind is not of course entirely useless. It allows literal communication, acquisition of information from others, requesting, ordering, etc. (Baron-Cohen 1999: 266, emphasis added; second quotation from footnote on p. 267)
The very existence of such linguistic abilities in individuals with autism serves as a kind of existence proof; it demonstrates the falsity of the claim that it is necessary to have ToM abilities in order to learn words. As the above quotation makes amply clear, at least some individuals with autism are surely capable of learning and using language competently to some extent, despite their ToM deficiencies. Presumbly they are able to achieve this because, even though they cannot jointly attend with others, they are supported by veteran language users who can. Without the normal forms of feedback and checking, their teachers must make even stronger assumptions than usual about what it is that the autistic child is attending to when the relevant associations are being forged. It is their job to ensure, to the best of their abilities, that the initiate is making the appropriate connections between items of reference and local labels (see Hutto 1999: 133–4, 2000: 31–5). Indeed this line of argument is made more implausible still when it is observed that full-fledged ToM abilities are not only unnecessary; to use them during basic word-learning situations would be downright unhelpful. Doing so would make it difficult even to establish basic referential triangles in the first place. Ironically, this fact is most evident in the cases of primitive lexicon forging and learning; it has been noted that:
First communions 267
Young children’s deficits in understanding intentional diversity may actually benefit word learning. As many theorists have noted, the task of interpreting novel words presents a complex inductive problem because, logically speaking, any given word could mean any of an infinite set of possible things (Markman 1989; (Sabbagh and Baldwin 2005: 172)17 Quine 1960).
It is clear that without a basic agreement in how to respond to things, symbol grounding and word-learning would not get off the ground. But it could be argued that something akin to the tendency for ‘default attribution’, in which children use the simple heuristic of attibuting beliefs to others based on what they take to be the case, is at play here (Leslie, Freidman and German 2004). This would obviate the need to decide between the countless possible ways of construing the other’s communicative intentions. Perhaps, so the thought goes, hominids had full-fledged ToM abilities after all, but these were only used at their lowest setting during acts of lexicon formation. But, for obvious reasons, this is a very poor strategy for arguing that hominids would have needed sophisticated ToM abilities. If anything, it shows that even if they had them they surely did not need them. After all, the question is not whether it is in some sense logically possible to imagine that hominids had sophisticated ToMs, we are trying to assess if there is any reason to think they actually had them. Rather than supposing that non-linguistic joint attention rests on prior facility with referential symbols, it is much more plausible that it – scaffolded by imitative abilities and the conventions of a mimetic culture – enabled the fashioning of the very first symbols. This would have permitted a kind of objective thinking – an awareness of sharing a world with others – of a kind that, when further supplemented with imaginative and mimetic abilities, could have been a bridge from perceptually-based forms of intersubjective interaction to amodal symbolic thinking (see Arbib 2003, 2005; Sinha 2004; Hutto 2006c). Being able to learn to use genuinely referring symbols is dependent upon our capacity to have a shared awareness of things, not the other way around. There are various proposals about the stages of this process on the market. It is plausible that most rudimentary forms of mime exploited shared ‘mimetic schemas’ – those that tap deeply into what lies at the base of our readily familiar embodied activities, the kind of activities that can be enacted and re-enacted (Zlatev 2005, 2007). These would have served as the initial points of connection with others, drawing on the common ways of acting in the world such as running, 17. For those with full-fledged interpretational abilities the early indicational phase is initially “extremely idiosyncratic: one or only a small number of interlocutors can understand what the child is indicating – i.e., grasp what aspect of the environment it is to which the child is drawing attention” (Rowlands 1999: 196).
268 Daniel D. Hutto
hitting, kicking, and so on. Mimetic schemas are more fundamental than their abstract cousins ‘image schemas’ (see Hampe 2005). Still, even the latter can be thought of as deriving from what is common to basic, embodied ways of responding in relation to certain normally encountered situations and activities. These yield certain familiar contrasts such as “UP-DOWN, IN-OUT, FRONT-BACK, LIGHT-DARK, WARM-COLD, MALE-FEMALE” (Lakoff and Johnson 1980: 58). Dichotomies of this kind have universal resonance because they feature in everyday ways of acting, reacting and interacting with the world and others. Moreover, it is plausible that material artefacts – tools, buildings, furniture, etc. – which can be the focus of joint attention, act as intersubjective anchors, since the activities they afford are non-arbitrary in important ways and these can be made canonical through convention (see Sinha 2005: 1542–1543). Indeed, our tendency to make use of mimesis has stayed with us even after the establishment of symbolic language. For example, we typically adorn purely linguistic speech acts with gestures – even when these are of no use to the hearer either for adding expression or for helping to establish significance, such as when one gestures whilst speaking on the telephone (Corballis 2003). This deeply ingrained way of connecting with others through mimesis looks to have stayed with us as something more than an inert cognitive vestige. Thus it has been convincingly argued that to a large and interesting extent, we only feel we have satisfactorily understood or ‘grasped’ something once we retreat to embodied schemas of some kind or other. However we ultimately choose to make sense of basic mimetic activities, it is from this sort of starting point that it is possible to sketch the gradual stages of likely linguistic development in hominids. For once the practice of jointly attending was well established it is plausible that a basic capacity for mime would have developed into a more rudimentary form of communication, one involving a kind of reference to common focal points (where these might be happenings, actions or objects) even in their absence. In such cases, communicators would need to bring the relevant objects ‘before the other’s mind’ by some means; the referents would have to be in some way ‘invoked’. It is plausible that early mimes might have achieved this by drawing on bonds of visuo-motor associations of a sort that would have been familiar to all participants. In doing so they would have had to tap into associations holding between certain worldly things (objects of potential co-attention) and certain salient aspects of mimetic acts which resemble these or would successfully remind the other of them. For example, this might be done by using highly stylised gestures, such as wriggling one’s arm in slithering fashion to mimic the movements of a snake.18 18. This is a description of what Zlatev calls “triadic mimesis” (Zlatev this volume).
First communions 269
Mimetic acts of this kind are not signals since they involve prior communicative intent. They are not attempts to effect more straightforward coordinations such as initiating imitation (i.e. getting the other to use their arms in a similar way) or to directly cue a certain kind of action routine on the part of the other (i.e. by inspiring characteristic responses that the presence of snakes normally calls forth). To use one’s arm so as to invoke thoughts of snakes is not an imperative act but an attempt at intentional communication, albeit of a crude and unstable kind. That this is possible is evident by the fact that games such as charades exist, but, of course, this sort of game has important structural supports that the imagined mimetic acts just described would have lacked (or would have lacked in the first instance). It should be clear that even a rudimentary capacity to use and appreciate mime would have brought unheralded degrees of freedom and new possibilities for communication. In key respects, its advent would have made the character of our ancestors’ first communicative efforts, quite literally, ‘dramatically’ different from the sorts of signals used for coordination by other animals, even those of our closest living cousins.19 Mimetic communication requires that others make the appropriate connections; they must recognise the significance of the communicative act. And, lacking established conventions, early mimetic acts would have depended on strong associations and resemblances, and these could hardly be relied upon. It is not easy to communicate by means of pantomime, even when using additional supports. To be sure it is a hit and miss affair: definitely more miss than hit; that is, unless the activity is structured and supplemented in important ways. Communicating by pantomime using only non-linguistic resemblance based modes of quasi-reference is a weak and highly ambiguous mode of communicating. Failures at successful indication would have been a spur to fall in line with publicly established norms. This is a matter of negotiating and adjusting one’s methods of communication to suit a public standard, as prompted by requests for clarification. Both participants forge a common understanding by recasting the communicative offerings in line with conventional requirements. This would have been a crucial step on the road from contextual, indicational communications to true predicative symbolic use.
19. Apes, for example, rarely use declarative as opposed to imperative gestures – and only those with extensive human contact do so at all. Hence, it has been speculated that “although apes can master the ‘referential triangle’ in their interactions with humans for instrumental purposes when they are raised in humanlike cultural environments, they still do not attain humanlike social motivations for sharing experience with other intentional beings” (Tomasello and Call 1997: 393).
270 Daniel D. Hutto
6.
Summary and conclusion
The principal aim of this chapter has been to remove a kind of aspect-blindness that is prevalent in much research on intersubjectivity – the idea that our basic social dealings necessarily rest on inherited theory-of-mind or mentalizing capacities. My strategy has been to cast doubt on the standard story about when such capacities were acquired in prehistory by challenging the familiar idea that hominids must have had mature theory-of-mind abilities in order to have (i) engaged in advanced tool-making, (ii) enjoyed social cohesion and (iii) formed and learned language. I have argued that close scrutiny of the available evidence gives no reason for believing this to be true. An alternative explanation, which I call the Mimetic Ability Hypothesis (or MAH), claims that growing recreative imaginative abilities which funded impressive technical skills and activities appears to have better prospects of accounting for the sophisticated social engagements of the hominids – even those implicated in their capacity to form and learn symbolic language. Much more needs to be said concerning the MAH with respect to exact character and level of hominid mimetic abilities – e.g. when and why they will have emerged and the kinds of activities they will have made possible at the various stages of hominid development. Making such refinements to the core thesis goes beyond the ambitions of this chapter but hopefully the sketch provided suffices to demonstrate the value of this research programme and the fact that mimetic abilities (and not ToMMs) – at least potentially – could account for the most important technical and social feats of our immediate ancestors. Most importantly, when the two main proposals are compared side by side, the abductive virtues of the MAH become evident and the suggestion that modern humans must have inherited mature mindreading devices from our nearest ancestors looks like a weak and somewhat incredible hypothesis. Even in sketch, the mere availability of the MAH reveals that we have no overriding reason to suppose that hominids must have had a sophisticated capacity for folk psychological understanding.
References Aiello, L.C. and Wheeler, P. 1995. “The expensive tissue hypothesis.” Current Anthropology 36: 184–193. Allen, C. and Saidel, E. 1998. “The evolution of reference.” In The Evolution of Mind, D.D. Cummins and C. Allen (eds.), 183–203. Oxford: Oxford.
First communions 271
Arbib, M. 2002. “The mirror system, imitation, and the evolution of language.” In Imitation in animals and artifacts, K. Dautenhahn and Nehaniv C.L. (eds.), 229–280. Cambridge, MA: MIT Press. Arbib, M. 2003. “The evolving mirror system.” In Language Evolution, M. Christiansen and S. Kirby (eds.), 182–200. Oxford: Oxford University Press. Arbib, M. 2005. “From monkey-like action recognition to human language: An evolutionary framework for neurolinguistics.” Behavioral and Brain Sciences 28 (2): 105–124. Arbib, M., Billard, A., Iacoboni, M. and Oztop, E. 2000. “Synthetic brain imaging: Grasping, Mirror Neurons and Imitation.” Neural Networks 13: 975–997. Baron-Cohen, S. 1999. “The evolution of Theory of Mind.” In The Descent of Mind: Psychological Persepctives on Hominid Evolution, M.C. Corballis, and S.E.G. Lea (eds.), 261–277. Oxford: Oxford University Press Berguno, G., and Bowler, D. 2004. “Understanding pretence and understanding action.” British Journal of Developmental Psychology 22: 531–544. Bermúdez, J. 1998. The Paradox of Self-Consciousness. Cambridge, MA: MIT Press. Bickerton, D. 2003. “Symbol and structure.” In Language Evolution, M. Christiansen and S. Kirby (eds.), 77–93. Oxford: Oxford University Press. Billard, A. and Arbib, M. 2002. “Mirror neurons and the neural basis for learning by imitation: Computational modelling.” In Mirror Neurons and the Evolution of Brain and Language, M. Stamenov and V. Gallese (eds.), 345–352. Amsterdam/Philadelphia: John Benjamins. Boesch, C. and Boesch-Achermann, H. 2000. Chimpanzees of the Tai Forest: Behavioural Ecology and Evolution. Oxford: Oxford University Press. Brinck, I. 2004. “Joint attention, triangulation and radical interpretation.” Dialectica 58: 179– 205. Brinck, I. and Gärdenfors P. 2003. “Co-operation and communication in apes and humans.” Mind and Language 18: 484–501. Call, J. and Tomasello, M. 1999. “A nonverbal false belief task: The performance of children and great apes.” Child Development 70: 381–395. Call, J. and Tomasello, M. 2005. “What chimpanzees know about seeing, revisited: An explanation of the third kind.” In Joint Attention: Communication and Other Minds, N. Eilan, C. Horel, T. McCormack and J. Roessler (eds.), 43–64. Oxford: Oxford Univeristy Press. Carpendale J.I.M. and Lewis, C. 2004. “Constructing an understanding of the mind: The development of children’s social understanding within social interaction.” Behavioral and Brain Sciences 27 (1): 79–151. Carruthers, P. 1998. “Thinking in language? Evolution and a modularist possibility.” In Language and Thought: Interdisciplinary Themes, P. Carruthers and J. Boucher (eds.), 94–119. Cambridge: Cambridge University Press. Carruthers, P. 2003. “Moderately massive modularity.” In Minds and Persons, A. O’Hear (ed.), Cambridge: Cambridge University Press. Corballis, M.C. 2003. “Gestural origins of language.” In Language Evolution, M. Christiansen and S. Kirby (eds), 201–218. Oxford: Oxford University Press. Currie, G. and Ravenscroft, I. 2003. Recreative Minds. Oxford: Oxford University Press. Dautenhahn, K. 2001. “The narrative intelligence hypothesis: In search of the transactional format of narratives in humans and other social animals.” In Proceedings of the Fourth International Cognitive Technology Conference: CT 2001: Instruments of Mind, M. Beynon, C.L. Nehaniv and K. Dautenhahn (eds.), 248–266. Berlin, Heidelberg: Springer-Verlag.
272 Daniel D. Hutto
Dautenhahn, K. 2002. “The origins of narrative.” International Journal of Cognition and Technology 1: 97–123. Donald, M. 1991. Origins of the Modern Mind: Three Stages in the Evolution of Culture and Cognition. Cambridge, MA: Harvard University Press. Donald, M. 1999. “Preconditions for the evolution of protolanguages.” In The Descent of Mind: Psychological Perspectives on Hominid Evolution, M.C. Corballis and S.E.G. Lea. (eds.), 138–154. Oxford: Oxford University Press. Donald, M. 2005. “Imitation and mimesis.” In Perspectives on Imitation: From Cognitive Neuorscience to Social Science, S.L. Hurley and N. Chater (eds.), 283–300. Cambridge, MA: MIT Press. Dunbar, R.I.M. 1992. “Neocortex size as a constraint on group size in primates.” Journal of Human Evolution 20: 469–493. Dunbar, R.I.M. 1993. “Coevolution of neocortical size, group size and language in humans.” Behavioral and Brain Sciences 16: 681–735. Dunbar, R.I.M. 2000. “On the origin of the human mind.” In Evolution and the Human Mind: Modularity, Language and Meta-Cognition, P. Carruthers and A. Chamberlain. (eds.), 238– 253. Cambridge: Cambridge University Press. Dunbar, R.I.M. 2003. “The origin and subsequent evolution of language.” In Language Evolution, M. Christiansen and S. Kirby (eds.), 217–234. Oxford: Oxford University Press. Dunbar, R.I.M. 2004. Grooming, Gossip and the Evolution of Language. London: Faber and Faber. Dunn, J. 1991. “Understanding others: Evidence from naturalistic studies of children.” In Natural Theories of Mind, A. Whiten (ed.), 51–61. Oxford: Blackwell. Dupré, J. 2001. Human Nature and the Limits of Science. Oxford: Oxford University Press. Eilan, N. 2005. “Joint attention, communication and mind.” In Joint Attention: Communication and Other Minds, N. Eilan, C. Horel, T. McCormack and J. Roessler (eds.), 1–33. Oxford: Oxford Univeristy Press. Gallagher, S. 2005. How the Body Shapes the Mind. Oxford: Oxford University Press. Gallagher, S. 2007. “Logical and phenomenological arguments against simulation theory.” In Folk Psychology Re-Assessed, D.D. Hutto and M. Ratcliffe (eds.), 63–77. Dordrecht: Springer. Gallagher, S. and Hutto, D.D. this volume. “Understanding others through primary interaction and narrative practice.” Gallese, V. 2003. “The manifold nature of interpersonal relations: The quest for a common mechanism.” Philosophical Transactions of the Royal Society of London 358: 517–528. Gallese, V. and Goldman, A. 1998. “Mirror neurons and the simulation theory of mind-reading.” Trends in Cognitive Sciences 2: 493–501. Gärdenfors, P. 2003. How Homo Became Sapiens: On the Evolution of Thinking. Oxford: Oxford University Press. Gauker, C. 2003. Words Without Meaning. Cambridge, MA: MIT Press. Gómez, J.C. 2005. “Joint attention and the notion of subject: Insights from apes, normal children and children with autism.” In Joint Attention: Communication and Other Minds, N. Eilan, C. Horel, T. McCormack and J. Roessler (eds.), 65–84. Oxford: Oxford University Press. Gopnik, A. 2004. “Finding our inner scientist.” Daedalus 133 (1):21–28. Gopnik, A. and Meltzoff, A.N. 1997. Words, Thoughts, and Theories. Cambridge, MA: MIT Press.
First communions 273
Goldman, A.I. 2005. “Imitation, mind reading and simulation.” In Perspectives on Imitation: From Neuroscience to Social Science. Volume 2: Imitation, Human Development and Culture, S.L. Hurley and N. Chater (eds.), 79–93. Cambridge, MA: MIT Press. Gordon, R.M. 1995. “Simulation without introspection or inference from me to you.” In Mental Simulation, M. Davies and T. Stone (eds.), 53–67. Oxford: Blackwell. Gordon, R.M. 1996. “Radical simulationism.” In Theories of Theories of Mind, P. Carruthers and P. Smith (eds.), 11–21. Cambridge: Cambridge University Press. Grice, P. 1989. Studies in the Way of Words. Cambridge, MA: Harvard University Press. Harris, P. and Want, S. 2005. “On learning what not to do: The emergence of selective imitation in tool use by young children.” In Perspectives on Imitation: From Neuroscience to Social Science. Volume 2. Imitation, Human Development and Culture, S.L. Hurley and N. Chater (eds.), 149–162. Cambridge, MA: MIT Press. Heiser, M., Iacoboni, M., Maeda, F., Marcus, J. and Mazziotta, J.C. 2003. “The essential role of broca’s area in imitation.” European Journal of Neuroscience 17: 1123–1128. Hobson P. 2007. “We share, therefore we think.” In Folk Psychology Re-Assessed, D.D. Hutto and M. Ratcliffe (eds.), 41–61. Dordrecht: Springer. Hobson, R.P. and Hobson, J.A. this volume. “Engaging, sharing, knowing: Some lessons from research in autism.” Hurley, S.L. 2005. “The Shared Circuits Hypothesis: A Unified Functional Architecture for Control, Imitation, and Simulation”. In Perspectives on Imitation: From Neuroscience to Social Science. Volume 1. Mechanisms of Imitation and Imitation in Animals, S.L. Hurley and N. Chater (eds). 177–194. Cambridge, MA: MIT Press. Hutto, D.D. 1999. The Presence of Mind. Amsterdam: John Benjamins. Hutto, D.D. 2000. Beyond Physicalism. Amsterdam: John Benjamins. Hutto, D.D. 2003/2006. Wittgenstein and the End of Philosophy: Neither Theory Nor Therapy. Basingstoke: Palgrave Macmillan. Hutto, D.D. 2004. “The limits of spectatorial folk psychology.” Mind and Language 19: 548–573. Hutto, D.D. 2006a. “Unprincipled engagements: Emotional experience, expression and response.” In Radical Enactivism: Intentionality, Phenomenology, and Narrative, R. Menary (ed.), 13–38. Amsterdam: John Benjamins. Hutto, D.D. 2006b. “Against passive intellectualism: Reply to Crane.” In Radical Enactivism: Intentionality, Phenomenology, and Narrative, R. Menary (ed.), 121–149. Amsterdam: John Benjamins. Hutto, D.D. 2006c. “Four Herculean labours: Reply to Hobson.” In Radical Enactivism: Intentionality, Phenomenology, and Narrative, R. Menary (ed.), 185–221. Amsterdam: John Benjamins. Hutto, D.D. 2007a. “Folk psychology without theory or simulation.” In Folk Psychology Re-Assessed, D.D Hutto and M. Ratcliffe (eds.), 115–135. Dordrecht: Springer. Hutto, D.D. 2007b. “The narrative practice hypothesis.” In Narrative and Understanding Persons, D.D. Hutto (ed.), 43–68. Royal Institute of Philosophy Supplement. Cambridge: Cambridge University Press. Hutto, D.D. 2008. Folk Psychological Narratives: The Socio-Cultural Basis of Understanding Reasons. Cambridge, MA: MIT Press. Jones, Susan. 2005. “Why don’t apes ape more?” In Perspectives on Imitation: From Neuroscience to Social Science: Volume 1: Mechanism of Imitation and Imitation in Animals, S.L Hurley and N. Chater (eds), 297–301, Cambridge, MA: MIT Press.
274 Daniel D. Hutto
Knoblich, G. and Jordan, J.S. 2002. “The mirror system and joint action.” In Mirror Neurons and the Evolution of Brain and Language, M. Stamenov and V. Gallese (eds), 115–124. Amsterdam/Philadelphia: John Benjamins. Lakoff, G. and Johnson, M. 1980. Metaphors We Live By. Chicago: Chicago University Press. Lamarque, P. 2004. “On not expecting too much from narrative.” Mind and Language 19: 393– 408. Leekam, S. 2005. “Why do children with autism have a joint attention impairment?” In Joint Attention: Communication and Other Minds, N. Eilan, C. Horel, T. McCormack and J. Roessler, (eds.), 65–84. Oxford: Oxford University Press. Leslie, A.M. 1987. “Pretense and representation: The origins of ‘Theory of Mind’.” Psychological Review 94: 412–426. Leslie, A.M. 1994. “Pretending and believing: Issues in the theory of ToMM.” Cognition 50: 211–238. Leslie, A.M., Friedman, O. and German, T.P. 2004. “Core mechanisms in ‘Theory of Mind’.” Trends in Cognitive Sciences 8: 528–533. Lieberman, P. 2000. Human Language and Our Reptilian Brain: The Subcortical Basis of Speech, Syntax and Thought. Cambridge, MA: Harvard University Press. Locke, J. and Bogin, B. 2006. “Language and life-history: A new perspective on the development and evolution of human language.” Behavioral and Brain Sciences 29: 259–325. Lovejoy, C.O. 1980. “Hominid origins: The role of bipedalism.” American Journal of Physical Anthropology 52: 250. Markman, E.M. 1989. Categorization and Naming in Children. Cambridge, MA: MIT Press. Marler, P. and Evans, C. 1995. “Bird calls: Just emotional displays or something more?” Ibis 138: 26–33. Meltzoff, A.N. and Moore, M.K. 1977. “Imitation of facial and manual gestures by human neonates.” Science 198: 75–78. Meltzoff, A.N., Moore, M.K. 1994. “Imitation, memory and the representation of persons.” Infant Behaviour and Development 17: 83–99. Mithen, S. 1996. The Pre-History of the Mind: A Search for the Origins of Art, Religion and Science. London: Thames and Hudson. Mithen, S. 2000a. “Paleoanthropological perspectives on the Theory of Mind.” In Understanding Other Minds, S. Baron-Cohen, H. Tager-Flusberg and D. Cohen (eds.), 488–502. Oxford: Oxford University Press. Mithen, S. 2000b. “Mind, brain and material culture: An archeological perspective.” In Evolution and the Human Mind: Modularity, Language and Meta-Cognition, P. Carruthers and A. Chamberlain (eds.), 207–217. Cambridge: Cambridge University Press. Mithen, S. 2002. “Human evolution and the cognitive basis of science.” In The Cognitive Basis of Science, P. Carruthers, S. Stich and M. Siegal (eds.), 23–40. Cambridge: Cambridge University Press. Mithen, S. 2005. The Singing Neanderthals: The Origins of Music, Language, Mind and Body. London: Weidenfeld and Nicolson. Moses, L.J. 2001. “Some thoughts on ascribing complex intentional concepts to young children.” In Intentions and Intentionality, B. Malle, L.J. Moses and D.A. Baldwin. (eds.), 69–84. Cambridge, MA: MIT Press. Myowa-Yamakoshi, M., Tomonaga, M., Tanaka, M. and Matsuzawa, T. 2004. “Imitation in neonatal chimpanzees (Pan troglodytes).” Developmental Science 7 (4): 437–42.
First communions 275
Nelson, K. 2003. “Narrative and the emergence of a consciousness of self.” In Narrative and Consciousness, G.D. Fireman, T.E.J. McVay and O. Flanagan (eds.), 17–36. Oxford: Oxford University Press. Papineau, D. 2003. The Roots of Reason: Philosophical Essays on Rationality, Evolution and Probability. Oxford: Oxford University Press. Povinelli, D.J. and Vonk, J. 2003. “Chimpanzee minds: Suspiciously human?” Trends in Cognitive Sciences 7: 157–160. Povinelli, D.J. and Vonk, J. 2004. “We don’t need a microscope to explore the chimpanzee’s mind.” Mind and Language 19: 1–28. Quine, W.V. 1960. Word and Object. Cambridge, MA: MIT Press. Richner, E.S. and Nicolopoulou, A. 2001. “The narrative construction of differing conceptions of the person in the development of young children‘s social understanding.” Early Education and Development 12 393–432. Rizzolatti, G. 2005. “The mirror neuron system and imitation.” In Perspectives on Imitation: From Neuroscience to Social Science, S.L. Hurley and N. Chater (eds), 55–76. Cambridge, MA: MIT Press. Rizzolatti, G. and Arbib, M. 1998. “Language within our grasp.” Trends in Neurosciences 21: 188–194. Rizzolatti, G., Graighero, L. and Fadiga, L. 2002. “The mirror neuron system in humans.” In Mirror Neurons and the Evolution of Brain and Language, M. Stamenov and V. Gallese (eds.), 37–62. Amsterdam/Philadelphia: John Benjamins. Rowlands, M. 1999. The Body in Mind. Cambridge: Cambridge University Press. Rowlands, M. 2003. Externalism: Putting the Mind and World Back Together. Chesham: Acumen. Sabbagh, M.A. and Baldwin, D.A. 2005. “Understanding the role of communicative intentions in word learning.” In Joint Attention: Communication and Other Minds, N. Eilan, C. Horel, T. McCormack and J. Roessler (eds.), 165–84. Oxford: Oxford University Press. Sinha, C. 2004. “The evolution of language: From signals to symbols to system.” In Evolution of Communication Systems: A Comparative Approach, D.K. Oller and U. Griebel (eds.), 217–235. Cambridge, MA: MIT Press. Sterelny, K. 2003. Thought in a Hostile World. Oxford: Blackwell. Tomasello, M. 2003. “On the different origins of symbols and grammar.” In Language Evolution, M. Christiansen and S. Kirby (eds.), 94–110. Oxford: Oxford University Press. Tomasello, M. and Call J. 1997. Primate Cognition. New York: Oxford University Press. Tomasello, M., Call, J. and Hare. B. 2003a. “Chimpanzees understand psychological states – The question is which ones and to what extent.” Trends in Cognitive Sciences 7: 153–156. Tomasello, M., Call, J. and Hare, B. 2003b. “Chimpanzees versus humans: It’s not that simple.” Trends in Cognitive Sciences 7: 239–240. Wellman, H. and Phillips A. 2001. “Developing intentional understandings.” In Intentions and Intentionality, B. Malle, L.J. Moses and D.A. Baldwin (eds.), 125–148. Cambridge, MA: MIT Press. Whiten, A., Horner, V. and Marshal-Pescini, S. 2005. “Selective imitation in child and chimpanzee: A window on the construal of other’s actions.” In Perspectives on Imitation: From Neuroscience to Social Science: Volume 1: Mechanism of Imitation and Imitation in Animals, S.L. Hurley and N Chater (eds.), 263–283. Cambridge, MA: MIT Press. Wittgenstein, L. 1953. Philosophical Investigations. Oxford: Basil Blackwell.
276 Daniel D. Hutto
Wittgenstein, L. 1983. Remarks on the Foundations of Mathematics. Oxford: Blackwell. Wohlschläger, A. and Bekkering, H. 2002. “The role of objects in imitation.” In Mirror Neurons and the Evolution of Brain and Language, M. Stamenov and V. Gallese (eds.), 101–113. Amsterdam/Philadelphia: John Benjamins. Woodward, A.L., Sommerville, J.A. and Guajardo, J.J. 2001. “How infants make sense of intentional action.” In Intentions and Intentionality, B. Malle, L.J. Moses and D.A. Baldwin (eds.), 149–170. Cambridge, MA: MIT Press. Wynn, T. 1991. “Tools, grammar and the archeology of cognition.” Cambridge Archeological Journal 1: 191–206. Wynn, T. 2000. “Symmetry and the evolution of the modular linguistic mind.” In Evolution and the Human Mind: Modularity, Language and Meta-Cognition, P. Carruthers and A. Chamberlain (eds.), 113–139. Cambridge: Cambridge University Press. Zlatev, J. 2005. “What’s in a schema? Bodily mimesis and the grounding of language.” In From Perception to Meaning: Image Schemas in Cognitive Linguistics, B. Hampe (ed.), 313–342. Berlin: Mouton de Gruyter. Zlatev, J. 2007. “Language, embodiment and mimesis.” In Body, Language and Mind. Vol 1. Embodiment, T. Ziemke, J. Zlatev and R. Roz Frank (eds.), 297–337. Berlin: Mouton de Gruyter. Zlatev, J. this volume. “The co-evolution of intersubjectivity and bodily mimesis.” Zlatev, J., Persson, T. and Gärdenfors, P. 2005. “Bodily mimesis as “the missing link” in human cognitive evolution.” Lund University Cognitive Studies 121.
part iii
Language
chapter 12
The central role of normativity in language and linguistics Esa Itkonen ‘Any natural language consists of rules which are inherently social and normative.’ It is the purpose of this chapter to establish the truth of this claim and to show that it is significant or non-trivial. The argument is based on the ineluctable place of normativity in any consistent account of language, as shown by Wittgenstein’s private-language argument. Furthermore, the chapter discusses the relation between semantics and pragmatics and elucidates the ontology of “the social”, showing that normativity implies a particular form of intersubjectivity: common knowledge. Finally, I spell out ramifications of the argument for the empirical study of language within diachronic linguistics, psycholinguistics and linguistic typology. I conclude by pointing to the possible sources of the anti-normative bias in much of theoretical linguistics.
1.
Introduction
The word ‘language’ can of course be used in many different senses, but it is reasonable to assume that one sense may be primary. Thus, when we speak of e.g. ‘English’, what kind of entity is it that we mean by this word? More specifically, is this entity social or non-social (in the sense of individual-psychological)? The ‘common-sense’ answer is that it is a social entity. It goes without saying (or so it seems) that e.g. a dictionary of English is about something that is common to or shared by all speakers of English, and whatever has these characteristics must be social by definition. But the ‘scientific’ answer (e.g. Chomsky 1965) is generally taken to be that linguistics is part of cognitive psychology, which entails that e.g. English is, at least primarily, an individual-psychological (and not a social) entity. I argue in this chapter that, on this particular issue, common-sense is right and ‘science’ is wrong, due to the irreducible place of normativity in any consistent effort to explain the nature of language, as shown by Wittgenstein’s private-language argument (Section 2). This has implications for the relation between semantics and pragmatics, which I touch upon in Section 3. Furthermore, I elucidate the
280 Esa Itkonen
ontology of “the social”, showing that normativity implies a special form of intersubjectivity – common knowledge – with implications for the theme of this volume. In Section 5, I spell out ramifications of the argument for empirical studies within diachronic linguistics, psycholinguistics and linguistic typology. Finally, I conclude by pointing to the source of the anti-normative bias in much of theoretical linguistics.
2.
The private-language argument
The primarily social nature of language can be shown in different ways. I have always preferred to rely on Wittgenstein’s so-called private-language argument, or PLA for short. PLA has spawned a huge number of publications, among which Saunders and Henze (1967) still stands out. Considered with all its ramifications, PLA is anything but simple. A ‘minimalist’ version of it will be presented in what follows (but see also Itkonen 1978: 91–113, 2003b: 120–125). PLA directs itself against the dominant tradition of Western philosophy, a tradition equally represented by Descartes, Hume, and Kant. According to this (‘Cartesian’) tradition, public things and qualities are reducible to subjective experiences, which constitute the ‘rock bottom’ of knowledge. Moreover, knowledge of other minds is supposed to be gained on the basis of the ‘argument from analogy’: When I perceive that bodies (constructed out of my sense-impressions and) resembling mine behave under similar circumstances in the same way as my body does, I may infer with a high degree of probability that these bodies are possessed by minds which think and feel in ways similar to mine. To start with, the incoherence of the Cartesian position may be demonstrated by a simple conceptual argument. The Cartesian ego, expressed as I or me, is supposed to be prior to other persons. But, just as there can be no ‘left’ without ‘right’, there can be no ‘I’ without ‘you’ and ‘we’: “If as a matter of logic you exclude other people’s having something, it loses its sense to say that you have it” (Wittgenstein 1958: §398). More elaborately, the Cartesian position may be reformulated in linguistic terms as follows. Since knowledge of the intersubjective or public world is supposed to be based on subjective or private experiences, the ordinary intersubjective or public language must – or could – have been preceded by a subjective or private language. Such a language is private in the twofold sense that it refers to subjective experiences and its rules are known to one person only. Wittgenstein (1958: §§ 243–277 and passim) argues that if a person constructs a private language and consciously tries to follow its (private) rules, he cannot know whether or not he has made a mistake. Because the notions of language and
Normativity in language and linguistics 281
rule presuppose the possibility of making a mistake, i.e. an aspect of the normativity of language, there can be no private language: “The test of whether a man’s actions are the application of a rule is ... whether it makes sense to distinguish between a right and a wrong way of doing things in connection with what he does” (Winch 1958: 58). Presented in outline, PLA goes as follows. Suppose that I am at this very moment going to (consciously) use some word X of my own private language. My use of X, i.e. what I mean (or intend to mean) by X, is based on my particular memory of how I have decided to use X, or how I have used X in the past. Maybe I wish to check this memory to make sure that I am not mistaken. But the only check I can rely on is the same memory; and of course it is no independent or genuine check – in fact it is no check (or basis for testing) at all. Therefore any ‘private’ rule-application that seems correct to me will be correct, which means that the notion of a private rule-application, and thus of a private language, ‘dissolves’ (cf. the Winch-quotation above). Written documents, for instance, do not get me out of this circle, because now the question arises whether I remember correctly the meanings of the written ‘private’ words. (Notice that on this reading of PLA the exact nature of the referent – thing or sense-impression? – is no longer of decisive importance.) Kenny (1973: 192–193) presents this argument exceptionally well. Genuine checks are provided only by other people’s memories, and more generally by their intuitions about the correct use of (public) language. Of course, there is no guarantee that these are always trustworthy. But at least they offer the possibility of genuine testing; and possible testing is certainly preferable to impossible testing (represented by exclusive recourse to my own memories or intuitions). This is nothing but the requirement of objectivity (in the sense of ‘intersubjectivity’), which is the cornerstone of scientific thinking. Some readers may still remain unconvinced. Therefore, to further clarify the issue let us deal with a concrete counter-argument which has kept reappearing in essentially the same form from the mid-50s onwards. Suppose that I formulate a ‘private’ rule according to which what is now called blue ought to be called mlue by me. I paint a blue patch on a piece of paper and write mlue under it, and on future occasions I will use this device to make sure that I am indeed following the rule correctly. Have I not proved that the notion of a private rule is a viable one? The answer is ‘No’, and here are some of the reasons why. Taken together, the blue patch and the word mlue constitute a ‘picture’. When composing this picture, I may have thought that its meaning is self-evident, i.e. that it can be interpreted in one way only. But this is wrong. One of Wittgenstein’s basic insights is that every picture or image can be interpreted in an infinite number of ways (and this is also true of mental images; cf. Blackburn 1984: 45–50; Heil 1992: 25–30). On the next occasion when I look at the picture, I may mistakenly think that the rule
282 Esa Itkonen
was meant to be not to say mlue when seeing something blue; or I may think that the blue patch was meant to remind be that I should check whether in any of the world’s languages ‘blue’ is called mlue; and so on. In other words, the human memory is notoriously fallible. It would be preposterous to assume that I am the only person in the world whose memory happens to be absolutely infallible. Now, what is true of memory is true of intellectual capacities more generally. Human beings may succumb to any kinds of aphasia, delusion or insanity. Today, with the well-documented spread of Alzheimer’s disease, this has become a near-certainty: everyone of us, unless released by a timely death, will become (more or less) insane. Let us keep this in mind, when we now return to the explication of PLA. Realizing the ever-present possibility of multiple interpretations, I may now wish to secure the unambiguous meaning of the picture by adding an explicit written instruction. If I use my own private language, the instruction will look something like this: “zmosh # glaark * mlue”. But nothing can guarantee that I will remember the meanings of these private words correctly, and if I attempt to avert this danger by further amplifying the instruction, infinite regress will ensue. If, on the other hand, I use English, the instruction will look like this: “I ought to say mlue whenever I see this colour!” But now I am cheating because my supposedly ‘private’ language is based on a public one. More importantly, however, this does not help me at all, because now any of the forms of human frailty alluded to above may attack me, either one by one or jointly. Perhaps I am colour-blind, but just do not know it; or perhaps I have become insane and think that, when looking at a blue patch, I am looking at my face in a mirror; or perhaps the moment when I lose the mastery of English has already arrived (but I just do not know it) and I either fail to understand the instruction or think that it says that I should go wash my teeth; and so on. The upshot is that my rule-following behavior needs checking by others. This is not fool-proof either. (Perhaps everyone is insane.) But at least it provides the possibility of genuine checking, which my private memory and understanding cannot provide. Wittgenstein assumes that in language, like in any other social institution, we are, or may become, conscious of the rules we either follow or break. Attempts to dispose of PLA are often based on redefining ‘private language’ as ‘unconscious psychological structure’, which makes it self-evidently true that everybody has his . Barresi and Moore’s (this volume) Intentional Relations Theory can be thought of as an empirical equivalent of PLA. In their requirement that, for a psychological concept to come about, “both the first-person, ‘inner’ aspect and the third-person, ‘outer’ aspect” are equally needed, they reproduce the insight that “[A]n ‘inner process’ stands in need of outward criteria” (Wittgenstein 1958: §580).
Normativity in language and linguistics 283
or her ‘private language’. But the redefinition is unjustified, in the first place. Just as well one might decide to call the internal structure of individual atoms their ‘private languages’. A useful up-to-date explication of what it is to follow a rule (of language) is provided by the doctrine of ‘response-dependency’ or ‘response-authorization’ (cf. Pettit 1996: 195–204; Itkonen 1997: 58–60, 2003b: 126–130, 165–168; Haukioja 2000). Nothing of what precedes entails in any way that language is exclusively social in character. Language has of course both a psychological and a biological aspect or, if you like, ‘substratum’. What the preceding discussion is meant to establish is that language is primarily a social entity.
3.
Semantics and pragmatics
Rules or norms do not just lie inertly there; rather, they only exist as rules or norms of acting. The social view of language, outlined above, suggests that the meaning of a linguistic expression is identical with its (conventionalized) use: “Look at the sentence as an instrument, and at its sense as its employment” (Wittgenstein 1958: §421). Here as elsewhere, the form of an instrument is a means to achieve different goals. Language use, i.e. speaking, is part of the same general meansends hierarchy as are all human actions and activities. Both linguistic meaning and its study are called semantics. More precisely, semantics is that part of (the study of) meaning which deals with meanings of words and sentences at the (‘general’) level of the conventional linguistic system, and not at the (concrete) level of single acts of speaking. However, the ‘actionist’ nature of language is present already in semantics. As a semantic entity, a sentence like I will come to see you at midnight encodes an act of asserting. The acts of requesting and asking are encoded in imperative and interrogative sentences. Language is not just action, but also interaction. In the case of requests and questions (codified as corresponding imperative and interrogative sentences) this is self-evident because they can only be conceptualized as being directed to someone different from the speaker himself. But the same is also true of assertions, codified as corresponding declarative sentences, as Sibawaihi, the founder of Arab linguistics, was perceptive enough to realize:
. The interdependence of normativity and consciousness has been explored in an illuminating way by Zlatev (2007). Zlatev’s (this volume) definition of ‘post-mimesis’ (language) as being essentially conventional-normative is also in consistence with the present argument.
284 Esa Itkonen
This is how we speak, even if the listener does not ask loud, because what you say follows the extent of the question he might pose if he were to ask you. (cf. Itkonen 1991: 155–156)
The same insight was achieved e.g. by Russell (1967 [1940]: 24): In adult life, all speech ... is, in intention, in the imperative mood. When it seems to be a mere statement, it should be prefaced by the words ‘know that’. We know many things and assert only some of them; those that we assert are those that we desire our hearers to know.
Pragmatics is that part of the study of meaning which deals with how the general meaning determined by the linguistic system becomes concrete or specific in single, either real or imaginary acts of speaking. This requires taking contextual information into account. In semantics, as noted above, the sentence I will come to see you at midnight has just the meaning of an assertion. (Which assertion? – this is evident from the lexical content and the grammatical structure.) In pragmatics, the same sentence (once uttered) becomes – depending on the context – either a promise (= Romeo is speaking to Juliette) or a threat (= a vampire is speaking to his future victim). This, in my view, is the relationship between semantics and pragmatics in a nutshell. It coincides with de Saussure’s (1962 [1916]) classic distinction langue vs. parole (see Section 5.1 below). It may seem natural to assume that pragmatics, concentrating on individual performance, pertains to psychology. In my view, however, pragmatics too is of social character. First, the performance is not individual but inter-individual, i.e. it necessarily takes place between speaker and hearer. Second, this inter-individual performance is publicly observable, and derives its identity from being (commonly) understood as a joint result of convention and context; just think of the Romeo vs. vampire contrast (cf. Leech 1983; Verschueren 1999). The truth of this statement remains unaffected by the fact that psychological explanations may of course be provided for any type of behavior (including linguistic interaction). In sum, semantics is the study of context-independent meaning whereas pragmatics is the study of context-dependent meaning. This ‘context-independent vs. context-dependent’ distinction was captured by Paul (1880 [1975]: Chapter 4) by means of his terminological dichotomy usuelle vs. okkasionelle Bedeutung (= ‘usual vs. occasional meaning’). Sometimes it has been claimed that the (inter)actionist nature of language becomes evident only in pragmatics. We have just seen that such a view is mistaken. At the level of semantics any sentence encodes a ‘frozen action’, and it is the task of pragmatics to ‘melt’ it (cf. Itkonen 1983: 152–164). It is
Normativity in language and linguistics 285
also clear that the ‘acts’ of referring and predicating belong already to semantics, and not just to (discourse) pragmatics. The relation between semantics and pragmatics is ‘dynamic’ in the sense that when context-dependent meanings recur, they may conventionalize and thus become part of the linguistic system. This kind of ‘ascent’ from speech (parole, okkasionelle Bedeutung) to language (langue, usuelle Bedeutung) is in general characteristic of language change (cf. Section 5.1 below). Having defended the social view of meaning (and of language in general), I may add a few words on why I find its opposite, i.e. the psychologist view of meaning, less convincing. To be sure, ‘psychologism’ may mean many different things, and in what follows I shall briefly deal only with one version of this doctrine. It is not uncommon to see meaning equated either with (unconscious) schema or with (conscious) mental image. First, let us assume that meanings are schemas. These are hypothetical entities: we do not know what they are, but only presume what they might be; and they may even be non-existent. (Implausible as this may sound, it is certainly possible.) In contrast, we do know the meanings of words like midnight and of sentences like I will come to see you at midnight; it makes no sense at all to assume that they are non-existent. Therefore meanings cannot be schemas. – It needs to be added immediately that we know the meanings of words and sentences only at the pre-theoretical level, i.e. we know them merely as ‘data’. We do not know how they should be theoretically analyzed (cf. Section 4.4). Second, let us assume that meanings are mental images. These are subjective or vary from one person to the next whereas meanings are intersubjective. (For instance, the sentence I will come to see you at midnight has only one meaning in the English language, not as many meanings as there are speakers of English.) Moreover, mental images may be non-existent. Even for a single speaker, there seems to be no mental image (or set of mental images) systematically and reliably connected e.g. with the word if. But if we accept the equation ‘meaning = use’, the meaning of if ceases to be a problem. It is enough to state (or list) its different uses: the transition from cause to effect (= If it is raining during the night, the streets will be wet in the morning) or from effect to cause (= If the streets are wet in the morning, it has been raining during the night), and so on. (But notice again that knowing the different uses of if does not entail knowing how they should be theoretically described).
. Thus, I am in broad agreement with the argument presented by Verhagen (this volume) that semantics (conventional meaning) is not only a matter of denotation, but also includes ‘argumentative’ aspects. I am less willing, however, to agree that these aspects constitute the core of lexical and grammatical meaning.
286 Esa Itkonen
In addition to these specific arguments against viewing meanings as schemas or mental images, we should heed the more general or philosophical admonition voiced by Wittgenstein (cf. above): Pictures or images (including ‘schemas’) are never enough. They must always come equipped with instructions about how they are meant to be interpreted. When the psychologistic conception of meaning amounts to equating meaning not with any specific mental image, but with subjective experience in general, it seems to be based on the following type of fallacy: In order to get clear about the meaning of the word think we watch ourselves while we think; what we observe will be what the word means! – But this concept is not used like that. (It would be as if without knowing how to play chess, I were to try and make out what the word mate meant by close observation of the last (Wittgenstein 1958: §316) move of some game of chess.).
Accepting the equation ‘meaning = use’ has a both clarifying and liberating effect. An enormous amount of time and energy has been wasted on trying to solve the problem of ‘how meaning exists’. But no one is – or need be – worried about how the use of a hammer or of a computer ‘exists’.
4.
The ontology of the social
4.1
Physical and social reality
The ontology of social entities is fundamentally different from the ontology of physical entities: There existed electrical storms and thunder long before there were human beings to form concepts of them or to establish that there was any connection between them. But it does not make sense to suppose that human beings might have been issuing commands and obeying them before they came to form the concept of (Winch 1958: 125) command and obedience.
The concept of ‘command’ is such as to be accessible to consciousness: commands exist only insofar as they are recognizable as, or known to be, what they are. This type of knowledge must be shared by all those who issue commands (and either obey or disobey them). In what follows it will be called common knowledge. This provides us with a preliminary definition of ‘social’: Social entities (unlike physical entities) exist if, and only if, they are commonly known to exist. For instance,
Normativity in language and linguistics 287
money ceases to exist, i.e. it is just pieces of metal and paper, as soon as people no longer know that it exists (qua money). This definition has some interesting consequences. Because a language like English exists if, and only if, it is commonly known to exist, it follows, among other things, that the correctness of correct sentences is a social fact, as elucidated by the following equivalence: (1) The sentence John is easy to please is a correct sentence (of English) iff the sentence John is easy to please is commonly known to be a correct sentence.
The formulation (1) is equivalent to the following formulation: (2) The sentence ‘John is easy to please is a correct sentence’ is true iff the sentence ‘John is easy to please is a correct sentence’ is commonly known to be true.
The sentence (2) instantiates the Tarskian ‘T-sentence’, which is of the following general form (cf. Itkonen 1983: 112): (3) X is true iff p
Here ‘p’ represents the truth condition of X. According to the received view, the truth value and the truth condition are two different things: we always know the truth condition of X, i.e. ‘p’, and we analyze it in a step-wise fashion, but this happens independently of whether we know X to be true or false. As far as physical facts are concerned, it is indeed the case that while we do know the truth condition of X, we do not know the truth value of X. Now, the example (2) refutes the received view as applied to social facts, because it shows that, in this crucial domain, it is impossible to know the truth condition of X without knowing the truth value of X (for discussion, cf. Itkonen 1983: 129–135). Thus, at the level of social facts, the T-sentence has the following form: (4) X is true iff X is (commonly) known to be true.
A declarative sentence X is used to make a statement (or assertion). In logical semantics, the truth-condition of X is equated with the meaning of X. This view is too restrictive, but it is certainly the case that knowing the truth condition of X is part of knowing the meaning of X. We have just seen that, in connection with social facts, knowing the truth condition (and, more generally, the meaning) of X entails knowing the truth value of X. But why should we think, in the first place, that we know the meanings of the words and sentences that we utter? Wittgenstein (1969: §370) suggests the answer: “I should stand before an abyss if I wanted so much as to try doubting their meanings...”
288 Esa Itkonen
4.2 The nature of common knowledge What does it mean to say that a social entity like the English language is an object of common knowledge? One way to answer this question, due to Lewis (1969), is to say that X is an object of common knowledge if, and only if, the three conditions given in (5) are true of X and of (practically) any two members of a community (where both ‘A’ and ‘B’ stand for each of the two): (5) A knows-1 X A knows-2 that B knows-1 X A knows-3 that B knows-2 that A knows-1 X
As abstruse as such a formulation may seem at first, it is quite easy to show that three-level knowledge of this kind necessarily occurs in all institutional encounters. Suppose I want to cash a check in a bank. The only reason why, when approaching the counter, I do not make soothing gestures or shout “I know what to do, you don’t have to tell me!”, is that I possess the relevant three-level knowledge: Not only do I know-1 what to do; and not only do I know-2 that the teller knows1 what to do; but I also know-3 that the teller knows-2 that I know-1 what to do. This type of ‘third-order mentality’ is also discussed and exemplified by Zlatev (this volume). From the logical point of view, there is no way to stop the infinite regress of different knowledge-levels (= ‘I know that he knows that I know that he knows...’). From the practical point of view, however, this is not a problem. People do not generally go beyond three- or four-level knowledge. Some people are able to do this; but nobody masters e.g. ten-level knowledge. The explication of ‘social’ in terms of many-level knowledge has sometimes been regarded as entailing some sort of philosophical idealism. Our example of check-cashing behavior should dispel this misunderstanding. The relevant common knowledge is ‘embodied’ not just in people’s behavior, but also in such physical artifacts as the bank building, its furniture, the clerks’ implements, and so on. Sinha (1988) rightly emphasizes the importance of taking into account the material grounding of institutions (including language). Our example is apt to illuminate another often-misunderstood aspect of common knowledge. My attitude vis-à-vis the bank teller is not invalidated if it later turns out that at the moment of our mutual encounter he happened, for instance, to be either unconscious or suffering from an attack of insanity, which means that he did not, as a matter of psychological fact, possess the requisite three-level knowledge about me. A’s three-level knowledge about B is not about what B knows in fact, but what A is entitled to expect B to know: Given the surroundings, I was entitled to expect that the bank teller whom I was approaching knew his
Normativity in language and linguistics 289
business, i.e. had the requisite three-level knowledge about me. Hence, common knowledge turns out to contain a crucial normative element. It is a ‘rational reconstruction’ of sociality, not a psychological description of what actually goes on in people’s heads in each and every case: For in most social situations, if not in all, there is an element of rationality. ...I refer to the possibility of adopting, in the social sciences, what may be called the method of logical or rational construction, or perhaps the ‘zero method’. ...The ‘zero method’ of constructing rational models is not a psychological but rather a logical method. (Popper 1957: 140–141, 158; for discussion, see Itkonen 2003b: 131–135)
Common knowledge is usually conceived of as being generalized, and conventionalized (cf. Section 5.1), out of single instances of (non-normative) third-order mentality, as described by Zlatev (this volume). But, in many accounts, the notion is not thought to contain an ineluctable normative element. The original version of the concept given by Lewis (1969) can be criticized for having ignored precisely this fact (cf. Itkonen 1978: 182–186). In sum, the social world (explicated by means of the notion of common knowledge) is permeated by normativity considerations through and through: It is perhaps the basic insight of Winch (1958) that we need criteria, whose use is governed by rules [= norms], to identify entities as same or different, and that as regards social entities, such criteria are internal to them. (Itkonen 1978: 185)
Clark (1996) too considers a language as an object of common knowledge, and he claims (pp. 75–77), more precisely, that a language qua commonly known is a set of conventions. This agrees perfectly with my view (even if I prefer the term ‘norm’). The conventions include those for ‘lexical entries’ and those for ‘grammatical rules’, i.e. norms for pairing (morphemic and lexical) forms with meanings and those for combining meaningful forms into phrases and sentences, as I would say. Common knowledge (like knowledge in general) must have a basis. In the simplest case, the common knowledge of a fact is based on its intersubjectively observable existence. For instance, the common knowledge that it is raining right now is based on the fact that (as everybody can see) it is raining right now. But remember that a physical fact, unlike a social fact, can exist, and typically does exist, even if it is not commonly known to exist. What is the basis for linguistic common knowledge, e.g. for (2) in Section 4.1? It cannot be pinpointed as easily as it can in the case of commonly known physical facts. It is not a particular happening, like someone uttering John is easy to please and no-one protesting its incorrectness. (To be sure, linguistic common
290 Esa Itkonen
knowledge must not – in general – conflict with such particular happenings.) The basis for common knowledge about the (in)correctness of sentences is ‘diffuse’, in the sense that it is constituted just by general facts about coming to master a language and by the concomitant common knowledge about those facts. In this respect linguistic common knowledge is just one instantiation of institutional common knowledge in general. The most important difference vis-à-vis common knowledge about physical facts resides in that the basis for linguistic common knowledge, though undeniably existent, cannot be used to strengthen or justify that which it is a basis for: And here the strange thing is that when I am quite certain of how the words are used, have no doubt about it, I can still give no grounds for my way of going on. If I tried I could give a thousand, but none as certain as the very thing they were (Wittgenstein 1969: §§ 306–307) supposed to be ground for.
4.3 A solution to the controversy between individualism and collectivism The definition of social ontology given in Section 4.2 dissolves rather than solves a long-standing controversy within the philosophy of the social sciences. One side has argued that there is an ontological level of social institutions distinct from the level of individual persons. The other side has argued that there is nothing but individual persons (cf. O’Neill 1973). Now we can see that they are both right. Indeed, there are nothing but individual persons; but what we have is not just an aggregate of individual persons endowed with arbitrary mental states and distributed in a random order; rather, we have individual persons endowed with quite specific mental attributes (namely many-level states of knowledge) placed in a quite definite structure or pattern (namely that characteristic of common knowledge). It is this structure that constitutes the ontological level of social phenomena. As an analogy, consider the distinction between a single line and a net. On the one hand, it can be argued that a net consists of nothing but lines, which means that the line is ontologically primary vis-à-vis the net. On the other hand, the net is not just a random heap of lines, but a quite specific structure or pattern of lines. When the lines constitute a net-like structure, then – and only then – there is this all-important difference that it is possible to catch fish with a net, but not with a line. This difference is important enough to be called ‘ontological’; and it shows how increasing complexity makes a new ontological level ‘emerge’ out of an ontologically simpler level. It could also be argued that in (dis)solving the controversy
Normativity in language and linguistics 291
between individualism and collectivism, we eo ipso show that the contrast between psychological and social, which was taken for granted in much of the previous discussion, is more apparent than real. In so doing, we have been forced to revise the meanings of these two words, i.e. ‘psychological’ and ‘social’, to some extent. The preceding discussion suggests that the metaphor of ‘social network’ should be taken seriously. The same analogy may also illustrate the distinction between (subjective) intuition and (intersubjective) norm, which may at first seem a little puzzling. Institutions are about norms. Norms are learned on the basis of observation, but once they are known, they can no longer be just a matter of observation because they are made use of to judge whether an observed (or imagined) action is correct or not: The correctness of a performance is not among its perceptual characteristics; it cannot be, since it is a relation between the performance and an adopted rule [= norm] – a relation which is more fully expressed by the statement that the (Körner 1960: 117) performance conforms to the adopted rule.
The subjective (non-observational) knowledge of norms is called intuition. It is a general truth, labeled ‘Hume’s guillotine’, that knowledge of norms (i.e. of what ought to be done) cannot be reduced to observation (of what is done). In the definition of common knowledge, it is the first level, i.e. ‘A knows-1 X’, which corresponds to that standard type of (subjective) linguistic intuition which is used in gathering the data that constitutes the basis for grammar-writing: ‘A knows that y is a correct sentence’. The second and third levels are also of ‘intuitional’ character; but more importantly, they bring out the interactional nature of language or of social facts in general. Moreover, there is also theoretical understanding about the three-level knowledge as a whole: Although I am just one knot in the social network, i.e. a single person qua member of an institution, whose knowledge and action constitute just a small contribution to its existence, it is nevertheless possible for me to reflect on the institution as a whole. The ‘social world’, understood as an object of common knowledge, is coextensional with Popper’s (1972) ‘world-3’, though without the latter’s Platonist overtones. The ineluctably interactional nature of all social facts was beautifully captured by Marx and Engels (1973 [1846]: 37): Es zeigt sich hier, dass die Individuen allerdings einander machen, physisch und geistig, aber nicht sich machen. (= So we see that in a physical and spiritual sense individuals make each other, but do not make themselves.)
292 Esa Itkonen
4.4 Normativity in language The fundamental distinction between linguistics and any genuine natural science consists in the fact that the subject matter of the former is inherently normative whereas the subject matter of the latter is inherently non-normative. Now the notion of normativity needs to be explicated more narrowly. First of all, we have to establish the distinction between a rule-sentence such as (6), which describes a rule (or norm), and an empirical hypothesis such as (7), which describes an (assumed) regularity. (6) In English, the definite article (i.e. the) precedes the noun (e.g. man) (7) All ravens are black.
The difference between (6) and (7) consists in the fact that (7) can be (and in fact has been) falsified by spatiotemporal occurrences, namely non-black ravens, whereas (6) is not, and cannot be, falsified. The utterance of a sentence (8) does not falsify (6). Why? – because this sentence is incorrect. Nor does the utterance of a sentence like (9) falsify (6). Why? – because this sentence is correct. Thus, (6) is unfalsifiable (on the basis of spatiotemporal occurrences). (8) *Man the came in. (9) The man came in.
The difference between rule-sentences and empirical hypotheses has been occasionally recognized in the philosophy of the social sciences, e.g. by Ryan (1970), who, to be sure, fails to distinguish between rules (= object of description) and rule-sentences (= description): A causal generalization has only one task to fulfil, namely telling us what will and will not happen under particular conditions, irregularities are thus falsifying counter-examples to the causal law. But rules [i.e. rule-sentences] are not falsifiable in any simple way – except of course that it may be false to say that there is a rule – and breaches of a rule are errors on the part of those whose behavior is (p. 141; emphasis added) governed by it.
In general, however, the distinction at issue has remained in some sort of methodological limbo. On the one hand, one may be willing to admit that perhaps – just perhaps – there may indeed exist something that resembles this distinction. On the other hand, one refuses to draw any methodological consequences from the (possible) existence of this distinction. What is at issue here is the normativity of language: sentences are normative (i.e. correct or incorrect) entities whereas birds are not (or, at least, not in the same
Normativity in language and linguistics 293
sense as sentences are). The normativity of language is ignored in traditional philosophy of language, as shown by the fact that the distinction between sentences and (e.g.) birds is ignored. At the face of it, this is a curious fact, because philosophy of language is brimming with talk about rules of language. In practice, however, no examples of these rules are ever given. Because the discussion is carried out at such a high level of generality, the distinction between sentences and (e.g.) birds is destined to remain hidden. – Among philosophers of language, to be sure, there are some laudable exceptions, for instance Cavell (1971a [1958], 1971b [1962]). In reality, the meanings of words are all based on corresponding rules: there are rules which determine that three designates a number, i.e. 3, and not a plant, whereas tree designates a plant and not a number; and so on for all words of all languages. These rules attach meanings to forms. And then there are rules that determine how meaningful forms have to be combined. One rule of this kind is described by our rule-sentence (6). Other such rules deal with facts of government (= ‘rection’) and agreement. It is correct to say I confided in him and incorrect to say I confided from him; it is correct to say I am upset and incorrect to say You am upset; and so on. As noted before, Clark (1996) assumes the existence of two corresponding types of rules. For any rule it is possible to construct a corresponding rule-sentence. The status of rules may be clarified by the following remarks: The problem for the grammarian is to construct a description ... for the enormous mass of unquestionable data concerning the linguistic intuition of the native (Chomsky 1965: 20; emphasis added) speaker (often himself). Few users of language know much in any systematic way about their language, though obviously they can discover any number of odd bits of correct information (Hockett 1968: 63; emphasis added) simply through self-observation.
Because of their trivial or pre-theoretical character, rules and corresponding rulesentences possess no linguistic (or scientific) interest whatever. However, their philosophical (or metascientific) significance is enormous. They show that, contrary to what is the case in the natural sciences, the basic data of grammatical description are not particular entities (= single spatio-temporal occurrences), but general entities (= norms) described, in principle, by general and unfalsifiable sentences. This insight constitutes the core of ‘response-dependency’ (mentioned in Section 1.1). The standard reaction to what precedes is to say that if the rules/norms of language are known in an unfalsifiable way, or with certainty, there is nothing left for the grammarian or linguist to do. But consider the case of Panini (c. 400 BC), “the greatest grammarian of all” (Dixon 2002: 145). At the pre-theoretical level, his contemporaries knew Sanskrit just as well as he did. But only he was able
294 Esa Itkonen
to construct the grammar that was to bear his name. Thus, once the data are in, everything still remains to be done. Similarly, Chomsky and Hockett clearly imply that there is a job for them to do, whatever odd bits of correct and indubitable information the average speaker may possess about his language. The same point can be made by briefly returning to the notion of truth condition. As Wittgenstein so eloquently put it, we stand before an abyss if we start to doubt whether or not we know the meanings of the words and sentences that we use. But of course we know them only at the pre-theoretical level. We know that John is easy to please is a correct English sentence (unlike e.g. *John is easy from please) and that it means the opposite of John is difficult to please, but we do not know the best theoretical description of this (or any other) sentence. Any theoretical description is falsifiable by definition. But falsification in grammatical description is not what it is in the natural sciences. There are many other ‘standard objections’ against the distinction between rule-sentence (= A) and empirical hypothesis (= B), for instance: – “If English were different, A would be falsified.” – “In English (as it is now), A is verified and any other formulation of the same facts is falsified.” – “The definite article does not (always) precede the noun (just think of Ivan The Terrible).” – “Maybe A is not falsifiable by simple observation, but neither are scientific theories.” – “The terms ‘definite article’ and ‘noun’ are theoretical, not pre-theoretical.” – “A and B are formulated in dissimilar ways.” – “Not all rules of English are of the same type as the one described by A.” – “The existence of the rule described by A is a contingent and not a necessary fact.” – “A is not an analytical sentence.” – “English has also statistical and experimental aspects not captured by A-type sentences.” Such and similar objections have been brought together and answered in Itkonen (2003b: Chaps 3, 6, 7); see also Section 5 below. It should also be pointed out that the mere existence of the normativity of language is enough to refute all varieties of physicalism (or ‘naturalism’), i.e. of the view that physical data is all there is. If you argue for this view, you must do so in the language of physics (and/or philosophy); and the language you use is not physical (or ‘naturalistic’), but normative.
Normativity in language and linguistics 295
4.5 Correctness vs. rationality In typical linguistic behavior, rational actions are performed by uttering correct sentences. It is quite possible, however, to perform irrational actions by uttering correct sentences, and to perform rational actions by uttering incorrect sentences, which shows that the dimensions of correctness and rationality are independent from each other. Since the use of language exemplifies the general means–ends hierarchy, as noted in Section 3, it is amenable to so-called rational explanation, which is a general explanatory model for human (and even animal) behavior: To explain an action as an action is to show that it is rational. This involves showing that on the basis of the goals and beliefs of the person concerned the action was the means he believed to be the most likely to achieve the goal. (Newton-Smith 1981: 241)
Even irrational behavior can be explained, if at all, only by means of rational explanation, namely by exposing the reason why it was performed. This involves coming to understand how behavior that is irrational in fact came to seem rational to the agent. The ‘transition’ from goals to means followed by the carrying-out of the means, as codified in rational explanation, can be seen as the causal force that brings about linguistic behavior investigated in such distinct linguistic subdisciplines as psycholinguistics, sociolinguistics, and diachronic linguistics (cf. Itkonen 1983). Using language must consist of the continuous making of linguistic choices, consciously or unconsciously, for language-internal (i.e. structural) and/or languageexternal reasons. (Verschueren 1999: 55–56; emphasis added)
This innocuous-looking statement, once its implications are spelled out, justifies the use of rational explanation.
5.
Normativity and beyond: Language change, language psychology and typology
5.1
Language change: The need for statistics
Language change entails that old norms (or rules) are replaced by new ones. Comparative Indo-European linguistics started with the idea of grammaticalization. Thus, Franz Bopp claimed in 1816 that, for instance, the endings of Sanskrit
296 Esa Itkonen
verbs had originally been full personal pronouns (cf. Arens 1969: 177). To give another example, let us consider the Modern French constructions venir de + INF and aller INF. Originally these had the concrete local meanings ‘come from INF’ and ‘go INF’. Then in some contexts these constructions were reanalyzed as having also the temporal meanings ‘recent past’ and ‘near future’. First, these meanings were more or less accidental or pragmatic; but later they became conventionalized or semantic. (As noted in Section 3, this ‘pragmatic vs. semantic’ distinction is just a reformulation of Paul’s (1975 [1880]) distinction between okkasionelle vs. usuelle Bedeutung.) That new conventions or norms had emerged, was evident as soon as the temporal meanings were extended to such contexts where the old concrete and non-temporal meanings are impossible, as shown in (10) and (11). (10) Il vient de mourir (‘he has just died’ < ‘he comes from dying’) (11) Il va s’éveiller (‘he will wake up’ < ‘he goes wake up’).
The mechanism of grammaticalization (= reanalysis-cum-extension) is discussed e.g. by Itkonen (2002). It is a curious fact that while in theoretical linguistics much attention has been devoted to the notion of conventionalization, the logically primary notion of convention (or normativity) has remained practically unknown. The (typical) linguist takes the existence of language for granted. He is not competent by training to answer the phylogenetic question concerning the origin of language. Nor is it his business to reconstruct the process through which norms may have emerged out of an attempt to coordinate originally non-normative actions (cf. Lewis 1969). This does not mean, however, that these are not worthwhile questions to be asked in an interdisciplinary framework. Traditionally, grammarians have been relying on self-invented example sentences, which means that traditional synchronic linguistics has been based on intuitional data (for extensive documentation, see Itkonen 1991). The use of intuitional data unites such otherwise dissimilar approaches as generativism (= Chomsky 1965; Jackendoff 1994), cognitive linguistics (= Lakoff 1987; Langacker 1987), and construction grammar (= Goldberg 1995; Croft 2001). The reliance on intuitional data is fully justified in so-called clear cases (exclusively focused upon by the six linguists just mentioned), but elsewhere one has to resort to observation of actually occurring utterances, which entails the use of statistics. Norms of language may be more or less binding, i.e. they may determine the correctness of expressions or sentences either in a discrete (‘either-or’) way or in a non-discrete (‘more-or-less’) way. In most languages, for instance, the norms of word order are non-discrete while the norms of affixal morphology are discrete. The norms of word meaning are open, in the sense that there is a discrete core
Normativity in language and linguistics 297
surrounded by a non-discrete periphery: “It is only in the normal cases that the use of a word is clearly prescribed...” (Wittgenstein 1958: §142). Even when the norms are discrete, the (normative) behavior they subsume is non-discrete, which is another way of saying that they may be broken (either deliberately or inadvertently). A much discussed example is the t/d deletion in today’s English (cf. Hudson 1997). The (discrete) norms determine the phonological form of the noun mist (‘fog’), the past tense left of the verb ‘to leave’, and the past tense missed of the verb ‘to miss’. But in actual practice, the word-final t/d may or may not be present, and in these three cases it is typically retained in the following proportion: 50%–65%–80%. There is the experience of this statistical pattern (based on observation), in addition to the (intuitive) knowledge of the above-mentioned discrete rules. This duality can be captured by assuming that what a discrete norm determines is a prototype: while a prototype is defined by its ‘typical’ features, any of these may be overridden in exceptional cases. The important thing is that this duality must not be explained away. In particular, it would be wrong to try to reduce the discrete norm to the corresponding non-discrete and statistical behavior. This follows from the fact, mentioned in Section 4.3: that ‘ought’ cannot be reduced to ‘is’. When the percentage of the norm-following behavior drops below 50%, at the latest, we are witnessing a diachronic process which turns a discrete norm into a non-discrete one and, in general, ultimately leads to its disappearance. This amounts to a change of the prototype, which in turn equals a language change. This can be a lengthy process. For instance, in one hundred years the correct pronunciation of today’s mist may actually be [mis]. To give a less speculative example, it took some 300 years (i.e. between 1450 and 1750) for the construction exemplified by (12) to be replaced by the construction exemplified by (13) as part of the emergence of the auxiliary system of Modern English. (12) Saw he the dragon? (13) Did he see the dragon?
First, the latter structure was totally incorrect, and in the end it came to be totally correct. In between, there was a gradual shift that can be described only in statistical terms (cf. Hudson 1997). In other words, language change is a prime example of less-than-clear cases. It is easy to see that Saussure’s terminological distinction between langue and parole captures the following dichotomy: on the one hand, language as a system of norms accessible to conscious intuition; on the other, actual spatio-temporally specifiable linguistic behavior that is accessible to observation.
298 Esa Itkonen
5.2 Language and the psychology of language: The need for experimentation “La langue est une institution sociale” (Saussure 1962 [1916]: 33). It is a general fact that an institution or, more generally, any rule-system S can be described or formalized in many different ways. This means that different people may view S from different perspectives and with different descriptive goals in mind. Thus, there is no a priori reason to assume that the description of S must aim at capturing the way that S has been internalized by those who have learned it. For instance, it is possible to describe S so as to achieve either a maximal degree of operational efficiency or a maximal degree of logical simplicity. The types of descriptions of S that result from adopting either one of these two perspectives will differ from each other, just as they will both differ from the type of description of S that sets the psychology of the users of S as its goal: But what would that grand success [of sequence-extrapolating algorithms] teach us about human perception, pattern recognition, theory formation, theory revision, and esthetics? Nothing – nothing at all. This ... brings out the vastness of the gulf that can separate different research projects that on the surface seem to belong to the same field. ... Today’s wonderfully powerful chess programs, for instance, have not taught us anything about general intelligence – not even about the intelligence of a human chessplayer! Well, I take it back. Computer programs have taught us something about how human chessplayers play – namely, how they do not play. And much the same can be said for the vast majority of artificial-intelligence programs. (Hofstadter 1995: 52–53)
This is a very clear formulation of the fact that there is a difference between a description of S, or D1, and a description of the psychology of S (= P-S), or D2. Thus, D1 and D2 refer to, and describe, two distinct entities, namely S and P-S. The understanding of this distinction has been made needlessly difficult by ambiguous terminology. On the one hand, P-S is often called ‘knowledge of S’. On the other hand, S is – by definition – commonly known. This creates the wrong impression that there is no difference between S and P-S nor, consequently, between D1 and D2. For the sake of clarification, consider the following analogy. If I describe the moon as I see it with the aid of a telescope, it is still the moon that I describe, and not my vision (enhanced by the telescope). If I genuinely wish to concentrate on my vision, and not on the moon, then I have moved from astronomy to the psychology of vision. Exactly the same remarks apply to the distinction between D1
Normativity in language and linguistics 299
and D2, as Hofstadter so well demonstrates. It is only D2 which aims at psychological reality whereas D1 has other desiderata (e.g. efficiency or simplicity). Once you have grasped this distinction, you realize that it applies practically everywhere. For instance, there is a difference between geometry and the perception of geometrical figures and shapes (cf. Itkonen 1983: 1–3). In just the same way, there is a difference between formal logic and psychology of logic (cf. Itkonen 2003a: Chap. XV). In linguistics, the matters may at first seem less clear. Therefore it is good immediately to point out that there are quite uncontroversial cases of non-psychological grammatical descriptions. For instance, it is a fact, pointed out by Paul Kiparsky (p.c.), that Panini’s grammar does not strive after psychological reality. Similarly, in arguing against the view that linguistics is psychology, Katz (1981) operates with the concept of ‘optimal grammar’: [There should be no] constraints that impose a ceiling on the abstractness of grammars by tying them down to one or another particular [i.e. physical or psychological] reality (p. 52). A grammar G is an optimal grammar for the language L, if ...G ...implies every true evidence statement about L ...and there is no grammar simpler than G... (p. 67; emphasis added). [O]n the most natural definition, an ‘optimal grammar’ is a system of rules that predicts each grammatical property and relation of every sentence in the language and for which there is no simpler (or otherwise methodologically better) such predictively successful theory. (Katz 1985: 201; original emphasis deleted)
However, Katz’s references to ‘optimal grammar’ remain rather unconvincing, because he is unable to exemplify this concept. Therefore it is important to emphasize that, within the ‘world history’ of linguistics, this concept has already been exemplified rather well, namely by Panini’s grammar: [Panini’s grammar] is the most comprehensive generative grammar written so far (Kiparsky 1979: 18). Modern linguistics acknowledges [Panini’s grammar] as the most complete generative grammar of any language yet written, and continues to (Kiparsky 1993: 2912) adopt technical ideas from it.
The same laudatory view of Panini’s grammar has been both documented and argued for in Itkonen (1991: Chap. 2, esp. pp. 68–70). In the present context it is important to understand that, in addition to being the best generative grammar, Panini’s grammar is – by Kiparsky’s own admission (cf. above) – also a non-psychological grammar, which means that it is indeed a serious candidate for being the Katz-type ‘optimal grammar’.
300 Esa Itkonen
The notion of non-psychological or ‘autonomous’ linguistics has been analyzed in Itkonen (1978) and Kac (1992). Katz (1981, 1985) gives it a Platonist interpretation, but there is really no reason to do so: The properties Katz assigns to abstract objects appear all to be possessed by the kind of conventions of mutual knowledge that Esa Itkonen argues are constitutive of linguistic rules (Itkonen 1978; not cited in Katz 1981).(Pateman 1987: 52)
While language is identical with a system of (social) norms, psychology of language is identical with the structures and processes involved in speech understanding and production as well as in the mental storage of linguistic units. In Itkonen (1983) this distinction was conceptualized as holding between (social) norms and (individual-psychological) internalizations-of-norms. It is in connection with the latter that the need for experimentation arises. This can be illustrated by means of what is probably the most famous example in recent decades. The ‘standard theory’ of generative linguistics, as expounded in Chomsky (1965), made use of a descriptive apparatus consisting of transformations that convert deep structures into surface structures. This is one possible method of presenting intuition-based data in a systematic way; indeed, it was already used by Apollonius Dyscolus, who wrote the oldest extant syntactic treatise of the Western tradition (cf. Itkonen 1991: 206–211). But is it also psychologically adequate? And how can this question be answered, in the first place? Experimentation provides the answer. If transformations are psychologically real processes, they must take time to be performed. Hence, the hypothesis is that there are longer reaction times connected with producing and/or understanding sentences that involve more (rather than less) transformations. Experimental data give this verdict: “[T]he hypothesis that the operations that the subjects performed were grammatical transformations is actually disconfirmed by the data” (Fodor et al. 1974: 241). That it is perfectly legitimate to use transformations in grammatical description (= ‘autonomous linguistics’) in spite of their psychological non-reality, shows that, in Hofstadter’s (1995) words, there is a “gulf ” that separates intuition-based autonomous linguistics from experimental psycholinguistics. More precisely, the data of the former is of pre-experimental character; it is a precondition for the data of the latter: “One cannot make experiments if there are not some things that one does not doubt” (Wittgenstein 1969: §337). The existence of pre-experimental linguistic knowledge has occasionally been acknowledged: “It is pointless to run an experiment which shows that if something is a pencil, appropriately motivated English speakers will call it ‘pencil’. Anyone who knows English knows that already” (Fodor et al. 1974: 399–400).
Normativity in language and linguistics 301
This type of experiment would be “a slightly absurd exercise, with the results a foregone conclusion” (Wason and Johnson-Laird 1972: 78). However, the larger implications have remained unexplored and poorly understood. The ambiguity of non-psychological vs. psychological study of language is well illustrated by the notion of analogy. On the one hand, analogy may be just a convenient descriptive device for presenting the data. On the other, analogy may be meant to capture the actual structure-cum-process that brings linguistic behavior about (cf. Itkonen 2005a).
5.3 The nature of typological linguistics Up to now we have come across three distinct types of linguistic data, namely intuitional, observational, and (observational-)experimental. The two latter types deal with frequencies of spatio-temporal occurrences and thus require a statistical mode of description. This division of labor between different linguistic subdisciplines was already set forth in Itkonen (1977) and (1980). What is the status of typological linguistics from the present perspective? An in-depth analysis of the reference grammars of ten more or less ‘exotic’ languages reveals a general lack of any statistical means of description (cf. Itkonen 2005b). This shows that, once again, we are dealing with intuitional data. In many cases, however, what we have is not the intuition of a (field) linguist, who, while writing his grammar, may still be in the process of learning the language to be described, but the intuition of his informant(s). In other words, we are dealing with elicitation. Haiman (1980: xi) gives an eloquent account of this method: I will always remember Kamani Kutane for his thought experiments: given a minimally contrasting pair of sentences, he would construct elaborate background stories which would be appropriate for only one of these sentences. Eventually I would understand one of these, and we could move on. It was by means of such continued thought experiments that he was able to make clear to me the meaning of that most mysterious of all Hua forms, the gerund -gasi.
As shown by this quotation, and as argued in Itkonen (2004), the study of ‘exotic’ languages is based on empathy as a form of intersubjectivity, or – in Collingwood’s (1946: 218) words – our capacity of “rethinking the same thought which created the situation we are investigating, and thus coming to understand this situation”. But once we have become aware of empathy in this context, we realize that we have been using it all the time. For instance, we can explain the grammaticalization of the constructions venir de INF and aller INF in the way we do (cf. Section 3.1), only because we understand the processes of reanalysis and extension
302 Esa Itkonen
that are involved here; and we understand them, because we can ‘re-enact’ them, i.e. we realize that we could have done the same thing. On reflection, this turns out to be an application of the model of rational explanation (cf. Section 2.5).
6.
Conclusion: The roots of the anti-normative bias in theoretical linguistics
Considering everything that has been said so far, one naturally wonders: Why has there been such a pronounced inclination to ignore the ineluctably normative character of language? There are many reasons, of which I mention here only two. First, there is sheer intellectual laziness: [It is wrong] to consider the salient features of an object as representative of its totality. In this way the evident concreteness of the sound of words leads one to ignore the extent to which use, however intangible, is necessary to word-hood. (Friedman 1975: 94, emphasis added; discussed in Itkonen 1978: 182–183)
Notice that it is the same, or very similar, fallacy that underlies the entire Cartesian tradition mentioned in Section 2. This is the Cartesian argument in outline: “I see, and hence I know, that this thing in front of me is a burning candle; but I do not see anyone else in the room; thus when I know what I know about the thing in front of me, I am alone; therefore my knowledge is not social but subjective; and what is true of my knowledge here and now is true of every type of knowledge.” Once this argument has been spelled out, one cannot help marvelling how simple, and simple-minded, it really is. Second, there is the temptation to replace the (normative) ‘correct vs. incorrect’ distinction by the (non-normative) ‘possible vs. impossible’ distinction. Thus, Jackendoff (1994: 49–50) claims that, unlike a sentence like Harry thinks Beth is a genius, a sentence like Amy nine ate peanuts is “not a possible sentence of English”. However, it is not only the case that this is a possible sentence of English. We see with our own eyes that it is also an actual sentence of English, namely incorrect English. It must be actual because (an exemplification of) it occurs in space and time (cf. Dretske 1974: 24–25; Itkonen 2003b: 142–144). But why should it be tempting, in the first place, to replace normative by nonnormative? – because of the prestige enjoyed by the natural sciences. The data of physics is inherently non-normative. From this, it has been wrongly inferred that the data of linguistics too must be non-normative, come what may. . The discovery of mirror neurons seems to have revitalized the notion of empathy, as shown in detail by Barresi and Moore (this volume).
Normativity in language and linguistics 303
Is there, then, no normativity in the natural sciences? Of course there is. Just think of protophysics which investigates the set of norms for measuring space, time, and mass (cf. Böhme 1976). But protophysics is not physics: “It is one thing to describe methods of measurement, and other to obtain and state results of measurement” (Wittgenstein 1958: §242). As argued in Itkonen (1978: 42–48) and elsewhere, protophysics is in a certain sense a methodological equivalent of autonomous linguistics. Still, this is an imperfect analogy because what protophysics deals with are norms of researchers, not of research objects. In sum, I have argued in this chapter that normativity is indispensable for the existence of language, and that it has been – often without self-awareness – pivotal for linguistics from its very dawn. To remain blind to this obvious fact, a strong bias has indeed been needed.
Acknowledgments I wish to thank Jordan Zlatev for comments and for his help in editing an earlier version of this chapter.
References Arens, H. 1969. Sprachwissenschaft: Der Gang ihrer Entwicklung von der Antike bis zur Gegenwart. Band I. Frankfurt a/M: Athenäum. Barresi, J. and Moore, C. (this volume) “The neuroscience of social understanding.” Blackburn, S. 1984. Spreading the Word. Oxford: Oxford University Press. Böhme, G. (ed.) 1976. Protophysik. Frankfurt a/M: Suhrkamp. Cavell, S. 1971a [1958]. “Must we mean what we say?” In Philosophy and Linguistics. C. Lyas (ed.), 131–165. London: Macmillan. Cavell, S. 1971b [1962]. “The availability of Wittgenstein’s later philosophy.” In Philosophy and Linguistics. C. Lyas (ed.) 1971, 166–189. London: Macmillan. Chomsky, N. 1965. Aspects of a Theory of Syntax. Cambridge, MA: The MIT Press. Clark, H. 1996. Using Language. Cambridge: Cambridge University Press. Collingwood, R.G. 1946. The Idea of History. Oxford: Clarendon Press. Croft, W. 2001. Radical Construction Grammar. Oxford: Oxford University Press. Dretske, F. 1974. “Explanation in Linguistics.” In Explaining Linguistic Phenomena. D. Cohen (ed.), 21–41.Washington, D.C.: Hemisphere Publishing Company. Dixon, R.M.W. 2002. Australian Languages. Cambridge: Cambridge University Press. Fodor, J., Bever, T. and Garrett, M. 1974. Psychology of Language. Cambridge, MA: The MIT Press. Friedman, H.R. 1975. “The ontic status of linguistic entities.” Foundations of Language 13–1: 73–94. Goldberg, A. 1995. Constructions. Chicago: University of Chicago Press.
304 Esa Itkonen
Haiman, J. 1980. Hua: A Papuan Language of the Eastern Highlands of New Guinea. Amsterdam: Benjamins. Haukioja, J. 2000. “Grammaticality, response-dependency, and the ontology of linguistic objects.” Nordic Journal of Linguistics 23: 3–25. Heil, J. 1992. The Nature of True Minds. Cambridge: Cambridge University Press. Hockett, C.F. 1968. The State of the Art. The Hague: Mouton. Hofstadter, D. 1995. Fluid Concepts and Creative Analogies. London: Penguin Books. Hudson, R. 1997. “Inherent variability and linguistic theory.” Cognitive Linguistics 8–1: 73–108. Itkonen, E. 1977. “The relation between grammar and sociolinguistics.” Forum Linguisticum I/3: 238–254. Itkonen, E. 1978. Grammatical Theory and Metascience. Amsterdam: Benjamins. Itkonen, E. 1980. “Qualitative vs. quantitative analysis in linguistics.” In Evidence and argumentation in linguistics. T.A. Perry (ed.), 334–366. Berlin: deGruyter. Itkonen, E. 1983. Causality in Linguistic Theory. London: Croom Helm. Itkonen, E. 1991. Universal History of Linguistics: India, China, Arabia, Europe. Amsterdam: Benjamins. Itkonen, E. 1997. “The social ontology of linguistic meaning.” SKY: The Yearbook of the Linguistic Association of Finland: 49–80. Itkonen, E. 2002. “Grammaticalization as an analogue of hypothetico-deductive thinking.” In New Reflections on Grammaticalization. I. Wischer and G. Diewald (eds.), 413–422. Amsterdam: Benjamins. Itkonen, E. 2003a. Methods of Formalization beside and inside both Autonomous and non-Autonomous Linguistics. University of Turku: Publications in General Linguistics 6. Itkonen, E. 2003b. What is Language? A study in the Philosophy of Linguistics. University of Turku: Publications in General Linguistics 8. Itkonen, E. 2004. “Typological explanation and iconicity.” Logos and Language V/1: 21–33. Itkonen, E. 2005a. Analogy as Structure and Process: Approaches in Linguistics, Cognitive Psychology and Philosophy of Science. Amsterdam: Benjamins. Itkonen, E. 2005b. Ten non-European Languages: An Aid to the Typologist. University of Turku: Publications in General Linguistics 9. Jackendoff, R. 1994. Patterns in the Mind. New York: Basic Books. Kac, M. 1992. Grammars and Grammaticality. Amsterdam: Benjamins. Katz, J. 1981. Language and Other Abstract Objects. Oxford: Blackwell. Katz, J. 1985. “An outline of Platonist Grammar.” In The Philosophy of Linguistics. J. Katz (ed.), 172–203. Oxford: Oxford University Press. Kenny, A. 1975. Wittgenstein. London: Penguin Books. Kiparsky, P. 1979. Panini as a Variationist. Cambridge, MA: The MIT Press. Kiparsky, P. 1993. “Paninian linguistics.” In The Encyclopedia of Language and Linguistics, Vol. 1(6). R.E. Asher (ed.), 2918–2923. Oxford: Pergamon Press. Körner, S. 1960. The Philosophy of Mathematics. London: Hutchinson. Lakoff, G. 1987. Women, Fire, and Dangerous Things. Chicago: University of Chicago Press. Langacker, R. 1987. Foundations of Cognitive Grammar, Vol. I: Theoretical Perspectives. Stanford, CA: Stanford University Press. Leech, G. 1983. Principles of Pragmatics. London: Longman. Lewis, D. 1969. Convention. Cambridge, MA: Harvard University Press. Marx, K. and Engels, F. 1973 [1846]. Die deutsche Ideologie. Werke, Band 3. Berlin: Dietz Verlag.
Normativity in language and linguistics 305
Newton-Smith, W.H. 1981. The Rationality of Science. London: Routledge. O’Neill, J. (ed.) 1973. Modes of Individualism and Collectivism. London: Heinemann. Pateman, T. 1987. Language in Mind and Language in Society. Oxford: Clarendon Press. Paul, H. 1975 [1880]. Prinzipien der Sprachgeschichte. Tübingen: Niemeyer. Pettit, P. 1996. The Common Mind, 2nd ed. Oxford: Oxford University Press. Popper, K. 1957. The Poverty of Historicism. London: Routledge. Popper, K. 1972. Objective Knowledge. Oxford: Oxford University Press. Russell, B. 1967 [1940]. An Inquiry into Meaning and Truth. Pelican Books. Saunders, J.T. and Henze, D.F. 1967. The Private-Language Problem. New York: Random House. Saussure, F. de. 1962 [1916]. Cours de linguistique générale. Paris: Payot. Sinha, C. 1988. Language and Representation: A Socio-Naturalistic Approach to Human Development. New York: Harvester. Verhagen, A. this volume, “Intersubjectivity and the achitecture of the language system.” Verschueren, J. 1999. Understanding Pragmatics. London: Arnold. Wason, P.C. and Johnson-Laird, P. 1972. Psychology of Reasoning. Cambridge MA: Harvard University Press. Winch, P. 1958. The Idea of a Social Science. London: Routledge. Wittgenstein, L. 1958. Philosophical Investigations, 2nd ed. Oxford: Blackwell. Wittgenstein, L. 1969. On Certainty. Oxford: Blackwell. Zlatev, J. 2007. “Language, embodiment and mimesis.” In Body, Language and Mind, Vol. I: Embodiment. T. Ziemke, J. Zlatev and R. Frank (eds.), 297– 337. Berlin: Mouton de Gruyter. Zlatev, J. this volume. “The co-evolution of intersubjectivity and bodily mimesis.”
chapter 13
Intersubjectivity and the architecture of the language system Arie Verhagen Certain lexical and grammatical units encode aspects of intersubjective coordination. On the basis of discourse connectives, and especially of negation and complementation, linguistic communication is argued to be inherently ‘argumentative’, a matter of influencing other people’s attitudes and beliefs. Intersubjectivity is built into the very structure of grammar, and systematic properties of grammar show that mutual influencing, rather than just ‘sharing information’ or ‘joint attention’ is at the heart of human language. Because of that, language can on the one hand be seen as a special case of animal communication systems, which basically involve management and assessment of other organisms, notably conspecifics. On the other hand, an important difference is precisely that this management and assessment is indirect, presupposing shared knowledge, and aimed at other minds.
1.
Introduction
Human languages have several features that are candidates for the status of ‘distinctive characteristic’ in comparison to communication systems of other animals (Hockett 1958). Some of these have a special connection to the concept of intersubjectivity, understood as the mutual sharing of experiential-conceptual content between subjects of experience. Thus, the basically conventional character of the relation between (observable) form and (unobservable) function in the symbols of human languages presupposes intersubjectivity: conventions are mutually shared solutions to coordination problems, rules that are followed because of the expectation that others will follow them and because one knows that others expect one to follow them (Lewis 1969; Keller 1998; Itkonen, this volume, traces the origins of this insight to Wittgenstein’s famous ‘argument against private language’ and argues that it entails that linguistic phenomena are inherently normative in a way that does not allow a reduction to strictly physical phenomena). Being ‘mutually shared’ is at the core of any definition of intersubjectivity (Zlatev, this volume),
308 Arie Verhagen
so that linguistic symbols, being conventional, are necessarily intersubjectively grounded. The way linguistic conventions emerge, change and are maintained thus provides a special window on human intersubjectivity. Another such feature is referentiality: the systematic use of a signal to make another individual pay attention to a specific phenomenon in the world. Systematic links between a signal and the external world have been shown to exist in other animal communication systems, but their character seems to differ systematically from that of linguistic symbols. With respect to the famous case of the different alarm calls of vervet monkeys (Cheney and Seyfarth 1990) – distinct for leopards, eagles, and snakes – Tomasello (2003: 10) comments: It seems as if the caller is directing the attention of others to something they do not perceive […], that is the calls would seem to be symbolic (referential). But several additional facts argue against this interpretation. First, there is basically no sign that vervet monkeys attempt to manipulate the attentional or mental states of conspecifics in any other domain of their lives. Thus, vervets also have different “grunts” that […] mainly serve to regulate dyadic social interactions not involving outside entities, such as grooming, playing, fighting, sex, and travel. Second, predator-specific alarm calls turn out to be fairly widespread in the animal kingdom. They are used by a number of species – from ground squirrels to domestic chickens – that must deal with multiple predators requiring different types of escape response (Owings and Morton, 1998), but no one considers them to be symbolic or referential in a human-like way.
The special kind of referentiality found in human communication by means of language thus also seems to be intimately tied up with intersubjectivity. It crucially involves a triadic relationship of sharing attention for an outside object with another individual. A similar comment applies to the discovery of individual vocal signatures, not dependent on voice characteristics of the caller, used by bottlenose dolphins (Janik, Sayigh and Wells 2006). These have been compared to ‘names’ in human languages, because of the fact that they identify an individual uniquely through the shape of the signal (not by voice characteristics), i.e. a dyadic relationship, in this case between language and the world. But humans only use their names themselves when introducing themselves to strangers; it is others who use names to address a specific individual and to talk about such an individual to others. In contrast, about half of a dolphin’s whistles in the wild consists of signature whistles (Janik et al. 2006: 8295), while in human language use, first person pronouns like English I, i.e. the same form for different individuals, belong to the most frequent words (e.g. De Jong 1979 for spoken Dutch). Thus, the use of deictic elements such as I as well as that of names in human language is dependent on understanding so-called role-reversal, which involves a triadic relationship of joint attention for an object (Tomasello
Intersubjectivity and the language system 309
1999: 103–107). Taking the common core of ‘intersubjectivity’ to precisely consist of ‘sharing’ (of some mental content) and ‘joint attention’ as a paradigmatic instance (Zlatev, this volume), intersubjectivity is precisely what distinguishes the dolphins’ signature calls from the way human names work, and from human ways of referring to oneself. Still, detailed studies of animal communicative behavior are highly relevant to understanding human behavior and human language, if only because they help to unwrap initially holistic concepts such as ‘referentiality’, ‘names’, and the like into different aspects, some of which have clear parallels in the animal kingdom (in the cases mentioned: picking out a specific category of phenomena in the world, or unique identification of individuals). In that way they contribute to linking human language to other phenomena in the natural world, to the prospect of a more complete understanding of language – a cultural phenomenon – as grounded in biology, and thereby to linking culture to nature. Now how about the concept of intersubjectivity itself? Can we distinguish different aspects of this phenomenon too, such that at least some of them can insightfully be regarded as comparable to aspects of animal communication? In this chapter, I want to argue that we can, in fact: that we should, given a proper understanding of crucial components of meaning and grammar.
2.
Is intersubjectivity something ‘completely different’?
Tomasello (2003: 12) lists the following points of difference between language and communicative signals of other primate species: 1. Language is socially learned and transmitted culturally. 2. Linguistic signals are conventional, i.e. understood intersubjectively (cf. Section 1). 3. Linguistic signals “are not used dyadically to regulate social interaction directly, but rather they are used in utterances referentially (triadically) to direct the attentional and mental states of others to outside entities” (ibid.). 4. Linguistic signals “are sometimes used declaratively, simply to inform other persons of something, with no expectation of an overt behavioral response” (ibid.).
. At this point, Tomasello refers to Dunbar (1996), who puts forward the hypothesis that language originated in the process of gossip, the sharing of information for purposes of social bonding.
310 Arie Verhagen
5. Linguistic signals “are fundamentally perspectival in the sense that a person may refer to the same entity as dog, animal, pet, or pest, or to the same event as running, fleeing, moving, or surviving – depending on her communicative goal with respect to the listener’s attentional states” (ibid.). Properties (2), (3) and (4) necessarily involve intersubjectivity. Property (5) may function, as Tomasello indicates, in an intersubjective way, but it certainly need not: different construals of the same entity or event may also be useful for a single individual’s interaction with the world, as different categorizations (e.g. as pet or as pest) invite different types of action. Property (2), conventionality, has already been discussed; intersubjectivity here provides the foundation for the way linguistic signals function in a community. It is in properties (3) and (4) that intersubjectivity enters into the character of the messages conveyed by linguistic signals themselves. Moreover, the two – directing someone’s attention to an outside entity, and informing someone of something – are obviously closely connected. With respect to these two features, Tomasello construes the specific character of human language in opposition to that of communicative signals of other primates; language involves joint attention and sharing information, whereas animal communication is dyadic, and consists of inducing behavior, such as an escape response in conspecifics. Owings and Morton (1998), to whom Tomasello refers in this connection, have developed this idea in great detail for many species using vocal communication. They describe their approach themselves in a programmatic way as follows: This book provides a discussion of animal vocal communication that avoids human-centered concepts and approaches, and instead links communication to fundamental biological processes. […]. Animals use signals in self-interested efforts to manage the behavior of other individuals, and they do so by exploiting the active assessment processes of other individuals. […] Communication reflects the fundamental processes of regulating and assessing the behavior of (Owings and Morton 1998: i) others, not of exchanging information.
Consider the vervet monkeys’ alarm calls mentioned by Tomasello (cf. above). Even if the call is species-specific, there is no reason to say that its meaning consists . In cognitive linguistics, this is known as the fundamental phenomenon of construal (Langacker 1987). For an overview, see Verhagen (2007). . Feature 1 does not imply intersubjectivity. For example, elements and structure of birdsong are culturally transmitted (transferred by learning, observation, memorization and copying of behavior; cf. Hultsch and Todt 2004), and intense interaction facilitates this learning and the quality of the result, but it certainly does not presuppose any mutual sharing of memory or experience.
Intersubjectivity and the language system 311
of reference to the predator (the individual, or the category). The meaning of the call is to induce predator-specific escape responses. The way Owings and Morton characterize animal communication presupposes that exchange of information does constitute the basic function of human communication by means of language, and as we have seen, Tomasello also construes some of the crucial differences between animal communication and human language in this way. But what if human language is also fundamentally a matter of regulating and assessing others, with exchange of information being secondary? No doubt, the descriptive power of human languages greatly exceeds that of animal communication systems (as far as we know), but that does not yet imply that linguistic meaning primarily consists in descriptive information and that regulatory effects are derivative; in principle, it may still be the other way around. Precisely this latter position is a crucial part of the conceptual framework developed in Verhagen (2005). It is this idea that I will develop and demonstrate further in this chapter. The evidence I will be considering consists of systematic characteristics of linguistic elements, especially from the domain of grammar, i.e. words and constructions that provide the scaffolding for sentences and discourse.
3.
Argumentativity: Concepts and methods
3.1
Argumentativity and conventional meaning
When one individual produces a linguistic utterance for another one, and this other individual understands it, the result is in systematic ways always more than the participants jointly focusing on the same object of conceptualization in the same way. It also consists in inducing, and engaging in, inferential reasoning. Normal language use is never just informative, but always ‘argumentative’, in the terminology of Anscombre and Ducrot (1989). Engaging in verbal communication comes down to, for the speaker/writer, an attempt to influence someone else’s thoughts, attitudes, and sometimes immediate behavior – even when a speaker simply says Over there in response to a Wh-question like Where is the bus stop? (cf. below, end of this section). For the addressee it involves finding out what kind of influence it
. Behavioral biologists may to some extent differ on the question whether a notion of ‘information’ has any role at all to play in explaining animal communication, but such differences are relatively marginal. Thus, although Bradbury and Vehrenkamp (2000) do not agree entirely with Owings and Morton (1998), their initial statement also reads: “It is widely agreed that animal signals modulate decision making by receivers of the signals” (Bradbury and Vehrenkamp 2000: 259, referring to the seminal work of Dawkins and Krebs 1978).
312 Arie Verhagen
is that the speaker/writer tries to exert, and deciding to go along or not. In terms of intersubjectivity: the process of verbal communication involves partially shared and partially divergent experiential-conceptual content, that communicating subjects attempt to coordinate on by means of (the speaker) attempting to influence the other’s inferences and (the addressee) assessing such attempts. In itself, this is not incompatible with an information view of linguistic meaning. The constant, conventional function of ordinary words and constructions might consist in the information they provide, with rhetorical effects coming on top of that, depending on the context, and thus being variable. However, Anscombre and Ducrot argue for the opposite position, which is therefore sometimes characterized as a theory of argumentativity ‘in’ the language system. The default condition for ordinary expressions, in this view, is that they provide an argument for some conclusion, and this argumentative orientation is what is constant in the function of the expression, while its information value is more variable. For example, in a commentary to the Dutch national Budget for the year 2001 – the most favorable one in many years – government officials from the Ministry of Finance wrote that there was a prospect of a “negative deficit”, thereby indicating that there were more reasons than ever to control the budget. A criterion of adequacy for a semantic theory is that it should explain why the effect of this expression on addressees is systematically different from that of the expression surplus, despite the fact that this is truth-functionally equivalent. The point is that the word . When pronominal reference to the roles of ‘speaker/writer’ and ‘hearer/addressee’ is called for, I will adopt the practice of using feminine forms (she, her) for the former, and masculine ones (he, his) for the latter. . This idea of argumentativity in natural language is to an important extent in agreement with the basic position adopted by Levinson (2000): most inferences associated with an expression, even if they are defeasible, are conventional, and not computed on-the-fly, contrary to certain traditional and newer approaches in (Gricean) pragmatics, notably Relevance Theory (cf. Sperber & Wilson 1986); i.e. they cannot be conceived of, in a cognitively realistic approach to semantics and pragmatics, as conversational implicatures that are derived on the basis of some strictly truth conditional content plus knowledge of the context. Rather, such inferences are derived because of the use of the expression itself. Thus, the argumentativity approach I advocate here and Levinson’s concept of generalized implicatures are in agreement about the idea that ‘strength’ is a normal and crucial part of semantics, and also with Itkonen’s (this volume), notion of conventional meaning as ‘frozen action’. An important difference between the argumentativity approach and these other pragmatic-semantic approaches to meaning in natural languages, resides in the distinction, in the argumentativity approach, between ‘orientation’ and ‘strength’. As we will see in the remainder of this chapter, precisely this distinction allows for a general treatment of some seemingly distinct phenomena. More generally, the argumentativity approach comprises, in a single conceptual framework, a number of notions that have been developed independently of each other in different fields of pragmatic research (cf. also Note 8).
Intersubjectivity and the language system 313
deficit is conventionally associated with warning, i.e. counts as an argument to cut spending. The use of the word negative does not reverse this argumentative status. On the contrary, it strengthens the point because it adds its own rhetorical force, which points in the same direction as deficit (that of warning). As with deficit, this must be considered an inherent part of the meaning of the word. It is the conventional meanings of these words that allowed the writers of this text to use them in attempting to regulate the attitudes of their readership in a way that is in their interests. If this effect were something that comes on top of the informational value of the utterance – inferred by readers in context, after having computed the information –, then it is impossible to explain why there is a systematic difference in signal value between negative deficit and surplus: the information value, which in this view is the starting point for readers’ inferences, is not different, so readers should be able to reach the same conclusions in both cases. But in fact, the inferences to be drawn from these two expressions are each other’s opposites. What we can now observe is that the information value of the term deficit is more variable than its argumentative value: in combination with the term negative, it is compatible with situations that can also be described as surplus, but even in such a combination, what remains constant is the argumentative value of the signal. Or consider a very simple sentence as in (1) (Ducrot 1996: 42). (1) There are seats in this room.
What are the properties of situations in which the utterance of (1) is appropriate? At first sight, the argumentative character of (1) may not be apparent, and one might think that understanding the utterance just consists of knowing how to check it against reality: are there seats in the room or not? Here it is crucial to take into account that understanding an utterance at least includes knowing how it fits into the ongoing discourse, i.e. how it relates to preceding and following utterances. People do not communicate by means of isolated sentences, but by means of discourse consisting of multiple utterances that enter into specific relations with each other, such as question-answer, cause-consequence, problem-solution, and the like. There are even special classes of elements that provide instructions on how to connect pieces of discourse, i.e. anaphora, and especially: different kinds of conjunctions (and, but, because, etc.) and connecting adverbs and adverbial phrases (so, as well, yet, etc.) – jointly: ‘discourse connectives’. So when investigating the meaning of an utterance containing the word seat, as in (1), we should not . This is not to say that the argumentative value of an expression can never be reversed, but this requires the use of special elements (e.g. a negative argumentative operator like barely); cf. Section 4.
314 Arie Verhagen
only look at how it relates to the/some world, but also at the kinds of discourse that it fits in a coherent way, and the kinds that it does not fit well. So consider what happens when the utterance following (1) is something like They are uncomfortable. How to connect this to (1)? The obvious way is to use a contrastive conjunction like but. Something like and moreover would be highly incongruous. Schematically (‘#’ indicating lack of coherence): (2) There are seats in this room. a. But they are uncomfortable. b. #And moreover, they are uncomfortable.
The reverse is the case if the next utterance is They are comfortable: (3) There are seats in this room. a. #But they are comfortable. b. And moreover, they are comfortable.
What (2) shows is that (1) as such induces an addressee to make positive inferences about the degree of comfort provided in this room. This is apparent from the need to use the contrastive conjunction but when the next utterance cancels this inference (because of uncomfortable), and from the strangeness of the additive connective in (2b). Saying that the seats are uncomfortable is not adding a simple piece of information to the information about the presence of the seats. When comfortable is used (rather than uncomfortable), the pattern is reversed, as shown in (3): here the inference induced by (1) is reinforced, so the additive connective in (3b) is appropriate, and the contrastive one in (3a) is not. Thus, an utterance like (1) counts as an attempt by the speaker to convince the addressee of some point that goes beyond the information provided. Moreover, this is part of the conventional function of the expression in (1). One simply does not know the meaning of seat if one can only distinguish objects as belonging to the class or not, but does not know that it licenses this kind of inferences. In view of this, we may say that the meaning of the word is its contribution to the argumentative value of utterances in which it occurs. On this basis, it can also be seen in what sense even an apparently simple piece of information such as the answer Over there to the Wh-question Where is the bus stop? is argumentative, too (cf. the beginning of this section). Observe the use of the contrastive connective in Over there, but the last bus has already left or Over there, but the line has been temporarily re-routed. Clearly, it is a matter of convention that providing information about the location of the bus stop counts as inducing the addressee to make certain inferences, probably about the
Intersubjectivity and the language system 315
ossibility to take a bus at that location in the near future (the question will also p have been taken as having such a desire as its background). The normal situation for linguistic meanings seems to be that their argumentative value is tied to a particular way of construing a situation or some aspect of it. Having only some conventional rhetorical strength constitutes a rather restricted type of language (Verhagen 2005: 18); it may be found in elements such as words for a greeting (Hello) or an apology (Sorry). Notice that these cases show that the expression of a positive attitude towards the addressee has the status of an interpretation. Using such expressions counts as a greeting or an apology, irrespective of the actual attitude of the speaker with respect to the addressee or the issue at hand, although, naturally, an inference about the speaker’s mental state is often justified. In the same vein, the meaning of That’s great!, for example as uttered in response to an interlocutor’s announcement of a job offer, is not primarily an expression of the speaker’s attitude, but a signal to the addressee that the speaker acknowledges the addressee’s (right to a) positive evaluation. Again, this normally licenses an inference about the speaker’s actual mental state, but it is not the primary meaning from which the rhetorical value is derived. Rather, it is the other way around: we infer the speaker’s personal mental state from the argumentative value of an expression.
3.2 The concept of ‘topos’ In exactly what way is the relevant argumentative dimension determined? In order to see how this works, recall that the inferential load of utterances is crucially involved in the way they relate to each other in connected discourse. Discourse consists of chains of inferential steps, including the possibility to reject one or more steps, and change direction. Consider the exchange in (4). (4) A: Do you think our son will pass his courses, this quarter? B: Well, he passed those of Winter Quarter.
In a purely information-oriented perspective, B’s utterance should be said not to address the question posed by A. So why can this be a coherent piece of discourse? The reason is, again, that every utterance is taken as orienting the addressee towards certain conclusions by invoking some mutually shared model in which the object of conceptualization figures, a ‘topos’ in Anscombre and Ducrot’s terminology. In our culture it is a rule, mutually known to the members of the
316 Arie Verhagen
c ulture, that passing some test normally licenses the inference that one will be able to pass other tests as well; in other words, the topos is that if someone passed a test, it is more likely that he will be able to pass other tests than that he will not. Notice the use of terms like ‘normally’ and ‘more likely’ in the formulation of this rule – it is a kind of default rule, not a universally valid one. Given such a topos, it is valid to infer from the statement He passed his courses, that he is probably capable of successfully performing certain tasks, like taking courses of this kind. In this way, B’s utterance can count as a coherent, in principle positive, answer to A’s question. That is, creating an argumentative connection is what appears to make a set of utterances into a coherent discourse. Again, an addressee takes an utterance not (just) as an instruction to construe an object of conceptualization in a particular way, but as an instruction to engage in a reasoning process, and to draw certain conclusions; it is typically not just attending to the same object, but understanding what the speaker/writer is getting at (what she wants you to infer), that counts as successful communication. And understanding what it is that your interlocutor wants you to infer, constitutes a move from a relatively indirect relation between coordinating minds (through shared attention for some object), to a more direct one; as we shall see, it is this more direct ‘inter-subjects’ connection that certain grammatical constructions operate on. The predicative use of ordinary adjectives, e.g. about size or quantity, also provides good illustrations (cf. Pander Maat 2006). Saying that someone is tall, in this view, does not primarily provide information about that person’s length, but counts as a recommendation of some kind (depending on the topos being activated), e.g. to select him for the basketball team, or not to select him as a jockey. Notice that a person being called tall in the jockey-selection situation may be shorter than a person rejected for the basketball team because he was ‘short’. Again, the constant value of the terms is in their argumentative orientation, not (just) in their information value. Of course, we are also getting some information about the world from the utterances, just like we are able to get information out of the expression negative deficit. In this case, knowing what the relevant topos is (e.g. the taller someone is, the better the chance that he will make a good basket ball player), and knowing something about the average length of persons in general and basket ball players in particular, we can make certain guesses about the range of possible sizes for the person involved. But that is not primary in the conventional knowledge activated by the word tall. Activation of a scale of length that allows inferences about a . A topos is thus a component of the ‘common ground’ in the sense of Clark (1996) (cf. Verhagen 2005: 7–16); cf. also Sinha (1999).
Intersubjectivity and the language system 317
person’s actual height is dependent on knowledge of the relevant argumentative scale, not the other way around.
3.3 Argumentative orientation and argumentative strength So far I have argued for the inherent argumentativity of language on the basis of phenomena in the domain of the lexicon. Other phenomena in this domain include several kinds of speech act verbs (see also Section 5), evaluative adverbs such as hopefully and unfortunately, and connectives like so and although. But arguably the most striking evidence for the fundamental argumentative character of human communicative intersubjectivity comes from the fact that it pervades core parts of grammar. That is what I will turn to in the remainder of this chapter.10 The methodology I will be using in order to demonstrate the precise argumentative character of certain parts of grammar is based on the appropriateness or inappropriateness of discourse connectives that was introduced in the previous section, cf. examples (2) and (3). These are taken as diagnostic cues for the argumentative value of the utterances being connected. An important distinction that can be elucidated in this way is that between argumentative orientation and argumentative strength. Consider the relation between the expressions a small chance and little chance. These may well refer to the same percentage of probability, for example 20%, but their roles in orienting an addressee to certain conclusions are systematically different. In their import, they are exactly opposite, as can be demonstrated with (5) and (6). Suppose someone is considering whether or not to perform a surgical operation on a patient who is in a serious condition; then it is coherent for this person to say (5a), but not (5b). (5) There is a small chance that the operation will be successful. a. So let’s give it a try. b. #So let’s not take the risk.
. This does not necessarily mean that such an informational component has to be so indirect for all words in a language, i.e. that it could never be conventional – languages are more flexible than that. For example, Anscombre and Ducrot (1989) propose the interesting hypothesis that numerical expressions in natural languages should be considered a special device, a kind of operator to remove the default argumentative orientation of ordinary expressions. Saying that someone’s height is 1.75 meter does not inherently display the argumentative orientation of saying that someone is tall or short. Using precise numeral specifications is obviously more artificial and elaborate than using words like tall, short, fast, slow, etc., which testifies to the default condition of the linguistic meaning of everyday expressions being inherently argumentative. 10. These are based on the analyses in chapters 2 and 3 of Verhagen (2005), respectively.
318 Arie Verhagen
What this shows is that saying There is a small chance orients an addressee to the same conclusions as the positive statement There is a chance. On the other hand, (6) exhibits the mirror pattern: (6a) is not coherent, but (6b) is. (6) There is little chance that the operation will be successful. a. #So let’s give it a try. b. So let’s not take the risk.
Saying There is little chance orients an addressee towards the same conclusions as the negative statement There is no chance. Notice that it makes no difference what the actual percentage of the chance of success is. Whatever turns out to be the case, a small chance basically orients the addressee to the same general kind of conclusions as a chance, while little chance orients one to the same sorts of conclusions as no chance.11 The expressions do not by themselves indicate positive vs. negative recommendations. Suppose the context is not that of a surgeon wondering whether or not to perform an operation, but of a policeman wondering whether or not to interrogate a seriously injured victim of a shooting, who is waiting to be operated. In that situation, it may very well be coherent to say There is little chance that the operation will be successful. So let’s give it a try., cf. example (6a), employing a topos of the kind ‘The more important certain information is, the more acceptable it is to take risks in obtaining it’. In that sense, the pragmatic import of the expression little chance is context dependent. But the significant point is that its effect is still the same, in this context, as that of the expression no chance, and the reverse of the effect of a small chance. In this context, it would precisely be coherent to say There is a small chance that the operation will be successful. So let’s not take the risk., cf. example (5a). Thus the conventional, context-independent linguistic meaning of an expression of the type little X is to reverse the orientation of the inferences associated with the predicate X (with less strength than no; cf. below), whatever topos is being employed. The equally conventional context-independent meaning of a small X is to maintain the orientation of the inferences associated with the predicate X, while their strength is less than with an unmodified assertion. Thus a generalization can be made over negation and expressions like little chance in terms of argumentative orientation: their use has the function of directing the addressee to infer that certain conclusions are invalid. The difference must 11. Therefore, as with the meaning of the expression negative deficit, there does not seem much prospect for deriving the difference between the distinct intersubjective functions of these expressions from a descriptive difference without somehow introducing the argumentative orientations in the derivation, i.e. in a non-circular way. The argumentative difference must itself be taken as part of the linguistic meaning of these expressions.
Intersubjectivity and the language system 319
Table 1. Argumentative orientation and strength of [operator]-chance a chance a small chance no chance little chance
orientation
Strength
+ + – –
High Low High Low
be characterized in terms of argumentative strength. Straightforward negation has maximal argumentative strength; its use relates a specific situation (the chance of success here and now) to a shared inferential model of a type of situations (the more chance of success, the more reason to operate) without any qualification: given the topos, the situation in the world provides the strongest possible argument for invalidating the conclusion ‘go ahead with the operation’. This is the locus of the difference with little X. The latter shares its argumentative orientation with negation (the chances of success are not optimal), but presents it as weaker: it is qualified, since the situation in the world comprises a feature that might be construed (in other circumstances, or by another person) as favoring an operation. Similarly, a small X shares its argumentative orientation with unmodified predication, but presents it as weaker. This is summarized in Table 1. The second column does not represent a binary distinction, but a scale on which expressions can also occupy in-between positions. For instance, the strength of no chance is arguably maximal, while that of a chance may easily be surpassed by that of, for example, every chance in the world. Operators with less than maximal strength leave room for discussion and negotiation. As we have seen before, an utterance that cancels inferences associated with the previous one, must be marked with a contrastive connective like but. This also applies to the cases we are considering here; for example, (7a) produces a coherent discourse, unlike (6a): (7) There is little chance that the operation will be successful. a. But let’s give it a try. b. #But let’s not take the risk.
But when the strength of the negative operator is maximal, there is no room for canceling the inference that it is not worth trying; uttering (8) always amounts to inconsistency: (8) #There is no chance that the operation will be successful. But let’s give it a try.
320 Arie Verhagen
A difference like this is a natural consequence of differences in strength, and of the scales involved having definite boundaries; it does not undo the parallel in argumentative orientation. Distinguishing argumentative strength from orientation precisely allows us to formulate in a natural way what straightforward negation has in common with other expressions that at first sight may not look like negation but nevertheless behave in highly similar ways, as we shall see in the next section.
4.
The negation system as an intersubjective coordination system
The claim is that both straightforward negation and the construction little X belong to a larger system of expressions that share an effect on the argumentative orientation of the utterances they are part of (though they may differ in their strength).12 To the degree that we can substantiate the claim that this grammatical system must be characterized in this way, we have provided evidence that the nature of intersubjectivity as built into the linguistic system, is basically argumentative. Consider the following set of expressions containing the expression let alone: (9) He didn’t pass Statistics-1, let alone Statistics-2. (10)
??He
didn’t pass Statistics-2, let alone Statistics-1.
(11) *He passed Statistics-1, let alone Statistics-2. (12) *He passed Statistics-2, let alone Statistics-1.
In view of these, let alone appears to connect two elements that are ordered on a scale in a specific way, witness the problematic status of (10): presumably, Statistics-2 is harder to pass than Statistics-1. Moreover, let alone is also a negative polarity item: since neither (11) nor (12) is fine, it appears to require the presence of a negation operator in the first clause (cf. Fillmore, Kay and O’Connor 1988). Next, notice that almost X has a negative entailment, whereas barely does not: (13) He almost passed → He did not pass (14) He barely passed → He passed
Nevertheless, almost cannot license the use of let alone, while barely can: (15) *He almost passed Statistics-1, let alone Statistics-2. (16) He barely passed Statistics-1, let alone Statistics-2. 12. It is especially because of this kind of systematicity that a linguistic analysis provides a powerful window on the mind. The significance of this point is easily overlooked by proponents of the information view, cf. Hinzen & Van Lambalgen (2008) and Verhagen (2008).
Intersubjectivity and the language system 321
Table 2. Argumentative orientation and strength of [operator]-passed passed almost passed did not pass barely passed
orientation
strength
+ + – –
high low high low
From an information-oriented point of view, this is a riddle. An explanation can be based on the insight that barely reverses the argumentative orientation of an utterance (as negation does), while almost preserves it (despite the facts, so to speak), as can be seen in the following set of connected utterances: (17) He passed Statistics-1. So there is hope that he may make it through the first year. (18) #He didn’t pass Statistics-1. So there is hope that he may make it through the first year. (19) He almost passed Statistics-1. So there is hope that he may make it through the first year. (20) #He barely passed Statistics-1. So there is hope that he may make it through the first year.
Naturally, as an instrument for reversing the argumentative orientation of an utterance, barely is weaker than straightforward negation, just like almost is a weaker positive argument for a conclusion than a statement without any hedges. Schematically, this is shown in Table 2. And just as we saw in the previous section, the weaker argumentative operators allow room for discussion, changing direction in the subsequent discourse (using a contrastive connective), while the strongest one is not defeasible: (21) He barely passed the test. But anyway, he did. (22) #He did not pass the test. But anyway, he did.
Recognizing barely and almost as operators and let alone as a connector at the argumentative level, allows for a unified explanation of the phenomena mentioned above, including the problem of barely licensing the use of let alone. What these elements have in common, and what determines the similarity of their grammatical properties, is the fact that their conventional meaning primarily functions at the level of the argumentative value of utterances, not on that of informational
322 Arie Verhagen
Figure 1. A schematic view of the intersubjective nature of the negation system (see the text for discussion)
content.13 Their common features in terms of intersubjectivity can be depicted as in Figure 1 (cf. Verhagen 2005: 57). The use of an element from the negation system, e.g. not or barely, sets up a configuration of two perspectives (‘mental spaces’ in terms of Fauconnier 1994), the first of which is that of the person responsible for the utterance (including the negative element), which contrasts in a particular way with a projected second perspective (by default: the addressee’s). The speaker/writer envisages that the addressee might entertain a thought q, for example, that there is hope for their son’s making it through the first year. This is represented as ‘?q’ in Space2. She furthermore believes that she shares the knowledge of a certain cultural model with the addressee, for example, that passing a statistics course in our culture normally provides some ground for the conclusion that one can also pass other sorts of courses. This is represented by the topos ‘P→Q’ in both Space1 and Space2. Both the use of not p and that of barely p invalidate q (given the topos), inviting the addressee to consider ¬q more justified than q, at least at this point in the discourse. What this explication shows is that intersubjectivity is built into the very semantics of natural language negation elements: they involve multiple distinct perspectives that are to be coordinated in a particular way, and they presuppose shared knowledge that can be invoked as a basis for inferential processes. The fact that it is this argumentative character that determines the coherence of a grammatical subsystem (in this case: 13. Cf. Verhagen (2005, Ch. 2) for a discussion of other phenomena relating to negation that can be illuminatingly (re)analyzed in such a perspective, such as the difference between sentential and morphological negation (by means of prefixes such as un-), and the reason why not impossible is not functionally equivalent to possible, despite the logical equivalence of these expressions.
Intersubjectivity and the language system 323
negation) is in turn strong evidence for the fundamental role of argumentativity in human communicative intersubjectivity. In fact, it is strengthened by the fact that similar properties are found in other subsystems of grammar.
5.
Complementation as an intersubjective coordination system
Traditionally, a sentence containing a clausal complement such as (23) is viewed as presenting one event or situation as part of another one, i.e. as some (special, structural) combination of two pieces of information about the world. (23) The envoy reported that the money had been delivered.
The situations being connected are seen as basically characterized by the verbs, i.e. report and deliver in (23). Thus, the structure of (23) is considered to be essentially the same as that of the simplex sentence The envoy reported something, with the slot of something being filled by another clause in (23) itself: this ‘subordinate clause’ fills the direct object slot in the ‘main clause’. However, children do not learn such complex syntactic constructs by combining two descriptions of events, i.e. by combining simplex clauses (Diessel and Tomasello 2001, cf. also Tomasello 2003). Rather, they start, at about 3 years of age, to add certain markers of subjective perspective like I think and you know to simple clauses of types they have already been producing before. Diessel and Tomasello show that, at least for the children, these are not complex structures that contain two propositions, but single-proposition utterances, the content of which is expressed completely by what would in a traditional structural analysis be regarded as the subordinate clause. It is only over a relatively long period of increasing linguistic experience that children gradually learn to use more verbs, sometimes also in the past tense and/or with third-person subjects, that in the end results in the emergence of a more general complementation construction that allows adults to say and understand things like (23). These facts cast very serious doubt on the validity of the traditional analysis for complementation in general, especially since a large part of complementation in spontaneous conversation of adults also consists of the elementary, singleproposition type with added perspective-marking that is first acquired by children. According to Thomspon (2002), this portion amounts to as much as 80%. It would of course be preferable to have an analysis that acknowledges the basic character of the perspective-marking function, and somehow incorporates cases with third person, past tense ‘main’ clauses as special cases.
324 Arie Verhagen
So let us reconsider the issue of the precise relationship between expressions such as (24) and (25) (24) I think it is raining. (25) John thinks it is raining.
According to two respectable and long-standing traditions, originating in linguistics (Benveniste 1958) and philosophy (Austin 1962, Searle 1969), these expressions belong to two completely different categories of utterances: either subjective/performative, or objective/descriptive, respectively. In modern linguistics, the performative/descriptive distinction is best known. In this view, the use of the following two sentences exhibits an important qualitative difference: (26) I promise that I’ll have the car up in front at 2 o’clock. (27) John promised that he’ll have the car up in front at 2 o’clock.
The point is, according to Austin, that it makes no sense to characterize (26) in terms of truth conditions, i.e. to treat it as a description of an act of promising. In uttering (26), one performs an act of promising, and the performance of an act can be felicitous or infelicitous, but not true or false. An utterance such as (27), on the other hand, constitutes a description of an act of promising, and thus its semantics can be characterized in terms of truth conditions. Accordingly, the two sentences belong to two wholly distinct categories of speech acts: (26) constitutes a ‘commissive’ one, (27) a ‘constative’ one. Benveniste (1958) had already classified speech act formulas like (26) and first-person present-tense uses of verbs of cognition like (24) together as ‘subjective’ utterances, and (25) and (27) as ‘objective’ ones. More recently, what is basically the same insight has also been formulated by others, e.g. Nuyts (2001), Diessel and Tomasello (2001: 103/4). According to Nuyts, expressions like It is probable that... and I think are used ‘performatively’ (in the sense that the speaker ‘performs’ an epistemic evaluation by uttering these expressions), while third person ones (Mary thinks that...) are used ‘descriptively’: “the speaker reports on someone else’s epistemic evaluation of a state of affairs without there being any explicit indication as to whether the speaker personally subscribes (i.e., is committed) to the veracity of the evaluation or not” (Nuyts 2001: 385). But from a grammatical and especially a functional point of view, such a dichotomy is unsatisfactory, as it implies a rather serious discrepancy between structure and function: Why are such dissimilar functions expressed in similar structures, i.e. complementation constructions? Also, the basis for children’s gradual extension of the use of complementation, observed by Diessel and Tomasello,
Intersubjectivity and the language system 325
remains a mystery, because under this analysis, it involves an abrupt shift from one category of communicative functions to a wholly different one. However, the argumentative perspective developed above precisely allows for a natural unification of these phenomena. Consider the way (26) and (27) function in the context of (28). (28) A: Can I be in Amsterdam before the match starts? B1: I promise that I’ll have the car up in front at 2 o’clock. [=(26)] B2: John promised that he’ll have the car up in front at 2 o’clock. [=(27)]
Both can count as an affirmative answer. Both can felicitously be followed by the explicit reassurance So don’t worry (notice the use of So). That is, both saying I promise that X as well as saying John promised that X count as arguments for an addressee to strengthen the assumption that X will happen, they have the same argumentative orientation.14 The difference is one of strength rather than argumentative orientation. Whereas the argumentative strength of the first-person, present tense utterance is maximal, the strength of the third person, past tense utterance is less, as the cognitive coordination between author and addressee is indirect, ‘via’ the onstage perspective of a third person; but it still functions to coordinate the perspectives of speaker and addressee, just like a first person utterance. According to this analysis, a difference between first person, present tense, and third person matrix clauses should be that the invited inference is defeasible in the latter case, but not in the former, which has a maximal strength. This is borne out, for both speech act verbs, witness (29a) and (29b), and verbs of cognition, witness (30a) and (30b): (29) a. John promised that he’ll have the car up in front at 2 o’clock. But he might have forgotten the route to your new home. b. #I promise that I’ll have the car up in front at 2 o’clock. But I might forget the route to your new home. (30) a. John believes that the mission has been successful. But in fact, it has failed. b. #I believe that the operation has been successful. But in fact, it has failed.
14. In Verhagen (1995), it is argued that it is precisely this constant argumentative orientation of the report of someone promising something that provides the basis for the development of the epistemic/evidential use of promise as in The debate promises to be interesting; in such cases the verb only functions as a speaker-oriented marker of argumentative orientation, and does not designate an act of promising.
326 Arie Verhagen
The fact that the contrastive conjunction but has to be used in (29a) and (30a) once again illustrates that the first sentence by itself has the argumentative orientation that I ascribed to it. The difference between performative/subjective and ‘constative’/objective use of verbs of communication and cognition turns out to be exactly parallel to that between maximally and less strong argumentative operators, observed in Section 4. Consider the parallel of the difference between the a and b cases in (29) and (30) with the difference between (21) and (22), repeated here for convenience: (21) He barely passed the test. But anyway, he did. (22) #He did not pass the test. But anyway, he did.
This parallel confirms the idea that third person matrix clauses of complementation constructions differ only in strength from first person ones, not in kind. In this analysis, the difference between these two types of uses appears to be not categorical, as they have the same argumentative orientation, but a matter of degree: they differ only in argumentative strength. This functional unification of first and third person matrix clauses of complementation constructions makes the discrepancy between structure and function inherent in the traditional approach disappear. The picture of the acquisition of complementation constructions also becomes more coherent: it starts with learning to add explicit markings of perspectives to utterances; initially these are completely grounded in the speech situation (I, you, present tense), they are formulaic, and have maximal strength; with experience, the child learns to understand and produce more and more indirect and general perspective markings, allowing for more nuances and for defeasibility. As with the system of negation, we find that the conventional meaning of complementation primarily functions at the level of the argumentative strength of utterances, not on that of informational value. Utterances that instantiate complementation do not consist of structural combinations of pieces of information, but of a constructed representation of some situation, structurally embedded in a perspective indicator (or more than one) that serves, sometimes in conjunction with other elements, to coordinate cognitive processes of speaker and addressee.15
15. And as with negation, this view is instrumental in solving a number of other long standing problems of grammatical analysis (cf. Verhagen 2005, Chapter 3). Among these are the issue of the precise grammatical analysis of sentences like The danger is that things will get out of hand (is the complement clause the subject or the predicate of the entire sentence?), the precise status of complements in sentences with copular predicates, like He is afraid that things will get out of hand (such predicates do not take direct objects, cf. *He is afraid a disaster), and the analysis
6.
Intersubjectivity and the language system 327
Discussion and conclusion
The evolution of cognition is to a considerable extent a story of subsequent generations of organisms interacting more and more indirectly with their environment (Dennett 1995: 370–400). The capacity for categorization and insight into causality allow an individual to act on the basis of prediction, i.e. selection of the hypothetically best course of action from two or more alternatives, without having to interact with the environment immediately; the potential advantages are obvious. The evolution of intersubjectivity fits into this picture: it provides a step beyond categorization and causality in that it allows individuals, among other things, to act on the basis of predicting what another individual will do, viz. by mentally putting oneself in the other’s position and consider what one would do oneself. The evolution of communication is a special case of increasing indirectness: it allows organisms to influence other organisms – conspecifics and others – by means of signals, without physical engagement and all its hazards. The evolution of language constitutes a further step of increasing indirectness, as a linguistic signal does not have a single directive nature, but one that is variable, since its contribution to the argumentative character of an utterance is dependent on the relevant topos, a shared cultural model; moreover, by means of argumentative operators, the argumentative orientation of a signal may be reversed, and its strength may be modified. This kind of indirectness presupposes a form of intersubjectivity: mutual recognition of the inferences the other is capable of making, given the shared model. This high degree of indirectness and this dependence on shared models provide room, in fact a basis, for referentiality: what different topoi associated with a signal have in common may be identified with the signal’s referent, the concept of some aspect of a phenomenon in the world that activates a topos. The difference with the functional referentiality of animal calls like the vervet monkeys’ predator-specific alarm calls (cf. Section 2), is that there is no one-to-one link between a specific category of phenomena and a particular type of response, but a one-tomany relationship (in modern, adult human beings). But this increased flexibility does not imply that human language is essentially informative or referential, and no longer a system for mutual management and assessment of senders and receivers of signals. In fact, as we have seen, mutual influencing is built into the very structure of grammar. Much of language use is only indirectly aimed at a behavioral response, and primarily a matter of manipulating the mental states and processes of others of Wh-questions like Who do you think pays the rent?, in which the question word seems to be extracted from its own clause (for the latter phenomenon, see also Verhagen 2006).
328 Arie Verhagen
(although such a manipulation may well constitute a particular way of eliciting a behavioral response). But a considerable portion is also conventionally aimed at immediate effects, such as making a request, asking a question, or issuing a warning, i.e. typically non-assertive speech acts. The latter point and its significance have also been noticed by Owings and Morton (1998). Discussing relations between communication and cognition in animals, they conclude “that animal knowledge structures are fundamentally pragmatic, i.e. about what to do about objects, events, and states [...]. According to this approach, signals are not statements of fact, that can be judged to be true or false, but are efforts to produce certain effects.” (Owings and Morton 1998: 211). They then notice a parallel with speech act theory, which is also specifically concerned with utterances that cannot be judged to be true or false – the discovery of which provided the original motivation for Austin’s (1962) proposals. As is well known (cf. Section 5), utterances of the type I promise to help you with your homework, Take your hands off of me!, I now pronounce you husband and wife are not to be understood as informative, they are not descriptions of states of affairs. Rather, their utterance constitutes the performance of an act; they are efforts to accomplish goals. To be sure, certain systematic conditions (called ‘preparatory conditions’ in the speech act literature) have to be satisfied in order for these utterances to count as a promise, a command, and a legitimate wedding. Owings and Morton notice that “a dedicated proponent” of the information view might try to construe these as the actual meaning of the expressions: Are these correlates, an information advocate would ask, not the information made available by these signals? No; this question confuses time frames of causation. Such correlates validate the cues as useful to assessing individuals [...], but are not the immediate cause of the assessing individual’s reaction to the statement. Note that [this] proposal turns the usual view of the role of information on its head. The correlates of signals (‘information’) are not immediate causes of the behavior of targets of signals, they are instead long-term validators of the signal’s (Owings and Morton 1998: 211) utility.
As we have seen in Section 5, the argumentative view of intersubjectivity in the language system allows for a substantial generalization of speech act theory. Standard speech act theory restricts its domain of application to non-assertive utterances, which simply cannot be understood as truth-functional. Nuyts (2001) has extended this domain to include epistemic verbs and predicates, but even then it is still limited to first person present tense use. But we have seen that the argumentative orientation of third person uses of such verbs is the same as that of first person ones. As instruments for influencing, they differ in strength, not in kind. More generally, assertive utterances which are in principle analyzable as
Intersubjectivity and the language system 329
truth-functional, such as There is going to be a negative deficit, There are seats in this room, John is tall (cf. Section 2) are also attempts to accomplish something, of the same type as standard speech acts (like warning, reassurance, or advice). In view of this generalization, we may broaden Owings and Morton’s view on the secondary character of referentiality in animal communication systems, to include human language: real-world correlates of signals are ‘long-term validators’ of argumentative cues. Human language may certainly be said to allow for distinguishing innumerably more distinctive ‘long-term validators’ of such cues than known animal communication systems. But that in itself does not turn human language into a system of information exchange rather than a system for mutual influencing. Since topoi are shared, they may usually remain implicit, and establishing joint attention suffices for a sender to get a receiver to make the desired inferences. But the way discourse units are systematically connected to each other, especially by means of linguistic elements, and systematic properties of the grammatical systems of negation and complementation reveal that the argumentative character of language is still basic. The parallel between the argumentative view of language and the modern ethological view of animal communication will now be obvious. It concerns the fact that human linguistic communication is primarily also a matter of influencing one another, by exploiting the cognitive capacities of others. Human language is more involved than (most?) animal communication with influencing mental states, with consequences for long term behavior, rather than with immediate behavioral effects, but this is a matter of degree. Since the intersubjective coordination of mental states and attitudes, according to this view, is a special form of mutual influencing, it is not to be put into opposition to the basic character of animal communication – rather, it is its specifically human variant.16
Acknowledgements I would like to thank Jordan Zlatev, as well as Chris Sinha, for insightful remarks and questions about a previous version of this chapter.
16. Beside involving a special kind of mutual influencing, human language also exhibits a number of other special features as mentioned in the beginning of this chapter. Some of these have their basis in intersubjectivity, such as the symbolic character of language (cf. Sinha 2004), its conventionality and normativity (Itkonen, this volume; Zlatev, this volume), and the exceptionally large size of the inventory of signals (due to, primarily, duality of patterning; cf. Martinet 1949; Hockett 1958), which allows for the emergence and survival of signals for other purposes than the immediate elicitation of a behavioral response.
330 Arie Verhagen
References Anscombre, J.-C. and Ducrot, O. 1989. “Argumentativity and informativity.” In From Metaphysics to Rhetoric, M. Meyer (ed.), 71–87. Dordrecht, etc.: Kluwer Academic Publishers. Austin, J.L. 1962. How To Do Things With Words. Oxford: Oxford University Press. Benveniste, É. 1971. Problems in General Linguistics. Translated by M.E. Meek. Coral Gables: University of Miami Press. Bradbury, J.W. and Vehrencamp, S.L. 2000. “Economic models of animal communication.” Animal Behaviour 59: 259–268. Cheney, D.L., and Seyfarth, R.M. 1990. How Monkeys See the World. Inside the Mind of Another Species. Chicago: University of Chicago Press. Clark, H.H. 1996. Using language. Cambridge: Cambridge University Press. Dawkins, R. and Krebs, J.R. 1978. “Animal signals: information or manipulation?” In Behavioural Ecology: An evolutionary approach, J.R. Krebs and N.B. Davies (eds.), 282–309. Oxford: Blackwell Scientific Publications. De Jong, E. (ed.). 1979. Spreektaal. Woordfrequenties in gesproken Nederlands. Utrecht: Bohn, Scheltema and Holkema. Dennett, D.C. 1995. Darwin’s Dangerous Idea. Evolution and the Meanings of Life. New York: Simon and Schuster. De Jong, E. (ed.). 1979. Spreektaal. Woordfrequenties in gesproken Nederlands. Utrecht: Bohn, Scheltema and Holkema. Diessel, H. and Tomasello, M. 2001. “The acquisition of finite complement clauses in English: A corpus-based analysis.” Cognitive Linguistics 12: 97–141. Ducrot, O. 1996. Slovenian Lectures/Conférences Slovènes. Argumentative Semantics/Sémantique argumentative. Editor/Éditeur I. Ž. Žagar. Ljubljana: ISH Inštitut za humanistične študije Ljubljana. Dunbar, R.I.M. 1996. Grooming, Gossip and the Evolution of Language. Cambridge, MA: Harvard University Press. Fauconnier, G. 1994. Mental Spaces. Aspects of Meaning Construction in Natural Language. Cambridge: Cambridge University Press. [First edition 1985, Cambridge, MA: The MIT Press.] Fillmore, C.J., Kay, P., and O’Connor, M.C. 1988. “Regularity and idiomaticity in grammatical constructions: The case of let alone”. Language 64: 501–538. Hinzen, W. and Van Lambalgen, M. 2008. “Explaining intersubjectivity: A comment on Arie Verhagen, Constructions of Intersubjectivity.” Cognitive Linguistics 19: 107–124. Hockett, C.F. 1958. A Course in Modern Linguistics. New York: The Macmillan Company. Hultsch, H. and Todt, D. 2004. “Learning to sing.” In Nature’s Music. The Science of Birdsong, P. Marler and H. Slabbekoorn (eds.), 80–107. San Diego, CA/London: Elsevier. Itkonen, E. this volume. “The central role of normativity for language and linguistics.” Janik, V.M., Sayigh, L.S. and Wells, R.S. 2006. “Signature whistle shape conveys identity information to bottlenose dolphins.” Proceedings of the National Academy of Sciences 103: 8293–8297. Keller, R. 1998. A Theory of Linguistic Signs. Oxford: Oxford University Press. Langacker, R.W. 1987. Foundations of Cognitive Grammar. Volume I: Theoretical Prerequisites. Stanford, CA: Stanford University press.
Intersubjectivity and the language system 331
Levinson, S.C. 2000. Presumptive Meanings. The Theory of Generalized Conversational Implicature. Cambridge, MA: The MIT-Press. Lewis, D.K. 1969. Convention: A Philosophical Study. Cambridge, MA: Harvard University Press. Martinet, A. 1949. “La double articulation linguistique.” Travaux du Cercle Linguistique de Copenhague V (Recherches Structurales 1949), 30–37. Nuyts, J. 2001. “Subjectivity as an evidential dimension in epistemic modal expressions.” Journal of Pragmatics 33: 383–400. Owings, D.H. and Morton, E.S. 1998. Animal Vocal Communication: A New Approach. Cambridge: Cambridge University Press. Pander Maat, H.L.W. 2006. “Subjectification in gradable adjectives.” In Subjectification: Various Paths to Subjectivity. A. Athanasiadou, C. Canakis and B. Cornillie et al. (eds.), 279–320. Berlin/New York: Mouton de Gruyter. Searle, J.R. 1969. Speech Acts. An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. Sinha, C. 1999. “Situated selves.” In Learning Sites: Social and technological resources for learning. J. Bliss, R. Säljö and P. Light (eds.), 32–46. Oxford: Pergamon. Sinha, C. 2004. “The evolution of language: from signals to symbols to system.” In Evolution of Communication Systems. A Comparative Approach. D. Kimbrough Oller and U. Griebel (eds.), 217–235 Cambridge, MA/London: The MIT Press. Sperber, D. and Wilson, D. 1986. Relevance. Communication and Cognition. Oxford: Basil Blackwell. Thompson, S.A. 2002. “‘Object complements’ and conversation: Towards a realistic account.” Studies in Language 26: 125–164. Tomasello, M. 1999. The Cultural Origins of Human Cognition. Cambridge, MA/London: Harvard University Press. Tomasello, M. 2003. Constructing a Language. A Usage-Based Theory of Language Acquisition. Cambridge, MA/London: Harvard University Press. Verhagen, A. 1995. “Subjectification, syntax, and communication.” In Subjectivity and Subjectivisation: Linguistic Perspectives, D. Stein and S. Wright (eds.), 103–128. Cambridge: Cambridge University Press. Verhagen, A. 2005. Constructions of Intersubjectivity. Discourse, Syntax, and Cognition. Oxford: Oxford University Press. Verhagen, A. 2006. “On subjectivity and ‘long distance Wh-movement’.” In Subjectification: Various Paths to Subjectivity, A. Athanasiadou, C. Canakis and B. Cornillie et al. (eds.), 323–346. Berlin/New York: Mouton de Gruyter. Verhagen, A. 2007. “Construal and perspectivisation.” In Handbook of Cognitive Linguistics, D. Geeraerts and H. Cuyckens (eds.), 48–81. Oxford: Oxford University Press. Verhagen, A. 2008. “Intersubjectivity and explanation in linguistics: A reply to Hinzen and van Lambalgen.” Cognitive Linguistics 19: 125–143. Zlatev, J. this volume. “The co-evolution of intersubjectivity and bodily mimesis.”
chapter 14
Intersubjectivity in interpreted interactions The interpreter’s role in co-constructing meaning Terry Janzen and Barbara Shaffer Introducing an interpreter into a discourse event affects the very nature of the interchange because in addition to the interlocutors’ intersubjective approach to each other, the interpreter necessarily bases her interpretation on assumptions she makes about each of the interlocutors’ shared and non-shared knowledge. Recently, many American Sign Language (ASL)-English interpreters have espoused what have been termed “expansions”, claimed to be grammatically required in ASL. But ASL has no such “explicitness” requirement; instead the interpreter must attend to the intersubjective domain of discourse interaction in order to attempt to more accurately represent what is in the minds of the interlocutors. This chapter examines triadic intersubjectivity in interpreted discourse and the role that “contextualization” plays in managing others’ shared and non-shared knowledge. [C]onversation is highly contextualized, filled with subtle cues at all levels marking the relation of utterances to contexts of prior discourse, to situational and cultural contexts, to contexts of social relations between speech event participants, and even to the mutual cognitive context within which the dialogic interaction is embedded. John Du Bois (2003: 52)
1.
Introduction
When we engage in discourse, various information types are encoded in linguistic structures we use. For example, discourse may be understood to include a combination of information that is already known to the interlocutors and information that is new. A certain amount of known information is required because it grounds what is new, while the new information is generally the point of the discourse. A balance of these two information types keeps the discourse from being either overly redundant or disconnected (Givón 1984). However, no two people
334 Terry Janzen and Barbara Shaffer
bring the exact same knowledge base or consciousness to the discourse event, so that when they express ideas in turn, they continually negotiate this balance of variables so as to best be understood by the other. This therefore constitutes an intersubjective view of co-constructed discourse. Because our focus is the interaction between adult language users, we are not concerned with intersubjectivity within an ontogenetic or developmental context, even though the emergence of mediated cognition described by Vygotsky (1978) and others is a rich area for investigation, and one that has greatly informed our discussion. Instead we take our cue from Per Linell (1995), who views the intersubjective relationship between speakers as a “continuous, collective process, where interactors mutually check understandings. What is said and understood gets continually updated on a turn-by-turn basis; each contribution to a dialogue displays (or can display) some understanding or reaction to the prior contribution” (Linell 1995: 193). Linell also believes that there must be some “meta-level management of interaction and understanding” (p. 183), consistent with Susswein and Racine’s (this volume) view that humans (but not other animals) possess an understanding that they attend to others, have intentions, and want things. Once speakers have a linguistic system in place, that system is by its very nature intersubjective, involving such elements as persuasion and conveying points of view – argumentativity, in Verhagen’s (2005, this volume) terms – and in our view, the on-going negotiation of meaning. How does this negotiation play out? How do speakers gauge that something they are conveying is understood as it is intended and whether the linguistic expression includes enough of the right information to be comprehended by the addressee? Du Bois (2003: 52) characterizes face-to-face conversational discourse as “pervasive, spontaneous, interactional, and contextualized”. We will argue that interpreted discourse is by nature a face-to-face enterprise, that the intersubjective views on the objects of discourse and on the discourse itself taken by interlocutors extends to the third-party interpreter, that the interpreter brings her own views and intentions to the discourse context, that these necessarily impact the direction and content of the discourse, and that the interpreter’s view does not always correspond to those of the primary interlocutors engaging in the discourse. Contextualization is always present in discourse as a necessary means to make our interactions coherent, both within the immediate discourse context and over the course of time. Contextualizing, where the speaker decides “on-line” the extent to which she must contextualize, is an inherently intersubjective action, taking place regardless of which language is being used, because of the negotiated nature of discourse. Further, it occurs whether the discourse participants share the same language or are attempting to communicate across a language boundary with, say, an interpreter. Contextualizing in discourse may be something that all language
Intersubjectivity in interpreted interactions 335
speakers do, but how they do it will depend on the discourse conventions of the particular language, and in turn on relevant aspects of the grammar that have evolved for that particular community of speakers. A question remains, however, as to what is required by the grammar as opposed to what is optional for participants when meaning is being negotiated. In on-line discourse where meaning is co-constructed, interlocutors navigate based on pragmatic factors and assumptions, choosing from lexical and grammatical options to construct utterances they believe will signal their intended sense. On the other hand if something is required by the grammar, the speaker or signer has no choice but to use it. For American Sign Language (ASL)-English interpreters, learning variable features of discourse and their corresponding grammar is no simple task. For many reasons interpreters experience inadequacies in their management of discourse due to insufficient language training, a lack of ready strategies, or inexperience. Nonetheless, the interpreter’s goal is to manage the information exchange and discourse packaging in not one, but two languages, hopefully with some finesse. In this paper, we suggest that the description of ASL “expansions” in Lawrence (1995) is misrepresentative of ASL grammar and of how ASL compares to other languages in terms of possible discourse strategies. The features that Lawrence describes are present to some degree in many languages, if not universally, including English. They are not formulaic as required grammatical expressions, but their use will instead be prompted by pragmatic principles having to do with negotiating the information exchange. If these principles are not understood by the interpreter there is the likelihood that she will make erroneous assumptions about shared knowledge and intentions on the part of consumers, and subsequently frame the information in ways that do not match the consumers’ discourse expectations (Janzen and Shaffer 2003; Shaffer and Janzen 2004). A more appropriate approach to information packaging may be what interpretation and translation theorists also have referred to as contextualization (e.g. Gile 1995) where contextualizing information is supplied by the interpreter based on situational factors rather than on assuming that the language requires something to be phrased in a certain way. Above all, coherent, discourse-appropriate interpretation necessitates that the interpreter have numerous language and overall discourse strategies within easy reach; any single strategy may work well in one circumstance, but fail in another. Thus managing these triadic interaction tasks – the activity of interpreting – tells us much about the very complex nature of intersubjectivity for adult speakers. In Section 2 below we review the notion of “expansions” in the literature and point out an early precursor to this idea. In Section 3 we introduce the role of pragmatics in interpreted discourse and discuss how an understanding of discourse pragmatics and grammar impacts the interpreter’s text. Section 4
336 Terry Janzen and Barbara Shaffer
f ocuses on contextualization in discourse as interlocutors negotiate meaning with one another, and in Section 5 we illustrate these principles with some examples. In Section 6 we summarize our discussion and draw some conclusions regarding the interpreter’s position in the complex intersubjective nature of three-party interpreted discourse.
2.
Expansions as “grammar”
In an attempt to better understand what constitutes native signers’ ASL, Lawrence (1995) identifies seven ASL elements that she refers to as “expansions”. Lawrence labels these as contrasting, faceting, reiteration, utilizing 3D space, explaining by examples, couching or nesting, and describe-then do. Here a description of two of these expansions should suffice to illustrate Lawrence’s claims. When “contrasting”, two ideas are juxtaposed, for example a positive and negative statement, to emphasize what is being asserted (what something is versus what it is not). Lawrence (1995: 208) gives as one example: (1) English: ASL:
Lenin’s tomb is austere. nod topic neg fs LENIN GRAVE PLAIN. FANCY // NOT. ‘Lenin’s tomb is plain, not fancy.’
For the English word austere, Lawrence suggests that the ASL construction would need two contrasting phrases. When couching or nesting, backgrounded information is added to clarify an idea. Humphrey and Alcorn (2001) suggest that couching/nesting is used “to provide information in an introductory expansion or ‘set up’ to ensure the listener has the schema or frame required to understand the upcoming discourse” . A persistent difficulty with authors’ representation of ASL on the page is that of differing methods of “glossing”, since ASL does not have a written system or standard means of notation. Also, not all authors provide a key to their transcriptions. In (1) we understand that “fs” means that something is fingerspelled. It is fairly common to find facial gestures represented by overlines, with “nod” meaning a concurrent positive head nod, “topic” (or “top”) representing facial topic marking, and “neg” meaning a negative head nod. ASL signs are given in upper case English words. . Our transliteration. . While Lawrence says that she is describing signers’ regular use of ASL, her examples appear to be ASL translations of English words. This is problematic because attempting to find crosslanguage equivalents confounds the issue of what might occur in a single language.
Intersubjectivity in interpreted interactions 337
(2001: 9.18). Inherent in this description is the premise that the use of some term in ASL cannot stand on its own. Humphrey and Alcorn (2001: 9.18) offer the following example for ‘allergy’: (2) MEDICINE-TAKE OR CREAM RUB-ON-SKIN OR FOOD EAT-FINISHITCH ALL OVER OR STOMACH UPSET OR HARD BREATH – A-L-L-ER-G-Y
Humphrey and Alcorn mean that the (fingerspelled) word A-L-L-E-R-G-Y is obligatorily “nested” within the example-rich phrasing that precedes it. Their intention is that without the “expansion” the word will not be understood by the addressee. Lawrence attributes expansions to the grammar of ASL specifically, meaning that in particular grammatical contexts, an expansion is required: In analyzing ASL discourse, it seems there are specific applications of language use and language phrasing in ASL that do not occur in spoken English. These unique applications are what I call EXPANSION. Although the word EXPANSION has many meanings in English, I chose this term because it is descriptive of (Lawrence 1995: 207, italics ours) what happens in native ASL signing.
When Lawrence specifies that expansions take place in “native ASL signing” she is suggesting that second language signers (i.e., many interpreters) are not using ASL structures in a way that first language ASL signers are. Lawrence sees expansions as a requirement of ASL in opposition to English, which suggests that, even though she speaks only about discourse and not grammar, they are grammatical constructions that must appear in ASL if language use is to be native-like: Isolating the features of EXPANSION may, in fact, give the ASL student and ultimately, the student of interpretation, the facility to produce a more natural form of ASL discourse that is not only more accurate, but also allows the Deaf consumer the ease of understanding the message in a more native-like form. (Lawrence 1995: 213)
The suggestion that expansions are required by the grammar of ASL is evident in Humphrey and Alcorn (2001), who claim that ASL demands a more explicit level of information coding than does English. They propose that: …an English presentation or exchange of information tends to deal with the specific issue at hand, avoiding a great deal of elaboration or detail. Thus unless the
. Upper case letters separated by hyphens indicates that the word is fingerspelled. . There is a lexical word in ASL for ‘allergy’ that might well suffice.
338 Terry Janzen and Barbara Shaffer
speaker is engaged in story-telling, acting, or some special form of discourse, it is likely s/he will not provide the rich variety of detailed and descriptive information required by ASL. (Humphrey and Alcorn 2001: 9.10; italics ours)
Aside from Humphrey and Alcorn’s disregard for different linguistic features associated with various genres of English discourse (over and above the so-called “special forms” they mention) and the Gricean maxim of “quantity”, that is, “Make your contribution as informative as is required (for the current purposes of the exchange)” (Grice 1975: 45), their claim is that English and ASL are fundamentally different in regard to expansion requirements. They do concede that “English sometimes uses similar techniques to convey information” (Humphrey and Alcorn 2001: 9:13) but include no elaboration of how this might take place. Instead their claim is that expansion “devices” are characteristic of ASL structure, and therefore must be inserted into ASL as the target language during English to ASL interpretation. We propose instead that if general principles of information exchange along with more specific principles of discourse structure in each language are understood, interpreters will better know how and when to contextualize propositions in either direction so that information is transferred cross languages accurately without compromising the intent. It may be, however, that these expansion techniques are put into practice because the interpreter has made certain assumptions on the part of the discourse participants she is interpreting for, assumptions which are perhaps unjustified. Conversely, she may have no pragmatic motivation at all, but is merely attempting to compensate for her own language and discourse inadequacies. Whatever the reason, it appears that ASL-English interpreters have come to believe that “expansions” are explicit grammatical features of ASL, meaning that in specific grammatical contexts, an expansion is the correct grammatical way to make an ASL sentence. This in fact is rarely, if ever, the case and yet the practice has become an expectation of proper grammar usage without regard to the complex pragmatic interaction of the primary participants in their discourse context. Instead, the interpreter is deciding what information should be filled in based on a prescribed notion of grammar rather than on cues from the setting, disregarding entirely the intersubjective negotiation that the principle interlocutors are attempting to engage in. As noted, both Lawrence (1995) and Humphrey and Alcorn (2001) state that the level of detail given in ASL discourse is much higher than that found in English discourse. As an example of incorporating expansion techniques in ASL, Humphrey and Alcorn cite an ASL narrative from a Deaf signer in which the narrator’s mother had planned to visit but had to cancel. Humphrey and Alcorn give the structure of the ASL story in a rough English approximation, but attempt
Intersubjectivity in interpreted interactions 339
to retain the level of detail contained in the original ASL version. Part of this narrative is as follows: After weeding the vegetables, she stood up and evidently tripped on a garden rake that had been left prongs-up on the sidewalk and broke her hip. (role-shift: standing up, not seeing rake prongs – tripping, sailing through the air with a panicked look on the face, landing on right hip with a great look of pain). (Humphrey and Alcorn 2001: 9.21)
The ASL narrative is represented by a total of 291 English words in three paragraphs that include phrases and sentences that Humphrey and Alcorn analyze as examples of various expansions. They then suggest an appropriate English interpretation of the narrative that is 66 words in length and includes none of the detail of the original ASL, just the main point along with only those phrases directly related to this point. They make two comments of interest. First, regarding the level of detail, “ASL … typically requires this degree of information [in the orignal version] in order to be linguistically correct” (2001: 9.22; italics theirs) and second, in contrast to the lack of detail in the English interpretation that “the absence of copious contextual information prevents a Deaf listener [addressee or “watcher”] from comprehending the point being communicated” (2001: 9.23). What do these comments represent? First, this example and the authors’ discussion of it illustrate the claim that expansions are part of the required grammar of ASL, implying that they must be included or the passage will be incomprehensible to the addressee. Second, if the truncated English version is recommended as the appropriate interpretation from ASL into English, what does the interpreter determine to be extraneous in the original text? The implication is that much of the detail in the ASL version is not appropriate in the “equivalent” English text. Third, if the two passages are considered as functionally equivalent, then in the case of interpreting from English to ASL, how would the interpreter decide on the specific details to add into the target text, especially given that she was not present to witness the aunt’s accident. If such details were added to the ASL target text (presumably because the grammar “requires” it), wouldn’t the addressee sense that the interpreter really did not know if the facts expressed by the details were actually true? Fourth, the discrepancy in the source and target texts in this example contradicts a statement that Humphrey and Alcorn have made, noted . By “role-shift” the authors mean that the signer has taken the perspective or “role” of the character and presents the action from that point of view. . This will be taken up once again in Section 4, where we address the appropriateness of the interpreter framing information from her own perspective, which may not reflect the level of shared information between the event participants.
340 Terry Janzen and Barbara Shaffer
above in Section 1, that in the genre of story-telling (or “acting” or other “special” kinds of discourse), English speakers are in fact expected to use a similar degree of detail inclusion. The observations we have made above show that some views held regarding the grammars of ASL and English, and what is necessary to interpret between the two, focus on requirements of form and not on negotiated aspects of interactions among speakers and signers. Such a “how-to” approach to interpretation has a solid history in the field; in the next section we discuss an earlier item-by-item approach that equally disregards the notion of shared knowledge.
2.1 An early perspective on the lexicon in ASL-English interpretation It is evident that even as the field of signed language interpreting began to formalize in the 1960s there were perceived non-equivalencies in the lexicons of English and ASL. Authors such as Quigley and Youngs (1965) attempted to compensate for this by suggesting interpretations for lists of English words, claiming that these English words had no ASL equivalent lexical item and that fingerspelling the English word was an option reserved for the “above average” Deaf ASL signer. For “low-verbal deaf persons” (1965: 37), an alternate approach was needed, referred to as paraphrasing. According to Quigley and Youngs, “[i]n interpreting for lowverbal deaf persons, the interpreter paraphrases, rephrases, defines, simplifies, and attempts to give the literal sense or conceptual essence of idiomatic expressions… The use of analogy, parallelism, and examples are helpful in this type of interpreting” (1965: 39). Some examples given are relatively simple, and may very well serve as good paraphrases in the target. The practice of paraphrasing in this manner is a widespread strategy in any interpreting when the target word is not known by the interpreter or doesn’t exist (e.g., Gile 1995: 198 lists “explaining or paraphrasing” as a reconstruction tactic). As English to ASL examples in the medical arena, Quigley and Youngs suggest the English word contagious be conveyed in ASL as EASY SPREAD SICKNESS (1965: 72) and far sighted as CAN SEE FAR, CAN’T SEE NEAR (1965: 73). These in fact may be reasonable paraphrases that do not stray significantly from the original (although note that each constitutes a . While the term “low-verbal” has often been used to mean persons with minimal language altogether, Quigley and Youngs define it in their context as follows: “In as much as there are relatively few deaf persons who have absolutely no verbal ability, the term low verbal will be used …to imply that mastery of the English language is either markedly deficient or totally absent on a functional level in ordinary conversation. These low-verbal deaf persons cannot understand or make themselves understood without the services of an interpreter in dealing with hearing people who are not fluent in the language of signs” (1965: 37).
Intersubjectivity in interpreted interactions 341
differently structured target phrasing which may or may not need to be taken into account in the overall text). Other examples stretch the meaning of paraphrasing. Quigley and Youngs’ ASL suggestion for cataract is THIN WHITE INSIDE EYE, COVER PART USE TO SEE, SLOWLY GET WORSE, CAN’T SEE, MUST REMOVE (1965: 72). Dope becomes INJECTION, BECOME HABIT, CAN’T STOP, DAMAGE BODY, MIND BECOMES CRAZY (1965: 73). Well beyond paraphrases, these examples explain, then take the condition to a concluding state, and finally dispense some advice or opinion. A doctor may be using the term in a sentence like ‘you have early signs of a cataract…’ which does not (yet) imply the inevitability of impending surgery. Equally, not all uses of the word dope entail addiction, damage, and insanity. Even though the original intent was undoubtedly meant as helpful for interpreting students, the formulaic nature of these translations assumed a single way to sign the item no matter what the situation or who the recipient was, disregarding the intersubjective principles of potentially shared context and shared linguistic knowledge or awareness (Linell 1995). Further problematic with Quigley and Youngs’ approach to transfer between the lexicons of English and ASL is that it assumes the absence of certain (and numerous) lexical items in ASL, when in fact what is not being considered is the possibility that English and ASL do not construct words using identical morphological and lexical processes (Janzen 2005b). The current practices of “expanding” and “compressing” (Finton and Smith 2004) have not resolved these issues because they are equally formulaic and once again do not take situational factors into account. Interpreters fall prey to formulaic work when they make assumptions about what the languages “are like” and what one group of language users will or will not be able to understand based on these assumptions. Thus, that ASL requires expansions and English requires compressions dictates the form of the target text without regard for the participants and their linguistic and extralinguistic knowledge stores and communicative goals.
3.
Pragmatics and discourse structure
Clearly, the interpreter’s objective is to construct a target message that is grammatical, and thus she must know and use the grammar that is available in the target language. In one sense, the interpreter is limited by the grammar of the target language. Gile (1995) refers to this as “Linguistically Induced Information” which in the source text, Gile comments, is of little importance to the interpreter because she cannot likely transfer the grammar of the source language to the target. It is more critical in terms of the target language, however, because the interpreter
342 Terry Janzen and Barbara Shaffer
c annot avoid the target language grammar when constructing her target text. This is highly problematic when the grammars of the source and target languages differ substantially, which we believe is the case for English and ASL. However, the interpreter is also limited by her knowledge of the grammars – if this knowledge is incomplete, her ability to choose from a set of grammatical possibilities is limited. But discourse among interlocutors is more than just grammar. Discourse participants manipulate their repertoire of structural elements to achieve optimum communicativeness in interactional contexts. To do this, interlocutors construct meaning that depends extensively on situational factors as well as by linking what is communicated at a given moment to knowledge that is carried forward from past events and experiences. “Communicative competence can be defined in interactional terms as ‘the knowledge of linguistic and related communicative conventions that speakers must have to create and sustain conversational cooperation,’ and thus involves both grammar and contextualization” (Gumperz 1982: 209). Thus beyond grammatical knowledge, discourse is always situated pragmatically, meaning that utterances must always be contextualized to some extent. Gumperz suggests that discourse participants constantly include contextualization “cues” to aid in building and maintaining communicative cooperation: “Roughly speaking, a contextualization cue is any feature of linguistic form that contributes to the signaling of contextual presuppositions. Such cues may have a number of such linguistic realizations depending on the historically given linguistic repertoire of the participants” (Gumperz 1982: 131). Fox (1994) points out that linguistic expressions underspecify meaning, so that a lexical item or construction only truly suggests a meaning when it linked to some specific context, and that meaning building is by default cooperative (cf. Seleskovitch 1976 on the “overspecification” of dictionary entries but the “underspecification” of words in context – words are tokens of meaning that the recipient is left to construct, in the final round). Itkonen’s (this volume) view on shared meaning is consistent with those of Fox and Seleskovitch in that while mental images (of events, for example) are subjective and vary from person to person, meanings are intersubjective because they are determined by the immediate cooperative exchange. Contextualization cues accompany propositions (i.e., the lexical material and associated grammar) so as to allow interlocutors to retrieve contextualizing information and thereby assess the communicative intent of utterances (Gumperz 1995). In fact, Verhagen (2005: 10) suggests that the function of discourse is not just to be informative, but to engage in “cognitive coordination”, that is, to influence another person’s thoughts, attitudes and behaviours, and for
Intersubjectivity in interpreted interactions 343
the addressee to ascertain what kind of influence the speaker might intend and whether or not to go along with it. Meaning, then, is not something objective found in the words and constructions of language, to be discovered and conveyed, but is co-constructed between discourse participants in an immediate social context (Wilcox and Shaffer 2005). This is the case for any interlocutors’ communication attempts (Coates 1995), and it is no less true for interpreted discourse. Thus, as Wilcox and Shaffer state, the interpreter cannot convey someone’s meaning, but must co-construct meaning along with the receiver in a specific situation for some specific purpose. This is the essence of “dialogic” discourse (Linell 1997; Wadensjö 1998), in fact, and in the case of interpreted discourse the interpreter and the receiver are co-participants in the co-construction of meaning. The significance of this is that a co-constructed target text will not be successful if the interpreter’s output is formulaic, that is, based on producing grammar without regard to the co-constructor’s contextualized participation in terms of their situational context, past knowledge and experience, and relationship with the originator of the source text. The interpreter must balance grammar and contextualization cues in co-constructing a meaningbased text. This is reflected in Enkvist’s (1991) position: Texts, so we may assume, are ordered, linearized, in ways which somehow optimize the order in which a speaker or writer wishes to eliminate the receptor’s uncertainties and expose to him the states of affairs that define the particular scenario or text world he wants to communicate. But as linearization patterns must be affected both by the syntactic constraints and by the cultural and rhetorical traditions associated with the specific language and text type which are involved, a given linearization pattern cannot always be transferred unchanged from source to target language. This lack of direct transfer is, then, caused partly by syntax, but also partly by cultural differences and rhetorical traditions linked (Enkvist 1991: 6) to types of discourse and texts.
Enkvist identifies three elements of texts that affect the transfer of material or “linearization patterns” from source to target: syntax (or grammar), cultural differences and rhetorical traditions. Interpreters may expect discrepancies between the source and target languages in each of these three areas, and thus must be prepared to encounter mismatches while interpreting. Regarding dissimilarities in grammar, ASL-English interpreters are faced with two quite distinctly structured languages. For example, a significant but fundamental difference is word order. English is a fixed SVO (subject-verb-object)
344 Terry Janzen and Barbara Shaffer
word-order language (Comrie 1981) whereas ASL has variable word order in the sense that more ordering options are open to the ASL signer. When the verb is final (e.g., SOV), there is a greater tendency for the verb to be morphologically complex (Liddell 2003). ASL verb structures that are highly complex morphologically may appear as single words that comprise a whole sentence with no overt subject or object nominals (Brentari 2002). As well, third person pronouns in ASL are not gender specific whereas they are in English (Liddell 2003). The use of space is prominent in the articulation of ASL such that numerous relationships between subject, object and verbal action necessarily include a spatial dimension, as do aspect marking and temporal relations (see, for example, Emmorey 2002), whereas these spatial features are not overt in spoken English. These and numerous additional differences between the grammars must be within the interpreter’s working repertoire, but she must also be able to distinguish between what is required in the grammar and what is a discourse strategy, given that discourse strategies involve some grammatical option chosen among several along with contextualization cues as appropriate. However, Enkvist also addresses cultural differences and rhetorical traditions, each of which can also motivate the interpreter’s linguistic choices. Distinct communities of language users exhibit cultural attributes that correspond to their history, traditions and experiences. Two things about culture are not quite so clear for interpreters, however. First, it is not always evident what the interpreter should do with a cultural difference. Should the interpreter compensate for the difference as Mindess (1999) maintains, or facilitate an “exchange” of cultures, that is, allow the interlocutors to experience each other’s culture (Simon 1995)? Second, it is often not obvious how cultural elements are reflected in specific linguistic constructions. How does the interpreter know that a particular linguistic or grammatical choice fits the supposed cultural experience of the recipient? Cultural differences may well be mediated by certain types of contextualization. This is not specific to one language or another, although the linguistic cues themselves will likely be language-specific. Rhetorical traditions refer to the types of text structure associated with particular genres that are frequently used within a community of speakers or signers. An example might be the typical way in which a visitor is introduced to members of the community. In the Deaf community a new Deaf person is most likely introduced by telling about her school history and by mentioning acquaintances who may be known to both people whereas a hearing person being . Liddell (2003), however, concludes that the basic order in ASL is SVO, with a leftward extra-clausal topic slot and a rightward post-clausal subject pronoun slot. Thus variability is more complex than simply reordering S, V, and O.
Intersubjectivity in interpreted interactions 345
introduced to others in a hearing community may be identified by her profession (Mindess 1999). Another example concerns the overall discourse structure sometimes attributed to ASL and English as the so-called “diamond” shape of ASL discourse (introduction of the topic or point of the discourse-elaborationconcluding reiteration of the topic/point) versus a “funnel” shape of discourse for English (initial general comments-narrowing to the eventual point of the discourse) discussed by Christie et al. (1999). For the interpreter, this example presents difficulties on a number of levels. Its application is often overgeneralized even though it is not yet clear how viable an analysis it is nor how broadly it can be applied to ASL and English genres. Further, these discourse structure-types may each appear in the other language, for example the diamond-shaped topic/ point introduction-elaboration-concluding repetition of the point is an excellent teaching strategy used by many English speaking teachers. In any case, deciding that one or the other discourse type is appropriate in a particular register says nothing about specific linguistic or grammatical structures that should be used; rhetorical traditions have more to do with framing information (Gile 1995) than with particular linguistic constructions.10 This is not to say that discourse strategies are never associated with particular lexical items or grammatical constructions. Topic negotiation in ASL, for example, frequently begins with items such as KNOW.THAT,11 KNOW, or UNDERSTAND along with facial topic marking (Janzen 1998), but note that here too the signer has lexical options whose choice depends on pragmatic factors within the immediate discourse situation. In sum, Lawrence (1995) and Humphrey and Alcorn (2001) discuss expansion elements as required at the level of grammar when in fact they are best understood as discourse strategies. Further, even though these authors attribute expansions to the grammar of ASL, they are not language specific strategies, but to one degree or another are all used among speakers or signers no matter what the language might be. This can equally be applied to so-called “compressions” (Finton and Smith 2004), which recently have been proposed as necessary when interpreting from ASL into English. It is true that certain strategies are favored by
10. Gallagher and Hutto (this volume) suggest that speakers have a lifetime of narrative practice and know a lot about what another speaker is taking about because of cultural and experiential knowledge that narrative frames provide. Interpreters need this knowledge in two languages, but may be missing the early practice Gallagher and Hutto speak about if they learned one of their working languages later as adults. 11. Some signs in ASL are commonly glossed using more than one English word. When this is the case we use upper-case words separated by periods.
346 Terry Janzen and Barbara Shaffer
one community of language users or another, but this in no way can mean that these are exclusive properties of a particular language itself and not others.
4.
Assumptions about shared knowledge
As we have stated, it appears that interpreters are making the assumption that expansions are an a priori feature of ASL grammar. This leads to the further assumption that expansions are an obligatory component of ASL signers’ discourse. As a result, they are being used as a kind of default without regard to other potential interpreting strategies, nor to the discourse pragmatics surrounding that particular interpreted interaction. It is not always obvious what strategy the interpreter should be using, but Janzen (2005a) makes it clear that strategizing constitutes a contribution that the interpreter makes to the resulting text. Leeson (2005) claims that additions to the interpreter’s target text have at times been wrongly labeled as “miscues” (Cokely 1992) when in fact the interpreter might from time to time believe that the text will be more clear if a strategic addition is made. Leeson makes the point, however, that this strategy is just one of many the interpreter must be skilled at, and when used, it must be a conscious decision that affects the resulting text but, the interpreter knows, is in addition to the original text. In other words the “expansion” belongs to the interpreter, not to the text.
4.1 Co-constructing discourse Coherence in communication depends on shared knowledge on numerous levels, for example shared experience, worldview, and linguistic knowledge. On these bases, interlocutors exchange clues to their intended meanings in the way that they choose linguistic items and constructions to represent pieces of meaning. It is worth reiterating here that meaning cannot be transferred directly, but must be constructed and re-constructed by the speaker or signer and the recipient, respectively (Wilcox and Shaffer 2005). In a sustained sense, this reflects Brinck’s (this volume) intersubjective notion of “interactivity” whereby agents act so as to directly affect each other, resulting in turn-taking. Interlocutors frame their discourse in terms of assumptions about shared knowledge, introducing pieces of information in such a way as to maintain an appropriate level of coherence in the interaction. According to Chafe (1994) information will either be active, new, or semiactive. Shared information is most obviously that which is already active in the discourse, in other words, it is information that is within the interlocutors’
Intersubjectivity in interpreted interactions 347
c onsciousness at the present moment in their interaction. New information is information that is newly activated at some point in the conversation. Semi-active information, however, is information that has been introduced at a previous point, thus has been active at some previous time but is no longer the immediate focus of the discourse. Shared information may be semi-active too in that interlocutors approach their discourse with a certain amount of shared context, whether or not these contextual aspects are ever expressed. Chafe suggests that how the speaker (or signer) encodes each type of information structurally will differ, but that this determination critically depends on the originator’s assumptions about the status of the information in the addressee’s consciousness at the time. Regarding semi-active information, the further back in time an item was mentioned, the less likely it will be active in the consciousness of the addressee. Importantly, this suggests that assumptions about the activeness of an item may not always be correct, considering that each interlocutor will not likely be thinking the exact same thing about pieces of information at all times. The speaker may be keeping something more active in her consciousness while the addressee has not retained it as a focus, for example. This is congruent with Givón’s (1995) discussion of referential distance (RD), in which he measures RD in terms of the number of intervening clauses between a past mention of an item and the most current mention. The greater the number of intervening clauses that do not include a mention, the more likely it is that the item does not retain its active (or even semi-active) status for the addressee. In co-constructing their discourse, interlocutors negotiate the relevance of topics, facts and observations about these topics, evaluations of statements, and numerous additional subjective elements (Scheibman 2002). Each interlocutor brings her view of the discourse event to that event, which includes assumptions about her addressee’s knowledge store, and she shapes her contributions based on ever-changing perceptions of the event as it progresses. Each interlocutor’s perceptions, contributions and reactions constitute interaction that is systemically intersubjective. When discourse is mediated by an interpreter there are several effects. Most importantly, even though the interpreter has often been conceived of as an objective, non-participating communication facilitator, when viewed as a co-participant co-constructing meaning, as Wadensjö (1998), Roy (2000) and Wilcox and Shaffer (2005) suggest, it is undeniable that the interpreter also contributes to the intersubjective relationship in the interchange. Second, the interpreter’s construction of meaning is complicated by the fact that she does not share the same context as the primary discourse participants given that she most likely spends the majority of her time outside the other participants’ environment(s). Third, the interpreter brings her own conceptualized view of the event to the discourse. And fourth, while the interpreter builds her own discourse relationships
348 Terry Janzen and Barbara Shaffer
with each of the other participants on-line, she must work to make sense of what she perceives to be the intersubjective relationship between those participants themselves, and further, she must attempt to linguistically represent each of their inputs into their discussion as if it were still their own. All of this has the potential to impose shared knowledge in an artificial way. The interpreter necessarily filters an input utterance through her own understanding of what it must mean, and chooses her target text coding based on what she understands the target recipient to know. The question remains, however, as to whether or not she can instead formulate her text based on what she ascertains the speaker to be assuming about what the target recipient knows. As a regular discourse strategy, interlocutors negotiate meaning, whereas it seems to be the case that interpreters often expect to be able to simply “take” a speaker’s message, reformulate it in the target language, and deliver it perfectly. Frishberg (1990) reflects this perspective by saying that accuracy means “rendering a message from the sender’s language into the addressee’s language” and “giving the receiver the complete message” (1990: 65). Hatim and Mason (1990) argue that the interpreter must necessarily be a negotiator of meaning too, which if true, implies a particular stance in the interaction: she must appeal to shared knowledge. As a part of negotiation, the interpreter should be able to contextualize as needed, listed as an interpreter’s “coping strategy” in Gile (1995). Contextualization is necessary to make a negotiated message coherent for the recipient, but it may be noted that contextualizing can occur in two phases of an interpreted discourse, that is, either the original speaker may contextualize her statements as a reflection of her beliefs about the addressee’s knowledge, or the interpreter may contextualize based on her own beliefs about the addressee. When the latter occurs, the target text shifts away from the originator’s actual message and becomes more subjective on the part of the interpreter. It should also be noted that not contextualizing is also a discourse strategy. If the speaker does not contextualize, she still is considered to be negotiating, and determining that no contextualization is needed. What then if the interpreter chooses to contextualize? First and foremost, the interpreter must be cognizant of what stance the original speaker has taken as a meaning negotiator, that she is now choosing a different route, and that the effect of this route on the addressee may be different from what the speaker intended. As the interpreter thus negotiates by contextualizing, the recipient will respond based on the interpreter’s framing, not necessarily based on the originator’s framing. Returning to the notion of “expansions” as required by ASL grammar, we see that they might occur because of the assumption that there is no other way to frame an idea rather than because of an assessment of discourse appropriateness and careful consideration of the originator’s purposeful choices. In other words,
Intersubjectivity in interpreted interactions 349
the interpreter expands without regard to the source text or the target recipient’s needs. Because the direction of interpretation that expansions tend to occur in is from English into ASL, it is the Deaf recipient who receives the expansion. Stratiy (2005), who herself is an ASL-as-first-language user, claims that this practice disregards the recipient in making the assumption that she does not share some context with her interlocutor, without any attempt at negotiation. Wadensjö (1998) documents similar examples in her study of Russian-Swedish interpreters in medical settings. Regarding an interpreter who has “over-explained” what a source speaker must have meant, Wadensjö says that there can be a “tendency to underestimate the patient’s ability to understand (which is sometimes considered patronizing)” (1998: 225). The interpreter has not understood the intersubjective dynamic between the speaker and addressee, and obstructs dynamic sharedness in her interpreted text. There are several effects here. First, by the interpreter believing she is required to “expand” at some point in the text, she sends a message to her recipients that they must be expanded to. Stratiy suggests that this is based on an assumption that the Deaf recipient will not be able to understand an “un-expanded” structure, which views the Deaf person as deficient or lacking. Stratiy questions why interpreters do not make the opposite assumption, that the recipient might in fact understand, but if not, will readily negotiate that point. Second, the interpreter can often block the discourse participants from their own meaning negotiation, during which they would otherwise learn about each other and what is shared and not shared between them. Finally, we would like to suggest that interpreters may use expansions at times when their own knowledge of lexicon or grammar fails them. Paraphrasing is a well-known and often used interpreter strategy (Gile 1995; Leeson 2005) that allows the interpreter to accommodate lexical gaps in the target language or lapses in recall; the problem with “expansions” as grammar is that the interpreter believes that the expansion is the correct, and perhaps only, grammatical structure and does not move beyond it to learn new lexical items or grammatical phrasing options.
5.
“Expansions”, contextualizing, and grammar
To reiterate our view on discourse negotiation in both dialogic discourse and interpretation, we take the stance that contextualizing is a cooperative principle that aids in the co-construction of shared meaning. To accomplish this, interlocutors necessarily take advantage of existing grammar in the language they are using – grammar both facilitates and constrains what interlocutors can do:
350 Terry Janzen and Barbara Shaffer
Selection of lexical items and grammatical decision making may be more difficult in one language than another because of differences in the variety of possible choices and in the flexibility of linguistic rules: a very wide vocabulary as opposed to a more restricted one, flexible or rigid lexical usage, or the number of possible escape routes in sentence structuring in case the source language leads to an unexpected segment and forces one to reconsider one’s options.” (Gile 1995:234–235)
ASL, like any human language, has numerous grammatical features that are useful when interlocutors contextualize as a discourse-level meaning negotiation strategy. Topic-comment constructions, identified as a basic sentence type (Ingram 1978; McIntire 1982; Janzen 1998), are sentences in which a topic constituent appears first followed by a comment. The topic provides some grounding information that situates a comment for the addressee. It is background information the signer assumes the addressee will recognize as the ground from which to “view” the comment. The comment constituent contains the actual point of the utterance or the new information. In these constructions, the topic constituent provides a kind of context for the main point found in the comment, and is chosen because the signer believes that the addressee will recognize it as something identifiable, that is, it is shared information. In choosing a topic as grounding information, the signer may not always be certain that it is in fact identifiable, thus a topic might be negotiated. Based on cues from the addressee that an intended topic is not equally shared, the signer will back up and fill in some contextual details to bring the addressee up to speed. This accomplished (and acknowledged as such by the addressee), the signer may reiterate the topic phrase she began with – now that it has shared status, it works as a grammatical topic phrase, and the discourse moves on to some comment(s) about that topic. Thus, in ASL, topic phrases represent an intersection of grammar at the sentence level and contextualization at the discourse organization level, and so may be counted as an overt marker of intersubjectivity. In English, interlocutors also choose constructions to negotiate in similar ways. Embedded relative clauses often serve this purpose, for example. In (3) the relative clause (in italics) contextualizes the guy – it supplies more information that the speaker believes to be identifiable to the addressee and which will assist in specifying who the guy is. (3) The guy who we thought we were meeting couldn’t come.
Another example, also from English, is the use of that, as in (4): (4) That plumber I called said I didn’t need a new water heater yet.
Intersubjectivity in interpreted interactions 351
That is used in (4) as opposed to the articles the or a as a signal to the addressee that the noun that follows is unquestionably shared, an intertextual reference (Hatim and Mason 1990) to a previously shared discourse event. The interpreter’s task is to recognize and make sense of contextualizing sequences in the utterances she is attempting to interpret, and relay that intertextual sense by choosing constructions that in essence do the same work, even if the construction in the target language is not of the same category as that chosen in the source. A problem that this presents is that the previous context is not necessarily shared by that interpreter, meaning that the same intersubjective motivation in choosing structures is not a part of the interpreter’s cognitive state (Shaffer and Jansen 2002). Even though some of these linguistic features may fit Lawrence’s rubric of so-called expansions, it is not the case that their use should be viewed a priori as required structurally. Rather, grammatical elements and constructions are chosen that fit the specific situational parameters at the moment, and most importantly, best facilitate the communicative goals of the speaker (Givón 2005). A relative clause is not chosen by a speaker if it is deemed unnecessary; a topic in ASL is not negotiated if there is no perceived reason to do so. Thus, even though the interpreter must use the grammatical items available in a given language and, as mentioned, is constrained by the grammar in that regard, contextualizing is clearly a discourse-level interplay in which the interpreter is an equally subjective player.
6.
Conclusions
In the intersubjective context of interpreted discourse, the interpreter’s participation in the negotiation of meaning includes an awareness of contextualizing on the part of discourse speakers and signers, and the intentional use of contextualization in her own target message formulation. We have noted that the interlocutors’ use of contextualization is purposeful and is part of their message framing. Accurate interpretation must take into account such framing choices along with message content. But secondly, the interpreter must understand that her own contextualizing choices may shape the target text in ways that are decidedly different from the original text, and the source speaker is likely not aware of these differences. In example (2) in Section 2 above regarding the interpretation of the English word allergy, the interpreter’s expanded rendition has two important consequences. First, it now enters the domain of the target recipient’s discourse context, even though the expanded detail was never a part of the source speaker’s negotiation. Second, the interpreter is developing her own intersubjective stance
352 Terry Janzen and Barbara Shaffer
with the recipient, which is not equally shared with the source speaker. Wadensjö (1998: 234) says that some interpreters’ “excessive willingness to explain for one of the parties, thus taking over the responsibility of the other party, may also turn out to be a ‘trouble source’ obstructing sharedness between the primary interlocutors” (italics added). In the case of ‘allergy’, the interpreter expands, or over-specifies, by adding details that were not in the source message (whether unconsciously or by design). The choice shows the interpreter’s construal of (1) what the recipient would not understand, and (2) what ‘allergy’ must mean. Unfortunately, by listing a particular set of sources of allergic reactions the interpreter has also defined the set – what if the doctor, when inquiring about allergies, had in mind an allergy to latex? Nonetheless, in interpretation generally, there are numerous points in the process where contextualizing is a beneficial interpreting strategy, and in fact, it is a regular feature of discourse that is particularly useful in interpretation. Contextualization occurs pervasively in discourse, and there is no reason to expect it not to occur in interpretation as well. We stress, however, that contextualization must be a part of the interpreter’s conscious decision-making process. The interpreter holds a unique position because it is her task to reformulate the intentions of one speaker into the linguistic terms and discourse framing of an entirely different language, and yet she must work to allow interlocutors to build their own intersubjective relationship. To do this, she makes myriad subjective linguistic and framing choices, including whether to contextualize or not and, depending on the nature of these choices, may either promote or obstruct the intentional interaction between those she is interpreting for. In this view, the interpreter never takes a neutral stance – it simply is not available to her. Regarding the intersubjective nature of discourse, interpreters have a dual role to fill. On one hand, as cross-language mediators, they form a discourse relationship with the speaker of language A and another with the speaker of language B. But on the other hand, they attempt to reconstruct the very subjective messages between the speakers of languages A and B without significantly interfering with the intersubjective relationship the two interlocutors are building with each other. The interpreter’s knowledge store will not match perfectly those of the interlocutors she is interpreting for, so she must be cognizant of the linguistic cues she encounters from each to help her in the mutual construction of meaning so that even though the two interlocutors do not share the same linguistic system, they are communicating meaningfully with each other.
Intersubjectivity in interpreted interactions 353
Acknowledgements First and foremost we wish to thank the late Donna Korpiniski, interpreter extraordinaire, for her forward thinking and cognizant reflection on her work, and for many fruitful discussions that have helped us formulate our own ideas presented here. Thanks also to the editors of this volume for their support, suggestions and thoughtful reviews.
References Brentari, D. 2002. “Modality differences in sign language phonology and morphophonemics.” In Modality and Structure in Signed and Spoken Languages, R.P. Meier, K. Cormier and D. Quinto-Pozos (eds.), 35–64. Cambridge: Cambridge University Press. Brinck, I. this volume. “The role of intersubjectivity in the development of intentional communication.” Chafe, W. 1994. Discourse, Consciousness, and Time: The Flow and Displacement of Conscious Experience in Speaking and Writing. Chicago: The University of Chicago Press. Christie, K., Wilkins, D.M., Hicks McDonald, B. and Neuroth-Gimbrone, C. 1999. “get-tothe-point: Academic bilingualism and discourse in American Sign Language and written English.” In Storytelling and Conversation: Discourse in Deaf Communities, E. Winston (ed.), 162–189. Washington, DC: Gallaudet University Press. Coates, J. 1995. “The negotiation of coherence in face-to-face interaction: Some examples from the extreme bounds.” In Coherence in Spontaneous Text, M.A. Gernsbacher and T. Givón (eds.), 41–58. Amsterdam/Philadelphia: John Benjamins. Cokely, D. 1992. Interpretation: A Sociolinguistic Model. Burtonsville, MD: Linstok Press. Comrie, B. 1981. Language Universals and Linguistic Typology. Chicago: University of Chicago Press. Du Bois, J.W. 2003. “Discourse and grammar.” In The New Psychology of Language: Cognitive and Functional Approaches to Language Structure, Volume 2, M. Tomasello (ed.), 47–87. Mahwah, N.J.: Erlbaum. Emmorey, K. 2002. Language, Cognition, and the Brain: Insights from Sign Language Research. Mahwah, NJ: Lawrence Erlbaum. Enkvist, N.E. 1991. “Discourse type, text type, and cross-cultural rhetoric.” In Empirical Research in Translation and Intercultural Studies, S. Tirkkonen-Condit (ed.), 5–16. Tübingen: Gunter Narr Verlag. Finton, L. and Smith, R.T. 2004. “The natives are restless: Using compression strategies to deliver linguistically appropriate ASL to English interpretation.” In CIT: Still Shining After 25 Years, Proceedings of the 15th National Convention, Conference of Interpreter Trainers, E.M. Maroney (ed.), 125–143. USA: CIT. Fox, B.A. 1994. “Contextualization, indexicality, and the distributed nature of grammar.” Language Sciences 16 (1): 1–37. Frishberg, N. 1990. Interpreting: An Introduction, Revised Edition. Silver Spring, MD: RID Publications.
354 Terry Janzen and Barbara Shaffer
Gallagher, S. and Hutto, D.D. this volume. “Understanding others through primary interaction and narrative practice.” Gile, D. 1995. Basic Concepts and Models for Interpreter and Translator Training. Amsterdam/ Philadelphia: John Benjamins. Givón, T. 1984. Syntax: A Functional-Typological Introduction, Volume 1. Amsterdam/Philadelphia: John Benjamins. Givón, T. 1995. “Coherence in text vs. coherence in mind.” In Coherence in Spontaneous Text, M.A. Gernsbacher and T. Givón (eds.), 59–115. Amsterdam/Philadelphia: John Benjamins. Givón, T. 2005. Context as Other Minds: The Pragmatics of Sociality, Cognition and Communication. Amsterdam/Philadelphia: John Benjamins. Grice, H.P. 1975. “Logic and conversation.” In Syntax and Semantics 3: Speech Acts, P. Cole and J. Morgan (eds.), 41–58. New York: Academic Press. Gumperz, J.J. 1982. Discourse Strategies. Cambridge: Cambridge University Press. Gumperz, J.J. 1995. “Mutual inferencing in conversation.” In Mutualities in Dialogue, I. Marková, C.F. Graumann and K. Foppa (eds.), 101–123. Cambridge: Cambridge University Press. Hatim, B. and Mason, I. 1990. Discourse and the Translator. London and New York: Longman. Humphrey, J.H. and Alcorn, B.J. 2001. So You Want to Be an Interpreter? An Introduction to Sign Language Interpreting (3rd Ed.). Amarillo, TX: H&H Publishers. Itkonen, E. this volume. “The central role of normativity for language and linguistics.” Ingram, R.M. 1978. “Theme, rheme, topic, and comment in the syntax of American Sign Language.” Sign Language Studies 20: 193–218. Janzen, T. 1998. Topicality in ASL: Information Ordering, Constituent Structure, and the Function of Topic Marking. Ph.D. Dissertation, University of New Mexico, Albuquerque, NM. Janzen, T. 2005a. “Introduction to the theory and practice of signed language interpreting.” In Topics in Signed Language Interpreting: Theory and Practice, T. Janzen (ed.), 3–24. Amsterdam/Philadelphia: John Benjamins. Janzen, T. 2005b. “Interpretation and language use: ASL and English.” In Topics in Signed Language Interpreting: Theory and Practice, T. Janzen (ed.), 69–105. Amsterdam/Philadelphia: John Benjamins. Janzen, T. and Shaffer, B. 2003. “Implicit versus explicit coding across two languages: Mismatches of cognitive domains during interpretation.” Paper presented at the 8th International Cognitive Linguistics Conference. La Rioja, Spain, July 20–25, 2003. Lawrence, S. 1995. “Interpreter discourse: English to ASL expansions.” In Mapping our Course: A Collaborative Venture, Proceedings of the Tenth National Convention, Conference of Interpreter Trainers, E.A. Winston (ed.), 205–214. United States: Conference of Interpreter Trainers. Leeson, L. 2005. “Making the effort in simultaneous interpreting: Some considerations for signed language interpreters.” In Topics in Signed Language Interpretation: Theory and Practice, T. Janzen (ed.), 51–68. Amsterdam/Philadelphia: John Benjamins. Liddell, S.K. 2003. Grammar, Gesture, and Meaning in American Sign Language. Cambridge: Cambridge University Press. Linell, P. 1995. “Troubles with mutualities: Towards a dialogical theory of misunderstanding and miscommunication.” In Mutualities in Dialogue, I. Marková, C.F. Graumann and K. Foppa (eds.), 176–213. Cambridge: Cambridge University Press.
Intersubjectivity in interpreted interactions 355
Linell, P. 1997. “Interpreting as communication.” In Conference Interpreting: Current Trends in Research, Y. Gambier, D. Gile and C. Taylor (eds.), 49–67. Amsterdam/ Philadelphia: John Benjamins. McIntire, M.L. 1982. “Constituent order & location.” Sign Language Studies 37: 345–386. Mindess, A. 1999. Reading Between the Signs: Intercultural Communication for Sign Language Interpreters. Yarmouth, ME: Intercultural Press, Inc. Quigley, S.P. and Youngs, J.P. 1965. Interpreting for Deaf People. Washington, DC: U.S. Department of Health, Education, and Welfare. Roy, C.B. 2000. Interpreting as a Discourse Process. New York: Oxford University Press. Scheibman, J. 2002. Point of View and Grammar: Structural Patterns of Subjectivity in American Conversation. Amsterdam/Philadelphia: John Benjamins. Seleskovitch, D. 1976. Interpretation, A psychological approach to translation. In Translation: Applications and Research, R.W. Brislin (ed.), 92–116. New York: Gardner Press. Simon, S. 1995. “Delivering culture: The task of the translator.” In Perspectives d’avenir en traduction, M. Aubin (ed.), 43–56. Winnipeg: Presses universitaires de Saint-Boniface. Shaffer, B. and Janzen, T. 2002. “Topic marking: What signers know and interpreters don’t.” Paper presented at the Association of Visual Language Interpreters of Canada national conference. Halifax, Nova Scotia, July 22–26, 2002. Shaffer, B. and Janzen, T. 2004. “Contextualization in ASL-English interpretation: A question of grammar or discourse strategy.” Paper presented at the Conceptual Structures, Discourse and Language conference. Edmonton, Alberta, October 8–10, 2004. Stratiy, A. 2005. “Best practices in interpreting: A Deaf community perspective.” In Topics in Signed Language Interpreting: Theory and Practice, T. Janzen (ed.), 231–250. Amsterdam/ Philadelphia: John Benjamins. Susswein, N., and Racine, T.P. this volume. “Sharing mental states: Causal and definitional issues in intersubjectivity.” Verhagen, A. 2005. Constructions of Intersubjectivity: Discourse, Syntax, and Cognition. New York: Oxford University Press. Verhagen, A. this volume. “Intersubjectivity and the achitecture of the language system.” Vygotsky, L.S. 1978. Mind in Society: The Development of Higher Mental Processes. Cambridge, MA: Harvard University Press. Wadensjö, C. 1998. Interpreting as Interaction. London and New York: Longman. Wilcox, S. and Shaffer, B. 2005. “Towards a cognitive model of interpreting.” In Topics in Signed Language Interpreting: Theory and Practice, T. Janzen (ed.), 27–50. Amsterdam/Philadelphia: John Benjamins.
chapter 15
Language and the signifying object From convention to imagination Chris Sinha and Cintia Rodríguez In this chapter we argue that intersubjectivity cannot be grounded in individual mental or representational content. Intersubjectivity, therefore, is not equivalent to “common knowledge”, rather common knowledge (indeed individual knowledge in the true representational sense) depends upon intersubjectivity. Intersubjectivity is the fundamental basis of what Durkheim (and Searle following him) have called “social facts”, which are irreducible to (though they depend upon) biological and individual psychological facts. Intersubjectivity is based upon participation in joint action, and such participation also implicates the shared material, interobjective world. Participatory engagement with signifying objects accompanies and underpins the child’s entry into the symbolic realm of language, and makes possible the development of subjectivity and cultural identity through participation in narrative practices. [It] is always difficult for the psychologist to think of anything ‘existing’ in a culture … We are, alas, wedded to the idea that human reality exists within the limiting (Bruner 1966: 321) boundary of the human skin! The body is our general medium for having a world … Sometimes the meaning aimed at cannot be achieved by the body’s natural means; it must then build itself an instrument, and it projects thereby around itself a (Merleau-Ponty 1962: 146) cultural world. Observation of O. at 2:4;5. Father goes to get him from the car seat. O. keeps his eyes closed, eyelids quivering slightly, with a slight smile. Then he opens his eyes and says “I’m sleeping”, laughing.
358 Chris Sinha and Cintia Rodríguez
1.
Intersubjectivity and the ontology of the social
This chapter has two primary aims. The first is to propose an account of the social nature of the shared mind. The second is to put forward arguments and evidence for regarding material objects, especially artefacts, as a crucial ingredient of intersubjectivity and its development. In this section we advance philosophical and psychological arguments for considering the shared mind to be fundamentally social. We critically assess the methodological individualism guiding the construal by most philosophers and psychologists of the notion of intersubjectivity, and propose an alternative construal of intersubjectivity which sees it as rooted in an ontology of the social, whose methodological counterpart is the Durkheimian concept of the “social fact”. There are two fundamentally different ways to conceive of the “shared mind”. The first of these is predicated upon the foundational status of individual mental or representational content, and in particular of intentional states such as beliefs. An intentional state is characterized, as Searle (1983) puts it, by its directedness to whatever it is about. Intentional states can be about anything at all: object, event or process, real or imaginary, and hence can also be directed at other intentional states, whether those of the subject or that of another subject. I can, for example, wish that I had thought of an idea before someone else did, or I can believe that my next door neighbour believes in fairies, and so on. It is on this basis that “theory theories” of intersubjectivity are constructed: intersubjectivity is considered to be a matter of knowledge (or belief, or intentional states in general). On this account, intersubjectivity is essentially a matter of “common knowledge” in the sense of Lewis (1969) (see also Clark 1996; Itkonen this volume). It is indisputable that normal adult human beings do indeed base much of their social reasoning on representations of other people’s mental states. There are also good arguments for viewing social institutions such as language as objects of common knowledge (Itkonen this volume). There are, however, at least three objections to regarding the “common knowledge” account as sufficiently foundational or inclusive to fully comprehend intersubjectivity. The first objection is logical. The “common knowledge” account is immediately vulnerable to Hume’s “Other Minds” problem: How can I know that the mental content that I ascribe to you is the mental content that you actually have, even excluding cases of mistaken or false ascriptions? In other words, if, for example, I (correctly) think that my neighbour believes in fairies, how do I know that whatever it is that my neighbour believes in is what I think they believe in? To know that, I have to be sure that what my neighbour’s mental content is about is the same as what my neighbour’s mental content under my representation is about. Without this guarantee of referential intersubjectivity, there can be no common knowledge. In other words,
Language and the signifying object 359
the “common knowledge” formulation of intersubjectivity presupposes, without explaining, the intersubjectivity of reference. Another way of putting this is to say that the “common knowledge” formulation presents an unsolved instance of the “Grounding problem”, which requires a logically prior appeal to intersubjectivity for its resolution (Sinha 1999). The second objection to the “common knowledge” account is that intersubjectivity is as much about feeling as knowing. As has frequently been pointed out, intersubjectivity is closely connected to the capacity for empathic identification. However, the affective phenomenology of intersubjectivity extends beyond empathy, in that there are some states of feeling that are constitutively intersubjective, in the sense that they implicate the experience of the feeling of another directed towards oneself. For example, for a couple to be in love it is necessary for each to be in love with the other. The experience of being in love with a lover is quite different from the experience of being unrequitedly or disappointedly in love, for no other reason than the intersubjectively shared nature of the former, as contrasted with the forlornly solitary nature of the latter. Although knowledge of the other’s feelings is important in this, knowledge is not all there is to it, since this intersubjective state also involves commitments and accountabilities – a quintessentially normative dimension that is fundamental to intersubjectivity, selfhood and to the social domain in general (Shotter 1984). The third (related) objection is that the “common knowledge” account of intersubjectivity is disembodied; it does not take into account the “intercorporeal” (Merleau-Ponty 1962) dimension of intersubjectivity, which manifests itself most clearly in the mimetic nature of primary intersubjectivity from the earliest stages of infancy (Zlatev this volume; Trevarthen 1979, 1998; Reddy, Hay, Murray and Trevarthen 1997). It is in the shared experience of corporeally expressed, emotionally rich states of the embodied mind that, as the French developmental psychologist Henri Wallon insisted, we should seek the roots of the instersubjective psyche (Netchine-Grynberg and Netchine 2002; Rodríguez 2006; Wallon 1984 [1925]; Zazzo 1975). We shall develop this argument further below, by demonstrating that intersubjectivity and, by extension, institutions and conventions also find material embodiment in artefactual objects. Despite these briefly-sketched problems afflicting the “common knowledge” account of intersubjectivity, it has been deeply influential, not only in philosophy of mind but also in developmental psychology. This is, firstly, because it accords with the tradition of reducing all realities existing “between” people to theories about what goes on “inside” individual minds; and secondly because it also accords with the mentalist emphasis in classical cognitivism on the primacy of mental representation. The ontogenetic version of this account seeks its explanation for certain representational capacities of the adult mind – the ability to represent
360 Chris Sinha and Cintia Rodríguez
representations – in the autonomous domain of representation itself. Whether “theory of mind” is proposed to be a consequence of meta-representational abilities first applied to the child’s own cognitive processes, or of an ability to “read” the intentions of others, the basic assumption of the paradigm remains in place: “mind” is an autonomous domain, and actions are secondary to the internal and private intentional states which they reveal. We turn now to the second, very different, conception of the “shared mind”, which has its roots in Durkheimian social theory. The object of social theory, for Durkheim, was the domain of social facts, which he described as “a category of facts which present very special characteristics: they consist of manners of acting, thinking, and feeling external to the individual, which are invested with a coercive power by virtue of which they exercise control over him.” (Durkheim 1982 [1895]). Social facts, for Durkheim, are not merely aggregates of the individual representations of them by the subjects that are regulated or “coerced” by the social facts, since for each individual subject the social fact presents itself as a part of an out-there objective reality. The objectivity of social facts consists in the fact they are independent of any single individual’s thoughts or will. As Jones (1986: 61) puts it, “it is precisely this property of resistance to the action of individual wills which characterizes social facts. The most basic rule of all sociological method, Durkheim thus concluded, is to treat social facts as things.” Durkheim’s treatment of social facts thus consists in, first, an ontological proposition, that social facts are irreducible to biological or psychological facts (or structures or processes); coupled with, second, an epistemological and methodological proposition regarding their treatment: as objects of a particular kind, whose determinate nature consists in their “coercion” of conduct. We shall return below to the question of what is implied by the treatment of social facts as “objects”. Durkheim, it must be said, has often been criticized for the breadth and vagueness of his notion of “social fact”. A particularly problematic aspect of his theory is that, in counterposing “social facts” to “individual conscience” (or mind), he sometimes identified the former with “states of the collective conscience”. Some social psychologists (e.g. Moscovici 2000) have followed this direction in constructing a theory of “social representations”, but critics have claimed that Durkheim sympathized with a view of society as a kind of super-organic “collective personality”. Whether or not Durkheim believed in the “collective mind”, such a concept is not only scientifically untenable, it is unnecessary. We propose that a social fact can most simply be defined as something regulating an aspect of conduct which requires the participation (Goodwin and Goodwin 2004) of more than one individual. This “something” may be a codified law, a norm, an institution, a rule in the Wittgensteinian sense, or a canon of interpretation. A natural
Language and the signifying object 361
language, therefore, qualifies as a social fact (indeed, as a social institution, see Itkonen this volume) under this reading of Durkheim’s theory. Let us now compare and (if possible) try to integrate Durkheim’s account with the “common knowledge” account. Social facts (like any other “facts”) are potential objects of intentional states. Individual beliefs about social facts (like any other beliefs) are also potential objects of intentional states (and hence of common knowledge). The efficacy of at least some social facts depends upon their being the objects of common knowledge (Itkonen this volume). However, claimed Durkheim, the social fact itself is not the sum, average or common denominator of all the individual beliefs of all participants. Rather, Durkheim insisted, the social fact is in some sense prior to these individual cognitions. This is at first blush puzzling, since the collectivity of participants (or some authority amongst them) can, in principle, change the social fact (e.g. the rules of a game) just by so deciding. To clarify this issue, let us compare Durkheim’s view with that of Searle (1995: 1–2): “There are things that exist only because we believe them to exist. I am thinking of things like money, property, government, and marriages … [such] Institutional facts are so called because they depend upon human institutions for their existence.” Durkheim, we suggest, would have agreed with the second, but not the first, of these propositions of Searle. How can we render this difference intelligible? The answer, we suggest, is to view social facts as constituting an emergent, normative ontological level existent only in the intersubjective field of joint action regulated by norms and commitments. Intersubjectivity is then essentially a matter of co-participation in joint action structures which, by virtue of their normative regulation, are conventionalized as social and communicative practices. Social practices, and the norms regulating them, can be objects of intentional states, including “common knowledge”, but they are not reducible to the aggregates of such states. This formulation helps us to understand the specific sense in which social facts are methodologically necessarily treated as “objects”; they are instances of the objectification, or reification, as conventional form, of intersubjectivity. As such, they are also potential epistemic objects of common knowledge, including scientific and theoretical knowledge. Our account of intersubjectivity, then, accords priority to co-participation, action and practice over individual mental states, both logically and ontogenetically. Note that this priority does not deny either the existence of individual mental states, or the reflexive structure of common knowledge. Rather, it regards . Of course, there must always have been some (mythic) inventor of a social fact, such as money, and at least one other participant to understand the intention behind the invention, but once invented, the social fact acquires a relative ontological autonomy.
362 Chris Sinha and Cintia Rodríguez
i ntersubjectivity as essentially social, and logically and ontogenetically pre-requisite to common knowledge. Indeed, we have argued that intersubjectivity is the fundamental condition of all social facts, a proposal which we suggests considerably clarifies Durkheim’s own formulations, while remaining true to his insight into the relative autonomy of social facts from psychological facts. Our proposal can now be compared with the following argument for the existence and role of “collective intentionality” advanced by Searle (1995: 25–26): The requirements of methodological individualism seem to force us to reduce collective intentionality to individual intentionality. [However] it does not follow from [the individual possession of intentional states] that all my mental life must be expressed in the form of a singular noun phrase referring to me. The form that my collective intentionality can take is simply ‘we intend,’ ‘we are doing so-and so,’ and the like … the intentionality that exists in each individual head has the form ‘we intend.’
Searle’s argument, however initially appealing, faces the problem of where this primitive or a-priori “we” comes from. The answer that we offer is that it is the lexico-grammatical expression of Intersubjectivity itself, deriving from and grounded in joint action (“we are doing so-and-so”), regulated by the normative social fact that makes recognizable the joint action as an instance of the practice “so-and-so”. This answer, similarly to our argument above with respect to the ontological status of social facts, preserves Searle’s recognition of the foundational status of intersubjectivity in joint intentions, while turning his own version of it on its head. It is not, we maintain, the intentional state “we intend X” that is constitutive of the practice X: rather, the intentional state is derived from the shared practice X, whose conventionalized “objectification” as a social fact is the object of the intentional state. Searle (who fails to reference Durkheim) goes on to write (ibid. p. 26): I will henceforth use the expression ‘social fact’ to refer to any fact involving collective intentionality. So, for example, the fact that two people are going for a walk together is a social fact. A special subclass of social facts are institutional facts … for example, the fact that this piece of paper is a twenty dollar bill is an institutional fact.
. This is an allusion to Karl Marx’s assertion, both laudatory and critical, that he had turned Hegel’s dialectical logic the right way up. “The mystification which dialectic suffers in Hegel’s hands by no means prevents him from being the first to present its general forms of motion in a comprehensive and conscious manner. With him it is standing on its head. It must be inverted, in order to discover the rational kernel within the mystical shell” (Marx 1976 [1873]: 103).
Language and the signifying object 363
It will be clear by now that, from our point of view, Searle’s proposal puts the cart before the horse. We would maintain, rather, that collective intentionality is based upon, not the source of, participation in joint action in an intersubjective field, regulated by social facts (norms, institutions etc.). What empirical evidence does developmental science offer for the existence of an “ontology of the social”, and how might this bear upon the difference between our account and Searle’s “collective intention” account? The classic experiment by Murray and Trevarthen (1986), who showed that infants were able to distinguish between a real-time video-mediated (CCTV) image of their mothers, and the same image recorded on videotape, unsynchronized with the infant’s own actions, is extremely illuminating. The experiment, we would argue, provides strong evidence of the reality of the “ontology of the social” as such (both as social fact and as psychological reality); and of the biologically based readiness of very young infants to participate. The important thing about such participation, which distinguishes it from mere coordination on the basis of the “stimulus situation”, is not only the temporal sequencing and rhythm of the interaction, but also the subjective recognition of being engaged in participation – indexed by the different emotional reactions of the infants to the two stimulus situations. Viewing primary intersubjectivity in terms of participation, rather than intention, has important consequences for developmental theory. The implausibility of attributing neonatal engagement to intentional mental contents has led some developmentalists to neglect the significance of primary intersubjectivity, and focus on secondary intersubjectivity (triadic joint attention) as the decisive achievement in the development of the shared mind (Tomasello 1999). We maintain, in contrast, that all later forms of intersubjectivity are predicated on primary intersubjectivity. It is, however, neither necessary nor correct to interpret the evidence for primary intersubjectivity in terms of “innate intentionality”, whether individual or collective. Rather, we prefer Trevarthen’s more recent formulation in terms of “motives for engagement”, while emphasizing that the constitution of engagement as intersubjective is effected as much by the structuring (by the caretaker) of participation, as by the biological predisposition and capacity of the infant to engage. Primary intersubjectivity, on this reading, is neither merely a psychological nor merely a biological fact, but a “proto-social fact” supported by human developmental psychobiology. Furthermore, without denying the developmental significance of the sharing of attention and of other individual intentional states, our prioritizing of participation in joint action enables us to conceptualize, similarly to Rakoczy (2006), the primary inter-mental dimension of intersubjectivity as being a normatively regulated commitment to the activity itself (see also Shotter 1978, 1995); and prompts
364 Chris Sinha and Cintia Rodríguez
us to re-examine the significance of the objects which in some sense “carry” or signify such norms and conventions.
2.
The object as a social-material signifier
Intersubjectivity is often conceived mentalistically, as a property of the “unmediated mind”. We reject the idea that intersubjectivity is to be considered as equivalent only to “inter-mental”, in that we stress that inter-corporeality extends beyond the body to encompass objects. Intersubjectivity is materially grounded in embodiment, and this embodiment extends “beyond the skin” to encompass its mediation by objects, or what we shall call, following Latour (1996), interobjectivity. Such mediation, we propose below, can be regarded as the ontogenetically first manifestation of semiotic mediation. We proposed above an account of intersubjectivity in terms of co-participation in joint action structures which, by virtue of their normative regulation, are conventionalized as social and communicative practices. This definition excludes actions which may be directed towards others, but which are not framed as part of an activity governed by a norm. It also excludes solitary activities which may be governed by norms of performance or of achievement, such as gardening or cooking a meal, which may properly be termed social practices, but which (when performed alone) do not involve social interaction. It includes both semiotically mediated discursive practices such as talking and gesturing, and socially organized non-discursive practices such as co-participation in games or in physical constructions. Primary intersubjectivity in infancy is a mode of co-participation in which the body of the infant is not so much the “vehicle” or “medium” of engagement, as the very engagement itself. Primary intersubjectivity is embodied in the strongest sense of the word. In semiotic terms, there is no distinction between the bodily movement as signifier, and the signified “meaning” that is communicated, between the inter-mental and the inter-corporeal. There is also, as yet, no differentiation between discursive and non-discursive co-participation. Inter-corporeal co-participation is not supplanted in development, but is elaborated and extended by semiotic mediation, most obviously in discursive practices employing conventionalized gesture and language. In this section, we explore the neglected role of objects (especially artefacts) in the constitution of intersubjectivity and subjectivity. The neglect stems not so much from a failure to recognize that the material world is an important dimension of co-participation, as from the tendency to downplay its semiotic status and regard it as mere
Language and the signifying object 365
“context” to language. Goodwin and Goodwin (2004: 222), for example, define participation as “actions demonstrating forms of involvement performed by parties within evolving structures of talk” [our italics], although they also recognize the need to “expand our notion of human participation in a historically built social and material world” by attending to “material structure in the environment” (ibid. p. 239). Our purpose in this section is to foreground the semiotic aspect of materiality, and the material aspect of meaning, and to analyze their role in the development of intersubjectivity and normativity. We owe the notion of semiotic mediation to Vygotsky, whose explanation of its operation in cognition, and in cognitive development, focused on the internalization of conventional signs originating in contexts of discursive practice. Although Vygotsky attributed great importance to the formative role of language in the emergence of “inner speech” and “verbal thought”, his employment of the concept of semiotic mediation also encompassed the use of non-systematic signs, including objects-as-signifiers. One of his most celebrated examples of semiotic mediation is that of a mother tying a knot in the handkerchief of her child, to remind him of the need to convey a message to the teacher – a social practice which was widespread, not only in Russia, until quite late in the 20th century. Vygotsky writes: When a human being ties a knot in her handkerchief as a reminder, she is, in essence, constructing the process of memorizing by forcing an external object to remind her of something; she transforms remembering into an external activity. This fact alone is enough to demonstrate the fundamental characteristics of the higher forms of behaviour. In the elementary form something is remembered; in (Vygotsky 1978 [1930]: 51) the higher form humans remember something.
The semiotic value of the knot is conventional, not by virtue of the knot being an element of a sign system, but because it is normatively framed by a social practice of “reminding”. It is this frame of practice which underpins the meaning signified by the knot on any given occasion, constituting the semiotic status of the knot as an example in miniature of what Searle (op cit.) calls an “institutional fact”. Vygotsky’s knot in the handkerchief, and Searle’s twenty dollar bill, are thus both institutional facts; and both are exemplars of the material semiotic mediation of social practices – the exchange of respectively information and goods. We may note both similarities and differences between the two cases. First, the similarities. There is no intrinsic property of the material substrate (cotton, paper and ink) which determines the semiotic or monetary value of the token, which is conventionally determined. Hence, the token is equivalent, for purposes of use, to any other type-identical token, which need not be made of the same material (a piece of string round the wrist, an electronic credit on a chip card). This
366 Chris Sinha and Cintia Rodríguez
i ndependence of semantic or monetary value from material substrate is, of course, a fundamental property of signs, a mark, as it were, of the domain of semiosis or signification. Now we may take note of the differences between the knot and the twenty dollar bill. If the monetary token is materially destroyed, the value that it signifies is also destroyed, whereas if the knot is untied, the information it signifies is not. The twenty dollar bill is “cashed out” or used up (for the purposes of the user) once exchanged, since it passes from the ownership and control of the user. However, its monetary value is preserved until it is withdrawn from circulation. Conversely, the knot can be “used” again by the user to recall another, different message, while the message signified by the knot no longer has any utility or communicative value once it has been exchanged. Finally, while it makes sense to say that the knot “stands for” the message, the twenty dollar bill does not “stand for” (say) twenty one-dollar bills, but is exchangeable for or equivalent to them. All these differences can be summed up by saying that while the knot is a sign of the message, the twenty dollar bill is its monetary exchange value, it is self-identical to that value. Nonetheless, although the twenty dollar bill is not a sign, its self-identity to its monetary exchange equivalents is not physical, but social and semiotic. Searle’s account of social or institutional facts (such as money) is that they depend upon collective agreement and knowledge that, under determinate rules, something counts as an instance of a social object. Hence, the general form of such rules is: “X counts as Y in context C” (Searle 1995: 28).
Note, here, that this definition is wider than, but subsumes: “S (a sign) stands for M (a message) in context C”.
For example, we could say that Vygotsky’s handkerchief counts as a sign for a message in its context of use, and the “standing for” relationship obtains between the handkerchief and the specific message in context C. So, on this interpretation, the sign relationship can be expanded into: “X counts as S, and S stands for M, in context C”.
The distinction between the “counts as” and the “stands for” relationship can now be used to distinguish between the grammatical acceptability and the semantic interpretation of a sentence (Itkonen this volume): “James eats meat” counts as a correct sentence in English, and stands for the proposition that James eats meat in context C.
Language and the signifying object 367
Note, now, that it is also in virtue of the combination of its formal arrangement and its context, that the sentence “James eats meat” counts as an assertion of the proposition; the sentence does not “stand for” the assertion, rather the act of uttering it in a particular context is that assertion, just as Searle’s twenty dollar bill is the twenty dollars, rather than standing for it. Hence, both the grammaticality and the illocutionary force of an utterance are aspects of what the utterance counts as (being) in its context, while its semantic interpretation is the interpretation that it stands for, in that same context. All of this is irreducibly normative, and it is this duality of normative structure, of “counting as” and “standing for”, that underlies the conditions on representation that are analyzed by Sinha (1988: 37): “To represent something … is to cause something else to stand for it, in such a way that both the relationship of ‘standing for’, and that which is intended to be represented, can be recognized.” The fact that the “standing for”, or sign relation, is embedded in the “counting as”, or institutional relation, also makes it clear why language must be viewed as primarily a social institution (Itkonen this volume). This account might suggest, too, that the institutional “counting as” relation is somehow cognitively simpler than the sign relation. This cannot be the case without qualification, since coined money was only invented in the period of 800-600 BCE, in Greece and China, at a time when we have ample evidence of written language. Indeed, Sohn-Rethel (1977) argues that it was the invention of coinage which simultaneously brought into existence both generalized commodity production and the very notion, fundamental to logic, of abstract equivalence and purely formal identity. Sociogenetically, then, institutional semiotic forms have continued to be historically elaborated along with symbolic forms (Sinha in press). Ontogenetically, however, we shall argue that the normative understanding of “counting as” precedes the development of symbolization and language. To make this argument, we briefly cite Searle once again, who points out that: “in order that something be a chair, it has to function as a chair; and hence, it has to be thought of or used as a chair. Chairs are not abstract or symbolic in the way that money and property are, but the point is the same in both cases.” And the point, of course, is normative. Let us examine more carefully the semiotics of material artefacts. Now, anything can be used as a chair, provided it has the affordances, in the sense of Gibson (1979), which permit it to be sat in or on. Such affordances are part of what Searle calls the “brute” or “natural” facts, as opposed to institutional or social facts. Is there, however, any sense in which something can be . This account also implies that Searle is wrong to characterize money as “symbolic”, inasmuch as symbolization involves denotation or representation (Sinha 2004). We can certainly agree, however, that the self-identical relation of “counting as” is inherently semiotic as well as social.
368 Chris Sinha and Cintia Rodríguez
said to properly “count as” a chair in Searle’s sense of an institutional fact? The answer, we suggest, is yes: an object counts as a chair if it is an artefact intended and designed to be used as a chair. The physical properties of the chair are then no longer merely “brute facts”, but socially constructed and normatively regulated affordances, which make possible the canonical function of the chair. The canonical functions of artefacts are therefore social facts, and the material world of artefactual objects is not one only of “brute facts” in their physical aspect, but also one of social meaning. We conclude that, in analogous fashion to the way that the twenty dollar bill signifies (without standing for) its normative identity as a representation of exchange value, the artefactual object (such as a cup, a chair, or a computer) signifies (also without standing for) its normative canonical function or use value. Objects, then, not only (as with Vygotsky’s handkerchief) can be signs for something else, but, when they are artefacts, as most objects we encounter in our everyday lives are, are also signifiers of their proper, socially standard, canonical functions in a context of social practices. Of course, a condition for the semiotic status of artefacts, as with any semiotic status, is that human subjects are capable of cognitively grasping it. As Searle says, for a chair to function as a chair, it has to be used as a chair and thought of as a chair. When do human infants begin to display such a cognitive grasp, and where does it come from?
3.
Early object use and exchange: Canonicality and normativity
In a series of experiments Walkerdine and Sinha (1978), Freeman, Lloyd and Sinha (1980), Lloyd, Sinha and Freeman (1981), Freeman, Sinha and Condliffe (1981), and Sinha (1982, 1983) investigated infants’ and young children’s understanding of object function, using infant search, action imitation and acting-out language comprehension paradigms. In an age range from 9 months to 3 years and 6 months, they found error patterns which were characterized by “canonicality effects”. Infants at the end of the first year of life were more successful in A-not-B search tasks (otherwise known as object permanence tasks) when the object was hidden in an upright than in an inverted cup. It seems that these infants understood that a cup is a “better” container when in an upright orientation than when inverted. Slightly older infants were generally unable to imitate the placement of a small block on the bottom of an inverted cup, preferring to turn the cup back . Expressed in an older philosophical lexicon, canonicality of object function is a normative phenomenon existing at the interface between “Erste Natur” and “Zweite Natur”.
Language and the signifying object 369
into an upright orientation and place the block inside the cup. In this response strategy, the infants showed that they were “locked” into a normative apprehension of the cup as a canonical container, which over-rode the “brute” affordance of the flat surface of the bottom of the inverted cup. Even after this response strategy disappeared in action imitation tasks, it re-appeared in language comprehension tasks: for example two year olds, when asked to place a block “on” an inverted cup, turned it to the upright position and placed the block inside it. These experiments can be interpreted as showing that, in the first place, objects are cognitively apprehended by infants, from an early age, in terms of their socially-imposed, normative and canonical function (the object “counts as” a container). In the second place, the emerging conceptualization of spatial relations between objects is also derived as much from the canonical functional relations which objects contract with each other as from purely perceptual-geometric information (for a discussion of the functional basis of spatial relational meaning, see Vandeloise 1991). Where does this understanding, on the part of the infant, of the canonical function of objects come from? This question is important, because of the intimate relationship between the physical properties of the artefact, and its socially “baptized” canonical function. In contrast with, for example, the monetary token (in which the relationship between the material from which the token is made, and its exchange value, has historically become increasingly attenuated, arbitrary and even, as money assumes the mantle of pure informational form, virtual), the physical structure of “traditional” artefacts such as cups is not only non-arbitrary, but essential to its fulfilment of its canonical function. Infants’ motivation to explore the physical world is well known, and it might be hypothesized that their apprehension of object properties in terms of function derives from an untutored, spontaneous sensori-motor engagement with the object as a purely physical entity (for example, the exploration of the cavity of a container giving rise to the dominance of this cavity in the early pre-conceptual representation of the object). We have several sources of evidence that this is not so. First, while there is evidence of understanding of containment as a physical relationship at 6 months (Hespos and Baillargeon 2001), we were unable to detect canonicality effects in search tasks below the age of 9 months. This may, however, be a consequence of a motor-involving against a violation-of-expectancies methodology. Second, when the perceptual-cognitive link between canonical orientation and canonical containment function of cups was broken, by painting schematic faces either upright or upside down on the cups, the canonicality effect in infant search was abolished (Lloyd et al. 1981). This finding reinforces the conclusion that the
370 Chris Sinha and Cintia Rodríguez
c anonicality effect is dependent upon socially cued expectations about the normative use of the object. Even more decisive experimental evidence for the role of joint action in establishing canonical object concepts comes from the experimental design used in Freeman at al. (1981), where the object was functionally “ambiguous”, consisting of a set of stacking / nesting cubes. The child was invited by the experimenter to play with the entire set of cubes, and the experimenter set up this pre-test game as either a nesting or a stacking activity. After successfully completing, as joint action, an activity of constructing either a nest of cubes, or a tower of stacked cubes, the experimenter extracted a medium-size cube and a small cube, and conducted either an action imitation task involving the placement of the smaller cube on top of/ inside/under the larger cube, or an acting-out language comprehension task with instructions to place the smaller cube “in”, “on” or “under” the larger cube. The results were dramatic. After playing a nesting game, childrens’ error patterns showed a response bias similar to the “canonicality effect” manifested in the same task using cups. In other words, there was a response preference for placing the small cube inside the larger cube. However, this effect was abolished in the stacking condition, in which there was a tendency to preferentially place the smaller cube on top of the larger cube. To conclude this review of experimental evidence, we emphasize that canonical function and orientation, though they are in some sense “intrinsic” to the object as a material entity with determinate structure and affordances for human action, are not essential object properties in the same way as object substance. The stacking / nesting cubes experiment showed that the framing of the object in terms of its normatively appropriate function and orientation can be “locally” taught and negotiated. There is also inter-cultural variation in the canonical orientation and function assigned to classes of objects which may be materially identical between the cultures. For example, in the indigenous agrarian Zapotec culture of Southern Mexico, baskets are commonly stacked, and are frequently used as covers for foodstuffs and in childrens’ games of catching chickens. As well as these differences between Zapotec and Euro-American cultural practices, the Zapotec language lexicalizes the different spatial relations that are lexically distinguished by English “in” and “under” using a single body-part term, translatable as the English word “stomach”. Young Zapotec children differed from their Danish counterparts not only in their response patterns in language comprehension tasks using baskets, but also in non-linguistic action imitation tasks. The Zapotec children clearly did not regard the relationship of what we consider to be canonical containment, and the orientation that we would regard as “upright”, as being
Language and the signifying object 371
canonical (Sinha and Jensen de Lopez 2000; Jensen de Lopez 2003; Jensen de Lopez, Hayashi and Sinha 2005). The experimental evidence we have reviewed supports the view, then, that it is the intersubjective structuring of the child’s participation in joint action, as much as (and indeed more so) than the affordances of the object “in itself ”, that enables the child, in a process of “guided reinvention” (Lock 1980), to appropriate the norms governing object use and to achieve an object representation in terms of canonical function. This process has a long developmental history, and the episodes of joint action are accompanied and mediated at every stage by the use of communicative signs by the adult participant, as is attested by the observations reported by Rodríguez and Moro (1999, 2002, this volume; see also Moro and Rodríguez 2005). Throughout this developmental process, “objects are invested with significance. They become, for the child, material representations and signifiers of the rules, norms, values, rituals, needs and goals of the entire … matrix within which they are embedded. In short, they become part of a meaningful system of signs” (Sinha 1988: 204).
4.
From signifying object to communicative symbol
Artefacts, we have argued, have an intrinsic meaning given by their canonical function or use value. What an object “means” on any given occasion, however, is dependent upon more than just canonical function. Not only can an artefact be used non-canonically, as when, for example, a cup is used as a paperweight, but there are also socially constituted meanings which are relatively (and conceptually) autonomous from the canonical use value of the object. Primary amongst these, at least from a developmental point of view, is the meaning of the object as an object of exchange. Give-and-take routines develop in our culture early in the second year of life. Such exchanges involve the super-imposition on the object of a semiotic status which is independent of its canonical function: that of an “abstract” signifier, and material embodiment, of a social relationship of exchange. Social and anthropological researchers from Marcel Mauss (2000 [1923–24]) onwards have posited exchange as a fundamental human universal (see Goux 1990). Object-exchange and the participatory induction of the infant into the normative knowledge of canonical object function are not interactively distinct in earlier triadic exchanges (Rodríguez and Moro this volume). However, the emergence of the give-and-take
372 Chris Sinha and Cintia Rodríguez
routine as a normative, reciprocal and mutually controlled format of co-participation lays the basis, we suggest, for the differentiation of signifier from signified that is necessary for mastery of the symbolic system of language. The object now becomes a signifier within a field constituted by differential, reciprocal and shifting subject positions: that of giver and that of recipient. From a communicative and symbolic point of view, the ability to negotiate these shifting subject positions constitutes a precursor of deixis. The object semiotically mediates the constitution of the triadic interaction as implicating “I” and “You”, a decisive differentiating moment in the construction of subjectivity. It has often been claimed that subjectivity is constructed in and through language. This may be so, in the sense that language provides the key symbolic support for the adoption of differential subject positions, but the “subject” that occupies these positions as simultaneously an “I” and a “Me for You” is, we suggest, constituted at the moment of entry into language through participation in the proto-institution of object exchange. There is also a psychoanalytic dimension of investment in this process of constitution of subjectivity, since the object signifies both power (to give or to withhold) and desire (the object represents a wish whose fulfilment is dependent upon the subjectivity of the Other, rather than being the immediate goal of a simple demand). Whether or not participation in give-and-take routines is a strict precondition for language acquisition, it is undoubtedly, in typical developmental trajectories, a precursor of it. Object exchange is usually co-terminous with the early stages of the development of language, and precedes the vocabulary explosion of the second half of the second year of life. We hypothesize, then, that it represents a fundamental step in the emergence of both subjectivity and the mastery of symbolization. The voluntary control in object exchange of the grasp and relinquishment of objects, governed by norms of communication rather than by immediate consummatory goals, prefigures the voluntary representational use of signs. Object exchange formats introduce into the triadic structure of joint attention a signifying element that is potentially extensible to the representational, “standing for” function of language. It also puts in place, in schematic and skeletal form, the perspectivally shifting dynamics of deictic identification of speaker and hearer. Early object exchange, we submit, like the guided appropriation by pre-linguistic infants of canonical object functions, is a neglected, fundamentally social, materially mediated aspect of the development of primary intersubjectivity towards symbolic intersubjectivity.
5.
Language and the signifying object 373
Beyond the dyad: Imagining communities and culture
In the preceding sections, we have focussed upon interobjectivity as semiotically mediating the development of participation by the infant and young child in joint actions based upon intersubjective and socially shared conventional meanings. We have also focussed upon triadic contexts of interaction, in which the object is the “third term” of a semiotic triangle constituted by the interactions between two individuals (the prototypical dyad of developmental accounts of intersubjectivity, and the ideal-typical speaker and hearer of linguistic theory). What is missing in our account so far is the “Social Third Person”: not the Object, but the community of practice and meaning that ultimately sanctions the norms governing the interactions between any two or more participants in their dealings with social reality. In confronting this construct – Society, Community or however it may be designated – we encounter a fundamental problem of the social and human sciences. How do we reconcile the agency of human subjects, their capacity for creating novelty, with the determining (though not strictly deterministic) structures and processes which permit the development of the encultured and socialized subject? In this section, we maintain our focus on the role of objects and “interobjectivity” in what Fogel, Valsiner and Lyra (1997) have called “the dynamics of indeterminism in developmental and social processes”; with particular reference to the article in that volume of that title by Smolka, de Góes and Pino. Smolka et al. (1997: 160) pose the following question: “In what way is [the development of the] sign related to the processes that generate or anchor creativity and individual resistance, the power of violating canonical rules?” They report and analyze an episode of socio-dramatic pretend play by a group of three 5–6 year old girls in the “house corner” of a primary school classroom, in which a cowboy hat played a crucial role as a prop in an enacted dramatic narrative. The hat, initially introduced into the play with an “extended” canonical meaning as a fashion accessory, later became a signifier of a new identity adopted by one of the girls as a feminine counterpart of a cowboy character who was a part of the background common knowledge of all the girls comprising the group. Crucial both to the investigators’ interpretation of this process, and to the children’s construction of their “play world”, was the creative linguistic designation of the character (signified by the hat as well as by the linguistic sign) as “Bete Carrera”, a grammatical feminization (in the Portuguese language) of the name of the male cowboy character “Beto Carrero” recruited from common knowledge (see Sinha 2005, for a fuller analysis).
374 Chris Sinha and Cintia Rodríguez
As Smolka et al. point out, the cowboy hat, qua artefactual object, remained throughout a hat, never used by the children as anything other than a hat. At the same time, it “became” – or, rather, came to signify – more than the canonical rules of object-usage that it embodied. Through language, the children created Bete Carrera (Turn 7), the feminine of Beto Carrero ... Language allows for this specific appropriation, for such a construction and transformation; it allows for a ‘performance’ that synthesizes old and new modes and models of acting. Through language, it is possible to become another, to become homo duplex … or, in fact, multiplex. In this consists the dra(Smolka, Gões and Pino 1997: 161) matic character of human experience.
The hat is thus simultaneously situated at two levels of meaning. At the first level, its canonical function is appropriated enactively by the participants (by putting it on and taking it off). At this first level, the construal of the hat is intersubjectively shared, non-contested and constant: the hat remains a hat. At the second level, the hat is invested with a “surplus meaning” which goes beyond canonicality. At this second level the hat comes to signify the subjective positionings and perspectives of the individual participants within a more comprehensive, discursively constituted and gendered frame, by means of which, say Smolka et al. (1997: 161), “the signifying aspect of the (inter)subjective actions … necessarily implies … immersion in language and meaning production.” The discursive frame is one of narrativity (Gallagher and Hutto this volume), through which, as Lightfoot (1997: 174) puts it, “temporal rhythm becomes history, and transitory meanings become forms of knowledge which linger long enough to be toyed with.” Through intersubjectively shared and constructed narrative, the world and the identity of the subject can simultaneously be explored, renewed and consolidated. As we emphasized earlier, this is a process in which emotional investment plays as important a role as cognitive structure, the two aspects being fused in what the cultural theorist Raymond Williams called “structures of feeling”. Here is what Williams (1977: 128) says about temporality in cultural activity and structures of feeling: If the social is always past, in the sense that it is always formed, we have indeed to find new terms for the undeniable experience of the present: not only the temporal present, the realization of this and this instant, but the specificity of the present being, the inalienably physical, within which we may discern and acknowledge institutions, formations, positions, but not always as fixed products, defining products.
Earlier, we drew upon Durkheim’s notion of social facts, emphasizing the “already there and formed” exteriority and objectivity of norms and institutions.
Language and the signifying object 375
Williams, in contrast, reminds us that it is through intersubjective agency in the present that social life and its normative institutions are enactively re-fashioned, permitting through the medium of shared narrative resources the construction of both the here-and-now and face-to-face shared mind, and the imagined community of unknown others whose history and identity we share (Anderson 1991). Indeed, it is through narrative deployment of the symbolic resources of language that our social reality becomes simultaneously actual and virtual, constrained by “objectively existing circumstances”, but pregnant with potentialities through the investment of the present by the horizon of the future. Williams also reminds us of the “inalienably physical” nature of participation and experience. Throughout this chapter, we have emphasized the neglected but vitally important role not only of the (inter)corporeal body, but of the (inter)objective materiality of shared “things at hand”; not merely in sustaining, but in developmentally constructing the shared mind.
References Anderson, B. 1991. Imagined Communities. London: Verso. Bruner, J.S. 1966. “An overview.” In Studies in Cognitive Growth, J.S. Bruner, R.R. Oliver, P.M. Greenfield, J.R. Hornsby and H.J. Kenney (eds.), 319–326. New York: Wiley. Clark, H.H. 1996. Using Language. Cambridge: Cambridge University Press. Durkheim, E. 1895. Les Règles de la méthode sociologique. Paris: Alcan. 1894a, with slight modifications, and a preface. Tr. 1982 as “The Rules of Sociological Method”. In The Rules of Sociological Method and Selected Texts on Sociology and its Method, S. Lukes (ed.), 29–163. London and Basingstoke: Macmillan. Fogel, A., Lyra, M. and Valsiner, J. (eds.). Dynamics and Indeterminism in Developmental and Social Processes. Mahwah, NJ: Lawrence Earlbaum Associates. Freeman, N., Lloyd, S. and Sinha, C. 1980. “Infant search tasks reveal early concepts of containment and canonical usage of objects.” Cognition 8: 243–262. Freeman, N., Sinha, C. and Condliff, S. 1981. “Confrontation and collaboration with young children in language comprehension tasks.” In Communication in Development, W.P. Robinson (ed.), 63–88. London, Academic Press. Gallagher, S. and Hutto, D.D. this volume. “Understanding others through primary interaction and narrative practice.” Gibson, J.J. 1979. The Ecological Approach to Visual Perception. Boston: Houghton Mifflin. Goodwin, C. and Goodwin, M.H. 2004. “Participation.” In A Companion to Linguistic Anthropology, A. Duranti (ed.), 222–244. Oxford, Blackwell. Goux, J.J. 1990. Symbolic Economies. Ithaca, NY: Cornell University Press. Hespos, S. and Baillargeon R. 2001. “Knowledge about containment events in very young infants.” Cognition 78: 204–245. Itkonen, E. this volume. “The central role of normativity for language and linguistics.”
376 Chris Sinha and Cintia Rodríguez
Jensen de Lopez, K., Hayashi, M. and Sinha, C. 2005. “Early shaping of spatial meanings in three languages and cultures: Linguistic or cultural relativity?” In Selected Papers from the LACUS Forum XXXI 2003: Interconnections, A. Makkai, W.J. Sullivan and A.R. Lommel (eds.), 377–386. Houston, Texas: Linguistic Association of Canada and the Unites States. Jensen de Lopez, K. 2003. “Baskets and Body-Parts: A cross-cultural and cross-linguistic investigation of children’s development of spatial cognition and language.” PhD dissertation, University of Aarhus. Jensen de Lopez, K., Hayashi, M. and Sinha, C. 2005. “Early shaping of spatial meanings in three languages and cultures: Linguistic or cultural relativity?” In Selected Papers from the LACUS Forum XXXI 2003: Interconnections, A. Makkai, W.J. Sullivan and A.R. Lommel (eds.), 377–386. Houston, Texas: Linguistic Association of Canada and the Unites States. Jones, R.A. 1986. Emile Durkheim: An Introduction to Four Major Works. Beverly Hills, CA: Sage Publications. Latour, B. 1996. “On interobjectivity.” Mind, Culture and Activity 3: 228–245. Lewis, D.K. 1969. Convention: A Philosophical Study. Cambridge, MA: Harvard University Press. Lightfoot, C. 1997. “Transforming the canonical cowboy: Notes on the determinacy and indeterminacy of children’s play and cultural development.” In Dynamics and Indeterminism in Developmental and Social Processes, A. Fogel, M. Lyra and J. Valsiner (eds.), 165–174. Mahwah, NJ: Lawrence Earlbaum Associates. Lloyd, S., C. Sinha and N. Freeman. 1981. “Spatial reference systems and the canonicality effect in infant search.” Journal of Experimental Child Psychology 32: 1–10. Lock, A. 1980. The Guided Reinvention of Language. London, Academic Press. Marx, K. 1976. Postface to the Second Edition, 1873. Capital vol. 1. Harmondsworth: Penguin Books. Mauss, M. 2000. The Gift: The Form and Reason for Exchange in Archaic Societies (Original publication 1923–1924: Transl. W.D. Halls). New York: W.W. Norton. Merleau-Ponty, M. 1962. Phenomenology of Perception. London: Routledge and Kegan Paul. Moro, C. and Rodríguez, C. 2005. L’objet et la construction de son usage chez le bébé. Une approche sémiotique du développement préverbal. Berne-New York: Peter Lang. Moscovici, S. 2000. Social Representations. Cambridge: Polity Press. Murray, L. and Trevarthen, C. 1986. “The infant’s role in mother-infant communications.” Journal of Child Language 13(1): 15–29. Netchine-Grynberg, G. and Netchine, S. 2002. “Vygotski, Wallon et les ‘mondes communs’”. In Avec Vygotski, Y. Clot (ed.), 85–104. Paris: La Dispute. Rakoczy, H. 2006. “Pretend play and the development of collective intentionality.” Cognitive Systems Research 7: 113–127. Reddy, V., Hay, D., Murray, L. and Trevarthen, C. 1997. “Communication in infancy: Mutual regulation of affect and attention.” In Infant Development: Recent Advances, G. Bremner, A. Slater and G. Butterworth (eds.), 247–273. Hove: Erlbaum Taylor and Francis Ltd. Rodríguez, C. 2006. Del ritmo al símbolo. Los signos en el nacimiento de la inteligencia. Barcelona: Horsori. Rodríguez, C. and Moro, C. 1999. El mágico número tres. Cuando los niños aún no hablan. Barcelona: Paidós. Rodríguez, C. and Moro, C. 2002. “Objeto, comunicación y símbolo. Una mirada a los primeros usos simbólicos de los objetos.” Estudios de Psicología 23–3: 323–33.
Language and the signifying object 377
Rodríguez, C. and Moro C. this volume. “Coming to agreement: Object use by infants and adults.” Searle, J. 1983. Intentionality: An Essay in the Philosophy of Mind. Cambridge: Cambridge University Press. Searle, J. 1995. The Construction of Social Reality. London: Allen Lane. Shotter, J. 1978. “The cultural context of communication studies: Theoretical and methodological issues.” In Action, Gesture and Symbol: The Emergence of Language. A. Lock (ed.), 43– 78. London: Academic Press. Shotter, J. 1984. Social Accountability and Selfhood. Oxford: Basil Blackwell. Shotter, J. 1995. “In conversation: Joint action, shared intentionality and ethics.” Theory and Psychology 5: 49–73. Sinha, C. 1982 “Representational development and the structure of action.” In Social Cognition: Studies in the Development of Understanding, G. Butterworth and P. Light (eds.), 137–162. Brighton: Harvester. Sinha, C. 1983. “Background knowledge, presupposition and canonicality.” In Concept Development and the Development of Word Meaning, T. Seiler and W. Wannenmacher (eds.), 269–296. Berlin: Springer-Verlag. Sinha, C. 1988. Language and Representation: A Socio-Naturalistic Approach to Human Development. Hemel Hempstead: Harvester-Wheatsheaf. Sinha, C. 1999. “Grounding, mapping and acts of meaning.” In Cognitive Linguistics: Foundations, Scope and Methodology, T. Janssen and G. Redeker (eds.), 223–256. Berlin: Mouton de Gruyter. Sinha, C. 2004. “The evolution of language: From signals to symbols to system.” In Evolution of Communication Systems: A Comparative Approach, D. Kimbrough Oller and U. Griebel (eds.), 217–235. Cambridge, MA: MIT Press, Sinha C. 2005. “Blending out of the background: Play, props and staging in the material world.” Journal of Pragmatics 37: 1537–1554. Sinha, C. in press. “Iconology and imagination in human development.” In Religious Narrative, Cognition and Culture: Image and Word in the Mind of Narrative, A.W. Geertz and J.S. Jensen (eds.). London: Equinox Publishing. Sinha, C. and Jensen de López, K. 2000. “Language, culture and the embodiment of spatial cognition.” Cognitive Linguistics 11: 17–41. Smolka, A.-L., Gões, M. de and Pino, A. 1997. “(In)Determinacy and the semiotic constitution of subjectivity.” In Dynamics and Indeterminism in Developmental and Social Processes, A. Fogel, M. Lyra and J. Valsiner (eds.), 153–164. Mahwah, NJ: Lawrence Erlbaum Associates. Sohn-Rethel, A. 1977. Intellectual and Manual Labor: A Critique of Epistemology. Atlantic Highlands, NJ: Humanities Press. Tomasello, M. 1999. The Cultural Origins of Human Cognition. Cambridge, MA: Harvard University Press. Trevarthen, C. 1979. “Communication and cooperation in early infancy: A description of primary intersubjectivity.” In Before Speech: The Beginning of Interpersonal Communication, M. Bullowa (ed.), 321–347. Cambridge: Cambridge University Press. Trevarthen, C. 1998. “The concept and foundations of infant intersubjectivity.” In Intersubjective Communication and Emotion in Early Ontogeny, S. Bråten (ed.), 15–46. Cambridge: Cambridge University Press.
378 Chris Sinha and Cintia Rodríguez
Vandeloise, C. 1991. Spatial Prepositions: A Case Study from French. Chicago: Chicago University Press. Vygotsky, L.S. 1978. Mind in Society: The Development of Higher Psychological Processes. Cambridge, MA: Harvard University Press. Walkerdine, V. and Sinha, C. 1978. “The internal triangle: Language, reasoning and the social context.” In The Social Context of Language, I. Markova (ed.), 151–176. London: Wiley. Wallon, H. 1925/1984. L’Enfant Turbulent. Etude Sur les Retards et les Anomalies du Développement Moteur et Mental. Paris: PUF. Williams, R. 1977. Marxism and Literature. Oxford: Oxford University Press. Zazzo, R. 1975. Psychologie et Marxisme. La Vie et l’Oeuvre d’Henri Wallon. Paris: Denoël/ Gonthier. Zlatev, J. this volume. “The co-evolution of intersubjectivity and bodily mimesis.”
Author index
A Adamson, L. 78, 97, 144, 192, 194, 198 Anscombre, J.-C. 311, 312, 315, 317 Arbib, M. A. 180, 217, 219, 220, 221, 222, 248, 250, 255, 264, 267 Astington, J. 29, 32, 216, 236 Austin, J. L. 121, 151, 172, 324, 328 B Bakeman, R. 79, 97, 144, 168, 176, 194 Baldwin, D. A. 20, 21, 22, 216, 267 Bard, K. 5, 6, 9, 96, 120, 121, 122, 145, 172, 176, 187, 190, 196, 198, 204, 205, 206, 224, 225 Baron-Cohen, S. 23, 91, 142, 175, 188, 190, 193, 202, 248, 257, 261, 266 Barresi, J. 6, 7, 19, 40, 42, 43, 47, 48, 60, 61, 76, 83, 96, 117, 126, 143, 215, 217, 218, 220, 282, 302 Bates, E. 120–123, 135, 152, 157, 158, 170, 171, 172, 175, 176, 178, 179, 190, 193, 198, 199, 201, 203, 205, 220, 252–256 Bateson, G. ix, 5, 188, 189, 190 Bateson, W. 189 Benveniste, É. 324 Bermúdez, J. L. 21, 249 Bickerton, D. 219, 221, 260, 262 Bloom, P. 216, 239 Bretherton, I. 146, 172 Brinck, I. 5, 6, 8, 115, 120–122, 128, 135, 136, 146, 152, 155, 158, 170–172, 189, 190, 191, 194,
201, 207, 215, 220, 229, 230, 257, 263, 346 Brooks, R. 21, 24, 117, 118, 126, 127 Bruner, J. 29, 30, 32, 33, 78, 109, 152, 171, 357 Bühler, K. 119, 139 Butterworth, G. 155, 178, 187, 188, 193, 195, 196, 198, 202, 206, 207 Byrne, R. W. 168, 174, 176, 200, 229, 231, 243 C Call, J. 123, 146, 167, 176, 177, 180, 187, 195–197, 200, 201, 205, 207, 217, 227–231, 234, 236, 245, 246, 252, 269 Camaioni, L. 125, 152, 175, 193, 220 Carpendale, J. 26, 91, 142, 146, 150, 158, 265 Carpenter, M. 142, 146, 156, 158, 178, 202, 217 Carruthers, P. 255, 262 Chafe, W. 346, 347 Cheney, D. L. 166, 308 Chomsky, N. 279, 293, 294, 296, 300 Churchland, P. 152 Clark, H. 233, 289, 293, 316, 358 Collingwood, R. G. 301 Condillac, E. B. D. 180 Corballis, M. C. 96, 180, 255, 260, 268 Costall, A. 91, 92, 94, 96, 109, 111 Csibra, G. 125, 127 Currie, G. 34, 265
D D’Entremont, B. 125, 156 Damasio, A. 54 Dapretto, M. 56, 60, 61 Dautenhahn, K. 64, 260, 261, 272 Dawkins, R. 311 de Saussure, F. 284, 297, 298 de Villiers, J. 216, 217, 236 de Waal, F. B. M. 168, 195, 196, 207, 217, 224–226, 228 Deacon, T. 219, 220, 232, 239 Decety, J. 32, 52, 53, 58, 59, 118 Dennett, D. 327 Donald, M. 217, 218, 220, 231, 238, 245, 248, 249, 252–256, 261 Dretske, F. 302 Du Bois, J. W. 333, 334 Ducrot, O. 311–313, 315, 317 Dunbar, R. 180, 258–261, 264, 309 Dunham, P. J. 146, 215 Dupré, J. 142, 148, 149, 151, 245 Durkheim, E. 5, 11, 357, 360, 361, 362, 374 E Emmorey, K. 344 Enfield, N. 196 Engels, F. 291 Enkvist, N. E. 343, 344 F Farroni, T. 125, 133 Feldman, C. F. 29, 33 Fillmore, C. 320 Fodor, J. 1, 300 Fogel, A. 142, 153, 373
380 The Shared Mind
Fouts, R. S. 195, 197, 200, 205, 232 Franco, F. 178, 193, 194, 195, 198, 207 Frith, U. 55, 61 G Gallagher, S. 3–7, 10, 17–21, 42, 48, 143, 191, 217, 221–223, 225, 246, 250, 251, 264, 265, 345, 374 Gallese, V. 19, 42, 45, 117, 217, 222, 224, 250–252 Gärdenfors, P. 128, 217, 218, 229, 247, 249, 257 Gardner, R. A. 176, 197, 201 Garfinkel, H. 153 Gibson, E. 94 Gibson, J. J. 3, 24, 25, 96, 367 Gile, D. 335, 340, 341, 345, 348–350 Givón, T. 333, 347, 351 Goldman, A. 17, 19, 42, 118, 251, 252 Gomez, J. C. 176 Goodall, J. 168 Goodwin, C. 360, 365 Gopnik, A. 17, 22, 41, 48, 223, 248, 249, 250 Gordon, R. 17, 42, 264 Gould, J. 95 Grice, P. 220, 259, 338 Guajardo, N. R. 29, 32, 265 H Hacker, G. P. 144, 145, 147, 150, 152 Hare, B. 226, 227 Harris, P. 42, 253 Hobson, R. P. 6, 8, 20, 22, 42, 67, 69–73, 77, 80, 83, 85, 126, 142, 143, 207, 217, 251, 264 Hobson (Meyer), J. A. 6, 8, 42, 67, 72, 73, 87 Hockett, C. F. 232, 293, 294, 307, 329 Hofstadter, D. 298–300 Honderich, T. 11
Hopkins, W. D. 5, 6, 9, 96, 120–122, 145, 175, 187, 190–197, 199, 201, 205 Hubley, P. 23, 42, 77, 127, 141, 146, 193 Hudson, R. 297 Humphrey, J. H. 336–339, 345 Humphrey, N. 42 Hurley, S. 19, 249 Hutto, D. D. 3, 5–7, 10, 12, 17, 18, 25, 28, 33, 42, 48, 143, 180, 191, 217, 218, 220, 222, 226, 236, 239, 245, 246, 251, 255, 257–260, 262, 265–267, 345, 374
Kita, S. 146, 234 Klin, A. 133, 138 Knoblich, G. 40, 42, 64, 250 Krause, M. 195, 197, 200, 205, 207, 210
J Jackendoff, R. 1, 296, 302 Janzen, T. 11, 333, 335, 341, 345, 346, 350 Jeannerod, M. 19, 118 Jenkins, J. M. 216, 236 Johnson, C. M. 188, 189, 190 Johnson, M. 152, 268 Johnson, M. C. 125 Johnson, S. C. 21, 22 Johnson-Laird, P. 301 Jones, R. A. 360 Jones, S. 249
L Lakoff, G. 152, 268, 296 Lamarque, P. 30, 37, 262 Lawrence, S. 333–338, 345, 351 Leakey, M. D. 204 Leavens, D. 5, 6, 9, 96, 120–122, 145–147, 152, 155, 175, 176, 187, 190, 195–202, 204–207 Lee, T. 22, 37, 69, 70, 72, 87 Leech, G. 284 Leeson, L. 346, 349 Legerstee, M. 21, 37, 97, 207 Leslie, A. M. 17, 37, 41, 58, 64, 91, 262, 267 Leung, E. H. L 193, 195, 198 Levinson, S. C 312 Lewin, R. 221, 232 Lewis, C. 26, 29, 37, 91, 158, 265 Lewis, D. K. 233, 288, 289, 296, 307, 358 Liebal, K. 169, 172–174, 176, 180, 200, 201, 207 Liebermann, P. 166 Lightfoot, D. 374 Linell, P. 334, 341, 343 Liszkowski, U. 156, 158, 179, 203, 207 Lloyd, S. 368, 369 Lock, A. 179, 193, 371 Locke, J. 256 Lohmann, H. 217, 236 Lou, H. C. 59 Lovejoy, C. O. 256
K Kac, M. 300 Kanner, L. 71, 84 Karmiloff-Smith, A. 95 Katz, J. 299, 300 Kendon, A. 195 Keysers, C. 42, 45, 55, 117 Kiparsky, P. 299
M MacKinnon, J. R. 168 Maestripieri, D. 168, 183 Malle, B. 2 Mandler, J. 92, 93, 113 Marler, P. 166, 167, 255 Martinet, A. 329 Marx, K. 291, 362 Matsuzawa, T. 37, 47
I Iacoboni, M. 50, 53, 56, 60, 264 Inhelder, B. 95 Itakura, S. 200, 201 Itkonen, E. 1, 4, 5, 10, 11, 109, 207, 216, 218, 229, 233, 239, 279, 280, 283, 284, 287, 289, 294–296, 299, 300–302, 307, 312, 342, 358, 361, 366, 367
Mauss, M. 371 McGrew, W. C. 168, 179 Meltzoff, A. N. 21, 22, 24, 37, 93, 94, 117, 118, 126, 127, 223, 248, 249, 250 Menyuk, P. 221 Menzel, E. W. 176 Merleau-Ponty, M. 3, 217, 222, 357, 359 Messer, D. 100, 195 Miall, C. R. 54, 58 Miles, H. L. 176, 195, 197, 203, 230, 232, 233 Mindess, A. 344, 345 Mitani, J. C. 167, 177, 195, 206 Mitchell, R. W. 190, 191, 194 Mitchell, P. 235 Moll, H. 48, 146, 217 Moore, C. 6, 7, 19, 39, 40, 42, 43, 47, 48, 60, 61, 76, 83, 92, 96, 117, 126, 143, 146, 156, 193, 203, 207, 215, 217, 218, 220, 282, 302 Moore, D. G. 22 Moore, M. K. 93, 94, 248, 249 Morissette, P. 157 Moro, C. 4, 5, 8, 89, 93, 94, 99, 109, 371 Morton, E. S. 166, 167, 308, 310, 311, 328, 329 Moses, L. J. 127, 130, 265 Mundy, P. 78–80, 84, 125 Murray, L. x, 359, 363 Myowa-Yamakoshi, M. 21, 223, 248 N Nelson, K. 26, 29, 31, 93, 98, 218, 221, 261 Nicolopoulou, A. 26, 31, 261 Nishida, T. 168, 179 Nuyts, J. 324, 328 O O’Neil, D. K. 194, 253, 290 Owings, D. H. 166, 167, 308, 310, 311, 328, 329
Author index 381
P Parker, T. 205, 243 Patterson, F. G. 176, 232, 235 Paul, H. 284, 296, 299 Perner, J. 83, 235 Persson, T. 217, 218, 249 Peterson, C. C. 29, 166, 216, 236 Petitto, L. 187, 188 Pettit, P. 283 Phillips, A. 31, 265 Phillips, W. 23 Piaget, J. 12, 90, 93–97, 170, 171, 193, 219, 220 Pika, S. 6, 9, 96, 165, 169, 172, 173, 176–179, 190, 195, 200, 201, 204, 206 Plooij, F. X. 168, 169, 172, 176 Popper, K. 289, 291 Povinelli, D. J. 175, 187, 190, 200–202, 245, 246 Premack, D. 176 Preston, S. 217, 224, 225 Prinz, W. 40, 139 Pyers, J. 216, 217, 236 Q Quigley, S. P. 340, 341 Quill, K. 221 R Racine x, 1, 5, 6, 8, 12, 68, 110, 137, 141, 142, 144, 146, 150, 155, 157, 190, 191, 201, 207, 334 Rakoczy, H. 79, 80, 128, 363 Reddy, V. 117, 118, 126, 137, 192, 193, 207, 223, 224, 359 Rheingold, H. L. 78, 193, 195, 198, 211 Ricard, M. 157, 199 Richner, E. S. 26, 31, 38, 261 Rizzolatti, G. 42, 45, 50, 117, 217, 222, 249, 250 Rochat, P. 97, 126 Rodríguez, C. 4–6, 8, 11, 89, 92–94, 99, 109, 110, 357, 359, 371 Roth, D. 91 Roth, R. R. 176
Rowlands, M. 259, 267 Rumbaugh, D. M. 167, 168, 176, 232 Russell, C. L. 198 Russell, J. L. 120, 121, 187, 190 Russell, B. 284 S Sabater-Pi, J. 176, 195, 203 Samson, D. 58, 59 Savage-Rumbaugh, S. 167, 168, 175, 176, 193, 197, 202, 203, 221, 232, 233 Scheler, M. 3, 4 Schutz, A. 3 Searle, J. 11, 324, 357, 358, 361–363, 365–368 Seibert, J. M. 78, 84 Senghas, A. 234 Seyfarth, R. M. 166, 308 Shaffer, B. 11, 333, 335, 343, 346, 347, 351 Shotter, J. 109, 359, 363 Sigman, M. 71, 78 Singer, T. 54, 55 Sinha, C. 1, 4–6, 11, 98, 109, 137, 180, 190, 197, 207, 217, 239, 248, 253, 255, 267, 268, 288, 316, 329, 357, 359, 367, 368, 371, 373 Smith, R. T. 341, 345 Smolka, A.-L. 373, 374 Sommerville, J. A. 46, 59, 265 Sonesson, G. 220, 221, 239 Spelke, E. 92, 114 Sperber, D. 312 Sterelny, K. 245, 251 Stern, D. 7, 8, 22, 116, 131, 132, 219, 220, 223 Stokoe, W. 232 Striano, T. 79, 95, 97, 125, 126, 156, 158 Sugarman, S. 78, 190, 193, 205 Susswein, N. 5, 6, 8, 68, 141, 190, 191, 201, 334 T Tager-Flusberg, H. 217, 236
382 The Shared Mind
Tanner, D. 168, 174, 176, 200, 231 Tomasello, M. 47, 48, 79, 97, 109, 123, 126–128, 142, 146, 156, 158, 165, 167, 169–180, 187, 188, 190, 193, 195–198, 200–202, 204, 205, 207, 216, 217, 226–231, 234, 236, 245, 246, 263, 269, 308–311, 323, 324, 363 Trevarthen vii–xi, 7, 17, 20, 23, 42, 68, 77, 116, 117, 126, 127, 141–143, 146, 190, 192, 193, 207, 219, 220, 223, 359, 363 Tutin, C. E. G. 168, 179 V Valsiner, J. 1, 373 van Lawick-Goodall, J. 168, 195, 204 Vandeloise, C. 369 Varela, F. 1 Vea, J. J. 176, 186
Verhagen, A. 6, 11, 285, 307, 310, 311, 315–317, 320, 322, 325, 326, 334, 342 Verschueren, J. 284, 295 Volterra, V. 152, 179, 190, 193, 220, 240 von Glaserfeld, E. 171, 186 Vonk, J. 245, 246 Vygotsky, L. S. 3–5, 8, 89–91, 96, 231, 334, 365, 366, 368 W Wadensjö, C. 343, 347, 349, 352 Wallon, H. 90, 96, 359 Watson, S. 29, 32 Wellman, H. 31, 38, 265 Whiten, A. 187, 195, 197, 229, 249 Wicker, B. 55, 64, 117 Wilcox, S. 343, 346, 347 Williams, E. 92 Williams, R. 374, 345 Wilson, D. 312
Winch, P. 281, 286, 289 Wittgenstein, L. 3–5, 8, 10, 23, 38, 81, 109, 110, 144, 158, 230, 279–283, 286, 287, 290, 294, 297, 300, 307 Woodruff, G. 176 Woodward, A. L. 46, 47, 127, 137, 265 Wynn, T. 247, 253 Y Youngs, J. P. 340, 341 Z Zahavi, D. 3 Zazzo, R. 359 Zlatev xi, 1, 5, 6, 9, 10, 20, 21, 25, 34, 38, 42, 43, 47, 56, 96, 111, 137, 166, 178, 215–218, 221, 222, 225, 226, 232, 249, 255, 259, 264, 267, 268, 283, 288, 289, 307, 309, 329, 359 Zuberbühler, K. 166, 167, 177
Subject index
A affective contact 73, 84 engagement 71, 72 intentional relations 34, 54 (see also intentional relations) response 203, 207 sharing 131, 134 states 22, 56, 215, 223 affordances 24, 136, 367–371 agent 21, 23–25, 30, 40–48, 52, 53, 94, 98, 148, 151, 152, 154, 175, 179, 295 agency 54, 190, 192, 373, 375 agreement adult-baby 8, 89, 90, 98, 100, 108–111 collective 267, 366 in grammar 293 in judgment 81 alarm calls 166, 167, 308, 310, 327 alignment 78, 132 allocentric representation 51, 52, 60, 132 (see also thirdperson information) American Sign Language (ASL) 11, 232, 235, 333, 335–346, 348–351, 353–355 amygdala 32, 56 analogy 221, 280, 301, 340 animal communication 11, 166–171, 231, 234, 307–311, 329 argumentative 11, 285, 307, 311–316, 322, 323, 325, 327–329, 334 orientation vs. strength 317–321, 326
artefacts 4, 91, 268, 288, 358, 359, 364, 367–369, 371 attribution of mental states/propositional attitudes 5, 26, 27, 30, 59, 127, 153, 175, 190, 193, 251, 265, 267 autism 2, 6–8, 43, 44, 60–62, 67–76, 81–86, 91, 92, 126, 142, 148, 202, 221, 224, 251, 266 autoscopic hallucinations 53 B background knowledge 26, 27 beliefs and desires 18, 20, 21–24, 91, 110, 265 attribute/understand beliefs 10, 19, 27, 41, 132, 191, 215–217, 222, 233, 234, 236–239, 245, 267, 295, 307, 348, 358, 361 (see also attribution) concept of belief 25, 26, 83, 234 false belief (tasks) 25, 26, 29, 39, 40, 49, 53, 58, 63, 91, 211, 217, 231, 235, 236 bipedalism 204, 205, 206, 207, 255 blindness 83 bodily intentionality 21, 22 (see also intentionality) mimesis (see mimesis) based 132, 217
body role for intersubjectivity 1, 3, 4, 10, 11, 21, 23, 51, 52, 54, 57, 82, 215, 219, 224, 234, 237, 238, 280, 357, 364, 375 body schema vs. body image 222, 223, 225, 250 body-centered 52 (see also first-person) brain 32, 49, 54, 56, 57, 60, 96, 117, 118, 189, 247, 250, 256, 258, 264 damage 53, 58 bridging modalities 45, 54 C canonical forms/activities 261, 268 function 368, 369, 370, 371, 374 uses 93, 107 caregiving 203, 204 Cartesian 280, 302 causal role/factor/force 5, 96, 159, 218, 222, 235, 236, 295 vs. definitional issues 8, 9, 141, 142, 147, 149, 159, 292 precondition 141, 147, 148 chimpanzees (Pan troglodytes) social interactions 47, 77 gestures 168, 170, 172–174, 176–179, 187, 194–201, 204, 205 intelligence 96 vocal signals 166, 167 neonatal mirroring 223 mutual gaze 224 consolation 226
384 The Shared Mind
understanding intentions 226, 227, 230 deception 228 language 232 false belief 236 cingulate cortex (CC) 54, 55 co-evolution 215, 218, 222 cognitive empathy 217, 225, 226, 237 (see also empathy) cognitivism 359 collective intentionality 362, 363 (see also intentionality) collectivism 10, 290, 291 common code 40, 46 common knowledge 5, 10, 216, 220, 233, 239, 279, 280, 286–291, 298, 357–359, 361, 362, 373 commonsense psychology 41 communication 5, 171, 174, 176, 179, 219, 269, 308, 310, 327, 328, 372 animal/primate (see animal communication) infant 8, 90, 95, 97, 111, 126, 170 intentional 8, 9, 115, 116, 120–131, 134, 135, 137, 171, 172, 174, 175, 188, 190, 192, 193, 198, 208, 218, 220, 228, 229, 230, 269, 353 linguistic/verbal 11, 180, 307, 311, 312, 329 (see also language) non-verbal/gestural 68, 71, 75, 98, 172, 175, 218, 231, 264, 266, 269 (see also gestures) communicative 44, 47, 69, 71, 74, 79, 84, 89, 91, 115, 120, 121, 123, 126, 130, 134–136, 171, 190, 192, 197, 198, 233, 269, 317, 323, 325, 341, 351, 361, 364, 366, 372 actions/acts 52, 111, 187, 188, 189, 193, 203 competence ix, 342 contexts 90, 98–100, 109, 110
cues/signals 42, 175, 179, 196, 200, 224, 309 intention 10, 161, 178, 179, 216, 220, 230, 231, 234, 237, 238, 239, 259, 267 (see also intentions) intent indicators 122, 124, 125, 127, 131 signs 166, 169, 172, 175, 179, 188, 195, 196, 200, 212, 215, 224, 230, 231, 309, 310, 371, 374 (see also signs) system 188 triangle 263 view of language 259 complementation 323, 324, 326, 329 compositionality 219 conceptual categorization 92, 94, 110 development 49 expanations (vs. causal/ empirical) 4, 142, 147, 149 understanding 8, 49, 55, 58, 81 consciousness 54, 217, 219, 220, 223, 225, 237, 280, 282, 283, 286, 334, 347 and intuition 297 and language 10 human vi landscape of (see landscape of consciousness) pre- 222, 223 self- 85, 223 visual vii contagion 56, 57, 132, 219, 220 contextualization and ASL 11, 334–336, 342, 343, 344, 348, 350–352 convention/conventional 96, 98, 111, 166, 216, 220, 221, 236, 238, 255, 261, 267–269, 284, 285, 307–310, 312, 318, 359, 364, 373 signs 196, 232, 365 (see also sign) symbols 97, 165
use of objects 93, 99, 100, 103, 105, 106, 108, 110, 111 -normative 219, 233, 283 (see also norm) conventionalization 165, 178, 289, 296, 361, 362, 364 conversation 27, 29, 153, 242, 258, 260, 262, 323, 334, 340, 347 analysis 153, 154 in autism 72, 75 in children 25 proto- viii, ix, 117, 132, 143 conversational implicature 312 cooperation 11, 229, 342 correctness 10, 220, 287, 290, 291, 295, 296 cross-modal mapping 117, 219, 220, 223 culture vi, viii, 1, 93, 191, 218, 231, 253, 309, 315, 316, 344, 370, 371, 373 mimetic (see mimetic culture) traditional 225 customs 261 D declarative behaviour 204 knowledge 251, 252 pointing 9, 122, 145, 147, 152, 155–158, 175, 176, 202, 203, 205, 218–220, 263 sentence 283, 287 signals 309 decoupling 58 decontextualisation 8, 135–137 definitional criterion 221 issues 8, 142, 159 question 215 deictic elements 308 identification 372 gestures 165, 178, 179, 193, 218 (see also gestures) desire 120, 158, 372
and beliefs 18, 20–24, 27, 91, 110, 191, 220, 265, 315, 372 -based psychology 257 developmental change x, 43, 79, 153, 172, 192 level/stage 9, 172, 239 model 219, 237 process/trajectory 82, 86, 116, 131, 133, 371–373 psychology/theory/science ix, 12, 18, 20, 30, 76, 82, 84, 95, 99, 116, 119, 141, 148–150, 155, 259, 263 diachronic 297 linguistics 279, 280, 295 (see also historical linguistics) differentiation 8, 100, 219, 237 expression-content 220, 223, 372 means-ends 170 self-other 7, 8, 223, 225, 207 discourse 11, 217, 236, 262, 285, 311–317, 319, 321, 322, 329, 333–338, 340–352 disgust 55 distributed 187, 188, 189, 190, 191, 290 dramatic 31, 373, 374 re-enactments 33, 261 dualistic 189, 191 dyadic exchange 71 interaction 44, 47, 62, 148, 178, 180 mimesis 47, 219, 220, 225–229 (see also mimesis) E ecological psychology 3 egocentric 135 representation 51–53, 60, 61 (see also bodybased, first-person representation) embodied capabilities 25 comportment 22
Subject index 385
intentional relation 43 (see also intentional relations) interaction/engagement 3, 257 mind/knowledge vii, 1, 222, 288, 359, 364 person/agent 48, 68, 76 practice/action 20, 23, 34, 52, 267 schemas 268 state 49, 55 stories 261 emotion viii, 4, 8, 20, 22, 25, 32, 56, 57, 60, 62, 67, 79, 80, 85, 117–120, 126–128, 130–132, 134, 192, 203, 207, 217, 223, 225, 226, 228, 233, 363 contact 70 expressions 22, 44, 56, 60, 224, 254 intentional relation 41, 46, 49, 58 states 142–145, 166, 226, 359 empathy 3, 10, 54–57, 62, 217, 223–226, 228, 237, 301, 302, 359 (see also sympathy) cognitive 217, 225, 226, 237 enactive 251, 374, 375 re-enactive 10, 261 social perception 21 encephalisation 247, 256 enculturation in apes 10, 155, 228, 230, 231, 236, 238, 239 Environment of Evolutionary Adaptiveness 245 evolution 10, 167, 189, 194, 195, 218, 224, 229, 232, 237, 238, 246, 251, 258, 260, 327 of intersubjectivity 9, 221, 238, 327 of language 9, 166, 264, 327 of symbolic gestures 179 of triadic interactions 100 of triadic mimesis 218 (see also mimesis) “zone or proximal evolution” 232, 238
expansions in ASL 11, 335–339, 341, 345, 346, 348, 349, 351 expectations 28, 30, 31, 120, 198, 234, 238, 335, 370 exteroception 219 eye contact 69, 70, 78, 84, 85, 97, 125, 129, 130, 131, 133, 134, 157, 171 (see also mutual gaze) F face-to-face 22, 23, 77, 122, 127, 202, 238, 334, 375 facial expressions 7, 32, 34, 117, 126, 130, 198, 224, 250 farewells 67–69 first-person experience/knowledge 45, 48, 58, 61 information 7, 41, 43–46, 48, 50–55, 60, 220 model/simulation 19, 20 perspective 57, 61, 116, 132 (see also egocentric) representation 40, 52 theory of mind 60, 264 (see also theory of mind) verbs 324, 325 fMRI 32, 54, 56, 58 folk psychology 7, 10, 20, 25, 29, 30, 150, 215, 236 competence/abilities 28, 29, 35, 245, 249, 266, 270 explanation 28, 29 narratives 26, 28–31, 33, 42 reasoning 239 functional permanence 94 G gaze 7, 24, 47, 58, 70, 73, 77–80, 85, 116, 121, 125, 126, 129, 131, 134, 155–157, 176, 192, 200, 201, 207, 216, 225 alternation 119, 122, 125, 130, 132, 188, 193, 198, 199
386 The Shared Mind
following 46, 119, 125, 127, 129, 130, 132, 134, 136, 145–149 mutual (see mutual gaze) reading 125, 133, 134 referential 131, 133 generative grammar/linguistics 299, 300 gestures 34, 116, 119, 196, 218, 219, 249, 250, 255, 264, 268, 336, 364 and signed language 234 in apes 9, 165, 166, 168–170, 172–180, 195, 196, 198, 199, 203, 231, 269 in autism 70, 75, 78, 84 in children ix, 8, 22, 23, 99, 102, 105–109, 111, 121, 122, 125, 129, 132, 135, 149, 155–158, 175, 178–180, 193, 194, 203 goal emulation 248, 249 goal-directed agents 47 behaviour/action 24, 40, 45, 46, 50, 127, 134, 135, 153 gaze 134 intentionality 116, 119, 135 movement 21, 46 gradual 47, 97, 126, 127, 136, 223, 246, 268, 297, 324 grammar 309, 311, 317, 323, 327, 335–346, 348, 349, 350, 351 development 235 evolution 235, 262 linguistic 291, 296, 299, 301, 307 Panini 293, 299 grammaticalization 295, 296, 301 greetings 67–69 H hand-eye coordination 247 head nods/shakes 70–72, 75 heterochrony 204 historical linguistics 342 (see also diachronic linguistics)
hominid evolution 10, 204, 229, 231, 245–250, 252, 254–258, 260, 262, 267, 268, 270 Homo ergaster/erectus 10, 254, 255, 256, 257, 258, 262 Homo habilis 247 Homo sapiens 237, 246, 256 I iconic gestures 176, 179, 218, 219, 235 (see also gestures) sign 220, 235 (see also sign) signed language 235 (see also signed language) identification of individuals/particulars 258, 289, 308, 309 with others 6, 8, 32, 55, 68, 71–76, 81, 82, 86, 118, 126, 217, 223, 226, 237, 264, 265, 359, 372 with personas 34 images mental 248, 281, 285, 286, 342 motor x visual 57, 93 imagination 10, 48, 55–57, 218, 252, 262, 265 imitation 3, 10, 20, 44, 50, 52, 53, 56, 60, 82, 147, 179, 218, 219, 224–226, 247–252, 263, 269, 368–370 in autism 68, 73, 74, 75, 80 neonate 21, 117, 118 (see also neonatal mirroring) imperative gaze 129 gestures 177, 179, 180, 203, 269 (see also gestures) pointing 122, 158, 171, 175–177 (see also pointing) sentence 283, 284 index finger 79, 121, 156, 157, 158, 178, 196, 197 indexical 128, 219, 220 sign 106, 107, 220
individual intentionality 362 sharing 119, 120 individualism 10, 290, 291 infants 21–23, 40, 42, 46, 47, 52, 78, 79, 92–94, 100, 101, 116, 120, 122, 124–131, 133, 135–137, 141–149, 153, 157–159, 166, 168–172, 178, 179, 195, 198–200, 204, 205, 223–225, 246, 249, 250, 263, 363, 368, 369, 372 inferior parietal (IP) 51–53, 57, 58 innate intersubjectivity viii, 142 (see also intersubjectivity) theory of mind 10, 41, 257 (see also theory of mind) mappings 22, 45 contagion 44 (see also contagion) body schema 223 (see also body schema) intentionality 363 (see also intentionality) institutional facts 362, 365, 366 instrumental action 80, 118, 119, 122, 127, 134, 203, 205, 206, 269 intentionality 135 (see also intentionality) insula 54–57, 60 intentions vi, x, 8, 10, 18, 20, 22–25, 27, 29, 34, 102, 104–108, 110, 111, 119, 128, 132, 134–137, 146, 151–155, 159, 170, 171, 179, 198, 217, 219, 225–229, 231, 237, 259, 263, 265, 266, 360, 361 communicative (see communicative intention) discourse 334, 335, 352 joint (see joint intention) in-acting 265 intentional 21, 23, 24, 27, 30, 38, 42, 49, 146, 169, 170, 178, 180, 191, 258,
264, 267, 351, 352, 358, 360–363 agent 47, 52, 53, 58, 94, 128, 179 (see also agent) attitudes 33, 220, 251, 265, 266 communication 8, 9, 115, 116, 120–131, 134, 135, 137, 160, 171, 172, 174, 175, 188, 190, 192, 193, 198, 213, 218, 220, 228–230, 269, 353 (see also communication) islands 47 relations 7, 39, 40, 41, 43– 48, 50–54, 57, 59–62 (see also Intentional Relations Theory) schema 43, 52, 53, 60, 61 stance 23 Essentially intentional behaviour 123, 124 Intentional Relations Theory (IRT) 7, 39, 41–43, 48, 49, 50–53, 55, 58–62, 282 (see also intentional relations) intentionality 9, 23, 36, 44, 47, 58, 116, 127, 128, 135, 140, 141, 165, 166, 193, 222, 226, 362, 363 bodily (see bodily intentionality) collective (see collective intentionality) individual (see individual intentionality) instrumental (see instrumental intentionality) intention-based semantics 10, 259 interaction 1, 3, 4, 10, 17, 20–23, 30, 32, 46, 57, 58, 71, 78, 92, 99, 103, 109, 116–118, 120, 122, 129, 131, 133–136, 141–143, 152, 153–158, 188, 190, 192, 193, 202, 207, 224, 231, 255, 258, 267, 283, 284, 309, 310, 333, 334, 338, 346–348, 352, 363, 364, 373
Subject index 387
dyadic 44, 47, 62, 148, 178, 180 triadic 8, 44, 45, 47, 62, 89, 90, 91, 97, 98, 100, 104, 110, 111, 126, 148, 149, 335, 372 Interaction Theory 17 interaffectivity 115, 131, 133, 134, 136, 137, 220, 223 interattentionality 131, 132, 134 (see also joint attention) interintentionality 131–133, 135, 137 (see also joint intention) internal states 54, 62 interobjectivity 357, 364, 373 interpersonal 7, 30, 42, 46, 48, 58, 72, 73, 77, 80, 85, 97, 223 co-ordination 68, 71, 82 engagement 70, 74, 76, 81–84, 224 understanding 76, 81 interpretation between languages 11, 335, 337, 338, 339, 340, 349, 351, 352 functional 251, 360 of others 21, 29, 52, 57, 315, 360 rich (vs. lean) 198, 265 semantic 367 inter-rater reliability 74 intersubjective engagement 9, 34, 68, 72, 73, 79–81, 83, 85, 86, 141–144, 146, 155, 245, 255, 263–265 experiences 22 system 72 intersubjectivity 1–12, 40, 41, 68, 72–74, 76, 79, 81–84, 116–119, 123–135, 137, 141–144, 149, 154, 157, 158, 172, 177, 179, 180, 190, 215–218, 220–225, 228–232, 237–239, 265, 279, 280, 281, 301, 307–310, 312, 317, 320, 322, 323, 327–329, 333–335, 350, 357–359, 361–365, 372, 373
primary (see primary intersubjectivity) secondary(see secondary intersubjectivity) intuition 291, 293, 297, 300, 301 J joint action 90, 95, 97–101, 104, 110, 111, 357, 361–364, 370, 371, 373 joint attention 23, 24, 30, 46, 78–80, 82, 84, 90, 97, 111, 120, 122, 127, 129, 132, 146, 147, 148, 153–155, 215, 219, 224, 226, 228–230, 237–239, 263–268, 307–310, 329, 363, 372 (see also interattentionality) joint communicative action 89 joint intention 362 (see also interintentionality) K kinesthetic 22, 50, 53, 219 know-how (vs. knowledge that) 252 knowledge base 334 L landscape of action 33, 34 of consciousness 32–34 language 2, 4–6, 9–12, 25, 31, 42, 43, 47, 48, 84, 93, 96, 99, 102, 109–111, 115, 116, 135, 149, 152, 153, 155, 166, 167, 174, 176, 179, 191, 196, 197, 200, 203, 215–218, 221, 222, 228, 230–239, 245, 255, 256, 259–262, 264, 266, 268, 279, 280–285, 287–297, 299–302, 307–312, 315, 317, 322, 327–329, 334–338, 340– 346, 348–352, 357, 358, 361, 364–370, 372–374 signed (see signed language) language change 285, 297 (see also diachronic linguistics)
388 The Shared Mind
langue 284, 285, 297, 298 (see also parole) layered model 22, 219 left hemisphere 52 (see also brain) linguistics 6, 10, 11, 12, 279, 280, 283, 292, 295, 296, 299, 300, 302, 310, 324, 354 generative (see generative linguistics) historical (see historical linguistics) love 41–43, 49, 359 M matching between self and other 19, 40–45, 52, 55–57, 59, 61, 62, 117, 126, 131, 132, 224, 249 of direction 125 material grounding 288 meaning viii, 9, 22–24, 33, 34, 41, 42, 62, 80, 81, 89, 90, 92, 98–111, 121, 123, 130, 142, 151–155, 167, 195, 229, 233, 235, 309, 310, 357, 364, 365, 368, 369, 371, 373, 374 as use 123, 158 iconic 235 (see also iconic signs) intended 175, 177, 179 linguistic 1, 216, 229, 233, 260, 281–287, 289, 291, 293, 294, 296, 311–315, 317, 318, 321, 326, 328, 333, 335–338, 341–343, 346–352 shared 94, 98, 99, 101, 105, 106, 110, 229, 342, 349 speaker’s 259 referential (vs. emotional) 224 mechanisms for reading gaze 134 Medial Prefrontal Cortex (MPFC) 56–59 mental agents 49, 190, 236 image 281, 285, 286, 342 space 80, 322
state 9, 10, 18–20, 22, 24, 25, 33, 39–42, 48, 49, 56–59, 91, 126, 131, 132, 141, 143–145, 150, 151, 153, 165, 175, 190, 191, 203, 228, 290, 308, 309, 315, 327, 329, 358, 361 metacognition 50, 58 metarepresentational 59, 60, 235, 238, 245, 258, 259, 262, 263, 265, 266 methodological 73, 74, 83, 84, 147, 191, 202, 225, 292, 303, 317, 360, 369 individualism 12, 358, 362 mimesis 10, 215, 226, 231, 232, 233, 234, 237, 252, 255, 256, 262, 268 bodily 9, 214, 217–225, 238 hierarchy 217–219, 221, 222, 237, 238 abilities 250, 252, 255–257, 267 mimetic culture 255, 256, 261, 262, 267 schemas 267 skills 218, 253, 255 Mimetic Ability Hypothesis (MAH) 10, 245, 257, 261, 270 miming 220, 255 mind sharing 39, 61 (see also sharing) understanding 39 (see also understanding) -reading 18, 22, 118, 272 mirror neurons 7, 17, 19, 21, 22, 39, 40, 50, 52, 60, 117–119, 222, 223, 251, 264, 302 systems 222, 250, 264 mirror self-recognition 48, 219, 226 monkey 40, 45, 58, 166, 226, 271 motivation ix, 26, 81, 82, 128, 142, 188, 192, 193, 203, 204, 227, 229, 230, 255, 269, 338, 351, 369
evolutionary 224 motivational intentional relations 53, 54 (see also intentional relations) multimodal 43, 45, 51 mutual engagement 117–119, 133, 134, 207, 225 gaze 219, 224, 225, 240 (see also eye contact) sharing 119, 120, 124, 127, 132, 133, 307, 310 (see also sharing) N narrative 7, 18, 20, 25, 27–34, 49, 191, 236, 261, 262, 338, 339, 345, 357, 373–375 practice 17, 28, 29, 30, 236, 345 narrative practice hypothesis 7, 12, 17, 28, 236, 273 narrativity 12, 374 negation 307, 318–323, 326, 329 negotiation of meaning 130, 153, 319, 334–336, 340, 347– 351, 370 neonatal mirroring 219, 220 (see also imitation) neuroscience 6, 7, 12, 17, 19, 22, 39, 40, 49, 61, 62, 138, 218, 221, 222 (see also brain) non-linguistic conventions 255, 261 (see also convention) non-verbal communication (see communication) nonverbal reference 115, 116, 120–126, 128–130, 135 (see also referential) norms 17, 27, 28, 31, 233, 236, 255, 256, 269, 283, 289–293, 295–297, 300, 360, 361, 363, 364, 371–374 normativity 10–12, 27, 109–112, 216, 219, 220, 238, 279, 280, 281, 283, 289, 292–294, 296, 297, 302, 303, 307, 329, 330, 354, 359, 361–365, 367–372, 375
O object exchange 46, 372 permanence 92–94, 110, 113, 368 -directed 39, 40, 44–47, 51, 147 okkasionelle Bedeutung 284, 285 (see also usuelle Bedeutung) ontology 10, 279, 280, 286, 358, 363 ostensive 132 actions 102 denotation 25 gestures 102, 107, 108, 109, 111 signs 99–108 uses of objects 8, 101, 111 other minds 2, 11, 26, 62, 139, 221, 280, 307 out-of-body experience 53 P pain 3, 54, 55, 57, 339 parietal cortex 50–54, 58, 60, 220 parole 284, 285, 297 (see also langue) participation 343, 351, 357, 360, 361, 363–365, 371–373 pedagogy 256 perception-based understanding 22 perceptual categorization 92 intersubjectivity 215, 244 (see also intersubjectivity) re-enactments 248 personal 19, 27, 31, 40, 42, 46–48, 55, 56, 73, 80, 82, 83, 222, 258, 296, 315 engagement 69 relatedness 67 persons 4, 8, 18, 21, 22, 29–31, 33, 42, 67, 68, 76, 81, 82, 96, 128, 132, 141, 142, 145, 146, 149, 150, 152, 181, 221, 274, 280, 290, 309, 316, 340
Subject index 389
perspective shifting 263, 265 phenomenology 3, 4, 7, 18–20, 132, 221, 359 phylogeny 61, 77, 194, 217, 222, 253, 259 physical grooming 258, 261 Piagetian 95, 172, 182, 205, 208 pointing 8, 9, 47, 97, 99, 105–109, 111, 121, 122, 125, 126, 129, 134, 135, 145–149, 152, 154–158, 176, 186–188, 190, 193–198, 201–207, 220, 230, 279, 280 (proto)declarative (see declarative pointing) (proto)imperative (see imperative pointing) “whole-hand points” 195 points of view 13, 25, 48, 76, 251, 334 post-mimesis 219–221, 232, 234, 237, 238, 283 (see also mimesis) pragmatic 24, 25, 89, 90, 93, 98, 103, 109–111, 151, 221, 296, 312, 318, 328, 335, 338, 345 contexts 7, 17, 23, 31 intersubjectivity 23 pragmatics 97, 98, 266, 279, 283–285, 312, 335, 346 premotor cortex 50, 57, 63 preparatory (attention getting) behaviour 122 pretend play 25, 31, 91, 261, 373 pre-theoretical 285, 293, 294 primary intersubjectivity (see intersubjectivity) primary sensory areas 54, 55 private-language argument (PLA) 4, 279, 280–282 propositional attitudes 220, 237, 251, 258, 260, 265 reasoning 228 form 233 instructions 249 knowledge 238 representation 233 thought 234, 253, 259
proprioception 7, 21, 45, 46, 53, 219, 222–224, 250 proto-conversation viii, ix, 117, 132, 143 protolanguage 219, 221–238, 262–264 proto-mimesis 20, 219, 220, 223, 225, 237 (see also mimesis) proto-sign 264 psychological reality 299, 363 psychology 2, 4, 9, 12, 18, 74, 76, 91, 94–96, 98, 99, 116, 119, 121, 122, 170, 188, 189, 191, 202, 222, 246, 257, 258, 279, 284, 295, 298, 299, 300, 359 commonsense (see commonsense psychology) folk (see folk psychology) of language 298, 300 R rationality 289, 295 readiness to interact 129 reality vii, 83, 89, 90, 226, 286, 300, 313, 357, 360, 373 objective 360 psychological (see psychological reality) social (vs. “physical”) 8, 89, 91, 94, 97, 98, 99, 103, 109, 111, 363, 375 reasons 7, 25–31, 33, 150, 265 reciprocal miming 255 recreative imagination 245, 262, 265 referential 9, 90, 91, 115, 116, 121, 125, 127, 129, 131, 133–136, 153, 155, 165–167, 169, 175, 177, 180, 187, 206, 207, 224, 232, 233, 266, 267, 269, 308, 309, 327, 329, 347, 358 behaviour 122, 124 reading 129, 130, 134 explicit reference 9, 76, 113, 139, 160, 187, 194, 195
390 The Shared Mind
reflective 51, 57, 81, 82, 215 intersubjectivity 8 understanding (vs. nonreflective/pre-reflective) 9, 40, 58, 81, 82 thinking 150 representations 40, 43–48, 51–54, 56–62, 82, 83, 92, 96, 99, 117, 220, 229, 233, 249–251, 326, 336, 358, 367–369 dialogic cognitive 128, 230 material 371 mental 49, 359 mimetic 218 shared (see shared representations) social 360 representational 3, 10, 48, 96, 128, 200, 203, 215, 218, 220, 226, 232, 357, 358, 359, 360, 372 artifacts 91 mental states 39, 49, 127 meta- (see metarepresentational) mind viii, 1, 87 non-representational 200, 223, 251 relation 83, 229 resonance systems 17, 22, 264 response priming 248, 249 right hemisphere 32, 51, 52 (see also brain) role-reversal 308 rule 125, 135, 281–283, 291–294, 298, 315, 316, 360 rule-sentence 292–294 S secondary intersubjectivity (see intersubjectivity) second-person 8, 19, 117 self vii, 3, 4, 21, 24, 27, 81, 83, 90, 97, 99, 107, 109, 111, 117, 175, 179, 213, 218, 222, 224, 228, 231, 237, 255, 264, 281–283, 293, 296
and other xi, 2, 5, 8, 39–55, 57, 59–62, 82, 118, 132, 223 conscious xi, 35, 85, 223 feeling self 54, 56, 57 self/other contrast/ differentiation 225–237, 250, 265 self/other-orientation 73–75 self-other equivalence 7, 39, 41, 43, 118, 119, 126, 129 semantic 147, 151, 186, 218, 220, 221, 233, 234, 236–238, 266, 283, 296, 366, 367 semantics 322, 324 and pragmatics 279, 283–285, 287, 312 intention-based 10, 259 musical xi semiotic capacity 219 mediation 4, 89, 364, 365 systems 89, 90, 98–100, 102, 106, 111 sensory-motor 20, 25 shared attention 10, 23, 102, 144, 187, 203, 215, 219, 226, 237, 316 (see also joint attention) knowledge 11, 307, 322, 333, 335, 340, 346, 348 meanings 94, 98, 99, 101, 106 (see also meaning) representations 17, 19, 21 (see also representations) situations 23 use 90, 98, 100 world 13, 78, 79, 83, 264 sharing ix, xi, 1, 3, 6, 8, 32, 39, 40, 42, 43, 45, 48, 55, 61, 62, 67, 73–84, 86, 87, 115–120, 123–128, 131, 132, 137, 154, 160, 161, 172, 175, 179, 215, 234, 245, 250, 256, 259, 267, 269, 308, 309, 363 affective 56, 133 attention 78, 175, 308
individual (vs. mutual) 119, 120 looks 68, 73–75, 79 mental states 9, 42, 141, 143, 144, 250 mutual 119, 120, 124, 127, 132, 133, 307, 310 sign 93, 101, 104, 106, 108, 110, 111, 135, 154, 182, 184, 220, 229–232, 234, 235, 238, 308, 341, 365, 366, 367, 373 signal 47, 54, 122, 123, 125, 129, 134, 156, 169, 171, 174, 176–178, 254, 308, 313, 315, 327, 328, 351 signed language 216, 219, 232, 234, 235, 340 Swedish Sign Language 232 Nicaraguan Sign Language 234 American Sign Language (see Americal Sign Language) social x, 1, 3–7, 10–12, 20, 22, 30, 31, 32, 39–44, 47, 49, 54, 57–62, 68–70, 74, 76, 77, 79–85, 89–99, 104, 106, 109–111, 119, 120, 125–137, 141–143, 145–151, 153–158, 165–169, 171, 172, 176, 178–180, 188, 190, 192–194, 198, 200–203, 207, 218, 229, 231, 237, 238, 246, 254–258, 260, 263, 269, 279, 280, 282–292, 300, 302, 308, 309, 333, 343, 357–368, 371–375 cognition 2, 11, 12, 17, 18, 25, 149, 161, 169, 190, 213, 217, 220, 221 cohesion 245, 258, 261, 270 facts 11, 96, 287, 291, 357, 360–363, 367, 368, 374 function of gaze 129, 133 learning 179, 229, 247, 248, 252 ontology 290, 304 reality (see reality)
referencing 79–82, 119, 125, 127, 129–132, 145–148, 198, 263 somatosensory 50, 54 spatial neglect 53 speech acts 233, 268, 324, 328, 329 stage in development x, 6–8, 130, 137, 148, 149, 152, 172, 180, 193, 205, 217–221, 224, 238, 260, 265, 371 still-face 77, 192 Stimulus enhancement 248, 249 subpersonal 19, 24, 32, 34, 47, 250 Superior Temporal Sulcus (STS) 35, 50, 52, 56, 58 symbolic 97, 99, 100, 132, 191, 216, 220, 221, 232, 245, 269, 308, 357, 367 art 262 cognition 232 culture 1 gestures 178–180 language 260, 268, 270, 329, 372 play 31, 81, 83, 261 uses of objects 8, 99, 100–102, 107, 111 reference 206, 219, 221 representations 83 system 221, 235, 236, 372 thinking 86, 267 sympathetic imagination 55 sympathy vii, viii, ix, x, 2, 35, 55, 57 (see also empathy) synchronic linguistics 296 synchrony 44 systematically 73, 219, 285, 308, 312, 317, 329 systematicity 221–234, 238, 239, 320
Subject index 391
T taxonomic concept 6, 9, 141, 143, 148, 149, 158 Temporal/Pariental Junction (TPJ) 51, 52, 53, 57, 58, 59 theory of mind (ToM) 2, 7, 8, 10, 11, 17–19, 25, 29, 32–34, 39, 40–43, 49, 52, 57, 58–60, 82, 83, 90, 91, 92, 109, 146, 175, 203, 215, 216, 217, 222, 235, 238, 239, 245, 248–252, 254–259, 261–263, 266, 267, 360 modules 2, 10, 25, 222, 238, 239, 245, 246 theory theory (TT) 7, 17–19, 20, 41–43, 48, 59, 61, 62, 250–252, 265 simulation theory (ST) 7, 17, 18–20, 36, 42, 43, 48, 61, 251, 252, 264, 265, 272 third-order mentality 218, 226, 228–231, 233, 234, 236, 238, 288, 289 third-person 19, 26, 39, 57, 128, 220, 282, 323 information 4, 7, 41, 43–46, 48–54, 56, 58–62, 220 representation 51, 52, 62 (see also allocentric) stance/perspective 19, 39, 128 topos 315, 316, 318, 319, 322, 327 touch 55, 158, 172, 178, 225, 260, 279 transduction 188, 189 Transitory Magnetic Stimulation (TMS) 50, 55 triadic interaction 8, 44, 45, 47, 62, 89, 90, 91, 97, 98, 100,
104, 110, 111, 126, 148, 149, 335, 372 mimesis 10, 47, 166, 215, 218–220, 228–231, 233, 235, 237, 238, 268 (see also mimesis) truth-condition 91, 287 truth value 287 typological linguistics 279, 280, 295, 301 U understanding attention 129, 145, 147, 148, 149, 152, 158 (see also joint attention) belief 25, 26, 40, 226 (see also belief) communicative intentions 10, 134, 178, 234, 238 (see also communicative intentions) intention 9, 128, 153, 159, 265, 267 (see also joint intentions) others 3, 7, 17, 18, 20, 24, 25, 40, 43, 47, 49, 91, 216, 217, 222 usuelle Bedeutung 285, 296 (see also okkasionelle Bedeutung) V vision 45, 54, 109, 133, 219, 224, 298 volition 220, 223 W wave 69, 70, 75 (see also gestures) Z Zapotec 370 Zone of Proximal Development (ZPD) 231, 238
In the series Converging Evidence in Language and Communication Research the following titles have been published thus far or are scheduled for publication: 12 Zlatev, Jordan, Timothy P. Racine, Chris Sinha and Esa Itkonen (eds.): The Shared Mind. Perspectives on intersubjectivity. 2008. xiii, 391 pp. 11 Lewandowska-Tomaszczyk, Barbara (ed.): Asymmetric Events. 2008. xii, 287 pp. 10 Steen, Gerard J.: Finding Metaphor in Grammar and Usage. A methodological analysis of theory and research. 2007. xvi, 430 pp. 9 Lascaratou, Chryssoula: The Language of Pain. Expression or description? 2007. xii, 238 pp. 8 Plümacher, Martina and Peter Holz (eds.): Speaking of Colors and Odors. 2007. vi, 244 pp. 7 Sharifian, Farzad and Gary B. Palmer (eds.): Applied Cultural Linguistics. Implications for second language learning and intercultural communication. 2007. xiv, 170 pp. 6 Deignan, Alice: Metaphor and Corpus Linguistics. 2005. x, 236 pp. 5 Johansson, Sverker: Origins of Language. Constraints on hypotheses. 2005. xii, 346 pp. 4 Kertész, András: Cognitive Semantics and Scientific Knowledge. Case studies in the cognitive science of science. 2004. viii, 261 pp. 3 Louwerse, Max and Willie van Peer (eds.): Thematics. Interdisciplinary Studies. 2002. x, 448 pp. 2 Albertazzi, Liliana (ed.): Meaning and Cognition. A multidisciplinary approach. 2000. vi, 270 pp. 1 Horie, Kaoru (ed.): Complementation. Cognitive and functional perspectives. 2000. vi, 242 pp.