Following Putnam's Trail: On Realism and Other Issues. (Poznan Studies in the Philosophy of the Sciences & the Humanities)

FOLLOWING PUTNAM’S TRAIL

POZNA STUDIES IN THE PHILOSOPHY OF THE SCIENCES AND THE HUMANITIES VOLUME 95

EDITORS Leszek Nowak (founding editor) Jerzy Brzeziski Andrzej Klawiter Krzysztof astowski Izabella Nowakowa Katarzyna Paprzycka (editor-in-chief)

Marcin Paprzycki Piotr Przybysz (assistant editor) Michael J. Shaffer Piotr Ziemian (assistant editor)

ADVISORY COMMITTEE Joseph Agassi (Tel-Aviv) Étienne Balibar (Paris) Wolfgang Balzer (München) Mario Bunge (Montreal) Nancy Cartwright (London) Robert S. Cohen (Boston) Francesco Coniglione (Catania) Andrzej Falkiewicz (Wrocaw) Dagfinn Føllesdal (Oslo) Bert Hamminga (Tilburg) Jaakko Hintikka (Boston) Jacek J. Jadacki (Warszawa)

Jerzy Kmita (Pozna) Theo A.F. Kuipers (Groningen) Witold Marciszewski (Warszawa) Ilkka Niiniluoto (Helsinki) Günter Patzig (Göttingen) Jerzy Perzanowski (Kraków) Marian Przecki (Warszawa) Jan Such (Pozna) Max Urchs (Konstanz) Jan Woleski (Kraków) Ryszard Wójcicki (Warszawa)

Address: dr hab. Katarzyna Paprzycka . Department of Philosophy . University of Warsaw Krakowskie Przedmiecie 3 . 00-927 Warszawa . Poland . fax ++48(0)22-826-5734 E-mail: [email protected]

POZNA STUDIES IN THE PHILOSOPHY OF THE SCIENCES AND THE HUMANITIES, VOLUME 95

FOLLOWING PUTNAM’S TRAIL ON REALISM AND OTHER ISSUES

Edited by María Uxía Rivas Monroy, Celeste Cancela Silva, and Concha Martínez Vidal

Amsterdam - New York, NY 2008

The paper on which this book is printed meets the requirements of "ISO 9706:1994, Information and documentation - Paper for documents Requirements for permanence". ISSN 0303-8157 ISBN: 978-90-420-2397-0 ©Editions Rodopi B.V., Amsterdam - New York, NY 2008 Printed in The Netherlands

CONTENTS

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Celeste Cancela Silva, Introduction: Putnam and the Notion of “Reality” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

7 9

PART I. HILARY PUTNAM’S PRAGMATISM

Hilary Putnam, What Makes Pragmatism So Different? . . . . . . . . Juan José Acero, Pragmatism, Pluralism, and the Peirce Principle . . Christopher Hookway, Dichotomies: Facts and Epistemic Values . . Jaime Nubiola, Dichotomies and Artifacts: A Reply to Professor Hookway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

19 35 55 71

PART II. PRAGMATISM AND HUMANISM

Maria Baghramian, Three Pragmatisms: Putnam, Rorty, and Brandom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 Ulvi Doguoglu, Putnam beyond Putnam: Understanding, Pragmatism, Humanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Russell B. Goodman, Some Sources of Putnam’s Pragmatism . . . . . 125 Chiara Tabet, Putnam and the “God’s-Eye View”: On the Logical Structure of Anti-Foundationalist Pragmatism . . . . . . . . . . . . 141 PART III. REALISMS

José Tomás Alvarado, Conceptual Relativity and Structures of Explanation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Celeste Cancela Silva, Conceptual Relativity and SpeakingSensitive Semantics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Matti Eklund, Putnam on Ontology . . . . . . . . . . . . . . . . . . . . . . Fernando González García, María Uxía Rivas Monroy, The Pragmatic Realism of Hilary Putnam . . . . . . . . . . . . . . . . . .

163 185 203 223

PART IV. FACT/VALUE DICHOTOMY

Ángel Manuel Faerna, Moral Disagreement and the “Fact/Value Entanglement” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245

6 Francisco Javier Gil Martín, Jesús Vega Encabo, Truth and Moral Objectivity: Procedural Realism in Putnam’s Pragmatism . . . . 265 Óscar L. González-Castán, The Fact/Value Entanglement as a Linguistic Illusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287 About the Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

PREFACE

The papers collected in this volume derive from the II Meeting on Pragmatism. Hilary Putnam’s Pragmatism, a conference held in Santiago de Compostela (Spain), from May the 13 th to May 14 th 2004. The conference was organized by the editors of this volume. This Meeting on Pragmatism was the second edition of a series of conferences dedicated to pragmatism. Two years before the I Meeting on Pragmatism was also held in Santiago de Compostela. The subject for that first edition was more general, namely, the strength, relevance and interest of the American classical Pragmatists as C.S. Peirce, W. James and J. Dewey. On that occasion a small number of well known specialists from England, USA and Spain were invited, specifically C. Hookway (University of Sheffield), N. Houser (Indiana University), Ángel Manuel Faerna (University of Castilla-La Mancha) and Ramón del Castillo (UNED). In this second edition the conference was devoted to Hilary Putnam’s pragmatism. He, as philosopher of our time, is doing a great philosophical job maintaining the philosophical arena alive, and connecting past and present. Something interesting about his work is how pragmatism and analytical philosophy are going together; sometimes to criticize the contributions of some analytical philosophers – as the members of the Vienna Circle, and sometimes to pay attention to the pragmatic insights of other analytical philosophers – as Austin or Wittgenstein. The content of this volume is divided in four parts. The first one – entitled “Hilary Putnam’s Pragmatism” – includes the papers and comments on them by the invited professors. The other three parts of the volume “Pragmatism and Humanism,” “Realisms” and “Fact/Value Dichotomy” respectively, include the eleven papers that conformed the II Meeting on Pragmatism. All the papers included in this volume are devoted to the analysis of some of the many aspects of Putnam’s pragmatism. The Area of Logic and Philosophy of Science of the Department of Logic and Moral Philosophy of the University of Santiago de Compostela organized this conference. The editors of the present volume served as the Organizing Committee for this event: Mª Uxía Rivas Monroy

8

Preface

(University of Santiago de Compostela), Concha Martínez (University of Santiago de Compostela) and Celeste Cancela (University of Santiago de Compostela). Juan José Acero (University of Granada), Christopher Hookway (University of Sheffield), Jaime Nubiola (University of Navarra), Eulalia Pérez Sedeño (Institute of Philosophy, CSIC), Hilary Putnam (Harvard University), and Juan Vázquez Sánchez (University of Santiago de Compostela), were the members of the Scientific Committee of the Conference. Without their help the selection process of the proposed contributions would have been even more difficult. This event was sponsored by the Ministerio de Ciencia y Tecnología, Xunta de Galicia (Consellería de Educación e Ordenación Universitaria), Xunta de Galicia (Consellería de Innovación, Industria e Comercio. Dirección Xeral de Investigación e Desenvolvemento), and Universidade de Santiago de Compostela. Without their support the conference would have not been possible. We would also like to thank the academic and organizational support provided by the Sociedad Española de Filosofía Analítca and the Sociedad de Lógica, Metodología y Filosofía de la Ciencia en España. Finally, we want to thank Hilary Putnam for contributing so exhaustively to this conference, since he not only gave a lecture, but also provided detailed comments to all the papers presented. We are also very greatful to C. Hookway, Juan José Acero and Jaime Nubiola who are always ready to collaborate generously with us in our initiatives, and who generously agreed to participate in various ways in the II Meeting on Pragmatism. We also want to show our gratitude to all the members of the Area of Logic of our Department for their availability to help us whenever we need them. Santiago de Compostela, November 2004 María Uxía Rivas Monroy, Celeste Cancela Silva, Concha Martínez Vidal

Celeste Cancela Silva INTRODUCTION: PUTNAM AND THE NOTION OF “REALITY”

ABSTRACT. The main aim of this brief paper is to present in which sense both Putnam’s internal realism and his natural realism are realist. In order to do that I will take into consideration the notion of “reality” that Putnam uses in his defense of his internalist perspective as well in his present natural realism. Although there exist significant differences between Putnam’s two different realist stances, the notion of “reality” present in each of them remains the same. In this paper I will clarify what Putnam understands by that notion.

1. The Many Faces of Hilary Putnam’s Realism I would like to distinguish two different periods in Hilary Putnam’s philosophical trajectory. The first period from 1976 to the 90’s decade, is characterized by internal realism. The second period is characterized by the most pragmatist of all his realisms and is called “natural realism” or “common-sense realism.” Although Putnam possesses certainly an amazing Russellean ability to change his mind, that should not lead us to considering that his philosophical trajectory is characterized by consecutive leaps that cause a certain discontinuity in his thinking. It would be more correct to see the various changes along Putnam’s philosophical route as an evolution of his philosophical journey rather than as a conversion from one view to another, i.e., that journey is a continuous evolution of his thinking. Putnam himself admits that when he writes: “I have, indeed, changed my mind on a number of other philosophical issues at various times, but I do not see any of those changes as large enough to constitute a genuine discontinuity” (Putnam 1992, p. 350).

This paper has been carried out with a Postgraduate Grant from the Fundación Caja Madrid.

In: M.U. Rivas Monroy, C. Cancela Silva, and C. Martínez Vidal (eds.), Following Putnam’s Trail: On Realism and Other Issues (Pozna Studies in the Philosophy of the Sciences and the Humanities, vol. 95), pp. 9-16. Amsterdam/New York, NY: Rodopi, 2008.

10

Celeste Cancela Silva

As I have said above, Putnam has evolved from being a metaphysical realist to an internal realist to calling himself presently a common-sense realist or natural realist. His internal realism 1 tries to be a view that rejects two opposed doctrines: metaphysical realism and cultural relativism. In (1999) Putnam summarizes his intelectual autobiography of the latest twenty years and explains how he reached the formulation of his internal realism. Then he goes on to describe briefly his position, defended under the label ‘internal realism’, and, alter pointing out where he lost his way, he begins the presentation of his present stance, i.e., natural realism. Nowadays there are, from my point of view, two reasons to keep considering Putnam as internalist: on one hand his insistence to avoid metaphysical realism and cultural relativism and on the other hand his defense of the thesis of conceptual relativity, which constitutes the core of his internal realism. Putnam’s current ideas about truth and perception mark the difference between his internal realism and his internal realism. He expresses his step from internal realism to natural realism as: I came to see that the sense of a “gap” between language and the world cannot be overcome unless we first overcome the idea of a gap between perception and the world. This interest in developing a form of what used to be called “direct realism” was also encouraged by my long involvement with the thought of Austin and Wittgenstein and, in the last ten years, with the work of William James, whose radical empiricism I regard as a brilliant, if ultimately flawed, attempt to reinstitute what he himself called “natural realism” with respect to perception. (Putnam 1992, p. 356)

1 The history of this term is explained by Hilary Putnam in his (1992, pp. 352-353) and in (1999, p. 182n). He introduces the term ‘internal realism’ in his paper in (1978) to refer to the position that he had already held in previous writings, scientific realism. However, his readers misunderstood that term as a new proposal of Putnam’s. He himself explains this saying: “Some readers, misled by a careless reading of a sentence in “Realism and Reason,” started referring to what they took to be my new position [. . .]. In “Realism and Reason” internal realism was not a term for my new position; it was rather a term for a kind of scientific realism I had already accepted for some years, for a position (I now argued) both realists and antirealists could accept. But I soon discovered that everyone else was using the term as a name for my new position (or for whatever they took that position to be) [. . .] in Reason, Truth and History I capitulated to the fashion of calling whatever Putnam’s new position happened to be “internal realism”” (Putnam 1999, p. 182n). Although, as Putnam himself recognizes, that the label ‘internal realism’ is highly confusing, I will use this term to refer to the philosophical perspective defended by Putnam in the first period mentioned above.

Introduction: Putnam and the Notion of “Reality”

11

In this paper I am interested in clarifying what Putnam understands as “reality” both when he calls himself an “internal realist” and when he calls himself a “natural realist.”

2. First Approach to the Notion of “Reality” in Putnam Quine opens his paper “On What There Is?” with the following words: A curious thing about the ontological problem is its simplicity. It can be put in three Anglo-Saxon monosyllabes: ‘What is there?’ It can be answered, moreover, in a word – ‘Everything’ – and everyone will accept his answer as true. However, this is merely to say that there is what there is. There remains room for disagreement over cases; and so the issue has stayed alive down the centuries. (Quine 1980, p. 1)

As Quine says the main question that ontology has to elucidate is “What is there?”. We could try to answer it as if we were lunatics enumerating the things that exist such as my niece Mayte, the computer screen that I have in front of me, immigration laws, sad events, red things, real numbers, black holes, the bipolar syndrome, radio programs, scientific theories, symphonies, García Márquez’s magical realism, Park Avenue, poems, twin prime numbers, Macondo, Piccadilly Circus, philosophers, clouds, computer viruses, mereological sums, etc. Most probably, answering the above question in that way would not lead us anywhere, and it would leave unresolved the matter of whether all the things listed exist in the same way, i.e., whether they exist in the same sense. For instance, it does not seem the case that real numbers exist in the same way that my niece Mayte exists, or in the same way that the town of Macondo exists, or in the same way that my computer screen exists. Furthermore, although from a theory with a given discourse universe we could admit the existence of some of those things enumerated, some others would not “exist” from the point of view of that very same theory. The term ‘exist’, just like a few other language terms, does not have a unique sense but many, so that saying things such as “Bugs Bunny exists” would leave us as ignorant as saying “Bugs Bunny does not exist” if we do not specify the sense of ‘exist’ in those sentences. Moreover, stating in an imaginary philosophy paper that Bugs Bunny exists or not does not have the same importance or relevance as the same assertion (or its denial) is used in a book about children’s characters. Explaining that Bugs Bunny does not exist to a child who not only watches him on TV

12

Celeste Cancela Silva

but also sleeps with him every night may be a hard task for the adult and a traumatic one for the infant. However, vagueness and imprecision are just as present in the sentences “Bugs Bunny exists” and “Reality does not exist independent of the human mind” (and this is a sentence that does have a certain relevance and frequency in philosophical texts). Merely stating that reality exists or not independently of the human mind without specifying in which sense the terms ‘exist’ and ‘independence’ are being used turns that sentence (both in the assertive and in the negative form) into an empty sentence. We will not know what we are talking about as long as we do not say what we are understanding by ‘exist’ and ‘independence’ (and assuming that we also ignore the question of what we understand by ‘reality’). In papers dealing with Putnam’s internal realism when presenting the key ideas of internal realism the following quote is frequent: ““Objects” do not exist independently of our conceptual schemes” (Putnam 1981, p. 52). The mistake that some authors who use that quote in their texts make (and I have myself fallen into it several times) is not taking into account the context in which Putnam uses the sentence in his work, therefore making the content of that quote not only ambiguous but also highly confusing. So, what does Putnam mean by saying that ““Objects” do not exist independently of our conceptual schemes”? If we analyze first of all the term ‘independence’, in which sense can we say that “objects” are not independent of conceptual schemes: is it true that “objects” depend empirically on conceptual schemes or is this dependence metaphysical, in which case we would have to open a new series of questions and consider in which sense we can say that something is metaphysically dependent? Putnam writes the following about “independent existence”: Talk of “independent existence” makes little sense when what is at stake is neither ordinary causal nor ordinary logical independence. That the sky is blue is causally independent of the way we talk; for, with our language in place, we can certainly say that the sky would still be blue even if we did not use color words [. . .]. And the statement that the sky is blue is, in the ordinary sense of ‘logically independent’, logically independent of any description that one might give of our use of color words. For these reasons, I have, especially in recent years, tried not to state my own doctrine as a doctrine of the dependence of the way things are on the way we talk. In any sense of ‘independent’ I can understand, whether the sky is blue is independent of the way we talk. (Putnam 1994, p. 301)

Introduction: Putnam and the Notion of “Reality”

13

3. Putnam’s Notion of “Reality” Putnam’s internal realism appears as an alternative to a philosophical stance that he had staunchly defended, i.e., metaphysical realism. The main defect affecting this perspective and that in Putnam’s point of view makes the doctrine unintelligible is its use of the term ‘reality’. One of the theses that defines metaphysical realism is considering that reality coincides with a fixed totality of objects that exist independently of our mind 2. Putnam’s internal realism is the result of rejecting the previous thesis as well as any others derived from it. However, at this point we have to be very cautious if we do not want internal realism to be as unintelligible as metaphysical realism. When we say that Putnam rejects the commitment of the metaphysical realist that there exists a reality in-itself that we are supposed to be able to account for, we do not mean that Putnam defends the rejection of this thesis; i.e., the assertion “noumenal reality does not exist independent of the human mind” could be included in the corpus of the main ideas of internal realism, since this sentence is as unintelligible as its negation. If the independent nature of reality is postulated in the metaphysical sense as the traditional metaphysical realist does, then the notion of reality at stake is a notion of reality as pure, transcendental reality. But what fails is not only every attempt to show the existence of this reality, but also every attempt to enunciate something about it, since that would imply describing that reality in a language alien to human beings. It is there where we encounter again the fundamental mistake, highlighted by Putnam, of thinking that if we cannot describe reality without using any conceptual scheme, then our descriptions of reality do not describe reality as it is either. And here we stumble again with the deluding shadow of a reality different from the one in our descriptions. The fact that we use our language to talk about reality does not make this reality less real. Our descriptions of reality deal with reality just as it really is. The notion of reality used by Putnam is one purifies of any metaphysical burden. The objects, the facts, that our descriptions refer to are no less real just because our descriptions are “shaped by our interests and nature” (Putnam 1995, p. 29). The inveterate metaphysical realist could tell us that we should not conclude that this Reality (with a capital R since I am referring to this reality as if it were a Superthing) does not exist from the fact that we cannot explain what it is like. But Putnam does not say it does not exist. 2

It is precisely the rejection of this thesis together with all the implications derived from it what constitutes the identity of internal realism.

14

Celeste Cancela Silva

The Kantian presence is fundamental at this point: both affirming and rejecting the existence of a noumenal reality is meaningless, which does not imply that, in some sense, the notion of noumenal reality is not intelligible. What we are saying is that the use that metaphysical realism makes of this notion is unintelligible. Metaphysical realism is unintelligible from its very first thesis, that Putnam understands as the “comfortable assumption that there is one definite totality of objects that can be classified and one definite totality of all properties” (Putnam 1999, p. 7). Nevertheless, abandoning the previous kin of realism does not mean we have to defend a diametrically opposed to and equally unsatisfying position: idealism. The starting point of Putnam’s internalism is a crossroads with a no-entry sign in each of them, the different roads being metaphysical realism and idealism. Internal realism is a middle position which, even if we assume that the world we can account for has already been conceptualized, does not fall into the clutches of a devastating constructivism. The philosophical perplexities provoked by the notion of “reality” have led us to the two already mentioned roads that Putnam avoids. Both roads constitute opposed and mistaken positions. It is as if the negation of one of them led irrevocably to the other, but in this case, using Putnam’s achievement has been knowing how to stop swinging between two disastrous pictures of the world, a pendulum movement to which philosophical thinking seems to be condemned to. What Putnam’s famous thesis of conceptual relativity shows is our incapacity to answer the question “How many objects are there?” independent of any theory or conceptual scheme. This incapacity is, in Putnam’s terms, impotence. And from our impotence to describe reality without using at the same time the only means we possess for speaking about it, our languages, the illusion or rather the fantasy that there is a reality beyond the one described by our languages, a reality which is different from the reality that our different theories talk about. It is as if we described reality with the help from our languages, but not reality as it really is. Beyond that reality polluted by our languages there is an aseptic reality, which would need, in turn, a special language to describe it, a language of its own kind, i.e., aseptic. And, from what should be aseptic? Here there are not many options but to answer that it should be aseptic from human beings. In “Sense, Nonsense and the Senses” Putnam admits that some terms like ‘reality’ are the “sources of deep philosophical puzzlement” (Putnam 1999, p. 9). At the same time he considers that “the notion that our words

Introduction: Putnam and the Notion of “Reality”

15

and life are constrained by a reality not of our invention plays a deep role in our lives and is to be respected. The source of the puzzlement lied in the common philosophical error of supposing that the term reality must refer to a single superthing instead of looking at the ways in which we endlessly renegotiate – and forced to renegotiate – our notion of reality as our language and our life develop” (Putnam 1999, p. 9). The philosopher “seems to be telling us of an Impotence, in the way the physicist tells us of an Impotence when he says “You can’t build a perpetual motion machine,” or of a barrier we can’t cross, but it turns out on examination that the barrier is a mirage, or even less than a mirage – that it is chimerical” (Putnam 1995, p. 40). The impotence about which the philosopher is talking is that “we can learn and change and invent languages, and in them we can state truths; that is describing reality. If you say, “Yes, but it is not describing reality as it is itself,” you are saying nothing” (Putnam 1995, p. 40).

4. Conclusions My interest in this paper has been to clarify the notion of “reality” in Putnam’s philosophy. I think that Putnam’s philosophy has been misunderstood very often due to the meaning of this notion. What I have intended to present here is that there is a common-sense realism shared by both the internal realism and natural realism defended by Putnam in the different periods of his thinking. This common-sense realism is in Putnam’s words: “the realism that says that mountains and starts are not created by language and thought, and are not parts of language and thought, and yet can be described by language and thought” (Putnam 1994, p. 303). But his native realism has nothing to do with a more metaphysical versions of realism, namely, with the versions that say that “there is – in a philosophically privileged sense of ‘object’ – a definite Totality of All Real Objects and a fact of the matter as to which properties of those objects are the intrinsic properties and which are, in some sense, perspectival” (Putnam 1994, p. 303).

Acknowledgements I am very grateful to Hilary Putnam for comments on a previous version and help.

16

Celeste Cancela Silva

University of Santiago de Compostela Department of Logic and Moral Philosophy Faculty of Philosophy 15782 Santiago de Compostela, Spain e-mail: [email protected]

REFERENCES Putnam, H. (1978). Realism and Reason. In: Meaning and the Moral Sciences, pp. 123-140. London: Routledge & Kegan Paul. Putnam, H. (1981). Reason, Truth and History. Cambridge: Cambridge University Press. Putnam, H. (1990). Realism with a Human Face. Edited by James Conant. Cambridge, MA: Harvard University Press. Putnam, H. (1992). Reply to Gary Ebbs. Philosophical Topics: The Philosophy of Hilary Putnam 20 (1), 347-358. Putnam, H. (1994). Words and Life. Cambridge, MA: Harvard University Press. Putnam, H. (1995). Pragmatism. Oxford: Blackwell. Putnam, H. (1999). Sense, Nonsense, and the Senses. In: The Threefold Cord. Mind, Body, and World, pp. 1-70. New York: Columbia University Press. Quine, W.V.O. (1980). From a Logical Point of View. Cambridge, MA: Harvard University Press.

PART I HILARY PUTNAM’S PRAGMATISM

Hilary Putnam WHAT MAKES PRAGMATISM SO DIFFERENT?

ABSTRACT. As the title of this essay indicates, my concern will be with the ways in which pragmatism is a unique metaphysical tradition. This is something I have written about before, 1 but in many quarters the idea still persists that pragmatism must be either the denial (à la Rorty) 2 that there is such a thing as an objectively warranted idea; or, on the other hand, just an outdated early twentieth century American movement with no real importance today. To show that it is neither of these, I propose to compare pragmatism with, on the one hand, the materialism or “naturalism” which dominates the thinking of most analytic philosophers who do metaphysics and epistemology today, and, on the other hand, the traditional understanding (which goes back to Aristotle) of what “metaphysics” is supposed to be. At the conclusion of this essay I shall also say a few words about the respects in which pragmatism also differs from European existentialism, a movement which shares pragmatism’s aspiration to break loose from the traditional project of metaphysics, without (pacé Rorty and Brandom)3 falling into relativism or nihilism.

1 This paper has a precursor in a lecture titled “The Uniqueness of Pragmatism” (unpublished) that I delivered on Dec. 13, 2004 at a conference on Pragmatism and the Philosophy of Technology at the Tokyo Center for Philosophy of the University of Tokyo (Komaba campus). See also my (1995). 2 In his (1993) Rorty writes, “I view warrant as a sociological matter, to be ascertained by observing the reception of S’s statement by her peers.” Strangely, he objects – in the same essay! – to being called a relativist. 3 Robert Brandom shares Rorty’s view that the pragmatists simply identified truth with belief that satisfies the desires of the believer. In my reply to Brandom in (Conant and Zeglen 2002), I show that this interpretation is “text free” – that is, it simply ignores what the pragmatists actually said.

In: M.U. Rivas Monroy, C. Cancela Silva, and C. Martínez Vidal (eds.), Following Putnam’s Trail: On Realism and Other Issues (Pozna Studies in the Philosophy of the Sciences and the Humanities, vol. 95), pp. 19-34. Amsterdam/New York, NY: Rodopi, 2008.

20

Hilary Putnam

1. Materialism Today a large majority of analytic philosophers call themselves “naturalists.” Although the term ‘naturalism’ was frequently used by John Dewey, Dewey did not mean by it either materialism or reductionism, as he made clear when he wrote: “The obnoxious quality of materialism is due to its depression of thought which is treated as an illusion or at most an accidental by-product” (Dewey [1929] 1981, p. 87, emphasis added). But what today’s “naturalists” mean by the word is a sophisticated (or “Post-Quinian”) materialism, and not merely the rejection of appeals to supernatural entities in philosophy. (For that reason, I shall usually refer to them in what follows as “materialists.”) The most sophisticated naturalist-cum-materialist of the last fifty years was not Quine, however, but the late Bernard Williams4 (who was, to be sure, deeply influenced by Quine). Williams’ most famous claim was that ideally a completed physics – and only a completed physics – can give us a complete and maximally non-perspectival view of the world as in is in itself (or as he preferred to say, “as it is anyway”). He called this hypothetical completed physics “the absolute conception of the world.” “The reason I write “a completed physics” and not simply “a completed science” here is that has elsewhere said explicitly that the notion of an “absolute conception” does not “look too pale” because we have a conception of “what an adequate physics might look like”” (Williams 1978, p. 247). One central difficulty with such a blatant materialism is the problem of accounting for semantical properties. Williams says that such properties are “perspectival” (1978), but what exactly is that supposed to mean? Well, one thing it means is that they aren’t “absolute” properties of the world; Williams says explicitly that that “the world itself has only primary qualities” (1978, p. 247). Given what Williams says about the absolute conception (e.g., that it contains only primary qualities – and, presumably, what is definable in terms of primary qualities using the apparatus of mathematical physics), and what he says about semantic relations, it follows from Williams’ premises that an account of them cannot figure in the absolute conception. And given his characterization of the absolute conception, it is not hard to see that he is right. What

4 Williams views were set out at most length in (1978) and (1985). I criticized Williams’ views in (1990a), (1992a, “Bernard Williams and the Absolute Conception of the World”), and (1994a). Williams replied to these criticisms in (2000), and I replied to this response in (2001).

What Makes Pragmatism So Different?

21

would an account of a semantical relation in the language of mathematical physics (or in terms of “primary qualities”) be? Thus, Williams, on my reading of him, found himself in the position of needing either to reduce semantical facts to purely physical facts or to become an outright denier of the reality of the semantical (an “eliminationist,” in the jargon of contemporary analytic philosophy).5 And in the book I just mentioned he did end up 6 by suggesting that eliminationism with respect to the semantical is probably the right line to take. For he wrote: [I]f the various sorts of considerations [Quinian and Davidsonian considerations] which have been summarily sketched here are correct, then we have to give up not just dualism but the belief in the determinacy of the mental. These considerations converge on the conclusion that there are no fully determinate contents of the world which are its psychological contents. (Williams 1978, p. 300, emphasis added)

Indeterminacy of “psychological contents” (in the sense of Quine and Davidson) is precisely indeterminacy of the semantical. And that indeterminacy, if their arguments are accepted, is far reaching indeed. (Quine says, astonishingly, for example, that there is no fact of the matter as to whether the name ‘Tabitha’ refers to his cat Tabitha or to the whole cosmos minus the cat [Quine 1990]!) It is true that Donald Davidson, while giving full credit to Quine for the arguments that allegedly establish “indeterminacy,” claimed that the extent of indeterminacy is much reduced in his theory, by his willingness to make “a more far reaching application of the principle of charity” than Quine (Davidson 1984, p. 228). (Perhaps this is the reason that Williams wrote “no fully determinate contents” and not simply “no determinate contents.”) But charity is a maxim of translation into the interpreter’s home language and it cannot bestow any additional determinacy on the 5 In current parlance, an “eliminationist” (Paul Churchland is the paradigm case) holds that propositional attitudes (e.g., belief) and semantic relations (e.g., reference) don’t exist, and that the idea that they do is a superstition (Churchland has compared it to belief in witches, or in phlogiston). Cf. “Activation Vectors vs. Propositional Attitudes: How the Brain Represents Reality,” in Churchland (1998). (In my view, Quine’s claim that reference is indeterminate to the extent that there is no fact of the matter as to which object ‘Tabitha’ refers to is just eliminationism under another name.) I criticize the cited essay of Churchland’s in my (1992b). Note that eliminationism is very different from reductionism. 6 I say “end up” because earlier in that work he favors the project of accounting for all the “perspectives” and the properties they project onto the world in absolute terms. By the closing pages of the book (Williams 1978, pp. 300ff ), however, this enterprise has been ruefully renounced as too ambitious.

22

Hilary Putnam

home language itself. It isn’t as if one could really argue that reference is partially, even if not “fully,” determinate, except in the sense (which even Rorty could agree with) of “determinate relative to a translation scheme.” Dewey’s remark that thought “is treated as an illusion or at most an accidental by-product” by materialists was right on the mark! For semantical properties are precisely the properties that we need to describe thoughts as thoughts. Indeed, the view that only the primary qualities of a completed future physics are genuine properties of the world (“the world itself has only primary qualities”) would banish much more than the semantical properties to the hazy realm of “indeterminacy.” (In Ethics and the Limits of Philosophy Williams says that ‘grass’ and ‘green’ are terms that would not occur in finished science!) As a consequence, this view would seem to banish such “special sciences” as economics. A similar criticism was made by the economist-philosopher Vivian Walsh, who wrote: Economists cannot afford to neglect the failure of an advertising campaign that tried to sell a shade of green which consumers rejected, or the devastating results of a record drought upon grasslands. The things consumers [and clients!] want, or buy, or have produced for them, are chosen or rejected in terms of features that arguably would not appear in “completed science” if it should ever arrive. They live, move, and have their being, just like those who make moral statements, on the “wrong” side of the dichotomy between “finished science” and everything else that anyone ever says. (Walsh 2000, p. 9)

2. Supervenience American materialists are usually more coy about the real content of their view than Bernard Williams was. Rather than say that the only terms that would appear in a complete and non-perspectival account of reality are terms for primary properties, they usually say that what Williams called the “perspectival” terms (including terms for psychological states) are “supervenient” on the primary qualities.7 But what is “supervenience”? The question is a difficult one, and the difficulty reminds me of an episode in the history of late logical positivist philosophy of science. As late as 1936, Rudolf Carnap still sought a way to “reduce” all the terms of physics to “observation terms” such as ‘blue’ and ‘touches’ (for 7

See my discussion of Jaegwon Kim’s highly influential views in my (1999).

What Makes Pragmatism So Different?

23

Carnap, as later for Bernard Williams, physics was the paradigm of serious scientific knowledge). But in 19398 Carnap gave up this project, and settled for taking the “theoretical terms” of physics as primitives, while claiming that there was still a profound difference between these terms, which, he said, were only “incompletely interpreted” and the “completely interpreted” observation terms. In a lecture titled “What Theories are Not” that I gave to an international congress in Philosophy of Science in 1960, I pointed out that The notion of partial interpretation has a rather strange history - the term certainly has a technical ring to it, and someone encountering it in Carnap’s writings, or Hempel’s, or mine certainly would be justified in supposing that it was a term from mathematical logic whose exact definition was supposed to be too well known to need repetition. The sad fact is that this is not so! The term was introduced by Carnap in a section of his monograph [Carnap 1939, pp. 61ff ], without definition (Carnap asserted that to interpret the observation terms of a calculus is automatically to “partially interpret” the theoretical primitives, without explanation), and has subsequently been used by Carnap and other authors (including myself)9 with copious cross references, but with no further explanation. (1962, pp. 244-245)

I believe that the situation is very similar with respect to the term ‘supervenience’, except that in the case of this term many definitions have been proposed. The problem, however, is that the term is used as if the definitions in question entailed that the relation they define had certain properties, and this is simply not the case! Of the many definitions which have been proposed, the following pair are by far the most common: (S global) Phenomena describable by means of predicates in a certain set B are globally supervenient upon the phenomena describable by means of predicates in a certain set A if and only if there do not exist two physically possible worlds, W 1 and W2 , such that the objects are the same in both worlds, and their A-predicates are the same in both worlds, but their Bpredicates are not the same. In short, global supervenience means that global sameness of the distribution of A-predicates necessitates global sameness of the distribution of B-predicates. 8

Carnap (1939). See especially §24, “Elementary and Abstract Terms,” pp. 61-67. I used the term myself without defining it, I am sorry to confess, in an early paper: (1957). 9

24 (S local)

Hilary Putnam

Phenomena describable by means of predicates in a certain set B are locally supervenient upon the phenomena describable by means of predicates in a certain set A if and only if there do not exist two physically possible objects O1 and O2 , such that the A-predicates of those objects are the same but their B-predicates are not the same. In short, global supervenience means that local sameness of the distribution of A-predicates necessitates local sameness of the distribution of B-predicates.

Although every philosopher who employs the term ‘supervenience’ would agree that there is a difference between saying that a type B of phenomena is supervenient upon another type A (usually described in the language of physics) and saying that B-phenomena are nothing but C-phenomena, or that B-phenomena are identical with certain C-phenomena, or that the language we ordinarily use to speak of B-phenomena can be replaced by (or “eliminated” in favor of) the language scientists use to speak of A-phenomena, in practice analytic metaphysicians take supervenience to have virtually the same metaphysical significance as identity or elimination or reducibility. In other words, once materialists have said that B-phenomena are “supervenient upon” A-phenomena, they often feel that they have, in effect, disposed of the question of the metaphysical status of Bphenomena, especially if they have argued to their own satisfaction that the supervenience of B-phenomena upon A-phenomena is “local” rather than global. As indicated above, the problem is that the term (or rather the pair of terms ‘gloabally supervenient’ and ‘locally supervenient’) are used by these analytic metaphysicians as if the definitions in question entailed that the relations they define had certain properties, but the relations defined by (S global) and (S local) are easily shown not to have those properties. The most important property ascribed to supervenience is this: if B-phenomena supervene on A-phenomena, then it is supposed to be the case that B-phenomena are determined by A-phenomena where ‘determined’ is obviously supposed to mean “explained.” Some years ago, the Dutch chemist-philosopher Jaap van Brakel protested against the unclarity of this notion of “determination” in a pair of important papers (van Brakel 1997 and 1999). In fact, it is not hard to see that the accepted definitions of supervenience, (S global) and (S local), do not even guarantee asymmetry!

What Makes Pragmatism So Different?

25

To see why they do not, consider the following completely imaginable possibility: suppose that it turns out to follow from General Relativity (or from quantum mechanics, or from a successor theory to both) that the global electromagnetic field couldn’t be different without there being some difference in the gravitational field, and suppose at the same time that the gravitational field couldn’t be different without there being some difference in the electromagnetic field. Does it follow without further argument either that the gravitational field “determines” (or is the reality at the “basis” of) the electromagnetic field or that the electromagnetic field “determines” (or is the reality at the “basis” of) the gravitational field? I See no reason why either should follow. Indeed, the possibility that we would immediately suspect is that both fields are physically determined by something more fundamental, some more basic field or some feature of the laws of Quantum General Relativity (a theory we have not yet succeeded in constructing). What this thought experiment reveals is that supervenience as conventionally defined simply means that there is a certain sort of functional relation between A-phenomena and B-phenomena. Since it is perfectly imaginable that there are functional relations in both directions (in the case of the example I just used, from gravitational properties to electromagnetic properties and vice versa), it is not surprising that “supervenience” does not always run in one direction. And more important, that a functional relation amounts to determination – and to “determination” in which sense – is something that has to be made out in each case, not something that the mere fact (or assumption) of “supervenience” does for us. Consider, now, the case of psychological phenomena. To say that psychological phenomena supervene globally on the “primary qualities” of physics is just to say that there could not be two physically possible worlds, which are identical in their physical descriptions – in classical physics, that would be to say that the distribution of “primary qualities” over fundamental physical objects is the same (or, these days of quantum mechanics, that the two physical universes have identical “state vectors” in Hilbert Space) – but in which different psychological phenomena take place. That psychological phenomena are indeed globally supervenient in this way on fundamental physical properties (or “states”) few would today doubt. But that the psychological states of individual organisms are locally supervenient on the fundamental physical properties of those organisms is false, if we accept “externalism with respect to the mental,” as I have argued we must in (1988) and elsewhere. By ‘externalism with respect to the mental’, I mean the thesis the content of psychological

26

Hilary Putnam

states is individuated by a history of interactions with one’s conspecifics and with the entire natural environment, and not just by what goes on inside one’s skull. (William James, by the way, was a fellow “externalist,” at least about perceptual states, as his [[1912] 1976] make clear [see my 1990b]). Today, the most common response of the materialist philosophers I am speaking of is to grant the truth of externalism with respect to such intentional states as thinking and believing, but to try to argue that our psychological states have a, so to speak, local core. In (1999), I argued that this claim is confused, and that, in fact, there are no isolated neural states that have the properties that the “narrow contents” of our content-bearing psychological states are supposed to have. To suppose, for example, that there is any one neural state that every human being who thinks the thought that there are many restaurants in Tokyo (that is, whose thought a good interpreter would interpret as the thought that there are many restaurants in Tokyo) must be in one and the same neural state, or one and the same “computational state,” independently of why she thinks that thought or what her knowledge state, interests, purposes, etc., are, is science fiction, not science. As I wrote there, “The futile search for scientific objects called “narrow contents” in the case of meanings and for “internal psychological states” in the case of beliefs are alike instances of the rationalist error of assuming that whenever it is natural to project the same words into two different circumstances there must be an “entity” that is present in both circumstances” (Putnam 1999, p. 125). As for “global supervenience,” as I and others have pointed out for many years, global supervenience of psychological phenomena of the global environment does not imply that psychological explanations are redundant. If someone lets the water run in the bathtub to take a bath, for example, the type phenomenon “deciding to turn on the water in a bathtub in order to take a bath” is not definable in physical terms. 10 And explanations, as Davidson rightly saw, connect events under types. A physical explanation of the trajectory of certain particles does not generalize to the same class of cases as the psychological explanation, that the subject decided to turn on the water in order to take a bath. Global supervenience does not mean that individual psychological states are correlated with individual physical states, or indeed, with definable set of physical states. Psychological explanations are still necessary, are 10

The question, famously raised by Donald Davidson, of the identity of the “token event” of, say, deciding to take a bath with a “token” physical event is beyond the scope of this lecture. My position (which I set out in [1999], p. XXX), is that “events” are much too vague for the notion of “token identity” to make any real sense.

What Makes Pragmatism So Different?

27

still valid, and have ranges of applications which are not the same as those of any physical explanations.

3. Pragmatist “Aristotelianism” Some years ago, Martha Nussbaum and I argued that, according to Aristotle, “the psychological activities of living beings, such as perceiving, desiring and imagining, are realized or constituted in matter, are in fact the activities of some suitable matter, and that the relation between form and matter is in fact one of constitution or realization not of identity of mere correlation” (Nussbaum and Putnam 1994, p. 28; see also Putnam 1994b). And we concluded by saying that “we can have nonreductionism and the explanatory priority of the intentional without losing sight of the natural and organic unity of the intentional with its constitutive matter that is one of the great contributions of Aristotelian realism” (Putnam and Nussbaum 1994, p. 5). Obviously, the positions we ascribed to Aristotle there are positions that John Dewey also argued for, and Human Nature and Conduct (Dewey [1922] 1988) has sometimes been characterized as a sort of “naturalized” (in Dewey’s sense, not the contemporary materialist sense!) Aristotelianism. The criticisms of the materialist school of “analytic metaphysics” I have been making might be restated in Aristotelian language thus: the materialists speak as if the only explanatory principles were the fields and particles of fundamental physics; what they entirely fail to see is that the world has many, many different levels of form, and that types of form are also explanatory principles. If you want to explain why, for example, why Kant wrote a certain passage in The Critique of Pure Reason, a knowledge of quantum mechanics and relativity theory won’t help you; as I put it in a paper I wrote many years ago, most of the structure at the level of physics is irrelevant from the point of view of [a] higher-level discipline (Putnam [1973] 1994, p. 432). But pragmatism also has serious differences with the Aristotelian tradition, differences that, as has often been pointed out, are connected with the fact the Pragmatists came after and were deeply influenced by the discoveries of Charles Darwin. For Aristotle, a given individual belongs to one and only one (lowest) species (which lowest species a given individual belonged to is supposedly determined by the essence of that individual), and the essence of a given individual or species is supposed to be perfectly clear cut (and, of course, unique). But for Darwin and those biologist who built on his work, what is important is

28

Hilary Putnam

variation. Species do not have sharp boundaries; indeed the criteria for specieshood are actually criteria for species difference (e.g., “populations” which are not cross-fertile belong to different species; populations which are geographically isolated and have sufficiently different phenotypes are normally classed as belonging to different species); and these criteria, as Ernst Mayr, the grand old man of today’s evolutionary biology, always emphasized, do not yield a “clean” division of organisms into disjoint species (Mayr 1976). Nor is this a defect in population biology: the basic teaching of Darwinism is that the line between species can’t be sharp – otherwise one species could not evolve from another! Variation is fundamental, and “essentialist” thinking is taboo. The anti-essentialism of which I speak is beautifully expressed in a letter that James wrote late in his life, in 1907, to a philosophical critic, Dickinson S. Miller: I got your letter about “Pragmatism,” etc., some time ago [. . .].I sent you a week ago a “Journal of Philosophy” with a word more about Truth in it, written at you mainly; but I hardly dare hope that I have cleared up my position. A letter from Strong, two days ago, written after receiving a proof of that paper, still thinks that I deny the existence of realities outside the thinker; and Perry [. . .] accused Pragmatists (though he doesn’t name me) of ignoring or denying that the real objects play any part in deciding what ideas are true. I confess that such misunderstandings seem to me hardly credible [. . .]. Apparently it all comes from the word Pragmatism – and a most unlucky word it may prove to have been. I am a natural realist. The world per se may be likened to a cast of beans on a table. By themselves they spell nothing. An onlooker may group them as he likes. He may simply count them all and map them. He may select groups and name these capriciously, or name them to suit certain extrinsic purposes of his. Whatever he does, so long as he takes account of them, his account is neither false nor irrelevant. If neither, why not call it true? it fits the beans-minus-him, and expresses the total fact, of beans-plus-him. Truth in this total sense is partially ambiguous, then. If he simply counts or maps, he obeys a subjective interest as much as if he traces figures. Let that stand for pure “intellectual” treatment of the beans, while grouping them variously stands for non-intellectual interests. All that [. . .] I contend for is that there is no “truth” without some interest, and that non-intellectual interests play a part as well as the intellectual ones. Whereupon we are accused of denying the beans, or denying being in any way. (James [1907] 1920, pp. 295-296)

What Makes Pragmatism So Different?

29

4. Philosophy without Essences From Aristotle to the great 17th century revolutions in philosophy, philosophy almost always posited a realm of “intelligible” entities, entities uniquely available to “Reason” (that is, to the reason of philosophers), with fixed and immutable essences. Those entities were, in one way or another, held to provide the reasons for being (the raison d’etre in a very literal sense) of all the “sublunar,” or mundane, and contingent entities that we encounter.11 In introductory courses in philosophy, the 17 th century revolutions I spoke of are often presented as if they completely overthrew this structure (or as if empiricism, at least, did so), but it is not hard to see that this is an exaggeration. As Dewey explains in The Quest for Certainty, classical empiricism took it for granted that the mind is confronted with “ideas” or “sensations” or Empfindungen, that these are mental objects, and that the general form of these objects and of their “qualities” was a pretty self-evident matter (Dewey 1960). Although, by the time he wrote The Analysis of Mind, Russell came to see that the possibility of dividing up every experience into a part that is simply given and a part that is contributed by conceptualization (a “mnemic” element, in his jargon) was contestable, he felt that he had to defend it, on virtually a priori grounds. By way of contrast, in Dewey’s view, if rationalism made the mistake of supposing that the most fundamental laws of nature, and hence the form of scientific explanations at least in physics, could be known a priori, still the empiricist belief that the most fundamental experiential objects and their properties (and hence the nature of all empirical “data”) could be known once and for all was a perfectly comparable mistake. Against this aspect of the empiricist tradition, Dewey, continuing a line of thought that James had begun, insisted that by creating new observation-concepts we “institute” new data.12 Modern physics (and of course not only physics) have richly born

11

Abe Stone has remarked that the twentieth century metaphysics of Husserl’s Ideas actually retains this traditional stucture; the novelty in Husserl lies in a new account of the relation of “psychology,” as the empirical science of sublunar minds, to philosophy (“phenomenology”) as the account of the intelligible beings and their essences. 12 “The new data thus instituted do much more than provide facts for confirming and refining old conceptions. They institute a new order of problems whose solution requires a new frame of conceptual reference. In particular, it was by the use of new instruments and techniques that changes and relations of change were disclosed in what had previously been taken as fixed; a process that has gone on at an accelerated rate since the 17 th century. This change in the nature of data was both the source and the product of the

30

Hilary Putnam

him out. A scientist may observe a proton colliding with a nucleus, observe a virus with the aid of an electron microscope, or observe genes or black holes, etc. Neither the form of possible explanations nor the form of possible data can be fixed in advance, once and for all. Like traditional metaphysics, contemporary materialism postulates a distinct realm of things which are the ultimate exlanatory principles of all beings, although it is falibilist to the extent of allowing physicists rather than philosophers to determine the contents of that realm. Following Quine, it often says that what we call “change” is a sort of illusion, and that the ideal description of this reality would be in what it calls a “tenseless” language. It does not think that realm can be known a priori, to be sure, but it does freely speculate about the form of an “absolute conception” of that realm, a finished science. And it often takes that speculation to be the only remaining function of philosophy. It is not hard to see that materialist metaphysics is a decidedly conservative affair. In rejecting the entire picture of a philosophical subject of metaphysics with a subject matter which consists of fundamental objects which are the explanatory principles of all “beings,” pragmatism is quite different.

5. Pragmatism and Existentialism Existentialists, from Nietzsche on, used scornful language to describe the traditional metaphysical enterprise that I have so sweepingly (but, I think, not inaccurately) described. Heidegger famously described that enterprise as “ontotheology.” Their alternative to what they saw as a fatally flawed conception of the task of philosophy was typically to diagnose of what they saw as the falsity at the root of (most of) our lives, and to recommend an alternative to that falsity. For Nietzche, fear of life is at the heart of our sickness, and the will to live (which Heidegger unsympathetically described as “the will to will”) is what has to be revived and strengthened. For Kierkegaard, addressing people who thought they were Christians, the problem was just that – that they thought they were Christians, that they knew what it means to be a “Christian,” and what they had to learn – not intellectually, because intellectualization is a principal symptom of their illness, but existentially – is what it really means to have a Christian relation to God.

universal adoption of the experimental method and the new order of conceptions demanded by its successful execution” (Dewey [1938] 1986, pp. 388-389).

What Makes Pragmatism So Different?

31

For Heidegger (even if he atempts to deny that “authenticity” is a normative concept), the sickness is inauthenticity, and the valorized life combines authentic acceptance of absurdity (the “Nothing”) with submissiveness towards Being in ways that Heidegger interpreters endlessly argue about. What I want to emphasize is that along with the abandonment of metaphysics, this search for God, in Kierkegaard’s case, or for a secular substitute for an existential connection to God, in the case of the atheist existentialists, led to either an abandonment of or an irresponsible relation to the political, as well as to a disinterest or a failure to see the philosophical significance of science, and, in Heidegger’s case, not just to disastrous politics, but to a disastrous philosophy of history.13 In contrast, Dewey famously declared that “Philosophy will recover itself when it ceases to deal with the problems of philosophers and addresses the problems of men” (Dewey [1917] 1976, p. 42). It is important to note that Dewey wrote “problems” in the plural. Pragmatists refuse to believe that there is just one problem of men, that is of persons, and just one solution. Dewey also wrote that philosophy “has no Mosaic or Pauline authority of revelation entrusted to it.” The pragmatist philosopher does not pretend to be a prophet or an oracle. But pragmatist philosophy does aspire to “the authority of intelligence, of criticism of [. . .] common and natural goods” (Dewey 1926, pp. 407-408). The pragmatist is willing to address existential concerns; but he does not think that one can responsibly address them, address them without falsifying them, without addressing quotidian concerns. If philosophy no longer has either a unique subject matter, nor a unique prophetic vision, then how can it continue at all? I sometimes heard students ask. Dewey’s answer was that philosophy has no need to be unique (an answer which is, however, itself unique!). James already argued in “The Moral Philospher and the Moral Life” “the ethical philosopher must wait on facts,” and furthermore he only knows if he makes a bad mistake, the cries of the wounded will soon inform him of the fact. In all this the philosopher is just like the rest of us non-philosophers, so far as we are just and sympathetic instinctively, and so far as we are open to the voice of complaint. His function is in fact indistinguishable from that of the best kind of statesman at the present day. His books on ethics, therefore, so far as they truly touch the moral life, must more and more ally themselves with a literature which is confessedly tentative and suggestive rather than 13

For a penetrating description and criticism of Heidegger’s philosophy of history, see Boyle (1998).

32

Hilary Putnam

dogmatic – I mean with novels and dramas of the deeper sort, with sermons, with books on statecraft and philanthropy and social and economical reform. Treated in this way ethical treatises may be voluminous and luminous as well; but they can never be final, except in their abstractest and vaguest features, and they must more and more abandon old-fashioned, clearcut, and would-be “scientific” form. (James [1897] 1956, pp. 158-159)

Here James exaggerates when he writes “indistinguishable.” Philosophy is distinguishable from statesmanship, spiritual exhortation, and literature. But the difference, Dewey would say is one of degree: philosophy at its best is simply more reflective, more critical, more wide ranging. ‘Criticism’ is a word Dewey loved, and, in fact, he once defined philosophy as “criticisms of criticisms” (Dewey [1925] 1981, p. 298). Although Stanley Cavell has been at times critical (unfairly critical, I believe) of Dewey (see Cavell 1998), I believe that Cavell’s characterization of philosophy as “education for grownups” is one that Dewey would have approved. A philosophy that renounces both the dreams of metaphysics and the self-importance of existentialism is a grownup philosophy; and only a grownup philosophy is fit to educate grownups. One last but essential remark. To call upon us, as I just have, to renounce both the dreams of metaphysics and the self-importance of existentialism, is NOT to join the logical positivists of yesteryear in calling both metaphysics and existentialism “nonsense.” There is MUCH of permanent value in the writing of both traditional metaphysicians and the great existentialists. It would be false to Dewey’s own spirit to deny that there is. But my self-imposed task today has been to bring out the uniqueness of pragmatism., and to do that I have had to emphasize what pragmatists see as the failures of those traditions.

Harvard University Department of Philosophy Emerson Hall 207 Cambridge MA 02138 USA e-mail: [email protected]

What Makes Pragmatism So Different?

33

REFERENCES Boyle, N. (1998). Who Are We Now? Christian Humanism and the Global Market from Hegel to Heaney. Notre Dame, IN: Notre Dame Press. Carnap, R. (1939). The Foundations of Logic and Mathematics. In: International Encyclopedia of Unified Science, vol. 1, no. 3, pp. 1-70. Chicago, IL: The University of Chicago Press. Cavell, S. (1998). What’s the Use of Calling Emerson a Pragmatist? In: M. Dickstein (ed.), The Revival of Pragmatism: New Essays on Social Thought, Law, and Culture, pp. 72-82. Durham, NC and London: Duke University Press. Churchland, P. (1998). On the Contrary: Critical Essays, 1987-1997. Cambridge, MA: The MIT Press. Conant, J. and U. Zeglen, eds. (2002). Hilary Putnam: Pragmatism and Realism. London: Routledge. Davidson, D. (1984). Inquiries into Truth and Interpretation. Oxford: Clarendon Press. Dewey, J. ([1917] 1976). The Middle Works, vol. 10 [Journal Articles, Essays, and Miscellany Published in the 1916-1917 Period]. Edited by J.A. Boydston. Carbondale, IL: Southern Illinois University Press. Dewey, J. ([1922] 1988). The Middle Works, vol. 14 [Human Nature and Conduct 1922]. Edited by J.A. Boydston. Carbondale, IL: Southern Illinois University Press. Dewey, J. ([1925] 1981). The Later Works, vol. 1 [Experience and Nature]. Edited by J.A. Boydston. Carbondale, IL: Southern Illinois University Press. Dewey, J. ([1929] 1981). The Later Works, vol. 4 [The Quest for Certainty]. Edited by J.A. Boydston. Carbondale, IL: Southern Illinois University Press. Dewey, J. ([1938] 1986). The Later Works, vol. 12 [Logic: The Theory of Inquiry]. Edited by J.A. Boydston. Carbondale, IL: Southern Illinois University Press. James, W. ([1897] 1956). The Moral Philosopher and the Moral Life. In: The Will to Believe, pp. 141-162. Edited by F. Bowers and F. Burkhardt. Cambridge, MA: Harvard University Press. James, W. ([1907] 1920). Letter to Dickenson S. Miller dated Aug. 5, 1907. In: The Letters of William James, pp. 295-296. Boston. The Atlantic Monthly Press. James, W. ([1912] 1976). Essays in Radical Empiricism. Edited by F. Bowers and I.K. Skrupskelis. Cambridge, MA: Harvard University Press. Mayr, E. (1976). Evolution and the Diversity of Life: Selected Essays. Cambridge, MA: Harvard University Press. Nussbaum, M. and H. Putnam (1994). Changing Aristotle’s Mind. In: Putnam (1994c), pp. 22-61. Putnam, H. ([1973] 1994). Reductionism and the Nature of Psychology. In: (1994c), pp. 428-440. Putnam, H. (1957). Mathematics and the Existence of Abstract Entities. Philosophical Studies 7, 81-88. Putnam, H. (1962). What Theories Are Not. In: E. Nagel, P. Suppes, and A. Tarski (eds.), Logic, Methodology, and Philosophy of Science. Proceedings of the 1960 International Congress, pp. 240-251. Stanford, CA: Stanford University Press. Reprinted in: Mathematics, Matter, and Method (Cambridge: Cambridge University Press, 1975), pp. 215-227. Putnam, H. (1975). The Meaning of ‘Meaning’. In: K. Gunderson (ed.), Language, Mind, and Knowledge. Minnesota Studies in the Philosophy of Science, vol. 7, pp. 131-193.

34

Hilary Putnam

Minneapolis: University of Minnesota Press. Reprinted in: Mind, Language, and Reality (Cambridge: Cambridge University Press, 1975), pp. 215-271. Putnam, H. (1990a). Objectivity and the Science/Ethics Dichotomy. In: (1990c), pp. 163-178. Putnam, H. (1990b). James Theory of Perception. In: (1990c), pp. 232-251. Putnam, H. (1990c). Realism with a Human Face. Cambridge, MA: Harvard University Press. Putnam, H. (1992a). Bernard Williams and the Absolute Conception of the World. In: Renewing Philosophy, pp. 80-107. Cambridge, MA: Harvard University Press. Putnam, H. (1992b). Truth, Activation Vectors, and Possession Conditions for Concepts. Philosophy and Phenomenological Research 52 (2), 431-447. Putnam, H. (1994a). Pragmatism and Relativism: Universal Values and Traditional Ways of Life. In: (1994c), pp. 182-197. Putnam, H. (1994b). Aristotle after Wittgenstein. In: (1994c), pp. 62-81. Putnam, H. (1994c). Words and Life. Cambridge, MA: Harvard University Press. Putnam, H. (1995). Pragmatism: An Open Question. Oxford and Cambridge, MA: Blackwell. Putnam, H. (1999). The Threefold Cord. Mind, Body, and World. New York: Columbia University Press. Putnam, H. (2001). Reply to Bernard Williams’ ‘Philosophy as a Humanistic Discipline’. Philosophy 76, 605-614. Quine, W.V.O. (1990). The Pursuit of Truth. Cambridge, MA: Harvard University Press. Rorty, R. (1993). Putnam and the Relativist Menace. The Journal of Philosophy 90 (9), 443-461. Van Brakel, J. (1997). Supervenience: The Priority of the Manifest Image. Behavioral and Brain Sciences 20, 167-232. Van Brakel, J. (1999). Supervenience and Anomalous Monism. Dialectica 53, 3-25. Walsh, V. (2000). Smith After Sen. Review of Political Economy 12 (1), 5-25. Williams, B. (1978). Descartes; The Project of Pure Enquiry. Hammondsworth, Middlesex: Penguin Books. Williams, B. (1985). Ethics and the Limits of Philosophy. Cambridge, MA: Harvard University Press. Williams, B. (2000). Philosophy as a Humanistic Discipline. Philosophy 75, 477-496.

Juan José Acero PRAGMATISM, PLURALISM, AND THE PEIRCE PRINCIPLE

ABSTRACT. This chapter examines Putnam’s views on Pragmatism and points out that, according to this philosopher, metaphysical pluralism, i.e. antiessentialism, is not only the distinguishing feature of this philosophical trend, but also a feature that makes impossible to reconcile Pragmatism with what Putnam calls the Absolute Conception of Reality, a view he attributes to Bernard Williams. After calling the reader’s attention towards how far is Putnam from adopting the Peirce Principle, which Dewey thought it to resume Pragmatism’s main substance, it is argued that this principle naturally fits in with two ingredients of the Absolute Conception, namely the transcendence requirement and the convergence requirement. Finally, the chapter claims that there are two different interpretations of the Peirce Principle, i.e. a verificationist reading and a non-verificationist one, and holds that the transcendence and convergence requirements were already present in Dewey’s work on education.

In his rich paper “What Makes Pragmatism So Different?”, Professor Putnam aims at bringing out the uniqueness of pragmatism within Western metaphysics. In order to do so, he compares pragmatism with three metaphysical traditions, namely contemporary naturalism, Aristotelian essentialism and existentialism. Besides tracing out analogies and differences, Putnam defends pragmatism against these three traditions. Thus, he reproaches existentialism – a label under which he mentions Nietzsche, Kierkegaard, and Heidegger – for having led contemporary philosophy along paths that withdraw her from the sorts of problems that demand common people’s attention. William James and John Dewey chose instead a view of philosophy that only gradually differed from literature, religion, politics and science. For them, philosophy brings with itself the means to equip not only the thinker but the grownup in general with a more reflexive and critical awareness of

In: M.U. Rivas Monroy, C. Cancela Silva, and C. Martínez Vidal (eds.), Following Putnam’s Trail: On Realism and Other Issues (Pozna Studies in the Philosophy of the Sciences and the Humanities, vol. 95), pp. 35-53. Amsterdam/New York, NY: Rodopi, 2008.

36

Juan José Acero

nature and society. It is in this sense that we are to understand Dewey’s happy characterization of philosophy as criticisms of criticisms. The main body of Putnam’s paper is not devoted to comparing pragmatism with existentialism, but to laying down and criticizing a metaphysical view of reality, the Absolute View of Reality, which results from combining both materialist and Aristotelian doctrines and which has had a heavy influence on current philosophy. Since Putnam’s attempt to systematically picture this view and to expose its weakness deserves careful reflection, it is this the part of his paper that will be focused on in what follows. It will be argued that in Putnam’s view metaphysical pluralism is pragmatism’s distinguishing feature, and pluralism entails the rejection of the Absolute View. In the first place, it will be pointed out that this view is a complex one, so that several ingredients could, and should, be set apart in shading light on it. In the second place, it will be argued that two of these components fit in with the spirit of pragmatism, namely the need for transcending any subject-centered view of reality and the need for converging with further world perspectives. In the third place, Putnam’s critical attitude towards the main insight of Pragmatism, i.e. the Pragmatic Maxim or the Peirce Principle, could be attributed to his not having distinguished what in the Absolute View is acceptable for a pragmatist and what such a kind of philosopher cannot stomach. In accordance with such a diagnosis, two interpretations of the Peirce Principle will be put forward: the verificationist reading Putnam endorses and a phenomenism-free reading of it. This second interpretation is compatible with the partial vindication of the Absolute View that will be carried out below. Finally, I will adduce that the transcendence and the convergence requirements have a neat pragmatist pedigree, their presence being detectable in John Dewey’s influential Democracy and Education (1966).

1. Putnam’s Metaphysical Pluralism The kind of naturalism Putnam deals with is not the supernaturalism that was the target of Dewey’s attacks, but materialism, a doctrine that Putnam finds paradigmatically put forward in Bernard Williams’ works. According to it, there is a conception, a belief system that pictures the world “with the highest degree of independence of our own perspective and its characteristic features” (Williams 1985, p. 178). It is the view from nowhere, as Thomas Nagel (1986) called it, a view from which

Pragmatism, Pluralism, and the Pierce Principle

37

every singular perspective would have been disposed of. I will refer to this doctrine as the Absolute View thesis or conception of reality. Not only for Williams but for Putnam as well the idea that “all fundamental properties and relations that actually occur are physical” (Lewis 1999, p. 292) – the thesis of materialism – is part and parcel of the Absolute View thesis. The view that the concepts recruited to make up the representation of the world from the zero point of view are concepts of material properties and relationships, coined within basic natural science, so that any other concept, whether intentional, psychological, moral or whatever, one might resort to in taking inventory of the world, will be a concept of either a property or a relationship that supervenes on the basic or primary ones. To be in a better position to discuss these views, I will distinguish the Absolute View thesis from the one that holds that any non-basic property (and relation) supervenes of one or more fundamental properties (and relations), which I will refer to as the Supervenience thesis. Reductive Physicalism, i.e. the result of applying the Supervenience thesis to the elucidation of mental properties, is the doctrine that holds that mental properties supervene on physical properties, and it is included in the Absolute View pack. Aristotelian essentialism supplies the third basis on which the view that Putnam rejects rests. It is the thesis that each and every individual belongs to one and lowest species and that the essence of either an individual or a species in unique and has perfectly clear sufficient and necessary conditions. The result of putting together all these ingredients is a metaphysical view according to which reality is made up of individuals that, given the fundamental properties they possess and the fundamental, i.e. supervenient, relationships they maintain, constitute categories with perfectly delineated borders. Therefore, any non-basic property supervenes on the fundamental ones. It follows from all this that any belief we are entitled to have about such a reality would be a piece of a representational system from which any singular perspective has been filtered out. Against such a view Putnam adduces a battery of arguments with which he tries to demolish the three pillars that have been briefly described. Among these arguments, the most explicitly developed ones have the Supervenience thesis as their target, and can be divided up into three groups. The arguments in the first group point out that the concept of supervenience does not satisfy the conditions that the program of reducing non-basic properties and relations to the fundamental ones should warrant. Those conditions are asymmetry and determination. Asymmetry demands that non-basic properties supervene on the

38

Juan José Acero

fundamental ones, but not the other way around, thus subjecting the program to a hierarchy requirement. As for determination, it is the demand that the metaphysical hierarchy rests on a level in which only primary properties and relationships are to be found. Now, Putnam briefly puts forward an ideal experiment that casts serious doubts about whether the usual definitions of the concept of supervenience are successful. I feel myself inclined to side with him. The second group of arguments is specifically addressed at undermining the basis of Reductive Physicalism. On the one hand, the well-known Twin Earth poisoned-pawn gambit (in Putnam 1975), one of the most outstanding findings in Putnam’s professional career, shows that people’s mental states have properties that do not locally supervene on the biological properties of their brains. And, on the other hand, Putnam argues that global supervenience does not substantially helps the physicalist in advancing the reduction program: that global supervenience “does not mean that individual psychological states are correlated with individual physical states, or indeed, with definable set of physical states” (pp. 9f ).1 A third group of arguments, in which the influence of Aristotle is clearly felt, is spot again on the Supervenience thesis. The main substance of those arguments lies in his proposal of abandoning the idea of supervenience in favour of that of constitution (or realization) and defending that mental life is nothing but activity of the matter and in the matter. To my mind, what Putnam seems to be advancing is a sort of Non-Reductive Materialism. It is Physicalism in so far as there is no mental life without matter; it is non-reductive because he holds that psychology is explanatory even if its concepts cannot be replaced by concepts provide by basic natural science.2 In other words, “most of the structure at the level of physics is irrelevant from the point of view of [a] higher-level discipline” (this volume, p. 27). 1

This is not new. The point had been made in Kim (1989). Among others, L. Rudder Baker and Tyler Burge have insisted on this point. See Baker (1993) and Burge (1986). As far as I know, Putnam had not employed this argument until now. 2 Putnam’s seems to be close to, if not identical with, what Lynn Rudder Baker has called Minimal Materialism, a doctrine she states with the following slogan: “Take away all the atoms and nothing is left.” See Baker (1995). The key idea is that, fundamental properties and relations aside, there are many levels of explanation. Putnam does not speak of levels of explanation, but of levels of form. However, he immediately adds, “types of form are also explanatory principles.” It follows from this remark that the ontology and the conceptual system of basic physical science, i.e. elementary particle physics, are irrelevant for psychology, semantics, economy and other high-level scientific disciplines. See Putnam (1994).

Pragmatism, Pluralism, and the Pierce Principle

39

Putnam’s criticisms not only aim at the Supervenience thesis. His pragmatism begins taking shape when he confronts Aristotelian essentialism and the Aristotelian tradition in metaphysics. The deep influence that Darwin’s evolutionary teachings had on James and Dewey (see Putnam 1990b, pp. 234ff ) runs parallel to Ernst Mayr’s – and, in general, Population Genetics – lessons that species are historic entities (Mayr 1982, Chs. 2 and 12) and that, as Putnam acknowledges, “[v]ariation is fundamental and “essentialist” thinking is taboo” (this volume, p. 28). (Thus, in the same way as intentional properties do not supervene on brain’s chemical properties, biological population properties do not supervene on DNA’s chemical properties.) This result provides Putnam with a basis on which to set up his own variety of pragmatist metaphysics, a variety which is defined by the following two theses: first, there is no fixed set of basic objects, properties and relations; and second, neither individuals nor species have sharp boundaries, i.e. essences. These two claims constitute Putnam’s metaphysical pluralism and anti-essentialism, Putnam’s own alternative to the Absolute View thesis.

2. A Partial Defence of the Absolute View of Reality The extent to which I accord to Putnam’s views is significant – it goes without saying that some of the ideas I share with him I have learned from his work. Among them I am glad to emphasize three of them: his externalism concerning the intentionality of mental properties, his conviction that those properties that basic natural science might include in the stocktaking of the world are of no help to higher-level disciplines, i.e. psychology among others, and his Darwinian rejection of essentialist thinking. However, I want to explore now the question that gives Putnam’s contribution its title, namely to what extent is pragmatism compatible with the Absolute View, which is the target of Putnam’s criticisms. More briefly put, where does the difference between a pragmatist and a pluralist lie in? In his manifesto Pragmatism, James thought that opting for pluralism would allow us to get rid of the kind of dogmatic strictness that friends of monism like to abide by (James 1907, pp. 554ff ). Since Reductive Physicalism is a kind of monism, i.e. of materialist monism, one way of interpreting Putnam is that of taking him as carrying out the sort of enlightened work that James recommended, a task much needed given the course adopted by contemporary analytic

40

Juan José Acero

philosophy. From this point of view, pluralism is enough to make of Putnam a pragmatist. Another way on interpreting Putnam in a pragmatist key consists in underlying those contents we find at the beginning of his paper. It is the key that has allowed Rorty to take a distinguished role in the current discussion of pragmatism and moved him to give Dewey’s thought a lot of coverage.3 On those pages, Putnam addresses Bernard Williams, the philosopher who personifies the Absolute View for him. If we decided to follow Rorty’s lead on this point, we might ask whether Putnam’s metaphysical pluralism is either the pragmatist alternative to a sort of monism implicit in natural science or a result of having become emancipated from a secular god, namely the way things are in themselves. Although Putnam does not echo this Rortyan theme in his paper, he comes close to doing it. Thus, whereas he denies that there is a view of the world from nowhere, he also insists, as his analysis of existentialism leads him to conclude, that philosophy has to be concerned with the common men and women’s problems and interests. I suggest these two ways of recognizing in Putnam’s paper a commitment to pragmatism because it is not clear to me that his is an entirely orthodox pragmatism. Specifically, I miss in his defense of pluralism any reference to the Pragmatic Maxim or, as I will call it, the Peirce Principle.4 In his argumentation no room is left for the principle that “there is no distinction of meaning so fine as to consist in anything but a possible difference of practice” (Peirce 1998, p. 44). I understand that if Putnam had resorted to it, he had not taken his attack on the Absolute View thesis so far as it in fact goes. One thing is his arguments against Aristotelian essentialism and the consequences that many different beliefs systems may be exploited to represent reality because the world “has many, many different levels of form” (p. 10). As it was pointed out above, this part of Putnam criticisms sounds plausible enough, and the same can be said about the claim that there exist a number of diverse conceptual repertoires with whose the help to describe how things are . . . even different ways of representing the very same things. However, once the Absolute View is purged of everything that makes it incompatible with pluralism, there still remain two ideas that, first, I judge necessary to retain; second, and more important, that the Peirce Principle validates. 3

See Rorty (1999, pp. 7-20; and 2000, Lesson One). Putnam (2000b) discusses Rorty’s metaphysical and epistemological ideas. 4 Rorty’s second characterization of pragmatism is committed to this principle. See Rorty (1982, pp. 163ff ).

Pragmatism, Pluralism, and the Pierce Principle

41

The first ingredient is the need for shaping our beliefs about the world by means of conceptual systems devoid of any particular perspective that might taint them, the need for articulating our concepts in such a way that they make possible to represent the world from without our own parochial interests and limitations. Far from considering futile the wish to transcend our own point of view, it is difficult to question the convenience, both theoretical and practical, of progressively freeing our conceptual system of such blind points. A view that transcends our own perspective reduces both the casual and contingent atmosphere that biases our understanding and gives us access to things in so far as they come into our span of vision. The struggle for transcendence is the struggle for widening our egocentric window while strolling through the world. When our understanding of it is the outcome of combining images exclusively obtained from our particular point of view, without coordinating them to others’ understanding, it is unavoidable to feel that things happen by accident and that a sense of chance pervades our inquires and explanations. Besides, our actions essentially become reactions to those circumstances that somewhat affect us, because our abilities to foresee them are severely limited. As our skills to intervene in the world develop, we manage to place our own vision of the world, or of parts of it, into a more objective, intersubjective, pattern, a task that gradually goes on and that requires continuous and piecemeal changes to better keep to the world’s scripts. As the feeling of reality unfolds, both our explanatory and predictive powers and our opportunities to change the world increase. The Peirce Principle, not only does not stand in the way of our recommending the need to transcend our point of view, but it helps making sense of it. From a pragmatist point of view, far from being after better copies of the world, we look forward to getting involved in more successful dealings with it, more skilful resources to get involved with things and to have them at our disposal. The Peirce Principle gives an enlighten expression to what the feeling of reality comes to. The Absolute View thesis’ second valuable insight gives a bonus to convergence, that is, to articulating shared conceptual systems that transcend singular perspectives, once again a task that asks for a program of piecemeal intervention. Our concept and belief systems would converge were they liable to be subject to others’ understanding and judgments and, as a consequence, it would gain acceptability for those who do not occupy our place among things. We make convergence possible in choosing to abide by general procedures for locating an event in space-time, replacing ‘It rains in Granada on April 04.28.04’ for ‘It rains here’, or an exchange economic system based on either currency or

42

Juan José Acero

a precious metal to an economic system based on barters. As before, the Peirce Principle makes it natural to infer that differences of practical involvement go hand in hand with different degrees of convergence. Of course, transcendence and convergence are not independent goals. The only way of looking at our own concept and belief systems from without and transcend them is through some else’s perspective, even though it is we who momentarily bracket our views and practices and catch them as it were from behind. The inseparability of transcendence and convergence, though it is not the whole story to tell about it, helps making Peirce’s proposal of identifying truth with the theory reached at the end of an indefinitely continued investigation intelligible. And it also gives its point to Williams’ idea of a belief system “with the highest independence of our own perspective and its peculiarities.” The core of such a proposal, as Putnam has rightly pointed out, is making truth the result of a procedure, and the point argued above is that the Peirce Principle requires that transcendence and convergence are built into the conditions that prevent investigation from coming off rails. This fact – if it is one – speaks in favour of the existence of something essentially human in the idea of an absolute truth, namely the need to go beyond our own perspective and accepting that truth and community are correlative (Putnam 1990, p. 31). On this basis, I gather that what might lead Putnam to flatly oppose to the Absolute View thesis is not the requirements of transcendence and convergence, but a conception of representing as copying. If all these components are factored out, no reason is left that blocks the way towards sharing Putnam’s criticism of that thesis while preserving those requirements. On the other hand, some writings of Putnam’s on pragmatism suggest an ambivalent attitude towards my diagnosis concerning what is right and what is wrong about the Absolute View idea. His papers on James and the pragmatist philosophy provide serious evidence that Putnam does not wholeheartedly subscribe to the Peirce Principle. He shares James’ view that Peirce’s theory of truth gives expression to a regulative ideal, and adds that if we cashed out such a theory, all we would get is the requirement of warranted assertability (Putnam and Putnam 1990, p. 223). This suggestion does not seem to me as convincing as many others of Putnam’s. Lacking an argument, I do not see that the concept of warranted assertability respects the requirements of transcendence and convergence that I deem worth retaining.

Pragmatism, Pluralism, and the Pierce Principle

43

3. Non-Conceptual Content and the Peirce Principle There is another topic, which Putnam does not deal with in his paper, though it has been extensively developed in other papers of his, about which some doubts, if not total disagreement, can be expressed. It is the problem of the nature of non-conceptual content and, more specifically, the doctrine, set out by James, according to which every perceptual experience has conceptual and non-conceptual aspects that can be sorted out. In The Threefold Cord Putnam has argued that there are forms of our experience that depend on our conceptual resources. Thus, he adduces that it is impossible to see x as being P unless we have got the concept of (being) P (Putnam 1999, Second Epilogue). Nothing will be said in what follows on this argument. More recently, in a long comment on John McDowell’s Mind and Word (Putnam 2002), Putnam has suggested that, without returning to the thought that some experiences involve no conceptual capacity at all, some experiences are preconceptual. He prefers the term ‘preconceptual’ to ‘non-conceptual’ in order to underlie that the assumption that there is a sharp boundary between conceptual and non-conceptual experiences makes no sense (Putnam 2002, p. 183). In this case, however, the argument Putnam rests on an unstable basis. According to McDowell, if the contents of our experiences were nonconceptual, they could not play any role at all in justifying our beliefs and judgments. Since McDowell thinks that experiences support as well as refute beliefs, he concludes that experiences must have conceptual content as well. Putnam puts some distance between McDowell and himself in setting perceptions apart from experiences. As Putnam understands them, while perceptions have conceptual constituents – his term is ‘aspects’ – experiences do not have any. 5 To be coherent, Putnam has to admit that the boundary between perceptions and experiences is fuzzy and that, because of it, it makes no sense to hold that there is a neat dividing line between perceptions-or-experiences that justify beliefs and perceptions-or-experiences that do not. No content without any justificatory role is his conviction. Some might find a bit uncomfortable living with this way of approaching the conceptual and non-conceptual dichotomy, because it mixes up the theory of content and the theory of justification. The first one is, if you want, a metaphysical theory about the constitution of mental states, e.g. the belief of such-and-such; the second one is an 5

It is interesting to note that Putnam says that successful perceptions have conceptual aspects “in some sense of ‘conceptual’ ” – a sense that the paper does not make it explicit. See Putnam (2002, p. 182).

44

Juan José Acero

epistemic theory about the goodness of reasons to belief that such-andsuch. Putnam’s distinction between perceptions and experiences leaves one wondering what experiences are for and in what do their contents consist in, if they are not involved in the providing-reasons business. Pragmatism should provide pragmatists with the means to answer these questions. To finish these critical comments an attempt to do it will be sketched. Well, in the same way as the Peirce Principle allows us to single out what there is in the Absolute View thesis that is worth saving, it is worth arguing now that the same principle can be used to vindicate a substantial notion of non-conceptual content. The insight that has guided the partisans of this kind of content is that perceptual states of many species of animals, very young children and deeply mentally handicapped aged people, e.g., Alzheimer syndrome patients, have content (see, e.g., Cussins 1990, p. 381). Any pragmatist should take the challenge to explain what the nature of non-conceptual content is, given his or her closeness to Darwin and evolutionary biology – a relationship Putnam reminds us of in his paper. Instead of construing the notion of nonconceptual content as hardly squaring with the pragmatist creed, since it contributes to strengthen a gap between experience and perception or, alternatively, between two kinds of content, it fits in with such a creed. This affinity has been clearly underlined by Evans: Such status [perceptual informational status endowed with nonconceptual content] – that is, states of a conscious subject. However addicted we may be to thinking of the links between auditory input and behavioral output in information-processing terms – in terms of computing the solution to simultaneous equations [. . .] – it seems abundantly clear that evolution could throw up an organism in which such advantageous links were established, long before it had provided us with a conscious subject of experience. (Evans 1982, pp. 157f )

This seems to me to be a deep remark. According to it, non-conceptual content is a matter of a direct correlation between an organism’s gathering information laying out there in the world, through the workings of its sensory organs, and the information gathered, without the mediation of any conscious judgment. A subject that hears a sound coming from a certain direction, Evans points out, need neither to say to herself “Its coming from over there” nor wondering what is going on, to lend her attention to the sound, to turn her head to it, or whatever. Those abilities are exercised regardless of whether that organism is a concept user, because what matters when it comes to non-conceptual content is activity. It is action what has to be taken into account in order to say

Pragmatism, Pluralism, and the Pierce Principle

45

what non-conceptual content amounts to. From this point of view, differences in non-conceptual content translate into differences in dispositions to do various things, because non-conceptual content captures what the psychologist James J. Gibson (1986) called affordances the environment provides to guide subjects’ actions, i.e. trails that afford them guidance. Imagine you get up in the middle of the night because someone had the ominous idea of phone you then. Not being entirely conscious of what you are doing, you feel for your slippers with your bare feet until you get the familiar feeling that you have put them on. Are concepts involved in your actions? Maybe there are. However, maybe this is not the right question to raise. By bringing it up, we loose sight of that fact it is not thought but action what world affordances should be mapped to. It is not the question “Where are my slippers?” what mediates my answering the call, but your slippers’ guiding your feet. 6 If there is something to this way of distinguishing non-conceptual from conceptual content, then the fact that experiences have nonconceptual content means an important difference, namely a difference in exploiting the environment’s affordances for those subjects whose biological endowment and learned abilities enable them to act upon those affordances. Being this difference one that makes a difference, the Peirce Principle cannot be blind to it. Therefore, this would leave Putnam in a position in which he should argue that the distinction between thinking and acting is another dichotomy to be abolished.

4. Putnam’s Rejection of the Peirce Principle The main goals aimed at in this chapter have been reached in previous sections. In what follows two objections to the partial defense of the Absolute Conception carried out above will be taken up and answered. To begin with, two interpretations of the Peirce Principle have to be distinguished. One of those interpretations makes it easy to understand why Putnam thinks there is nothing in the principle worth retaining and why the idea of a non-conceptual content makes no sense at all for him. The other reading makes it possible not only to avoid the difficulties with which Putnam saddles the principle but also to give support to the thesis that Putnam’s pragmatism is essentially Metaphysical Pluralism. 6

These last remarks closely follow not only Evans’ insights but Cussins’ elaborations on them. See Cussins (1992, 2002).

46

Juan José Acero

Putnam has made his reasons for rejecting the Peirce Principle explicit. Summed up, his objection amounts to the following: once you abide by it you are committed to a verificationist conception of meaning and truth. The formulation of the Peirce Principle Putnam considers is the well-known following one: Consider what effects, which might conceivably have practical bearings, we conceive the object of our conception to have. Then, our conception of these effects is the whole of our conception of the object. (Peirce 1998, p. 45)

In his opinion, the key idea in this statement lies in how one understands those effects our concepts or ideas have. Peirce states the maxim in a context that makes it natural to conclude that the term ‘effects which might conceivably have practical bearings’ means sensible effects, sensible qualities. Therefore, the principle identifies a concept’s (or a thought’s) content, as well as the meaning of an expression, with its sensible effects; and any difference between concepts (or thoughts) with a difference between the sensible effects that their use would bring out. Seen to this light, the Peirce Principle would come close to, if not coincide with, the Verification Principle, and Putnam has an interesting objection to raise against either the Peirce Principle or the Verification Principle.7 This objection opens up the following alternative for us. One option, the one Putnam choices, takes on the verificationist interpretation of the Peirce Principle and distinguishes a liberal and a strict reading of Verificationism. The pragmatist philosopher would then be a liberal verificationist. Pragmatism denies that knowledge claims’ ultimate touchstone consists of sense data presenting themselves in the theatre of the knower’s consciousness. Pragmatism is neither phenomenism nor committed to the requirement that a thought or proposition can be completely verified by appealing to incorrigible sensuous items. On the other hand, Pragmatism coincides with Verificationism in at least one point, which Putnam strives to state. He concedes the verificationist philosopher that “our grasp of empirical concepts depends on our perceptual verification abilities” (Putnam 1995, p. 305), but immediately adds that those abilities are deeply rooted in our life forms – a 7

For reasons of space, I won’t put forward Putnam’s objection to Verificationism here. I will only point out that the objection relies on a distinction between sense and significance he does not articulate in his paper. See Putnam (1995, pp. 301f ). According to such a distinction, ‘system of star arranged at the vertices of a 100-gon (in a region of space otherwise free of stars) outside of our light cone’ would have significance or meaning but lack any sense.

Pragmatism, Pluralism, and the Pierce Principle

47

Wittgensteinian touch. Putnam cares to underline that Peirce and James anticipated the sort of Verificationism that Carnap and Reichenbach made popular in contemporary philosophy and makes the best to remind the reader that the defenders of the Verification Principle gave the pragmatists the credit of having hit on it. As far as James is concerned, this historical note may be not disputed. Dewey (1988) pointed out it that rather than a pragmatist James was an empiricist – and James’ insistence on particular sensible experiences and possible differences in perceptual flux hardly leaves any room to dispute this claim. The second way of interpreting the Peirce Principle is adopted by those who think that the principle does not run the risk of being assimilated to Verificationism. From this point of view, which Dewey not only found in Peirce’s writings but shared himself as well, the crux of the matter lies in the link between concepts’ (and thoughts’) contents and linguistic expressions’ meaning, on the one hand, and the consequences of their use, on the other hand. ‘Consequence’ means here the subsequent actions, both real and possible, that the deployment of concepts and the use of terms and sentences give rise to. In Dewey (1988) it is openly acknowledged that ideas and concepts, in addition to being devices that allows us to register past experiences, are also means to anticipate future experiences and to guide possible human action. Human reason, said Dewey, is constructive. Thus, from this point of view the sensible quality of experience makes little showing. Peirce himself challenged the right to connect Pragmatism with Phenomenism, a link that James turned into a central doctrine of his philosophy. Peirce’s strategy for purging Pragmatism from empiricist principles is a straight one, namely the need for distinguishing in the sensible effects that concepts’ (and thoughts’) content consist of – in other words, their value – their purely qualitative features, their sensuous aspect, from their rational purport (see Dewey 1916, p. 711). This second aspect really is the relevant one. It is not differences in the qualitative features of experience, but differences in rational purport, what differences in content and meaning amount to. In “How to Make Our Ideas Clear” a variant of this strategy is explicitly brought up (Peirce 1998). Peirce begins by identifying the sensible effects of our concepts and thoughts with beliefs. In more exact terms, the sensible effects that concepts, thoughts and linguistic expressions have emerge into consciousness as beliefs: Here, then, let us apply our rules. According to them, reality, like every other quality, consists in the peculiar sensible effects which things partaking of it produce. The only effect which real things have is to cause

48

Juan José Acero

belief, for all the sensations which they excite emerge into consciousness in the form of beliefs. (Peirce 1998, pp. 43f )

No matter which maneuver we resort to, the final result is the same: what fixes the content identity of concepts and thoughts and the meaning of words and sentences is the consequences they have for the subsequent action. There is no distinction of either content or meaning that cannot consist in “anything but a possible difference of practice” (Peirce 1998, p. 44). All this lead us to conclude that, though Putnam holds that Peirce provided the Pragmatic Maxim with a verificationist reading, the only support that can be given to Putnam’s claim comes from Peirce’s application of his principle to the doctrine of transubstantiantion of wine and bread into Christ’s blood and flesh, in order to argue that the dispute between Protestants and Catholics lacks any base (cf. Peirce 1998, pp. 44f ). Nevertheless, the example should not turn the anecdote into a category. It is perfectly reasonable to literally interpret Peirce’s mention of sensible effects as referring to the taste and smell effects that drinking wine produces in the wine drinker, but it would be to go too far to assimilate any other dispute to this one. Putnam himself acknowledges that pragmatists applied the Peirce Principle, i.e. the pragmatic method, to the analysis and comparison of different metaphysical systems and rightly claims that this use of the principle proves to what extent Pragmatism gave up the sort of Phenomenism that goes hand to hand with Verificationism.

5. Transcendence and Convergence in Dewey’s Views of Education It has been argued above that a pluralistic view of reality is compatible with two requirements that are built into the Absolute View: a requirement of transcending subjective or particular perspectives on the world – the Transcendence Requirement – and a requirement of convergence of those perspectives with further human and institutional viewpoints – the Convergence Requirement. It has been held that a pragmatist philosopher is not obliged to abandon either of those two demands. However, it is possible to go beyond this conciliatory remark and defend that they can be found in the classical pragmatist curriculum. More specifically, Dewey’s views on the philosophy of education, as put forward in his book Democracy and Education, give those two desiderata of the Absolute View a central place in explaining the emergence of thought.

Pragmatism, Pluralism, and the Pierce Principle

49

As far as convergence is concerned a brief comment will be enough. The question, What is education, says Dewey, is the question of how to achieve that the child’s beliefs and further psychological attitudes converge on their elders’ beliefs and attitudes and come to share them. In other words, “our problem is to discover the method by which the young assimilate the point of view of the old, or the older bring the young into like-mindedness with themselves” (Dewey [1916] 1966, p. 11). Dewey thinks that education is a matter of creating a new environment, of changing the young’s environment into the elder’s environment, thus bringing about the convergence of the former on the latter. A living being’s environment consists of every condition that stimulates or inhibits its characteristic activities. When we come to a human being’s education, the environment she has to share is a social environment, one in which those characteristic activities cannot develop without taking the activities of other human beings into account. To educate a human being requires, first of all, giving rise to those habits that control the ways in which natural responses to stimuli occur. This amounts to the passage from a more natural environment to a less natural one. Any success is a step forward in training, i.e. in replacing those stimuli that, in virtue of their natural tendencies, have influence on human beings’ activities by new stimuli. However, it is not the ability to be successfully trained what separates human beings from the rest of living beings. To train a living being is one thing, to educate a human being is another one. To educate a young is to create a social environment by arousing in her an interest for what Dewey describes as “the social use to which his action is put” (Dewey [1916] 1966, p. 13). The idea is to modify the subject’s intellectual and emotional attitudes in order for her to share both the thoughts and emotions that prevail in the community the subject belongs to. In educating a person “[sh]e not merely acts in a way agreeing with the actions of others, but, in so acting, the same ideas and emotions are aroused in h[er] that animate the others” (Dewey [1916] 1966, pp. 13f ). Convergence on a group’s beliefs and emotions is the main desideratum of education. It is not only convergence of environments, that is, convergence of beliefs, desires and, in general, goals and values, but convergence of emotions and motivations as well. For an individual to be in line with the members of her group, it is necessary to behave the way the others behave, and this requires sharing the same environment with them. To converge, therefore, brings with itself truly significant advantages for the one who achieves it, since not doing it implies not being part of the group. One of those prizes, not to be overlooked just now, is that of “free intercourse and communication of experience”

50

Juan José Acero

(Dewey [1916] 1966, p. 99), the present experience and the one that accumulated over the generations. The need for transcending our more-or-less parochial perspectives on reality also plays an important role in Democracy and Education. In fact, the significance of this requirement goes beyond Dewey’s strictly educational worries and deeply touches the fundamental questions of pragmatist philosophy. There is no exaggeration in the idea that the Transcendence Requirement is the key to the analysis of the origin of thought and the nature of our experience that this work of Dewey’s counts on. In our experiences Dewey finds two elements, one passive and one active. In virtue of its passive ingredient, any experience is an undergoing. The active ingredient is the one that gives any experience the character of a trying or an essay. Thus, there is in any experience a representational ingredient, and this means a cognitive factor. Now, experiences are not what they are if its active element is ignored, because both elements are not disconnected. We manage to learn from our experiences, because they are linked in the appropriate way. “We do something to the thing and then it does something to us in return: such is the peculiar combination” (Dewey [1916] 1966, p. 139). We try out things and, as a result of our attempts, we suffer or feel their effects on us. Our actions are experiments we perform on the world (or on the right part of it), and our joys and sufferings become patterns of action that relate them to further joys and sufferings. Learning from experience is relational in nature. In learning there are backwards-looking relationships, thanks to which we are often able to correct the missed shot and to validate the right try; and there are forward-looking relationships that help us to anticipate future experiences by connecting our actions with the effects on us that result from our having intervened in the world. From this analysis Dewey distills two philosophically interesting remarks. First, experience is only secondarily a cognitive affair. Second, that an experience’s value “lies in the perception of relationships or continuities to which it leads up” (Dewey [1916] 1966, p. 140). It follows that the fundamental form of thought “is the discernment of the relation between what we try to do and what happens in consequence” (Dewey [1916] 1966, pp. 144f ). The step from the fundamental way of thinking to the derived forms of thought is the step between thought based on the trial-and-error method and thought due to what Dewey calls reflection. In the kind of thought that follows the trialand-error method, the minute details of the link that connects action and its effects are opaque for us. We simply ignore why our action brings about the affects. On the contrary, reflection is in the origin of action

Pragmatism, Pluralism, and the Pierce Principle

51

because reflection makes us aware of those minute details, i.e. of why actions’ effects occur. Reflective thought turns its possessor into a kind of agent whose abilities amount to much more than “a piece of registering apparatus” (Dewey [1916] 1966, p. 147) of our experiences’ effects. The ability to reflect allows transcending, first, the human agent’s insertion among things to see them from a wider perspective, a viewpoint progressively less dependent on his or her particular contingencies. Thus, we typically reflect, that is, we plainly think, when we calculate what it might happen, though it has not happened (cf. Dewey [1916] 1966, p. 147). Such a capacity speaks of us, human agents, as having got a view of our insertion in reality as an active part of the world causal traffic. Campbell (1994) has argued that the emergence of human subjectivity, i.e. the creation of a gap between a living being that sees him or herself as acting on the world, on the one hand, and the world itself, on the other, correlates with the awareness of such a insertion. In the second place, and as the culminating moment of the process, reflective thoughts result from the need that agents have for transcending their immediate involvements with those situations they live in the middle of. Dewey describes this interesting point by saying that the reflective agent has to get a “certain detached impartiality,” a separation that demands “keeping one’s self out of the data” (Dewey [1916] 1966, p. 147). This is not the only basis on which it can be attributed to Dewey a deep commitment to the need for transcending our own perspective on the world. It was claimed above that the requirements of transcendence and convergence are not independent from each other. Dewey was perfectly aware of such a nexus when he held that human beings’ becoming part of a social environment decisively pushes them to keep on transcending their particular stances: There is, however, no incompatibility between the fact that the occasion of reflection lies in a personal sharing in what is going on and the fact that the value of the reflection lies upon keeping one’s self out of the data. The almost insurmountable difficulty of achieving this detachment is evidence that thinking originates in situations where the course of thinking is an actual part of the course of events and is designed to influence the result. Only gradually and with a widening of the area of vision through a growth of social sympathies does thinking develop to include what lies in beyond our direct interests: a fact of great significance for education (Dewey [1916] 1966, pp. 147f ).

52

Juan José Acero

No doubt can be cast on Dewey’s recognition of the transcendence requirement. This desideratum is built into the very possibility of human thought.

Acknowledgements I express my gratitude to Hilary Putnam for his comments on the first version of this chapter. I also thank Christopher Hookway and Russell Goodman for having suggested me how to develop some ideas I bring up in the second and third sections of this chapter. The last two sections below contain my own way of developing those suggestions. However, neither Hookway nor Goodman is responsible at all for the conclusions arrived at. The research that has made possible to obtain them belongs to the research project HUM-2004-118, financed by the Spanish Ministerio de Educación y Ciencia.

University of Granada Department of Philosophy 18011 Granada, Spain e-mail: [email protected]

REFERENCES Baker, L. Rudder (1993). Metaphysics and Mental Causation. In: J. Heil and A. Mele (eds.), Mental Causation, pp. 75-95. Oxford: Clarendon Press. Baker, L. Rudder (1995). Explaining Attitudes. A Practical Approach to Mind. Cambridge: Cambridge University Press. Burge, T. (1993). Individualism and Psychology. Philosophical Review 95, 3-45. Campbell, J. (1994). Space, Time and Self. Oxford: Oxford University Press. Cussins, A. (1990). The Connectionist Construction of Concepts. In: M.A. Boden (ed.), The Philosophy of Artificial Intelligence, pp. 368-440. Oxford: Oxford University Press. Cussins, A. (1992). Content, Embodiment and Objectivity – The Theory of Cognitive Trails. Mind 101, 651-688. Cussins, A. (2003). Content, Conceptual Content, and Nonconceptual Content. In: Y. Gunther (ed.), Essays on Nonceptual Content, pp. 133-163. Cambridge, MA: The MIT Press. Dewey, J. ([1916] 1966). Democracy and Education. New York: The Free Press. Dewey, J. (1916). The Pragmatism of Peirce. Journal of Philosophy 13, 709-715. Reprinted in: Peirce (1998), pp. 301-308.

Pragmatism, Pluralism, and the Pierce Principle

53

Dewey, J. (1988). The Development of American Pragmatism. In: The Later Works of John Dewey, vol. 2, pp. 3-22. Carbondale and Edwardsville, IL: Southern Illinois University Press. Evans, G. (1982). The Varieties of Reference. Edited by J. McDowell. Oxford: Oxford University Press. Gibson, J.J. (1986). The Ecological Approach to Vision. Hillsdale, NJ: Lawrence Erlbaum. James, W. (1907). Pragmatism. A New Name for Some Old Ways of Thinking. New York: Longmans, Green, and Co. Revised edition in: William James: Writings 1902-1910, pp. 478-625 (New York: The Library of America, 1987). Kim, J. (1989). The Myth of Nonreductive Materialism. Proceedings of the American Philosophical Association 63, 31-47. Lewis, D. (1999). Reduction of the Mind. In: Papers in Metaphysics and Epistemology, pp. 291-324. Cambridge: Cambridge University Press. Mayr, E. (1982). The Growth of Biological Thought. Cambridge, MA: Harvard University Press. Nagel, T. (1986). The View from Nowhere. Oxford: Oxford University Press. Peirce, C.S. (1998). How to Make Our Ideas Clear. In: Chance, Love and Logic: Philosophical Essays, pp. 32-60. Lincoln, NE: University of Nebraska Press. Putnam, H. (1975). The Meaning of ‘Meaning’. In: Philosophical Papers, vol. 2 [Mind, Language and Reality], pp. 215-271. Cambridge: Cambridge University Press. Putnam, H. (1990a). Realism with a Human Face. Edited by J. Conant. Cambridge, MA: Harvard University Press. Putnam, H. (1990b). James’ Theory of Perception. In: Putnam (1990a) , pp. 232-251. Putnam, H. (1994). Reductionism and the Nature of Psychology. In: Words & Life, pp. 428-440. Edited by J. Conant. Cambridge, MA: Harvard University Press. Putnam, H. (1995). Pragmatism. Proceedings of the Aristotelian Society 95, 291-306. Putnam, H. (1999). The Threefold Cord. Mind, Body and World. New York: Columbia University Press. Putnam, H. (2000b). Richard Rorty on Reality and Justification. In: R.B. Brandom (ed.), Rorty and His Critics, pp. 81-87. Oxford: Basil Blackwell. Putnam, H. (2002). McDowell’s Mind and McDowell’s World. In: N. Smith (ed.), Reading McDowell, pp. 174-190. Oxford: Blackwell Publications. Putnam, H. and R.A. Putnam (1990). William James’ Ideas. In: Putnam (1990a), pp. 217-231. Rorty, R. (1982). Pragmatism, Relativism and Irrationalism. Consequences of Pragmatism. Brighton, Sussex: The Harvester Press. Rorty, R. (1999). Pragmatism as Antiauthoritarism. Review Internationale de Philosophie 207, 7-20. Rorty, R. (2000). El pragmatismo, una versión. Antiautoritarismo en epistemología y ética. Barcelona: Ariel. Williams, B. (1999). Ethics and the Limits of Philosophy. Cambridge, MA: Harvard University Press.

Christopher Hookway DICHOTOMIES: FACTS AND EPISTEMIC VALUES

ABSTRACT. The paper explores Putnam’s denial of the “fact/value dichotomy.” After attempting to identify the main themes in this aspect of Putnam’s thought, I explore its implications for our understanding of epistemic evaluation and also consider its relations to some similar moves by other philosophers in the pragmatist tradition. The final section examines an argument of Putnam’s which is sued to suggest that such a dichotomy can be self defeating when applied to epistemic evaluation.

1. Dichotomies Since the early 1980s, Putnam has emphasised one important point of continuity between his philosophy and that of the pragmatists. This involves challenging, and rejecting, some fundamental philosophical or “metaphysical” dichotomies. In Reason, Truth and History (1981, Ch. VI), there was emphasis on the need to reject the dichotomies (or dualism) of subjective and objective, of the epistemically given and that which is constructed on the basis of what is given, and of fact and value. In a more recent book, The Collapse of the Fact/Value Dichotomy (2002), these ideas are developed, with the main emphasis on this last dichotomy – although there is also an illuminating discussion of the analytic/synthetic and a priori/a posteriori dichotomies. I shall begin by making some general remarks about this notion of a (metaphysical) dichotomy or dualism and then address some questions about the implications of Putnam’s claims about the interpenetration of fact and value of our understanding of what we would normally think of as “descriptive,” “empirical” and “scientific beliefs.” His most recent book, Ethics without Ontology (2004) takes these points further, offering important clarifications of the metaphysical implications of what he is saying. This further formulation suggests that we do not need an

In: M.U. Rivas Monroy, C. Cancela Silva, and C. Martínez Vidal (eds.), Following Putnam’s Trail: On Realism and Other Issues (Pozna Studies in the Philosophy of the Sciences and the Humanities, vol. 95), pp. 55-69. Amsterdam/New York, NY: Rodopi, 2008.

56

Christopher Hookway

ontological account of the metaphysical differences between facts and values, understood as fundamental elements of reality. My primary interest, then, lies in what happens when one abandons the “fact/value” dichotomy. However, I shall follow Putnam in beginning with a more familiar case where philosophers have abandoned an entrenched dualism: this is Quine’s rejection of the dualism of analytic and synthetic propositions. It is helpful to start in this way. This case is easier to grasp and more widely endorsed – indeed Putnam comments on the number of philosophers (including Quine himself) who have been ready to give up this dichotomy but have had no hesitation in retaining a fundamental distinction between (objective) facts and subjective (values). Hence examination of the analytic/synthetic dichotomy will prepare us for appreciating the force of similar considerations in connection with fact/value. The first section of the paper will introduce the notion of a “dichotomy” that is in play here and consider what is involved in rejecting such a dichotomy. Then, in section two, we comment on some of Putnam’s claims about the interplay of the factual and the evaluative in our understanding of some distinctive concepts that are used in making evaluations and, in section three, we turn to some models for thinking about what might be called factual concepts, in the light of this: how are evaluative considerations involved in their use and understanding. The final sections introduce some arguments, endorsed by Putnam which appeal to the role of epistemic values in scientific inquiry in order to launch an attack on the fact/value dichotomy. In the last two sections (three and four) I shall compare Putnam’s ideas with some similar themes in classical pragmatism. There are many valuable distinctions we can draw which can answer to different cognitive purposes. We can distinguish mental states such as pains from paradigmatic physical states such as the mass of a piece of matter or the charge of a particle. We can distinguish paradigmatically ethical or aesthetic beliefs from everyday descriptions of the objects in our surroundings. We can distinguish those propositions whose truth can be established through proof or calculation, from those that must be tested experimentally. We can distinguish those propositions that must be accepted by anyone who understands them (“All bachelors are unmarried”) from those which must normally be tested through observation (“It has just started to rain”) and so on. Putnam, as I understand him, has no quarrel with our drawing most such distinctions and he recognizes that doing so may be valuable in philosophy and elsewhere. The errors, which he, along with other pragmatists, wishes to

Dichotomies: Facts and Epistemic Values

57

reject arise only when such distinctions are transformed into “dichotomies.” Indeed Putnam agrees with Dewey that, in these cases we need both to understand “the importance of respecting the distinction” while recognizing the damaging effects of inflating it into a dichotomy. The first question I want to consider concerns what happens when one of these distinctions is “inflated into a dichotomy” (Putnam 2002, p. 7). In the first chapter of The Collapse of the Fact/Value Dichotomy, Putnam provides an answer to this question, in connection with the example of the analytic/synthetic distinction. First, once it is a dichotomy, it is: (a)

[Wielded] as if it comprised an exhaustive classification of all possible judgments.

It is also treated as “omnipresent”: (b)

The distinction was conceived as something that could be applied to absolutely every meaningful judgment in absolutely every area.

(Putnam 2002, p. 10)

Once this is done, and the distinction becomes a “metaphysical bogey,” then we respond to cases where it is unclear whether propositions are analytic or synthetic by assuming that they are either ambiguous or semantically flawed or unclear. Issues about whether fundamental postulates of scientific theories are analytic or synthetic become pressing in a way that is an obstacle to philosophical progress. Quine’s achievement was to recognize (in his later work anyway) that while there may be some statements that are easily identified as analytic, and some others that were easily recognized as synthetic, the vast majority of the statements with which we deal – especially the philosophically important ones – could not be easily forced into this straitjacket. To describe them either as “analytic” or as “synthetic” would be a mistake, for both sensitivity to evidence and something analogous to conventional decision can be inextricably involved in their acceptance. If we do accept the dichotomy, then, according to Putnam, we are required to treat “analytic” and “synthetic” as both capturing natural kinds of statements or propositions, and, indeed, as providing a system of kinds which exhaustively classifies well behaved statements. Note that what we do here is classify kinds of statements: typically philosophers will look to the theory of meaning to tell us what it is for a statement to be analytic and to tell us which statements are analytic, and so on. Different accounts of the distinctions can be provided. Although the details are not important for this paper, it is striking that, both in connection with the analytic/synthetic distinction and the fact/value

58

Christopher Hookway

dichotomy, there has been a tendency to treat one side of the dichotomy as substantive and objective (synthetic, factual), and the other side as lacking objective content, as empty or tautological (analytic) or as subjective (value). What is important is that, typically, the defender of the dichotomy requires a semantic account of what it is for a statement to be analytic (or synthetic). This positive account of analyticity must explain the modal properties of analytic statements (why they are necessary) and also explain their epistemological properties (why they are knowable a priori). And, in the same way, the positive account of syntheticity must explain why (at least some) synthetic statements are contingent, and why at least some are testable empirically. The view of the distinction that Quine was attacking urged that analytic statements owed their truth to the semantic rules or conventions that governed their use, while the meaning of a synthetic statement would typically specify which experiences would count as confirming or refuting it. In recent analytic philosophy, we can see rejection of such a dichotomies as passing through two stages. A famous passage of Quine’s suggests that the meanings of most statements incorporate both conventional and testable elements: The lore of our fathers is a fabric of sentences [. . .]. It is a pale gray lore, black with fact and white with convention. But I have found no substantial reasons for concluding that there are any quite black threads in it, or any white ones. (Quine 1966, p. 132)

This seems to suggest that all statements somehow share semantic features with paradigmatically analytic statements as well as sharing features with typically synthetic ones. As Putnam emphasises, this is an unattractive picture and, indeed, another passage from Quine suggests that he, too, would not want to be interpreted in this way. We may adopt some proposition as true “by convention,” to be held true come what may at a particular stage of inquiry. But this characterizes an “act” and not its “enduring consequences”: conventionality is “a trait of events and not of sentences” (Quine 1966, pp. 112-113). It thus identifies a way in which we can use sentences on particular occasions, but it says nothing about their enduring meanings or the grounds of their truth. As time goes by, describing this statement as held true by convention is, at best, very misleading, describing it by reference to an episode in of the history of its use which has no lasting significance. At most we can investigate why some statements are more suited for this sort of use (in particular kinds of circumstances) than others. But that does require a semantic account that determines whether the statement is analytic or synthetic. Similarly we can treat a statement as analytic at some time: anyone who did not

Dichotomies: Facts and Epistemic Values

59

accept it would be identified as failing to understand some of the words it contains. But our semantic account of the sentence need not entail that the statement must always be treated as analytic in this way. And such a change need not involve anything that would be seen as a conventional change of meaning. So we have a distinction we is reflected in different uses we make of statements, and a few statements may be such that all their uses fall on one or other side of this distinction, but there is no deep semantic dichotomy here. And something like this, I take it, is Putnam’s view when he rejects the analytic/synthetic “dichotomy.” It is one of Putnam’s claims that we face a fully parallel position with respect to the fact/value distinction. Indeed, as noted above, he lines up with those who think it an inconsistency in Quine’s work that he wants to retain a dichotomy between (subjective) values and (objective) facts while abandoning that between analytic and synthetic statements. We can distinguish ethical judgments from statements of scientific theory; maybe we can also distinguish “thin” expressions of goodness or badness from simple observational reports of our surroundings. But we need to recognize that “factual” and “evaluative” judgments to not represent two fundamental natural kinds into which all judgments can be made to fit. And this is partly for the same reason that we cannot elevate the analytic/synthetic distinction into a dichotomy: we lack a theoretical grasp of a notion of a “factual” or “synthetic” statement which can be used as a foil for recognizing the distinctive (probably subjective) character of values (Putnam 2002, p. 19ff ). Many judgments display a mix of “evaluative” and “descriptive” features, and these elements cannot be “disentangled” or “extricated” one from another. As in the previous case, description and evaluation capture some different uses we make of judgments, but there is no reason to think (as many philosophers have supposed) that our semantic theories must identify these as distinctive kinds of judgments. Sometimes, saying that an act as cruel is to describe it; and sometimes it is to make an evaluation. And asking whether the judgment is descriptive or evaluative simply makes no sense (1981, Ch. VI; 2002, Ch. II; McDowell 1981). Although there may be a few purely statements that can only be used to make evaluations, and perhaps a few that can only be used for nonevaluative purposes, and although there may be a variety of nonevaluative uses of language of which “describing” is only specialised example, it does seem that adopting this kind of picture does have implications for our understanding of our language. We might follow the

60

Christopher Hookway

suggestion of Quine’s passage about the lore of our fathers, expecting accounts that explain: 1. 2.

3. 4.

In what respects are (e.g.) mathematical claims more like ordinary descriptive/empirical claims than the “dichotomy” allows? In what respects are ordinary empirical claims (say) more like what are traditionally thought of as analytic judgments than the dichotomy allows? In what respects are paradigmatic ethical or evaluative claims more like ordinary descriptive claims than the dichotomy allows? In what respects are paradigmatic descriptive claims more like evaluative claims than the dichotomy allows?

Even if, like Putnam, we do not like that way of putting things, we need to be able to understand how statements can be used in these different ways. In what follows, I shall pursue some Putnamian themes, and some pragmatist themes, which have a bearing on how there can be statements that can be used both evaluatively and non-evaluatively.

2. Entanglement A number of different arguments are used to show that the distinctions that can be drawn in these areas should not be inflated into dichotomies. One, is found in Putnam’s writings since Reason, Truth and History, and applies most directly to question 3. As Putnam has expressed this recently, it holds that so-called “thick” ethical concepts “are counterexamples to the claim that there is an absolute fact/value dichotomy” (2002, p. 35). Deep “entanglement of fact and value” becomes evident, he says, when we examine words like ‘cruel’. This word clearly has normative, indeed ethical, uses: “If one asks me what sort of person my child’s teacher is, and I say, “He is very cruel,” I have both criticized him as a teacher and criticized him as a man. I do not have to add “He is not a good teacher” or “He is not a good man”” (2002, p. 34). And Putnam continues that I cannot intelligibly say of someone “He is a very cruel teacher and a very good man.” Yet, Putnam continues. ‘Cruel’ can be used wholly descriptively, as when I describe a monarch as cruel. (And presumably, there would be no linguistic error in describing someone as “wonderfully cruel.”) Putnam concludes: “‘Cruel’ simply ignores the supposed fact/value dichotomy and cheerfully allows itself to be used sometimes for a normative purpose and sometimes as a descriptive term.” Once we take note of words or concepts that have this

Dichotomies: Facts and Epistemic Values

61

sort of mixed descriptive/normative function, he suggests, we come to see that the factual/evaluative straitjacket does not yield a happy system of categories for classifying judgements or statements. We can sometimes distinguish different kinds of uses of statements involving this word; and often it may be very unclear whether we should regard a use as normative or descriptive. There is no metaphysical dichotomy here. We may go further, as I think both Putnam and McDowell do, by arguing that although this word can be used in a largely descriptive way, it can only be understood, as used descriptively, by someone whose interests and attitudes enable them to use the word normatively, or, at least, by someone who can understand the use in ethical evaluations. Even what seems to be issued as a normatively neutral descriptive characterisation of a monarch my involve descriptions that cannot be wholly divorced from the possibilities implicit in the normative use. Someone with absolutely no sense of what was wrong with cruelty could not understand how the varieties of phenomena we describe as cruel belong together. The evaluative use is somehow implicated in the way that the extension of the term is determined. I do not intend to get involved in the debates about how far this set of claims can be defended. Detailed consideration of examples would be needed to show that the normative use of ‘cruel’ is not somehow secondary, drawing on common attitudes towards a kind of behaviour that can be described using a purely “descriptive” concept of “cruelty,” or that the meaning of ‘cruel’ cannot be, as it were, pulled apart into an evaluative and a properly descriptive component, it being a contingent matter that these two components have been somehow fused into the use of one word.

3. Entanglement and “Descriptive” Terms It is interesting to ask how far this strategy extends. Is it applied solely to words (like ‘cruel’ and ‘kind’, for example) that really do seem to have a centrally ethical or evaluative use? The question is: if this argument works for think ethical concepts, how far does its application extend? Are there concepts that we would superficially think of as “descriptive” to which it applies? It does seem plausible that this can be done. We might think of “lamp” as a wholly descriptive concept – certainly it does not embody any ethical evaluation. But identifying something as a lamp involves regarding it as having a distinctive use, as meeting some need or purpose.

62

Christopher Hookway

Nobody who failed to grasp how lamps are used, nobody who could not evaluate how suitable something was to meet the needs which lamps are designed to need, could properly apply the concept of a lamp as we do: a sensitivity to how we evaluate things as meeting the needs which lamps are designed to meet seems to be required for applying the term and assessing what falls in its extension. Evaluation may be involved in a case like this in several ways. First, describing something as a lamp (or as any other kind of artifact) identifies it is as fit for some purpose: if we want illumination to help with reading, then we can make use of the object in question, for example. The description of something as a lamp is action-guiding: in appropriate circumstances, given appropriate information about our needs, it can yield a defeasible prescription about what we should do. Second, we need to take note of the fact that lamps serve a range of purposes and whether we should describe something as a lamp may be sensitive to our judgments about what sorts of uses are paramount in any given conversation. And, third, there are uses that are less sensitive to the normal purposes of lamps: our judgment may depend upon whether the manufacturer intended some flawed artefact as a lamp. When using concepts like lamp, we are sensitive to a wide range of pragmatic considerations, drawing the boundaries of its extension in ways that are sensitive to purposes and valuations that influence our interests in lamps or that shape the course of the conversation. Unless I can make evaluations about which of the many uses of judgments is currently most important, and about how well objects will serve the uses at issue, then I cannot properly apply the concept. I take it that most of the classical pragmatists were responsive to these kinds of questions: the example of the lamp, for example, is taken from Peirce (1931-1958, 1.205). One route that can lead to an insistence on the role of evaluation in primarily descriptive discourse is simply to insist that artefactual terms like ‘lamp’ are plainly descriptive and their use evidently depends upon evaluations of things as fit for important purposes. But it is not clear how far this route alone can lead. Even a defender of the fact value dichotomy can allow that many of our evaluations supervene on factual patterns and regularities and insist that all that we have shown is that words like ‘lamp’ (and other artefact terms) are mixed evaluative/descriptive terms. This does nothing to show that the majority of paradigmatic descriptive language is infused with evaluations. It may still be possible to argue that what goes for the semantics of artefact terms goes for descriptive predicates in general.

Dichotomies: Facts and Epistemic Values

63

When Dewey claims that “the conception that scientific judgements are to be assimilated to moral is closer to common sense than is the theory that validity is to be denied to moral judgements because they do not square with a preconceived theory of the nature of the world to which scientific judgements must refer” (Dewey [1908] 1980, p. 132), he may have something like this in mind. The key characteristic of moral judgments is that they relate to “changes to be made,” and his instance that scientific judgments all relate to “changes to be made,” that all scientific inquiry transforms the situations in which we find ourselves, making them complete rather than problematic. The style of argument we have been considering here is one that is directed at attempts to establish a fact/value dichotomy by reference to the existence of distinctive kinds of terms: both in ethics and in ordinary “descriptive/discourse” we find words being employed which are used to make evaluations and also used for non-evaluative descriptive purposes. And in each case, the “descriptive” uses are cannot be characterised in ways that are wholly independent of the evaluative ones: descriptive uses of terms such as ‘lamp’ – and the integrity of the extensions of such terms – cannot be specified independently to our needs, interests and values; and the applications of such terms frequently depends upon the use of capacities which are involved when we evaluate things such as lamps etc. We just can’t separate words into “descriptive” and “evaluative” and this is both because paradigm evaluative terms have “descriptive” uses and because paradigm “descriptive” terms have “evaluative” uses.

4. Facts and Epistemic Values: Pragmatism I now want to begin by looking at another argument, one that Putnam makes extensive use of in his book. Once we recognize that epistemic values are values, and take note of the importance of acts of evaluation within our practice of theoretical inquiry, we can see that the fact/value dichotomy cannot be drawn (Putnam 2002, Chs. 2 and 8). I want to explore just how this argument is meant to work, identify some reasons for being sceptical about it and, then, come round to defend it. There are two stages to Putnam’s claim here. First, he endorses a claim from classical pragmatism: “value and normativity pervade all experience” (Putnam 2002, p. 30). He takes it to be a consequence of this that “normative judgments are essential to the practice of science itself ” (Putnam 2002, p. 30). His examples of the normative vocabulary

64

Christopher Hookway

employed in the practice of science are varied: we use words like ‘coherence’, ‘reasonableness’, ‘simplicity’, ‘beauty’ and we discuss ‘oughts’ that govern reasoning. And he concludes that, since such value judgments are involved in the practice of science, we cannot coherently hold that values are subjective while science is objective. The second move is direct against the idea of a fact value dichotomy: Since evaluation is central to the practice of science, then the results of science are objective only if values (well, some values) are objective (2002, pp. 30-32). We cannot hold to the kind of dichotomy which treats facts as solid, objective and real; and values as subjective and unreal. I want to start by exploring the first stage a little, comparing Putnam’s use of this idea with the forms it takes in the classical pragmatist tradition. William James’s pragmatist clarifications of truth as what it is good to believe or as what puts us into a satisfactory relation to our experience provide clear illustrations of this. In this section, I want to use some of Peirce’s writings in order to distinguish two distinct themes that may be involved. Although I shall not explore this here, I think that similar points could be made on the basis of the work of both James and Dewey. The first of these themes can be identified by reference to Peirce’s famous pragmatic maxim. This is a rule for clarifying concepts and propositions. Peirce’s own formulations do make the position entirely clear, as he acknowledges. However it is important that it receives two different kinds of formulation in his work. The first, which I have called “verificationist” suggests that we can clarify what it means for some object to fall under a concept by, for example, identifying the sensible effects of this being so (see Hookway 2004). The other kind of explanation, which I have called “pragmatist” claims that we clarify our concept by identifying the implications for practice of its applying to something: we look for the effects which have “practical bearings.” In some writings (for example, “How to make our ideas clear”), Peirce appeals to some distinctive features of his theory of action to argue that the effects which have practical bearings can be identified with the sensible effects. After 1900, Peirce often used a different kind of formulation. Suppose we want to clarify what it means for some object to be F. Then we identify consequences of this claim which have a distinctive form: they are presented in conditional propositions; the antecedent of the conditional concerns “desires and circumstances,” and the consequent is in the imperative mood or identifies a “general rational mode of conduct.” So, as an example, consider the proposition:

Dichotomies: Facts and Epistemic Values

65

That object is an umbrella. We provide a pragmatist clarification of this by deriving from it conditional propositions such as: If the weather forecast predicts rain and you want to stay dry when you go out, you ought to take that with you. If we provide a complete clarification by listing conditionals of this form (it is likely to be a long list), then the content of a “factual claim” is exhausted by its implications for conduct, by the conditional imperatives that result from it. The content of the factual claim can thus be cashed out in terms of its action guiding potential, in terms of conditional claims about what we ought to do in different sorts of circumstances. This provides quite a strong sense in which the normative is implicated in the factual. The second theme, which heads in the same direction, is more difficult to make clear. In Peirce’s later work, we clarify concepts like truth, reality, and fact in the discipline of logic; and one way to understand the role of logic is as an account of the norms of inquiry – and inquiry is a sort of activity. Indeed, in Peirce’s own work, logic is represented as depending on ethics: the latter is an account of what sort of ends can be adopted unconditionally, and logic applies its results to the case of inquiry. And ethics, in turn, depends upon aesthetics, the study of what it is possible to admire unconditionally. Thus norms and values are fundamental to our understanding of concepts such as fact and reality. The standards we should follow in trying to settle what to believe are defended by showing that they are required for the responsible normative regulation of inquiry. This yields a further respect in which our grasp of the normative informs our grasp of the factual. Facts and values are intertwined in the means of factual propositions; and norms of inquiry are presupposed by our ways of settling beliefs and recognizing propositions as true.

5. Putnam’s Arguments This argument has immediate appeal, but I should like to explore in a little more detail just how it is supposed to work. How might someone resist it? Here is one possibility, a very crude one, but a useful starting point for discussion. Suppose that when I evaluate a belief as “reasonable” it is generally true (or generally one that will stand the test of time and inquiry, etc.). My evaluating it as reasonable is reliably

66

Christopher Hookway

correlated with its being a belief that will some meet specified aim in inquiry. Even a defender of the fact/value dichotomy must allow that many of our evaluations have a means/ends character, our evaluations attach to things in virtue of their empirical correlations. (We can easily envisage there being circumstances in which we do not value beliefs we judge to be reasonable – we have had too much experience of judgments of reasonability letting us down in certain circumstances. It can be a difficult question whether we are reasonable to belief what strikes us as reasonable.) So, the crude view holds: these terms of (apparent?) epistemic evaluation are, at root, judgments of reliability, or judgments whose epistemic important stems from their being reliably associated with (e.g.) truth. An externalist story of epistemic evaluation, perhaps, but it is also a useful reminder that we can use descriptive judgments to make evaluations without those judgments thereby becoming evaluative (at least for a defender of the dichotomy). It is unsurprising that we value reliability but, once again, a defender of the dichotomy is not required to hold that all our evaluations are capricious. The idea that evaluations are subjective does not require that they are capricious or that they differ wildly between people. As I said, this is a crude picture – epistemic evaluation is much more complicated than that. But the interesting question is how far a story of this kind (perhaps a more developed one) is incoherent and, if it is incoherent, why this is. It is as much an error to be too rapidly persuaded that this sort of story won’t work as it is to too crassly assume that it does. So let us ignore all the need for complexities and try to identify a deep error with this story. There are different versions of this crude s story. One is internalist: what is required is both that we rely upon reliable methods and we also know we do. A slightly less internalist version holds that we must just be properly confident that the indicators we rely upon are reliably linked to achievement of cognitive goals (e.g., truth). A third version is much more externalist: all that is required is that the indicators we employ are reliably correlated with achievement of these goals. Perhaps we must also believe that they are – in order to avoid a disabling sense of alienation – but that need not involve any further beliefs with an evaluative epistemic content. What I want from the argument so far is: it is not obvious that the use of terms of epistemic evaluation in the course of (scientific) inquiries and investigations undermines the fact/value dichotomy. How does the role of values in such inquiries undermine objectivity? There are several strategies we could follow in trying to find the source of incoherence here. One is to note the references to our cognitive

Dichotomies: Facts and Epistemic Values

67

goals in the talk of reliability: perhaps it could be objected that we talk of methods of belief formation being reliably correlated with obtaining truth and truth is an evaluative concept. But if that is the argument being employed, it is hard to see why we need the detour through justification, coherence, reasonableness and the other evaluative terms used in inquiry. The strategy that Putnam himself offers emerges of page 32 of his book: [. . .] it is not that we have some way of telling that we have arrived at the truth apart from our epistemic values and can, so to speak, and run a test to see how often choosing the more coherent, simpler, and so on, theory turns out to be true without presupposing those very standards of justified empirical belief. The claim that on the whole we come closer to truth about the world by choosing theories that exhibit simplicity [etc.] [. . .] [is itself an empirical hypothesis] that we choose because we have been guided by the very values in question in our reflection upon records and testimonies concerning past inquiries. (Putnam 2002, p. 32)

This is an attractive argument. We cannot test the reliability of various indicators of belief empirically because all we can do is see whether the belief that those indicators are reliable possesses one or more of those very indicators. Simplicity is a good indication of truth because it contributes to the simplest view of he world to believe this. Putnam’s favourite example, about which similar points could be made, is the value of being reasonable: we accept propositions on the basis of considerations that we take to be reasonable (2002, p. 144). The mere fact that we rely upon values (make evaluations and use “evaluative” vocabulary) in identifying facts does not seem to undermine the fact/value dichotomy automatically. But there clearly are circumstances in which the kind of circularity that Putnam worries about is indeed troubling. Let me mention two cases. The first which may be suggested by Putnam’s remarks cited above, becomes pressing if we combine three claims: 1. 2. 3.

Facts are real and objective Whether something is a fact is determined by whether it can be evaluated in a distinctive way. Values are subjective and unreal.

If the evaluations that are involved in identifying facts are instrumental ones, and can be defended because they are a means to identifying facts, there is no circularity. If, on the other hand values are somehow

68

Christopher Hookway

constitutive of facthood, then facts lose their metaphysical robustness unless the values display it too. This needs clarification and exploration, but it is easy to see how it could threaten the fact/value dichotomy. A similar difficulty is suggested by an argument against positivistic materialism due to Jean Hampton in her book Authority of Reason (1998). This has a parallel structure but is employed in a distinctive intellectual context. She is concerned with what she calls “naturalism” or materialism. This is a view that rests on special claims about the authority of scientific knowledge. If this naturalistic or materialist view is correct, then it could dwell turn out that the best explanation we can give of our values (especially ethical values) is a dismissive one: value judgments express feelings or reflect our social practices. The naturalistic metaphysical framework encourage a metaphysical view which can make little sense of objective values, the sort of position defended by John Mackie or Alan Gibbard. But this view of values is acceptable only if we concur with the claim that scientific knowledge possesses this special authority in settling metaphysical questions. And giving such authority to scientific knowledge is to make an evaluation of it. If this evaluation is merely the expression of a feeling, then naturalism lacks the sort of intellectual authority that it needs. So unless the evaluations that underlie the authority of science are objective, naturalism cannot be defended. So the arguments that are used to dismiss objective values can be accepted only if there are, at least some, objective values. I have described two different issues that arise when philosophers challenge what Putnam calls the fact/value dichotomy. The first concerns the interpenetration of “descriptive” and “evaluative” elements in the ways in which we understand predicate expressions. And the second concerns the apparently self-destructive ways in which some philosophers have argued for a fundamental metaphysical distinction between (objective) facts and (subjective) values.

University of Sheffield Department of Philosophy Sheffield S10 2TN, UK e-mail: [email protected]

Dichotomies: Facts and Epistemic Values

69

REFERENCES Dewey, J. ([1908] 1980). Does Reality Possess a Practical Character. In: John Dewey Middle Works 1899-1924, vol. 4, pp. 125-143. Carbondale: Southern Illinois University Press. Hampton. J. (1998). The Authority of Reason. Cambridge: Cambridge University Press. Hookway, C. (2004). The Principle of Pragmatism: Peirce’s Formulations and Examples. Midwest Studies in Philosophy 28, 119-136. McDowell, J. (1981). Non-Cognitivism and Rule-Following. In: S. Holzman and C. Leich (eds.), Wittgenstein: To Follow a Rule, pp 141-162. London: Routledge and Kegan Paul. Peirce, C.S. (1931-1958). Collected Papers of Charles Sanders Peirce. 8 vols. Edited by C. Hartshorne, P. Weiss, A.W. Burks. Cambridge, MA: Harvard University Press. Putnam, H. (1981). Reason, Truth and History. Cambridge: Cambridge University Press. Putnam, H. (2002). The Collapse of the Fact/Value Dichotomy. Cambridge, MA: Harvard University Press. Putnam, H. (2004). Ethics without Ontology. Cambridge, MA: Harvard University Press. Quine, W.V.O. (1953). From a Logical Point of View. Cambridge, MA: Harvard University Press. Quine, W.V.O. (1966). The Ways of Paradox. Cambridge, MA: Harvard University Press.

Jaime Nubiola DICHOTOMIES AND ARTIFACTS: A REPLY TO PROFESSOR HOOKWAY

ABSTRACT. In this reply to Professor Hookway’s lecture the comments are focused, first, on the topic of what dichotomies really are, since it is an illuminating way of understanding pragmatism in general and Putnam’s pragmatism in particular. Dichotomies are artifacts that we devise with some useful purpose in mind, but when inflated into absolute dichotomies they become metaphysical bogeys as it is illustrated by the twentieth century distinction between fact and value. Secondly, a brief comment on the so-called “thick” ethical concepts and artifact terms is presented, and finally it is added a word on John L. Austin, whose approach to dichotomies is aligned with pragmatism and Putnam.

The familiar contrast of “normative or evaluative” as opposed to the factual is in need, like so many dichotomies, of elimination. J.L. Austin, How to Do Things with Words (1962), Lecture XII.

Professor Hookway’s lecture concerns a key issue, perhaps the key issue of Hilary Putnam’s reflections in recent years: the putative dichotomy of fact and values – inherited from the empiricism of the Vienna Circle – that has distorted the vision of a good number of philosophers of the twentieth century. In his brilliant lecture, Hookway pays particular attention to the first and second chapters of The Collapse of the Fact/Value Dichotomy, and to its final chapter: “The Philosophers of Science’s Evasion of Values” (Putnam 2002). In my brief comments I will deal first with the topic of dichotomies, which is an illuminating way of understanding pragmatism in general and

In: M.U. Rivas Monroy, C. Cancela Silva, and C. Martínez Vidal (eds.), Following Putnam’s Trail: On Realism and Other Issues (Pozna Studies in the Philosophy of the Sciences and the Humanities, vol. 95), pp. 71-80. Amsterdam/New York, NY: Rodopi, 2008.

72

Jaime Nubiola

Putnam’s pragmatism in particular; second, a comment on the so-called “thick” ethical concepts and artifact terms, and finally a word on John L. Austin, from whom I have selected a quote as a motto for my paper.

1. Dichotomies First of all, I want to say in advance that I am in agreement with Hookway’s general presentation of Putnam’s approach to distinctions and dichotomies. I will add here some comments of a historical and lexicographic character in order to gain a clearer view of the whole issue that may help to understand better pragmatism in general and Putnam’s pragmatism in particular. In the introduction to Classical American Philosophy, a collection of writings of Peirce, James, Royce, Santayana, Dewey and Mead, John J. Stuhr identifies seven characteristics that constitute a “family resemblance” amongst the six major American pragmatist philosophers. The first such characteristic is the rejection of modern philosophy’s “dualism,” i.e. its habit of articulating and defending its positions in terms of dichotomies: subject/object, mind/matter, appearance/reality, theory/practice, facts/values, individual/community, and so on. Stuhr explains: Classical American philosophers did not refuse to use these terms; instead their point was that these notions refer to distinctions made in thought rather to different kinds of being or levels of existence. That is, these terms have a functional rather an ontological status; they stand for useful distinctions made within reflection, and not for different kinds of being, discrete and separate prior to reflection. (Stuhr 1987, p. 5)

In The Collapse of the Fact/Value Dichotomy Hilary Putnam aligns himself with this honorable tradition, crediting John Dewey’s lifelong battle against dualisms as a source of inspiration. For Putnam, distinctions are not dualisms or dichotomies (Putnam 2002, p. 9). “The errors [. . .]” – Hookway writes – “arise only when such distinctions are transformed into “dichotomies.” Indeed Putnam agrees with Dewey that, in these cases we need both to understand “the importance of respecting the distinction” [Putnam 2002, p. 10] while recognizing the damaging effects of inflating it into a dichotomy” (Hookway 2004, pp. 1-2). On Putnam’s account, harmless and innocent distinctions become – at least in some cases – metaphysical bogeys when inflated into absolute dichotomies. “One difference between an ordinary distinction and a metaphysical dichotomy [is that] ordinary distinctions have ranges of

Dichotomies and Artifacts: A Reply to Professor Hookway

73

application, and we are not surprised if they do not always apply” (Putnam 2002, p. 11). But let us stop here for a while. What is a dichotomy? In order to answer this type of question I always love to go to Baldwin’s Dictionary (1901-1905), which from time to time gives the gift of an entry by Charles S. Peirce. In the case of “dichotomy” the entry is by Robert Adamson, professor of Logic at Glasgow University, but Peirce is the coauthor of a related entry “Dualism (in Philosophy).” Here is Peirce’s definition of ‘dualism’: (1) A general tendency to divide any genus of objects of philosophical thought into two widely separate categories, as saints and sinners, truth and falsehood, &c.; opposed to the tendency to look for gradations intermediate between contraries. Especially (2) any theory which explains the facts of the universe by referring them to the action of two independent and eternally coexistent principles. (Baldwin 1901-1905, I, p. 298)

In his entry on “dichotomy,” Adamson explains that a dichotomy is “a form of logical division in which, at each step, the genus is separated into two species, determined by the possession and non possession, the presence and absence, of a mark or attribute. The species so determined satisfy the rules of division: they exclude one another, and they exhaust the extent of the genus divided.” Adamson describes cursorily the use of dichotomy by Plato and Aristotle, and suggests that the method of division by dichotomy “lies at the foundation of Jevons’ and indeed of all symbolic logic,” because it “expresses the fundamental distinction in thought between position and negation” (Baldwin 1901-1905, I, p. 279). The point I want to stress is that a “dichotomy” is mainly a form of logical division, of classification; that dichotomies appear when and where there are human beings thinking, classifying, establishing classes, collections and categories. To assert a dichotomy is to cut something into two parts which are “mutually exclusive and jointly exhaustive” (Brody 1967, p. 60). This cut is a logical one, made by our reflection: it could be said that if were nobody to do the cutting there would be no dichotomies. I like to tell my students that in the middle of the sun there are no problems; that problems only appear where there are human beings thinking. The same could be said of dichotomies. In the world there are no dichotomies, but only continuity and interaction. Along this way, the only two ontological usages of ‘dichotomy’ found in dictionaries – for instance the Oxford English Dictionary – come from astronomy (“That phase of the moon, (or of an inferior planet) at which exactly half the

74

Jaime Nubiola

disk appears illuminated”) and from botany (“A form of branching in which each successive axis divides into two; repeated bifurcation”). But the phenomenon that Putnam has detected is the metaphysical inflation of a distinction that transforms a logical division into a metaphysical dichotomy, “thought to be philosophically obligatory, both sides of the distinction were thought to be natural kinds, each a category whose members possessed an “essential property” in common” (Putnam 2002, p. 13). This could be a new usage of the term ‘dichotomy’. We human beings are very fond of making this type of division into two parts, and then promptly to forget that the division was made by us, with some practical purpose in mind in each case. We are tempted to believe that reality is divided in two according our dichotomy. As Putnam shows in detail, the transformation of a distinction into a metaphysical dichotomy turns the distinction into a Procrustean bed (cf. Putnam 2002, p. 27). Let’s recall the story from Greek mythology: Procrustes was a bandit from Attica who had a bed which he invited passersby to lie down in. Those who were shorter than the bed he would stretch until they fit, while those who were too tall would have their legs chopped short.

2. Ethical Concepts and Artifact Terms In his book, Putnam demonstrates convincingly that “the fact/value dichotomy is, at bottom, not a distinction but a thesis, namely the thesis that “ethics” is not about “matters of fact”” (Putnam 2002, p. 19). Putnam identifies three reasons why we are tempted to establish the fact/value dichotomy. The first (and more dangerous) reason is that, like the Procrustean bed, this dichotomy is a way of cutting off not only rational argument about values, but even thought: “The worst thing about the fact/value dichotomy is that in practice it functions as a discussionstopper; and not just a discussion-stopper, but a thought-stopper” (Putnam 2002, p. 44). As Putnam has written extensively elsewhere “a central insight in Dewey’s ethics (and all of Dewey’s work is in one way or another connected with “ethics”) is that the application of intelligence to moral problems is itself a moral obligation” (Putnam 2000, p. 10). This is the core of pragmatism and the distinctive mark of philosophy itself, and has been in the heart and in the texts of Putnam for at least the last two decades. Philosophy, as understood and lived by him, is “to do what Socrates tried to teach us: to examine who we are and what our deepest convictions are and hold those convictions up to the searching test of reflective examination” (Putnam 2002, p. 44).

Dichotomies and Artifacts: A Reply to Professor Hookway

75

In his lecture, Chris Hookway pays special attention to the entanglement of fact and value that appears in the use of words like ‘cruel’, that have a mixed descriptive and normative function. Putnam holds that “the so-called thick ethical concepts are counterexamples to the claim that there is an absolute fact/value dichotomy” (Hookway 2004, p. 4; cf. Putnam 2002, p. 35), and he adds that this has long been pointed out by Philippa Foot, Iris Murdoch, John MacDowell and David Wiggins in their criticisms of the dichotomy. Putnam discusses with some detail the answers of R.M. Hare and John Mackie to those criticisms, but before doing so he announces that a possible Humean response would be just to banish all thick ethical concepts to the limbo of the “emotive” or “noncognitive” right along with the “thin” ones (“good,” “ought,” “right” et al.). This would be a highly impoverished language. Putnam continues, For a Spanish reader this historical reference. Not even David Hume would be willing to classify, for example, “generous,” “elegant,” “skillful,” “strong,” “gauche,” “weak,” or “vulgar” as concepts to which no “fact” corresponds (Putnam 2002, 35). [And in a footnote Putnam adds] The words quoted in this sentence were listed as examples by José Ortega y Gasset, who noticed the phenomenon of entanglement very early. See his Obras Completas, vol. 6 (Madrid: Revista de Occidente, 1923). (Putnam 2002, pp. 317 and 320-321)

For a Spanish reader this historical reference – which already appears in Putnam’s “Pragmatism and Relativism: Universal Values and Traditional Ways of Life” (Putnam 1994, pp. 188-189) – makes a lot of sense. The quotation from Ortega proceeds from his “Introducción a la estimativa” of 1918: it was his address for entering into the “Real Academia de Ciencias Morales y Políticas” of Spain. I have defended for years – with small success until now! – the deep affinity between American pragmatism and Hispanic philosophers of the early decades of the twentieth century, such as Unanumo, Ortega and d’Ors (Nubiola 1998). In the case of Ortega, John Graham published in 1994 a careful study in which, after noting Ortega’s hostility to American pragmatism, he reveals “many basic connections, similarities and points of identity, so that concrete influence and dependence seem more plausible than “coincidence” between Ortega and James.” Graham gives evidence that Ortega read James early in his career, and that Ortega was aware that James had anticipated the notion central to his of “razón vital” (Graham 1994, pp. 145 and 147-152; see also Barzun 1984, p. 299). His evidence of James’ influences on Ortega by German sources themselves influenced by James is specially convincing (Donoso 1995, p. 499).

76

Jaime Nubiola

Back to the main line of my comment: Putnam’s discussion of these thick ethical concepts show clearly the deep entanglement of fact and value or – what in the end is the same – that the dichotomy, as a sharp division, is really impossible, as it is said by Claudine Tiercelin in her delightful book on Putnam: The border between fact and value is faint: the terms considered to have a function of ethical evaluation have also a descriptive explicative and predictive function. When somebody is qualified as “scrupulous,” at the same time one evaluates his or her character, one describes it. Understanding a human being is a mixture of estimation of his or her character and of explication and prediction of his or her actions’. (Tiercelin 2002, p. 50)

In his lecture, Hookway skillfully describes the problem of the mixture of evaluation and description that appears in these thick ethical concepts. However, for me the most interesting aspect is the unexpected turn that his reflection makes when he asks himself how far this strategy extends: Is it applied solely to words (like ‘cruel’ and ‘kind’, for example) that really do seem to have a centrally ethical or evaluative use? The question is: if this argument works for thick ethical concepts, how far does its application extend? Are there concepts that we would superficially think of as “descriptive” to which it applies? (Hookway 2004, p. 5).

Hookway chooses an excellent example in order to test the extension of the “semantic strategy” against the fact/value distinction: a ‘lamp’. This is the name of an artifact that surreptitiously – it is the same with all artifacts – reintroduces in the scene a mixture of description and evaluation: We might think of a “lamp” as a wholly descriptive concept – certainly it does not embody any ethical evaluation. But identifying something as a lamp involves regarding it as having a distinctive use, as meeting some need or purpose. Nobody who failed to grasp how lamps are used, nobody who could not evaluate how suitable something was to meet the needs which lamps are designed to need, could properly apply the concept of a lamp as we do: a sensitivity to how we evaluate things as meeting the needs which lamps are designed to meet seems to be required for applying the term and assessing what falls in its extension. (Hookway 2004, p. 5)

What is it to be a lamp? Is ‘lamp’ in fact a pure descriptive term, or does not its use involve some evaluation about the extension of the concept? As most of the readers of this volume know, Putnam’s view on

Dichotomies and Artifacts: A Reply to Professor Hookway

77

natural-kind terms as indexical can be extended also to artifacts. ‘Pencil’, ‘chair’ or ‘bottle’ were the terms used by Putnam as paradigmatic examples in “The Meaning of ‘Meaning’”: [. . .] ‘pencil’ is not synonymous with any description – not even loosely synonymous with a loose description. When we use a word ‘pencil’, we intend to refer to whatever has the same nature as the normal examples of the local pencils in the actual world. ‘Pencil’ is just as indexical as ‘water’ or ‘gold’. (Putnam 1975, p. 243)

According to Putnam, the nature of a pencil or of a lamp is not some mysterious structure underlying the surface characteristics of the object (for a discussion of this issue, see S.P. Schwartz 1978 and H. Kornblith 1980). With Aristotle we can say that the essence of an artifact is extrinsic to the artifact, is its utility, while the matter of its composition is accidental (Aristotle, Physics II, 2, 1929b 8-20): the function or utility of the artifact is the nature referred to by the user of the term ‘lamp’ or ‘pencil’. Moreover, when we think about what it is to be a “proper lamp” or a “good lamp,” we realize that there is some hidden evaluation that governs the pragmatic application of the term, as Hookway’s argument suggests. I agree with that, but my minor reply is that there are no sharp boundaries about what it is to be a lamp: a broken lamp or a miner’s lamp are also lamps, in spite of the fact that I cannot use them on my desk. (By the way, Hookway has credited Peirce for the example of the lamp, and looking in Peirce’s Collected Papers I found that there are eight occurrences of ‘lamp’ there, and one of them is “Aladdin’s lamp”! [Peirce 1931-1958, 6.552, 1887]) Let me quote here one of Wittgenstein’s comments about lamps and their boundaries in his Philosophical Investigations: One may say of certain objects that they have this or that purpose. The essential thing is that this is a lamp, that it serves to give light; – that it is an ornament to the room, fills an empty space, etc., is not essential. But there is not always a sharp distinction between essential and inessential. (Wittgenstein 1953, p. 62)

Having all this in mind, the last point I want to make in this section is that any dichotomy, like the fact/value dichotomy (which has been consistently rejected by Putnam as a metaphysical bogey), is also an artifact, a product of the human mind. It seems to me really important to realize that “dichotomies” are tools like lamps, drawn to get a clearer view of some area of philosophical discussion or in this case – as it was said earlier – unfortunately to stop any discussion at all. The making of

78

Jaime Nubiola

dichotomies was perhaps seen by the vast majority of our colleagues in the past century as a useful tool to make philosophical progress; thanks, however, to Putnam’s work we now realize that this distinction, especially when metaphysically inflated, blocks inquiry and distorts our view of rationality and human flourishing. We know now what the classic pragmatists already knew: that facts and values interpenetrate, that “value judgments are essential to the practice of science itself ” (Putnam 2002, p. 135). From a practical point of view, we have learned that the dichotomy between value judgments and judgments of facts was not really a dichotomy, i.e. a division of the judgments into two sub-classes mutually exclusive and jointly exhaustive. Rather, it was an ideological thesis made in order to avoid philosophical discussion about values. This provides ample reason for the negative overtones of the term ‘dichotomy’ in Putnam’s writings.

3. A Final Word on J.L. Austin Almost thirty years ago I did my licentiate thesis on John Austin’s analysis of truth, working closely on his William James Lectures (delivered at Harvard in 1955 and published posthumously as How to Do Things with Words). Over the last ten years I have been teaching regularly Austin’s philosophy of language, and I am persuaded of the deep affinity between Austin and the pragmatist tradition. In a footnote to chapter seven of The Collapse of the Fact/Value Dichotomy, Putnam writes: “Most of the points I have attributed to “American pragmatism” in this chapter are also made by Austin – who, like Murdoch, appears not to have read the pragmatists” (Putnam 2002, p. 171, n. 24). My impression is that the view of language that Austin presents in the last lecture of How to Do Things with Words fits pretty well with a pragmatist approach to language. But the point I want to highlight here is not only that John Austin, like Putnam, advocates the elimination of the contrast between the normative or evaluative and the factual, but he characterizes of this dichotomy in lecture XII as a “fetish.” According to the dictionary, a fetish is something “that is worshiped, because a spirit is believed to live in it,” or “to which more respect or attention is given than is normal or sensible” (Cowie 1990, p. 449). What Hilary Putnam has shown is that the “scientific spirit” that was embodied by that fetish is simply unscientific, and not only does not deserve any respect at all, but on the contrary has to be denounced systematically as a treason to philosophy

Dichotomies and Artifacts: A Reply to Professor Hookway

79

and an attack on human flourishing. This is one of the treasures of the heritage of American pragmatism, and Putnam has been instrumental and influential in its rediscovery.

Acknowledgements I want to give thanks to Concepción Martínez and to Maria Uxia Rivas for their kind invitation to take part as a discussant of professor Hookway in the II Meeting on Pragmatism held in Santiago de Compostela, on May 13-14, 2004. I want to express my enormous gratitude, both personal and intellectual, to Hilary Putnam. For the revision of my English text I am indebted with Erik Norvelle.

Universidad de Navarra Depto. Filosofía E-31080 Pamplona, Spain e-mail: [email protected]

REFERENCES Austin, J.L. (1962). How to Do Things with Words. Edited by J.O. Urmson. Oxford: Oxford University Press. Baldwin, J.M., ed. (1901-1905). Dictionary of Philosophy and Psychology. 3 vols. New York. Macmillan. Barzun, J. (1984). A Stroll with William James. Chicago: University of Chicago Press. Brody, B.A. (1967). Logical Terms, Glossary of. In: P. Edwards (ed.), Encyclopedia of Philosophy, vol. 5, pp. 57-77. New York: Macmillan & The Free Press. Cowie, A.P., ed. (1990). Oxford Advanced Learner’s Dictionary of Current English. Oxford: Oxford University Press. Donoso, A. (1995). Review of Graham’s A Pragmatist Philosophy of Life in Ortega y Gasset. Hispania 78, 498-499. Graham, J.T. (1994). A Pragmatist Philosophy of Life in Ortega y Gasset. Columbia, MI: University of Missouri Press. Hookway, Ch. (2004). Dichotomies: Facts and Epistemic Values . Typescript distributed in the Conference. Kornblith, H. (1980). Referring to Artifacts. Philosophical Review 89, 109-114. Nubiola, J. (1998). C.S. Peirce and the Hispanic Philosophy of the Twentieth Century. Transactions of the Charles S. Peirce Society 24 (1), 31-49. Ortega y Gasset, J. (1923). Obras Completas. Madrid: Revista de Occidente. Peirce, C.S. (1931-1958). Collected Papers of Charles Sanders Peirce. 8 vols. Edited by C. Hartshorne, P. Weiss, A.W. Burks. Cambridge, MA: Harvard University Press.

80

Jaime Nubiola

Putnam, H. (1975). Mind, Language and Reality (Philosophical Papers, vol. II). Cambridge: Cambridge University Press. Putnam, H. (1994). Words and Life. Edited by J. Conant. Cambridge, MA: Harvard University Press. Putnam, H. (2000). Dewey’s Central Insight . Democracy and Education in the XXIst Century: John Dewey’s Challenge and Influence. Cosenza, Italy. Typescript available at ‘http://staff.washington.edu/dalexand/Putnam.htm’. Putnam, H. (2002). The Collapse of the Fact/Value Dichotomy and Other Essays. Cambridge, MA: Harvard University Press. Schwartz, S.P. (1978). Putnam on Artifacts. Philosophical Review 87, 566-574. Stuhr, J.J., ed. (1987). Classical American Philosophy. New York: Oxford University Press. Tiercelin, C. (2002). Hilary Putnam, l'heritage pragmatiste. Paris: Presses Universitaires de France. Wittgenstein, L. (1953). Philosophical Investigations. Oxford: Blackwell.

PART II PRAGMATISM AND HUMANISM

Maria Baghramian THREE PRAGMATISMS: PUTNAM, RORTY, AND BRANDOM

ABSTRACT. Over the last several decades an increasing number of philosophers have announced their sympathies for or have become affiliated with what has become known as neo-pragmatism. The connection between the various strands of pragmatism, new and old, however, remains quite unclear. This paper attempts to shed some light on this issue by focusing on a debate between Hilary Putnam and Robert Brandom on classical and contemporary pragmatisms. Using the Brandom-Putnam debate as my starting point, I examine the relationship between the pragmatisms of Putnam and Rorty, two of the most influential neopragmatists, and argue that differing conceptions of the normative are at the heart of their disagreement. I further argue that this disagreement has similarities to, and can be illuminated by, two differing conceptions of norms in Wittgenstein’s work. I conclude that Brandom does not delineate the differences between various strands of pragmatism convincingly.

The term ‘neo-pragmatism’ has been used rather indiscriminately to characterise the philosophical affiliations of a large number of twentieth century American philosophers - the best known among whom are Quine, Sellars, Putnam, Davidson, Goodman, Stich, and Brandom. It is unclear in what sense, if any, this new pragmatism may be seen as a unified doctrine; furthermore, the connections between neo-pragmatism and the classical American pragmatism of the turn of the twentieth century are also unclear. These difficult questions have been thrown into sharp relief through a recent exchange between Hilary Putnam and Robert Brandom (in Conant and egle 2002). In this paper, I try to disentangle some of the prominent strands of contemporary pragmatism by looking at the Brandom-Putnam debate. Ironically, the debate, and hence this paper, turns out to be not so much about Brandom’s views as about Rorty’s brand of neo-pragmatism,

In: M.U. Rivas Monroy, C. Cancela Silva, and C. Martínez Vidal (eds.), Following Putnam’s Trail: On Realism and Other Issues (Pozna Studies in the Philosophy of the Sciences and the Humanities, vol. 95), pp. 83-101. Amsterdam/New York, NY: Rodopi, 2008.

84

Maria Baghramian

because, as Putnam maintains with good evidence, despite claims to the contrary, Brandom’s real target is Richard Rorty.1 I begin the paper by looking at how each member of this discordant triumvirate defines the supposedly common starting point of the debate: pragmatism.

1. The Many Faces of Pragmatism According to Robert Brandom the term ‘pragmatism’ is “a generic expression that picks out a family of views asserting various senses in which practice and the practical may be taken to deserve explanatory pride of place” (Brandom 2002, p. 41). He distinguishes between narrow and broad versions of pragmatism. Pragmatism in the narrow sense is “a philosophical school of thought centered on evaluating beliefs by their tendency to promote success at the satisfaction of wants, whose paradigmatic practitioners were the classical American triumvirate of Charles Peirce, William James and John Dewey” (Brandom 2002, p. 40). The basic idea behind the classical pragmatism, Brandom argues, is “that one can understand normative assessments of the truth of beliefs as assessments of the extent to which the holding of that belief would contribute to the satisfaction of desires. Beliefs are true insofar as they are good tools or instruments for getting what one wants” (Brandom 2002, p. 51). Brandom calls this “instrumental pragmatism.” Pragmatism, defined broadly, on the other hand, is a “movement centered on the primacy of the practical, initiated already by Kant, whose twentiethcentury avatars include not only Peirce, James and Dewey, but also the early Heidegger, the later Wittgenstein and such figures as Quine, Sellars, Davidson, and Rorty” (Brandom 2002, p. 40, emphasis added). 2 Hilary Putnam has vehemently objected to Brandom’s characterisation of classical pragmatism. He expresses dismay that despite the efforts of serious students of pragmatism to rebut this type of interpretation for almost a century, the view still receives an airing from such an eminent philosopher. According to Putnam, the position attributed to the classical pragmatists by Brandom is one that should rightly be ascribed to Richard Rorty.3 He maintains that Brandom shows a deep misunderstanding of the 1 Putnam’s suspicions are backed by a rather careless footnote in Brandom’s paper. See footnote 3 below. 2 Presumably Brandom should also be included in this list. 3 As Putnam notes, the very last footnote of Brandom’s paper more or less gives away the game. Brandom concludes: “For this reason, I think one ought to reject the global form of instrumental pragmatism, as well as the local one” (p. 58), and then in the footnote he

Three Pragmatisms: Putnam, Rorty, and Brandom

85

work of Peirce, James and Dewey (Putnam 2002a, pp. 59-65). More specifically, Putnam denies Brandom’s claims that the classical Pragmatists: Either (1) identified what is true with what promotes success in the satisfaction of wants; or (2) thought that we should forget about truth and just concentrate on finding what promotes success in the satisfaction of wants; or (3) thought that what promotes success in the satisfactions of wants is more important than what is true. (Putnam 2002a, p. 60)

It is useful to point out, in support of Putnam, that James’s own account of the genesis of pragmatism emphasized the significance of the practical and in doing so it readily fits in with Brandom’s definition of broad pragmatism. To remind ourselves, James, in his classic statement of the doctrine, credits Peirce with having coined the term ‘pragmatism’ and recounts: Mr. Peirce, after pointing out that our beliefs are really rules for action, said that, to develop a thought’s meaning, we need only determine what conduct it is fitted to produce: that conduct is for us its sole significance. And the tangible fact at the root of all our thought distinctions, however subtle, is that there is no one of them so fine as to consist in anything but a possible difference of practice. To attain perfect clearness in our thought of an object, then, we need only consider what conceivable effects of a practical kind the object may involve – what sensations we are to expect from it, and what reactions we must prepare. Our conceptions of these effects, whether immediate or remote, is then for us the whole of our conceptions of the object, so far as that conception has positive significance at all. (James 1907, p. 86)

As James’s account shows, the core of classical pragmatism is the emphasis on the lived experiences and practices we engage in our encounters with the world. Putnam’s characterisation of his brand of pragmatism is scattered in his writings of the past twenty years and amounts to a nuanced and complex picture. One way to understand Putnam’s pragmatism is to see it as the negation of a philosophical position he rejects – the position he calls “metaphysical realism.” He defines metaphysical realism as the philosophical perspective according to which:

adds: “Accordingly, I find a major tension in Rorty’s thought, between his robust appreciation of the transformative potential of new vocabularies and his continued appeal to instrumental models for thinking and talking about them” (p. 215).

86 (a)

(b) (c)

(d)

(e)

Maria Baghramian

The world consists of some fixed totality of mind-independent objects, (a view defended, among others, by Wittgenstein in the Tractatus), and it has a fixed totality of properties. There is exactly one true and complete description of “the way the world is.” There is a sharp distinction between properties we “discover,” i.e., the world, and the properties we “project” onto the world. Similarly, there is a sharp distinction between factual judgements and value-judgements. Truth involves a relation of correspondence between words, thought-signs, or propositions and external things and sets of things. It is possible to take an externalist, or God’s eye point of view, on reality. The empirical sciences describe such a conceptindependent and perspective-independent reality. A “finished science” will provide us with a full account of what there is. (Based on Putnam 1981, pp. 49-56; 1987, p. 4; and 1990.)

Putnam finds these assumptions of metaphysical realism, incoherent, rather than merely false. Contra the metaphysical realists, he argues that we cannot have a view of the world that does not reflect our interests and values – and in this he echoes sentiments expressed by James. But contrary to the relativistic interpretations of pragmatism, implicit in Brandom’s definition of instrumental classical pragmatism, he maintains that we are “committed to regarding some views of the world – and, for that matter, some interests and values – as better than others” (Putnam 1990, p. 210). Putnam believes that Brandom’s conception of classical pragmatism, or pragmatism in the narrow sense, comes close to the relativism he wishes to reject but is an accurate description of Rorty’s views on the subject. Pragmatism as a positive doctrine, for Putnam, comes down to subscribing to the following theses: (I)

The rejections of various unhelpful and pernicious dualisms, including, most importantly, the subjective-objective dualism and its close relation, the fact-value dichotomy (e.g., Putnam 1985 and 2002). Like the classical pragmatists, Putnam believes that all knowledge of fact presupposes value judgements (Putnam 2002b, p. 60). In this he echoes the view of Dewey who denied the existence of a dividing line between moral and scientific knowledge or facts and values and who argued, “to frame a theory of knowledge which makes it necessary to deny the validity of

Three Pragmatisms: Putnam, Rorty, and Brandom

(II)

(III) (IV)

(V)

87

moral ideas, or else to refer them to some other and separate kind of universe from that of common sense and science, is both provincial and arbitrary” (Dewey 1908, p. 81). Fallibilism: All beliefs are open to revision and all interpretations and methods of enquiry have a provisional authority only (Putnam 1994, p. 152). This is a philosophical position also common to the classical pragmatists, Dewey, James and Peirce (Putnam 2002a, p. 60). Antiscepticism: “pragmatists hold that doubt requires justification just as much as belief does” (Putnam 1994, p. 152). The thesis that practice, including practical reason, is primary in philosophy (Putnam 1994, p. 152; and 1995). This thesis may be seen as the core of pragmatism, both new and old. Jamesian pluralism: the position that our views of the world reflect our interests and values and that our interpretations of the world are correct given the interests relevant to the context of those interpretations, and hence there could be more than one correct conception or interpretation of a given situation.4

However, Putnam has also explicitly stated that he is not a pragmatist (Putnam, R.A. 2002, p. 7) because he rejects the pragmatist theory of truth which may be susceptible to the charge of relativism.5 Over the last decade, Putnam has come to embrace, once again, a robust noninstrumental and non-epistemic conception of truth, this more recent reaffirmation of the role of truth places him in the realist rather than pragmatist camp, even though he sees his new realism as continuous with the dictates of the common sense rather than with metaphysical realism that he has rejected since the early 1970s. Rorty shares Putnam’s distaste for dichotomised thinking6 and rejects the divisions between descriptions and evaluations. Unlike Putnam, however, Rorty embraces a “sociological view of truth” whereby truth, rationality, and objectivity are to be defined in terms of the practices of a community. According to this view, the terms ‘knowledge’ and ‘truth’ are compliments paid to beliefs that we think to be well-justified, for the moment, so that no further justification is needed (Rorty 1998a). In other words, ‘true’ for Rorty is a term of commendation for exactly those beliefs that we consider well-supported at a given time. Pragmatism in 4

For this point see my (2004, Ch. 7) and Russell Goodman in this volume (pp. 125-139). This was not true of an earlier Putnam who supported the Deweyan truth-like notion of warrant or warranted assertibility (see Putnam 1987, for instance). 6 Or what, in Baghramian (2004), I call philosophical Manichaeism. 5

88

Maria Baghramian

Rorty’s hand comes down to the view that “there is nothing to be said about either truth or rationality apart from descriptions of the familiar procedures of justification which a given society – ours – uses in one or another area of enquiry” (1991, p. 23). Rorty rejects the accusation that his view is tantamount to the “self-refuting” variety of relativism he condemns, because, according to his brand of pragmatism, there is a distinction between better and worse beliefs, beliefs that are justified and those that are not. However, he claims that pragmatism entails ethnocentrism the view that “we must, in practice, privilege our own group, even though there can be no non-circular justification for doing so” (Rorty 1998a, p. 29). He argues: “there is no truth in relativism, but this much truth in ethnocentrism: we cannot justify our beliefs (in physics, ethics, or any other area) to everybody, but only to those whose beliefs overlap ours to some appropriate extent” (Rorty 1991, p. 30, n.13). He defends the inevitability of ethnocentrism by arguing that terms such as ‘warranted’ and ‘rationally acceptable’ always invite the question “to whom?”. Just as the terms ‘better’ and ‘worse’ invited the question “by what standard?”. The answer, he claims, is always “us, at our best,” where the relevant group, the us, consists of the educated, sophisticated, tolerant, wet liberals, the people who are always willing to engage in debate and keep an open mind (Rorty 1998b, p. 52). Despite Rorty’s protestations, his ethnocentrism sails perilously close to relativism because of the strong links it forges between truth and the cognitive and social practices of a community of enquirers. If we accept that there are diverse communities whose cognitive practices vary substantially and at times are in conflict with each other, then the sociological account of truth ends up giving us differing and incompatible conceptions of truth, rationality and objectivity where each of these conceptions is legitimately backed by distinct social and cognitive practices. Rorty’s ethnocentrism ultimately is a relativistic doctrine for all relativists are condemned to be ethnocentric.7 The relativist, through her claim that truth and knowledge, standards of justification and criteria of right and wrong are relative to their sociocultural background embraces a type of determinism that makes ethnocentrism inevitable. If truth is decided by the local norms of our culture, then we are condemned to believe what our culture tells us to be true. The relativist of course accepts that members of other cultures are similarly entrapped by their own cognitive and ethical norms, but that acknowledgement does not free the relativist from the bind of 7

But obviously not all ethnocentrics are relativists.

Three Pragmatisms: Putnam, Rorty, and Brandom

89

ethnocentrism, or thinking that his ethnos has the best hold on truth and justification. Rorty tries to distance himself from relativism because he wishes to disassociate from the pernicious conclusion that every point of view is as good as every other. But it’s not really clear that his ethnocentrism would achieve this goal. A similar point can be made regarding the suggestion that truth is a matter of intersubjective agreement. By tying the idea of truth to that of justification, and by making justification audience and context dependent, Rorty embraces relativism. Rorty readily admits, “justification is relative to an audience” (Rorty 1998a, p. 22), and that he cannot give “any content to the idea of non-local correctness of assertion” (Rorty 1998b, p. 60), non the less he continues denying that his views amount to relativism. But to explicate truth in terms of local correctness is to concede to one of the crucial posits of relativism: that truth is dependent on local and changing norms and conceptions.8 Rorty also argues that we can make sense of the notion of objectivity only in terms of intersubjective agreement – “there is nothing to objectivity except intersubjectivity” (Rorty 1998c, p. 72). However, given that we can cite many instances when different communities of enquirers have agreed on beliefs incompatible with ours, then to rely on the authority of intersubjective agreement is to accept that there are as many legitimate epistemic authorities as there are internally united communities of enquirers. To make truth a matter of intersubjective agreement is, in effect, to deny the distinction between truth and falsity for history is resplendent with examples of false beliefs which were accorded intersubjective agreement – where a false belief was seen as “true” and “justified,” by a community of enquirers. Rorty also claims that to be a pragmatist is to accept that truth simply is the best idea we currently have about how to explain what is going on. One problem is to decide who the relevant and significant “we” is. Given a conflict between differing scientific explanations or ethical viewpoints how are we going to decide which “we” has the best idea? Unless we assume either that, there always is a consensus on what the best idea for an explanation is, or that there is a method of grading various explanations, we are left with divergent and conflicting “truths” and relativism looms large. The point is how to arbitrate between conflicting explanations or vocabularies, if what decides whether a statement is warranted, or good to believe, depends only on local knowledge, historic and cultural conditions, and the presuppositions that inform the epistemic 8

For a discussion of different types of relativism see Baghramian (2004, Introduction).

90

Maria Baghramian

judgments of a group of thinkers. Once we start using the term ‘warranted (to us)’ rather than ‘true’ the specter of relativism begins to haunt us. It may be objected that the rupture between truth and warrant comes about only in cases where we mistakenly had assumed that we had a warrant for a belief and not in those instances where we actually have genuine warrant. The problem with this reply is that the distinction between thinking that one is warranted and actually being warranted is the distinction between true (warrant) and false (warrant) and hence the distinction between truth and falsity.9 Rorty comes close to conceding that a substantive notion of truth is eliminable by arguing that the only use of truth which could not be eliminated from our linguistic practices with relative ease is the cautionary use. That is the use we make of the word when we contrast justification and truth, and say that a belief may be justified but not true. According to him, this cautionary use is used to contrast less-informed with better informed audiences, past audiences with future audiences. (Rorty 2000, p. 4). This is the use of ‘true’ in statements such as “although your statement satisfies all our contemporary norms and standards, and I can think of nothing to say against your claim but still, what you say might not be true.” Rorty thinks that this cautionary use is a gesture toward future generations or the “better us” as he calls them (Rorty 1998c, pp. 60-61). So even this cautionary use is interpreted in sociological terms, and relies on an ethnocentric approach to what may count as true. There is much in common to Putnam and Rorty’s conceptions of pragmatism. Both authors reject metaphysical realism, the correspondence theory of truth and an ontology of facts; they deny the intelligibility of there being “a view from no-where” and argue against absolutist conceptions of truth and knowledge; they emphasize the role of lived experiences; they also tend to trace their philosophical genealogy to classical American pragmatism in general and Dewey (and to varying degrees James) in particular, and acknowledge the formative role of later Wittgenstein on their thought. And yet, Rorty and Putnam are distanced by a considerable philosophical gulf. It is to their differences that we now turn.

9

For a discussion of relativism about truth in general and a more detailed treatment of some of the above points see Baghramian (2004, Ch. 4).

Three Pragmatisms: Putnam, Rorty, and Brandom

91

2. Norms and Their Standing A central issue at the heart of the varying conceptions of pragmatism I have discussed is the status and role of norms and values. Cognitive or epistemic norms, truth, rationality, justification, coherence and such like, provide us with rules or at least guideline as to how we ought to reason, infer or conceptually engage with the world. Similarly, ethical norms in the moral domain provide us with guidelines for our interactions with other people and the world. The status and the source of the authority of cognitive or epistemic norms are central to the disagreement between Rorty and Putnam’s differing conceptions of neo-pragmatism and are also crucial to Brandom’s discussion of the topic. According to Brandom, as we saw, the classical pragmatists “endorse a normative pragmatics” (Brandom 2002, p. 58), but they emphasise the instrumental role that norms play in our conceptual and epistemic economy. Cognitive performances can be seen as better or worse, correct or incorrect, insofar as they contribute to the agent’s success in securing some end or achieving some goal. Brandom’s own view of norms is noninstrumental. Norms are implicit in discursive practices. They are commitments that can be understood as social statuses, instituted by practical attitudes of participants in an essentially social linguistic practice (Brandom 2002, p. 54). Both Putnam and Rorty wish to emphasize the central role of norms and epistemic values in our cognitive economy. For Putnam, normative discourse is indispensable not only to what scientists do, or to scientific inquiry, but also to our social and personal life, for we “cannot escape making value judgements of all kinds in connection with activities of every kind. Nor do we treat these judgements as matters of mere taste, we argue about them seriously, we try to get them right . . .” (Putnam 1994, p. 154). Rorty agrees with Putnam that the terms ‘knowledge’ and ‘truth’ have a normative role, but unlike Putnam, he deems their authority to be merely local – to be confined to the linguistic and social practices of a given community. Putnam and Rorty, then, part company in their constructions of the nature and sources of norms, the vexed question of whether their authority is based on intersubjective agreement only or whether they have some form of objective authority. A strong feature of Putnam’s philosophical outlook – throughout the many changes it has undergone – is a belief in the possibility of a robust conception of right and wrong, truth and falsehood. The difficulty facing him now, as in the past twenty years, is how to reconcile this quest for objectivity with his insistence

92

Maria Baghramian

that all such judgements are also context-dependent. Putnam’s initial solution was to appeal to the epistemic notion of idealised rational acceptability, a view that he came to reject in the 1990s. For a period, in Realism with a Human Face for instance, he appealed to the Deweyan notion of warrant (short for the Dewey’s technical notion of “warranted assertibility”) and argued that: In ordinary circumstances, there is usually a fact of the matter as to whether the statements people make are warranted or not [. . .]. Whether a statement is warranted or not is independent of whether the majority of one’s cultural peers would say it is warranted or unwarranted. (Putnam 1990, p. 21)

In his most recent book, Putnam has appealed to the core but nebulous Kantian idea of the authority of reason. According to his current thinking objectivity is a matter of “judgments of the reasonable and the unreasonable” (Putnam 2004, p. 71). Logical statements, methodological value judgments in science and most ethical judgments are objectively true or false in this sense of objectivity. Reasonableness is not the outcome of the activities of a transcendent metaphysical faculty, rather what is and what is not reasonable is decided in the context of the concerns of the specific enquiry at hand. For Rorty, on the other hand, objectivity, in so far as it can be made sense of, is a question of socially sanctioned warrant, there is no more to objectivity than the achievement of intersubjective agreement by members of an ever-expanding community of enquirers. For pragmatists, he argues: The desire for objectivity is not the desire to escape the limitations of one’s community, but simply the desire for as much intersubjective agreement as possible, the desire to extend the reference of “us” as far as we can. Insofar as pragmatists make a distinction between knowledge and opinion, it is simply the distinction between topics on which such agreement is relatively easy to get and topics on which agreement is relatively hard to get. (Rorty 1991, p. 23)

Rorty’s version of objectivity, as we saw, leads to relativism – a philosophical cul-de-sac that Rorty himself wishes to avoid. Rorty also emphasises what he sees as his Darwinian or naturalist conception of normative concepts such as truth. Language is a tool for coping with our environment, rather than representing it, he argues, and the normative elements of language are rules for enabling us to use this tool more effectively. Rorty does not see any tension between his sociological account of warrant and the naturalist, Darwinian approach, but the two

Three Pragmatisms: Putnam, Rorty, and Brandom

93

strands of his thought are not fully compatible. For instance, it is unclear how Rorty will explain the prevalence of irrational (and in that sense non-adaptive) ways of thinking and dealing with the world.10 Putnam’s view on the status of norms, on the other hand, leaves us with the perennially problematic questions: whose reason? and which standard of reasonableness? As Putnam himself accepts, reason is historically embedded and informed by the context in which embodied human beings find themselves. Even in logic, and not just in ethics, there are disagreements on what counts as reasonable. Consider, for instance, the disagreements between classical logicians and relevant and paraconsistent logicians on the status of the law of excluded middle and even the principle of non-contradiction. Whose norms of reasonableness are we going to accept? Moreover, there is ample empirical evidence to show that human beings deviate quite substantially from accepted norms of reason in their thinking. For instance, as various studies have shown, experimental subjects tend to reason very badly in standard tests which supports the common sense belief that human beings do not reason according to a single standard norm of rationality (see Stich 1990, p. 4). Therefore, Putnam cannot rely on empirical arguments to establish the universality of reason. Without a more secure footing for his appeal to the authority of reason and reasonableness, is he condemned to Rortyan relativism? Do all good pragmatists – new and old – end up either as relativists or as ethnocentrics?

3. Pragmatism, Relativism, and Wittgenstein The issues facing Putnam and Rorty’s pragmatism has parallels in competing interpretations of the views of later Wittgenstein – a philosopher who has been a source of inspiration for both Putnam and Rorty. To appreciate Wittgenstein’s relevance to the present discussion we need to go back to Brandom’s position paper. Brandom directs his critical remarks towards what he calls the “narrow conception of pragmatism” but believes the broader version to be both important and interesting. Wittgenstein as well as Rorty, according to Brandom, belong to this broader tradition. Putnam, on the other hand, believes

10

Steven Jay Gould and Richard Lewontin have, with good reason characterised the attempts to use natural selection as an explanation for every human trait as “Panglossian” for they resemble Dr. Pangloss’s argument, in Voltaire’s Candide, that humans have noses so that their glasses can be kept in place.

94

Maria Baghramian

Wittgenstein to be fundamentally non-Rortyan (see for instance Putnam 1992, 1995, and 2002).11 Which of these interpretations is correct? Wittgenstein may be seen as a pragmatist insofar as he emphasises the role of socially regulated practices and public use in creating meanings. No rules, including rules of logic, Wittgenstein argues, exist independently of the occasions of using them. In language, as in reasoning, “in the beginning was the deed” (Wittgenstein 1968, p. 31). Instead of looking for abstract relations between language and the world, we are asked to pay attention to the lived contexts in which language is used. Linguistic communication is a rule-governed social activity which takes place within the framework of a whole host of other social activities (or what he calls “language-games”). All human life, including our conceptual life, takes place within a cultural, social and biological context, or a “form of life.” Language, thought and action cannot be understood in isolation from the activities, the goals and the needs of the players of specific language-games within the background of their form of life. And at all times we should remember that “What has to be accepted, the given, is – so one could say – forms of life” (Wittgenstein 1958, p. 226e). Rorty and Putnam’s differing conceptions of norms have strong parallels with two seemingly distinct strains in Wittgenstein’s attempts to give an account of the rule-governed nature of our conceptual dealings with the world. According to Rorty, the later Wittgenstein, like James and Dewey, belongs to the group of philosophers who have “kept alive the suggestion that, even when we have justified true belief about everything we want to know, we may have no more than conformity to the norms of the day” (Rorty 1980, p. 367, emphasis added). For Rorty’s Wittgenstein, as Putnam also notes, what is true or false in different language-games is determined by criteria internal to those games, according to this conception, “there is no such thing as one language game being better than another except in the sense of ‘better relative to certain interests’” (Putnam 1995, p. 33). This is the Wittgenstein who explains “the normativity of rule – following in terms of conformity to the standards of a community” (Putnam 1996, p 243). This reading turns the later Wittgenstein into a relativist, admittedly a sophisticated and subtle one, but a relativist non-the-less. 12 11

Wittgenstein’s own relationship with Pragmatism has been the subject of much debate (see, for instance, Russell Goodman 2002). Wittgenstein did read James and seemed to have had a high opinion of him. However, I am going to put aside these historical connections for the purposes of this paper. 12 See Blackburn (2004) and O’Grady (2004) – two treatments of the question of

Three Pragmatisms: Putnam, Rorty, and Brandom

95

Several lines of thought in Wittgenstein’s work are conducive to this relativist reading. The emphasis he places on the role of the communal and shared nature of the form of life that informs all conceptual activities finds a sympathetic ear among the cultural relativists. Wittgenstein seems to be arguing that all judgments can meaningfully arise only from within the context of a form of life, and that there is no possibility of standing outside all forms of life in the hope of making objective, external comparisons or offering criticisms of the comparative merits of various belief systems. Furthermore, in several places Wittgenstein seems to be envisaging alternative language-games or forms of life which embody concepts and conceptions very different from the ones we find familiar. For instance, he describes communities with different approaches to measuring length or quantity (e.g., using rubber rulers), or ones with alternative ways of counting (e.g., counting the same object twice). In the same way, he suggests, there could be forms of life that use rules of logic and processes of reasoning substantially different from the ones we take for granted. Alternative forms of life can have alternative conceptions of reason and reasonableness, or alternative cognitive norms. Even the rules of logic and standards of reasonableness, he seems to say, are relative to the form of life from which they emanate and in which they are embedded. In Zettel he tells us: “I want to say: an education quite different from ours might also be the foundation for quite different concepts. For here life would run on differently” (Wittgenstein 1972, p. 387). The Rortyan Wittgenstein emphasises the contingency of all human activities. He wants to show that the role of thinking and inferring in our life is both defined and circumscribed by our social and cultural activities, for as a good pragmatist he has to accept that in the beginning was the deed and the deed was bounded only by the dictates of societal conventions. The Rortyan Wittgenstein, in Philosophical Investigations, announces: “To obey a rule, to make a report, to give an order, to play a game of chess are customs (uses, institutions)” (1958, §199). Since customs, frequently, are entirely culture-specific, one natural reading of this passage is that Wittgenstein delineates and interprets the act of rulefollowing, the quintessential manifestation of normativity, in purely culturally contextualised terms. This is Wittgenstein engaged in Rortyan pragmatism, but it is not at all clear if this is the pragmatism that Brandom wishes to defend. Almost simultaneously, Wittgenstein appears to be distancing himself from relativism. In On Certainty, for instance, he claims that a man cannot make a mistake unless he already judges in conformity with Wittgenstein’s relativism. I have discussed Wittgenstein’s relativism in my (2004, Ch. 3).

96

Maria Baghramian

mankind (Wittgenstein 1968, §156), a comment that seems to emphasize the universal features of human judgment. This is the Wittgenstein Putnam champions. Putnam agrees that for Wittgenstein, a form of life is a background to all speech and thought. A form of life is constituted by our physical nature and particular socio-linguistic conventions, but not merely by them for the normative elements governing reason and thought go beyond the merely natural and conventional. In a telling passage in Remarks on the Foundations of Mathematics he argues: Then according to you everybody could continue the series as he likes; and so infer anyhow! In that case we shan’t call it “continuing the series” and also presumably not “inference.” And thinking and inferring (like counting) is of course bounded for us, not by an arbitrary definition, but by natural limits corresponding to the body of what can be called the role of thinking and inferring in our life. (Wittgenstein 1978, §116)

This is the Putnamian Wittgenstein. This Wittgenstein is denying that rules can exist outside those practices which invoke them, or that they can be understood independently of their applications. He accepts that rules cannot exist independently of the social occasions of using them, but, yet this emphasis on the social character of thought should not be taken as a form of relativism. According to Putnam, “the Wittgensteinian strategy [. . .] is to argue that while there is such a thing as correctness in ethics, in interpretation, in mathematics, the way to understand that is not by trying to model it on the ways in which we get things right in physics, but by trying to understand the life we lead with our concepts in each of these distinct areas” (Putnam 1996, p. 263). The question for Putnam’s Wittgenstein, and indeed Putnam himself, is how to cash out the idea of correctness so that while retaining the pragmatist ideal of the primacy of practice we do not succumb to the temptations of Rortyan relativism. This very question, I believe, is at the heart of Wittgenstein’s selfquestioning. He asks: So you are saying that human agreement decides what is true and what is false?

And responds: – it is what human beings say that is true and false; and they agree in the language they use. That is not agreement in opinions but in form of life. (Wittgenstein 1958, §241)

It is difficult to know how exactly to interpret this passage. What does Wittgenstein mean by “agreement in form of life”? One possible, but unsatisfactory, reading is that what is true and false, in some sense, is

Three Pragmatisms: Putnam, Rorty, and Brandom

97

decided by the (agreed) conventions of a shared form of life, and in that sense it is relative to it. A second possible reading is to see “agreement” as a limit concept which places specific constraints on all (human) forms of life. The constraints may operate in two distinct but complimentary ways. The agreement on form of life, firstly, may arise out of the natural and biological dimensions of human existence. Putnam acknowledges the importance of this type of constraint on the limits of a form of life and in doing so highlights an interesting connection between Dewey and Wittgenstein. According to Putnam: Like John Dewey, [. . .], Wittgenstein has a naturalistic (but not a reductionist) view of man. We are not mere animals, but our capacities for understanding and for reasoning are capacities which grow out of more primitive capacities which we share with animals (Dewey spoke of “biosocial continuity” in this connection). (Putnam 1992, p. 175)

Rorty, who advocates a Darwinian naturalism but sees its role as one of the main points of his disagreement with Putnam, would agree with much of this. However, while according to Rorty we are mere animals, for Putnam there are further constraints at play – constraints that belong more firmly to the space of reason. The further constraint operating on form of life, the constraint that imposes agreement within a given form of life, has to do with the connection between foundational activities – such as thinking and inferring – within a language game and the norms that govern these games.13 The role of thinking and inferring in our life sets limits to what can be seen as intelligible instances of these cognitive activities just as the role of the activity of measuring in a form of life, to take a favourite Wittgensteinian example, limits what may be seen as an accurate and useful systems of measurement. The role or function of taking measurements provides us with norms that are constitutive of and foundational to the very activity of measurement and simultaneously define what would count as a usable system of measurement. Such constitutive norms are non-contingent in the limited sense that the activity under consideration is defined by them and achieves its coherence and applicability from them. They are also foundational in the Wittgensteinian sense that “one might say these foundation walls are carried by whole house” (Wittgenstein 1968, §248; also in Putnam 2004, p. 63). Such norms are also universal to the effect that they are binding 13

I am not sure if Putnam would endorse this position, but I think the various things he has said on Wittgenstein over the last decade are not out sympathy with the position canvassed above.

98

Maria Baghramian

on all those who engage in activities of that particular kind. A similar point can be made regarding the norms governing the activity of counting, on the one hand, and the role of counting in a form of life. In Remarks on the Foundations of Mathematics, for instance, Wittgenstein says: For what we call “counting” is an important part of our life’s activities. Counting and calculating are not – e.g. – simply a pastime. Counting (and that means: counting like this) is a technique that is employed daily in the most various operations of our lives. And that is why we learn to count as we do: with endless practice, with merciless exactitude; that is why it is inexorably insisted that we shall all say ‘two’ after ‘one’, ‘three’ after ‘two’ and so on. – “But is this counting only a use, then; isn’t there also some truth corresponding to this sequence?” The truth is that counting has proved to pay. – “Then do you want to say that ‘being true’ means: being usable (or useful)?” – No, not that; but that it can’t be said of the series of natural numbers – any more than of our language – that it is true, but: that it is usable, and, above all, it is used. (Wittgenstein 1978, pp. 37-38)

The pragmatist overtones of the passage, both in Brandom’s narrow and broad senses, are quite striking. But what is also noticeable is the emphatic sounding claim that we call “counting” is an important part of our life activities. To be able to count is to be able to follow a set of rules and to follow them the same way that others do. Wittgenstein is denying that norms can exist outside the actual practices which involve their application, this is part of his anti-platonism, norms have no reality independently of their actual applications. He allows that there can be many possible interpretations for any given rule. This is a part of his purported relativism. But he also emphasises that the possibility of a plurality of interpretations does not imply that we will be correct in following rules according to our individual interpretation, because such rules possess a social character and their correct application is determined in a communal, rather than solipsistic setting. Acknowledging this social character is a crucial aspect of the social pragmatism of Wittgenstein. This, I think, is non-controversial. But to allow that the correct application of norms is a communal affair should not lead to the conclusion that their authority is only social as well. To know how to count, to know that “we shall all say ‘two’ after ‘one’, ‘three’ after ‘two’ and so on” is to follow the norms and rules that define, and are constitutive, of the act of counting within our form of life. To paraphrase Wittgenstein, “Counting is of course bounded for us, not by an arbitrary definition, but by natural limits corresponding to the body of what can be

Three Pragmatisms: Putnam, Rorty, and Brandom

99

called the role of counting in our life” (based on Wittgenstein 1978, p. 20). What goes for counting also goes for other, core and universal, language games. Cognitive norms often have both constitutive and foundational roles. The very act of thinking presupposes logic, to debate or even to have a coherent discussion with someone is to respect the law of non-contradiction, to do science is to engage with the world within the sphere of cognitive norms that in part define science. This seems even true of moral norms, to be a human being (rather than a psychological aberration) is to have ethical concerns – for our ability to have moral engagements with others and the world is part of what makes us human. Thus to be able to participate in a (human) form of life and to engage with others within it is to be governed by a set of norms. The binding force of these norms, as well as the constraints imposed by nature, brings about agreement in form of life and mitigates against mere conventionalism. Both Rorty’s and Putnam’s Wittgenstein are real enough. Rorty is right in emphasising the contingency of all norms. Since human being and thinking inevitably have a social dimension Rorty’s emphasis on the social character of norms and values is well judged. However, there is a sense of the “normative,” emphasised by Putnam’s Wittgenstein, which by virtue of being constitutive and not merely regulative of certain types of activity goes beyond the socially defined contingent rules and norms that Rorty emphasises. Such norms transcend the merely local and yet are not independent of all human activities and interactions. On the face of it, there is a real tension between the two Wittgensteins, just as there is a real tension between the Pragmatisms of Putnam and Rorty. The tension, however, is a result of the inherent complexity of the role of norms in our thought and action. The tension might somewhat dissipate once we realize that at least some key cognitive norms have both social and constitutive roles. Norms become concrete, and in that sense actual, only through their applications in concrete and specific social conditions. They have no reality outside of such applications. Yet, some norms, both epistemic and moral, also have a constitutive role which is prior to their usage in specific social conditions. In this limited sense, one could say, norms transcend the activities that make them manifest. This conception of the normative assessments of the truth and reasonableness of beliefs is far removed from the one Brandom attributes to pragmatism narrowly defined. The issue of the contribution that beliefs make to the satisfaction of desires, from this perspective, although not wholly irrelevant, is subsumed under the wider consideration of the pre-

100

Maria Baghramian

conditions of the intelligibility of foundational cognitive practices such as thinking, reasoning, inferring, etc. If the above is correct then the lines delineating different strands of neo-pragmatism are to be drawn quite differently form those suggested by Brandom. Putnam and Rorty, separated by their distinct conceptions of the normative, fall into distinct groupings which may be called objective versus social pragmatism. Wittgenstein, whom Brandom groups with Rorty, hovers between the two pragmatisms, but it would certainly be wrong to group him with Rortyan sociologism and to ignore his Putnamian objectivism. Wittgenstein’s work shows the signs of the real tension between the objectivist and the social interpretation of norms but it also contains the first signs of a solution to this real tension. What of the classical neo-pragmatists? Is Brandom correct to see them as advocating a view that ultimately emphasises the contingent and situational features of norms, most significantly, the role they play in the satisfaction of desires? The question is too large to be addressed here. Suffice it to say that, as in the case of Wittgenstein, we can find both the objective and sociological features of the normative in the work of James and Dewey. However, this should not be seen as a sign of philosophical sloppiness on their part, objective and social pragmatisms are not necessarily philosophical rivals; they are the two sides of the coin minted by the human condition.

University College Dublin School of Philosophy Dublin 4, Ireland e-mail: [email protected]

REFERENCES Baghramian, M. (2004). Relativism. London: Routledge. Blackburn, S. (2004). Relativism and the Abolition of the Other. International Journal of Philosophical Studies 12 (3), 245-258. Brandom, R. (2002). Pragmatics and Pragmatisms. In: Conant and egle (2002), pp. 40-58. Conant, J. and U.M. egle, eds. (2002). Hilary Putnam: Pragmatism and Realism. London: Routledge. Dewey, J. (1908). Does Reality Possess a Practical Character? In: Goodman (1995), pp. 79-91. Goodman, R.B. (2002). Wittgenstein and William James. Cambridge: Cambridge University Press.

Three Pragmatisms: Putnam, Rorty, and Brandom

101

Goodman, R.B., ed. (1995). Pragmatism: A Contemporary Reader. New York: Routledge. O’Grady, P. (2004). Wittgenstein and Relativism. International Journal of Philosophical Studies 12 (3), 315-337. Putnam, H. (1981). Reason, Truth and History. Cambridge: Cambridge University Press. Putnam, H. (1987). The Many Faces of Realism. La Salle, IL: Open Court. Putnam, H. (1990). Realism with a Human Face. Cambridge, MA: Harvard University Press. Putnam, H. (1992). Renewing Philosophy. Cambridge, MA: Harvard University Press. Putnam, H. (1994). Words and Life. Cambridge, MA: Harvard University Press. Putnam, H. (1995). Pragmatism. Oxford: Blackwell. Putnam, H. (1996). On Wittgenstein’s Philosophy of Mathematics. Proceedings of Aristotelian Society, Supplementary Volume 70, 243-264. Putnam, H. (2002a). Comments on Robert Brandom’s Paper. In: Conant and egle (2002), pp. 59-65. Putnam, H. (2002b). The Collapse of the Fact/Value Dichotomy. Cambridge, MA: Harvard University Press. Putnam, H. (2004). Ethics without Ontology. Cambridge, MA: Harvard University Press. Putnam, R.A. (2002). Taking Pragmatism Seriously. In: Conant and egle (2002), pp. 7-11. Rorty, R. (1980). Philosophy and the Mirror of Nature. Oxford: Blackwell. Rorty, R. (1991). Philosophical Papers, vol. 1 [Objectivity, Relativism, and Truth]. Cambridge: Cambridge University Press. Rorty, R. (1998a). Is Truth a Goal of Inquiry? In: (1998d), pp. 19-42. Rorty, R. (1998b). Putnam and the Relativist Menace. In: (1998d), pp. 43-62. Rorty, R. (1998c). John Searle on Realism and Relativism. In: (1998d), pp. 63-83. Rorty, R. (1998d). Philosophical Papers, vol. 3 [Truth and Progress]. Cambridge: Cambridge University Press. Stich, S. (1990). The Fragmenation of Reaosn. Cambridge, MA: The MIT Press. Wittgenstein, L. (1958). Philosophical Investigations. Oxford: Blackwell. Wittgenstein, L. (1964). Philosophical Remarks. Oxford: Blackwell. Wittgenstein, L. (1968). On Certainty. Oxford: Blackwell. Wittgenstein, L. (1972). Zettle. Oxford: Blackwell. Wittgenstein, L. (1978). Remarks on the Foundations of Mathematics. Cambridge, MA: The MIT Press.

Ulvi Doguoglu PUTNAM BEYOND PUTNAM: UNDERSTANDING, PRAGMATISM, HUMANISM

ABSTRACT. It seems natural to think that one of the businesses of the philosophy of language is to provide an explanation of the understanding of language. It is the aim of the present essay to show that and why this is not so. Instead, I shall hold that both meaning and understanding go beyond the scope of traditional philosophy of language. Philosophers widely agree today that the meaning of linguistic expressions is to be explained by recourse to the use of words and sentences; an explanation of this use does not only take recourse to linguistic but to any kind of human practices, and according to Wittgenstein these in turn can neither be fully described from outside these practices nor can they be theoretically captured once and for all. On the account suggested here a model of understanding in general yields a model of the content of linguistic expressions. Such a model of understanding calls for a renewal of philosophy as well as a recasting of philosophical argument and procedure. To sketch this project I shall review Putnam’s struggle for an adequate theory of meaning and reference. In trying to disconnect understanding from meaning as far as psychological aspects are concerned, Putnam catches a “problem of reference.” A comparable problem finds reflection in Kripke’s influential reading of Wittgenstein on rules and private language. In both cases the problem grows from an inadequate use of two central notions, namely objectivity and subjectivity. An alternative reading of Wittgenstein along the lines of McDowell and later Putnam allows a new, pragmatic view on understanding, its subjects, and its objects.

It seems natural to think that one of the businesses of the philosophy of language is to provide an explanation of the understanding of language. It is the aim of the present essay to show that and why this is not so. One line of argument to this conclusion could be to hold that meanings go beyond what can be grasped in an understanding of

In: M.U. Rivas Monroy, C. Cancela Silva, and C. Martínez Vidal (eds.), Following Putnam’s Trail: On Realism and Other Issues (Pozna Studies in the Philosophy of the Sciences and the Humanities, vol. 95), pp. 103-124. Amsterdam/New York, NY: Rodopi, 2008.

104

Ulvi Doguoglu

linguistic expressions and that therefore understanding is not among the subjects of philosophy of language. This is, however, not the line taken here, even though I shall pass by this idea in a detour to Hilary Putnam’s initial version of semantic externalism. On the contrary, I shall uphold that both meaning (in a broad sense including reference) and understanding go beyond the scope of traditional philosophy of language. Here is why: philosophers widely agree today that the meaning of linguistic expressions is to be explained by recourse to the use of words and sentences; according to Wittgenstein, an explanation of this use does not only take recourse to linguistic but to any kind of human practices and these in turn can neither be fully described from outside these practices nor can they be theoretically captured once and for all. In short, along these lines an account of meaning transgresses the boundaries of philosophy of language (at least in any traditional sense); its very enterprise fails and becomes obsolete. On the account suggested here, then, a model of understanding in general yields also a model of the sense or meaning of linguistic expressions. Since any practical human performance presupposes understanding, and philosophy as an activity not only depends on such practices but as an endeavour takes them as the subject of reflection and survey, the notion of understanding becomes central to philosophy. What is more, in disrupting traditional disciplinary boundaries understanding calls for a renewal of philosophy as well as a recasting of philosophical argument and procedure. To sketch this rather broad project I shall review Putnam’s struggle for an adequate theory of meaning and reference. In trying to disconnect understanding from meaning for its psychological aspects, Putnam trades in a “problem of reference”: in shunning reference from subjectivity, mind’s understanding grip on objective reality goes by the board. The same problem finds reflection in Kripke’s influential reading of Wittgenstein on rules and private language. In both cases the problem grows from an inadequate use of two central notions, namely objectivity and subjectivity. An alternative reading of Wittgenstein along the lines of McDowell and later Putnam allows a new, pragmatic view on understanding, its subjects, and its objects.

1. A Problem of Understanding In his early writings Putnam proposes a non-psychological notion of linguistic understanding in terms of intelligibility: a sentence, Putnam

Putnam beyond Putnam: Understanding, Pragmatism, Humanism

105

claims, is, given that it is more or less grammatically correct, “clearly intelligible” for a certain group “if almost any speaker (in the relevant group) can think of discourses in which [it] could occur without any kind of linguistic or situational inappropriateness, and can paraphrase it readily in those discourses” (Putnam 1975, p. 308). On the other hand, a sentence is not intelligible if competent speakers cannot imagine contexts for this very sentence, if they cannot or not readily paraphrase it, or if it is simply grammatically ill-formed. There are two levels of understanding at work here which I shall discuss in turn: 1 grammar and the lexical parts of semantics. A third level, implicit in the quoted passage in the talk of appropriateness of occurrence, is pragmatics. This level determines the irreducibly pragmatic part of semantics, namely reference and truth-evaluable content.2 On the level of grammar the condition for intelligibility is grammaticality or grammatical wellformedness requiring the elements of an expression to be grammatically congruent. In most cases wellformedness is a matter of degree due to the fact that in everyday language we are often confronted with slightly agrammatical expressions that remain perfectly intelligible. Thus, if someone said “I like your sister, he is a nice person,” we would usually assume that the correct pronoun is supposed to be “she.” Appearing in a book, most people would simply take the wrong grammatical sex as a typo not obstructing understanding in any relevant respect. Even though full grammatical wellformedness is not a strictly necessary condition for understanding, grave violation of the syntactic congruence of words renders an expression unintelligible. For purpose of illustration consider the wellknown example: (1)

Furiously sleep ideas green colourless (Chomsky 1968, p. 15).

This sentence is grammatically ill-formed and thus affects intelligibility severely – understanding already fails on the level of grammar here.3 Grammaticality, then, is a first condition of understanding, but in point of fact is not sufficient for a sentence to be intelligible. If a 1 The distinctions regarding grammar, semantics and pragmatics draw from Schulthess (2001). 2 Another relevant level that is hardly ever mentioned and often neglected is phonology. See Kittay (1987, Ch. 6) for a slight radicalization of Quine’s radical translation along such lines. 3 Note, by the way, that grammar and semantics are always tied to a certain language. Thus a linguistic expression is only meaningful relative to a certain language.

106

Ulvi Doguoglu

linguistic expression is grammatically well-formed it possesses a syntactic meaning constituted by the syntactic composition of its parts. Together with the lexical meaning of the categorematical words the linguistic expression or sentence has, according to Putnam, a meaning entirely detached from any particular pragmatic situations. In “The Meaning of ‘Meaning’” this distinction between syntactic and lexical meaning of words is captured by the notions of “syntactic markers” on the one, and “semantic markers” and “stereotypes” on the other hand (Putnam 1975c, p. 269). These parts of the meaning of a linguistic expression are just as independent of their linguistic surroundings (a wider text and other sentences, for example) as they are of particular pragmatic situations, even though they are not intelligible parted from every discourse and situation. For our concerns, this second level of relevance for intelligibility which is traditionally part of semantics might be called “lexical semantics,” or simply “lexicon.” To clarify this issue the following three sentences may serve as examples: (2)

Colourless green ideas sleep furiously.

Even though we cannot complain from a purely grammatical point of view, the meanings of the words in (2) do not fit the grammatical form for semantic reasons in at least three respects: (i)

(ii) (iii)

talk of “colourless green” is inconsistent (except used metaphorically in order to point out that something displays a pale green and looks almost colourless); the notion of ideas is neither compatible with colour-talk nor with the uses of ‘sleep’; ‘furiously’ is an adverb that belongs to (intentional) actions whereas sleeping is not such an action.

Comparable entanglements are directed at in the next example: (3)

Dead linguists smoke buildings (Putnam 1975b, pp. 307f ).

Actions are – apart from a certain genre of frightening stories and movies – not reconcilable with dead persons. Further, even though humans smoke things you would never dream about, buildings are certainly not among these. These oddities notwithstanding, a meaningful expression can easily be construed (at least as regards speech) by taking “Dead Linguists” to name a punk-band and “Buildings” to refer to a new brand of cigarettes. This different understanding of the sentence is brought about by alterations in the grammatical syntax of parts of the expression

Putnam beyond Putnam: Understanding, Pragmatism, Humanism

107

and by changes of their referents. Thus, even though the name of the band still means “linguists who are no longer alive,” it does not denote all and every dead linguist but designates a group of musicians. Hence it does not play the same role in (3) as initially supposed. Finally an example of Carnap’s: (4)

Caesar is prime (Carnap 1931, p. 227).

In this case subject and predicate do not fit because of categorical incompatibility; their logical syntax does not match, as Carnap would say. Yet ‘Caesar’ could apart from designating a person be taken to be the name of a number (as ‘Euler’) thereby smoothing out the incompatibility. To sum up, on this semantic or (more precisely) lexical level intelligible expressions have their content constituted by the lexical meanings and the syntactic configuration apart from any particular pragmatic context. While on the level of grammar the necessary condition is congruence, the decisive requirement on the semantic level is consistency. The sketched line of argument might raise the following rejoinder: why not just say that these sentences are perfectly intelligible and hold that someone attributing an action to a dead linguist is plainly saying something false? Dead linguists don’t do anything, hence they do not smoke. For the line taken here it is central that these sentences are held to be unintelligible instead. The question of compatibility of lexical semantics is related to the traditional question of analyticity. Rejecting Kant’s analytic-synthetic distinction, Putnam (1975a) has shown that statements traditionally held to be analytic do actually fall into one of two classes. Either they do not only depend on lexicon and grammar but include pragmatic aspects as well (as in the case of statements containing socalled law-cluster concepts); or their truth indeed draws on lexicon and grammar solely. This latter type of statements is not exceedingly interesting and a philosophical theory dealing with it is of philosophical value only in contrast to the first category revealing what we might, in allusion to Wittgenstein, call the “grammar of practice.” The relevance of such disclosure lies in its exclusion of certain combinations of words from language: To say “This combination of words makes no sense” excludes it from the sphere of language and thereby bounds the domain of language. But when one draws a boundary it may be for various kinds of reason. If I surround an area with a fence or a line or otherwise, the purpose may be to prevent

108

Ulvi Doguoglu

someone from getting in or out; but it may also be part of a game and the players be supposed, say, to jump over the boundary; or it may shew where the property of one man ends and that of another begins; and so on. So if I draw a boundary line that is not yet to say what I am drawing it for. When a sentence is called senseless, it is not as it were its sense that is senseless. But a combination of words is being excluded from the language, withdrawn from circulation. (Wittgenstein 1958, 4 §§499-500)

By this token these sentences are of philosophical importance and are to be distinguished from sentences which are simply false. But the problem with such sentences goes beyond the levels sketched so far and calls for further scrutiny. Besides the two levels introduced in the foregoing, a third level is implicit in Putnam (1975b); it was loosely termed “pragmatic” above. This label is not quite appropriate for Putnam’s early writings, though. On the contrary, the irreducible and untameable aspects of understanding are taken to be rooted in psychology and subjectivity, and consequently give reason to exclude understanding from philosophy of language. Individual understanding is displayed in one’s ability to use an expression appropriately or in one’s versatility to use expressions in various linguistic or situational contexts. Albeit these contexts are specifiable for understanding in particular situations, they are neither determinable in general nor once and for all, since linguistic intelligibility depends “on matters quite other than what we can or cannot “settle with certainty”: e.g. on grammar, on the ability to occur in coherent and appropriate discourses, on paraphrasability” (Putnam 1975b, p. 309). The reason that linguistic understanding cannot possibly be formalized or computationally modelled lies in the various subjective skills or abilities depending on the individual speakers themselves. In order to avoid the latent psychologism that inheres these abilities, Putnam takes three steps leading to semantic externalism. The first of these is only cosmetic and a matter of language-use: understanding gets superficially expelled from philosophy of language by way of no longer speaking of the individual’s understanding but rather of “the sentence” that is supposed “to be intelligible,” “to be able to occur in discourses,” “to be paraphrasable.” The second step paving the way to semantic externalism consists in a rejection of verificationism in the style of Malcolm (1959), seeking to fix the meaning of terms and sentences in the following way: the meaning of a sentence is tied to the criteria of its verification and the meaning of a term is determined by the methods by means of which the correct 4

Hereafter I will use ‘PI’ for Wittgenstein (1958).

Putnam beyond Putnam: Understanding, Pragmatism, Humanism

109

application of the term is verified. According to Malcolm (1959), we have for instance two criteria for the application of ‘sleep’: the concept is correctly applied if adult human beings look as though being asleep and if they say so after their waking up. In the case of babies only the first criterion is applicable. Since a difference in application criteria entails a difference in the meaning of a term, the term ‘sleep’ as regards babies and adults differs in meaning. 5 The main problem of verificationism is that according to this view “it is impossible to refer to a thing (or kind of thing) if in no case do we have better than indications of its presence or absence” (Putnam 1975b, p. 306). Further, if we do not have any criteria but only indications, our talk is meaningless. Thus Putnam aims at showing that our understanding of linguistic expressions does not include reference to objects and that therefore “the theory of language understanding and the theory of reference and truth have much less to do with one another than many philosophers have assumed” (Putnam 1979, p. 97). To illustrate this point it is helpful to consider a case of the description of a recently discovered disease, Bovine Spongiform Encephalopathy (BSE), say, in which certain infectious particles, namely prions (“proteinaceous infectious particles”), figure as causally effective. We take it that talk about prions was perfectly intelligible from the very start even though we did not have any clear ideas about their structure and their causal efficacy, that is, even though we did not have any hard and fast criteria but merely indications.6 Hence, Putnam concludes, the reference of a term is irrelevant to understanding. Consequently in Putnam (1975b) the object of understanding is restricted to the meaning of terms and sentences, while reference and truth are not part of understanding but entirely disconnected from it. The third step to avoid psychologism and the uncertainties related to understanding is, in effect, again more of a cosmetic rearrangement: by definition, the objects of reference are included into the meaning of a term. Individual understanding emcompasses no longer the whole meaning of a term but only parts of it, namely the syntactic and semantic markers, and stereotypes, while reference and truth are left out of it (Putnam 1975c, p. 269). While Putnam claims that “‘meanings’ [. . .] ain’t in the head” (Putnam 1975c, p. 227), understanding remains just there.

5

See the delineation of Malcolm’s views in Putnam (1975b). Hilary Putnam’s example is multiple sclerosis (Putnam 1975b, pp. 310f ). Thanks to Karim Bschir for discussion on this issue.

6

110

Ulvi Doguoglu

Being tied to understanding, semantics would be precluded from being theoretically captured because of its subjective and psychological aspects. Conceiving of meaning as largely independent from understanding, on the other hand, eventually allows philosophy of language to be neatly separated from psychology – or so it seems. The notion of understanding is slimmed down to the ability to use or manipulate signs, be they representational or linguistic and is relevant to philosophy of language only insofar as it comes into play via the notion of use. While Putnam conceives of “use” as being fully describable in functional or computational terms – describable (that is, from a third person point of view) – the aspects concerning first person or subjective matters (“ability” in the first place) at philosopher’s option drop out of philosophy of language as irrelevant (Putnam 1983). The conception of understanding developed so far bears certain tensions which are addressed in Putnam (1981, Chs. 1 and 2) as two related sceptical paradoxes: the story of Brains in Vats (BiV) and the model theoretic argument (MTA). In BiV, Putnam claims that if understanding is modeled along the lines mooted above, then, for all we know, we all could be Brains in a Vat: assuming that a computer were able to trigger our nerve endings in such a way we got the impression of living in the world we know, our experience of the world could be as it is, yet the world we actually live in might be completely different. Therefore we cannot, this story maintains, be certain that our mental representation or our experience tell us something substantial about the world we live in. MTA is aimed at showing that a comparable problem arises in language under certain presuppositions. If language is taken to be a formal system and semantics is modelled correctly by model theoretic semantics, then, given the truth conditions of all possible sentences in all possible worlds, the extension of our predicate-signs is not determined except up to isomorphism and the truth-evaluable content of thought and language is left notoriously underdetermined.7 All things considered, both of these paradoxes – the first displaying a problem of understanding or representation, the second a problem of reference – bring about a problem of access of mind or language to the world and show that the use of linguistic and representational signs does not determine the reference of these signs and hence neither the truthevaluable content of the sentences they occur. In other words, if understanding is ultimately reduced to a functionalist notion of use, it 7

For MTA see also Putnam (1983 and 1994a).

Putnam beyond Putnam: Understanding, Pragmatism, Humanism

111

does not manage to reach out to the world, and reality remains hidden behind its causal effects confronting us with ontological relativity. In both cases Putnam concludes in the face of the paradoxes that there must be something wrong with their presuppositions. Not only is the subject of understanding lost by cutting it down to a scientifically describable “use,” but also the objects of understanding remain inscrutable. In order to avoid these problems, Putnam proposes a rejection of the idea that “use” can be described from an all-embracing third person or God’s eye view conclusively and for all times (Putnam 1999, Part I, Lect. 1). In the German original of the Philosophical Investigations (1958) Wittgenstein uses two words for what is indifferentially translated as “use”: Verwendung and Gebrauch. While Verwendung usually means a single, particular use (or use-token), Gebrauch is a type of use, or a practice. The alternative, irreducibly pragmatic account of use Putnam calls for to avoid the sketched problems coincides with Wittgenstein’s Gebrauch which is central to the considerations on understanding, rule-following and practice in the PI.

2. Subjectivity and Objectivity in Practice 8 In his Philosophical Investigations, Ludwig Wittgenstein opens the discussion of rule-following in PI §138 with a question concerning understanding. After opting for a conception of meaning as use in the preceding paragraphs (roughly PI §§1-43), the question arises of how, within such a conception, understanding can be possible if we grasp the meaning of a word “in a flash” while the use constituting this meaning “is extended in time” (PI §138). In the subsequent paragraphs, Wittgenstein rejects various proposals until in PI §202 he gives the solution to the problem by claiming that rule-following is a practice. If the use of linguistic expressions is a form of rule-following, then this use, too, is a practice and understanding it is possible if I master the practice or technique (PI §199). The normativity of rule-following practices has been one of the most intensely discussed and analyzed topics in theoretical philosophy in the last twenty years. The controversy on normativity centres on two issues: firstly on the notion of objectivity of rule-following practices – how can there be a difference between seeming right and being right, between thinking one is following a rule and in fact following it?; and secondly on 8

The ideas in this part of the paper are developed at further length in Doguoglu (2004).

112

Ulvi Doguoglu

the question of how these practices can be normative for the actor – how the subject autonomously can follow a rule. In the first place, the objective condition of rule-following is that it is determinedly this and no other rule I am following, while it is not up to the individual subject or to a community’s shared belief which rule one is following. The subjective condition is that it is an individual, autonomous subject following it; a subject that principally could step back and act otherwise – her doings are neither accidentally or merely grounded in (intrinsic or extrinsic) coercion or instinct, nor need they be interpreted as such by others (a community, e.g.). Against this background it is the main task as regards the clarification of practices to give an adequate account of how these two conditions can be fulfilled: how can a subject’s following of a rule be objective? This is the very problem of the normativity of (rule-following) practices. It is remarkable that the loss of objectivity and subjectivity we found in Putnam’s early notion of understanding finds its correspondence in Kripke’s influential reading of Wittgenstein on rules and private language (Kripke 1982). Kripke claims that Wittgenstein proposed a sceptical solution to a sceptical paradox concerning the objectivity as well as the subjectivity of rule-following. Since rule-following is a practice, there are neither any objective (natural) nor any subjective (mental) facts determining the correctness of rule-following. When Wittgenstein in PI §201 invokes practice as the ground for rulefollowing, Kripke takes him to put rule-following into a Procrustean bed, chopping down its objectivity to mere intersubjectivity, while analogously stretching its subjective aspects to this same measure. The point of Kripke’s argument is that an individual is not able to follow a rule on her own, but that she is necessarily leaping blindly in the dark while illumination is warranted by a community only. Nor can following rules as a social practice ever attain proper objectivity. In short, from Kripke’s point of view it is wholly impossible to see how rules can be followed. Kripke mainly derives his reading from PI §§201-202 where Wittgenstein draws on the fact that the grasping of a rule is exhibited in what we, that is, my community of language users, call following a rule. Kripke takes this allusion to a community as being the reason for Wittgenstein’s conclusion: “hence [. . .] “to follow rules” is a practice” (PI §202),9 in which the appeal to practices is taken to be an appeal to community. In these lights the claim is supported by what immediately 9

I rectified Anscombe’s translation.

Putnam beyond Putnam: Understanding, Pragmatism, Humanism

113

comes after in PI §202: if following a rule is a practice and practices are tied to communities, it is trivially clear that “one cannot follow rules privately.” For Kripke, it is only within a community that questions of correctness arise, and it is only from this point of view that they can be decided. In his interpretation, Kripke relies on the following notional field: Wittgenstein is taken to use notions like “privately,” “individually,” “subjective” as belonging to one notional sphere opposed to and absolutely detached from two other notional spheres containing “objective,” “real,” “natural” on the one hand and “intersubjective,” “practice,” “communal” or “social” on the other. “Practice” for Kripke is supposed to belong to another notional sphere than “objective” and “subjective” right from the start. According to this grid, practices can strictly be neither objective nor subjective, but only intersubjective and communal. Instead of aiming at an explanation of how objective rulefollowing for individual subjects is possible, Kripke is content with merely showing the strict impossibility of these aspects of rule-following practices. The account then amounts to a reduction of the subjective and objective aspects of rule-following to a socially restricted practice rendering it in actual fact impossible to attain objectivity or subjectivity and by this absolute separation from the objective and the subjective human practices lose contact with objective reality as well as with the individual subject. Kripke’s notional grid is moulded by a naturalist outlook, still it is not naturalism in general that is not compatible with normativity, but rather it is the specific form of (Quinean) scientific naturalism Kripke presupposes and applies in his discussion of the issue. There is another reading of Wittgenstein,10 however, which does not accept this tripartite notional structure and promotes a single notional texture instead. The basic idea is that subjectivity does preclude neither objectivity nor intersubjectivity. Even though language and rulefollowing practices are social arts, this does not disqualify individual subjects to follow rules all by themselves (once they have learnt to follow them) and perform such practices on an objective basis of correctness. Rule-following practices do not solely depend on us individual subjects being members of a society but just as much on us living with physical bodies in a physical environment. The notion of 10

As regards such a reading see Putnam (1992, 1998, and 1999); the essays on Wittgenstein in Putnam (1994b, Part I); see also McDowell (1998b, esp. Chs. 11-14; 1998a, esp. Chs. 13-16), Travis (2001), and Haugeland (1998, esp. essays 1, 9, 10, and 13).

114

Ulvi Doguoglu

practice involves a community of practitioners without precluding an objective world in which it is performed or individual practicing subjects. The objective side of human practices is vividly illustrated by Wittgenstein’s example of weighing cheese: if it happened often that a piece of cheese grew or shrank without evident cause, the procedure of weighing cheese would lose its sense (PI §142). Thus the community cannot be the rock bottom where the spade turns;11 objectivity collapses if it is reduced to a mere substitute like “being accepted by a community (or by most members of it).” Practice and objectivity do not, as Kripke contends, exclude but complete each other. As to the subjective aspects, the point of Wittgenstein’s is not, negatively, that we are not able to follow the rule but blindly, but, positively, that we are able to follow rules blindly – under normal circumstances an individual does not (have to) make up her mind. In a well-rehearsed practice there is no need for interpretation; one can follow a rule without having to choose or to decide consciously. In such a case, one is able to act with eyes closed (and not, as Kripke contends, as blind as a bat). In the reading preferred here, the notion of practice in Wittgenstein does not stand in opposition to the concept of individual rule-following. Kripke for his part, however, cannot see these finer grains in Wittgenstein’s considerations due to his tripartite notional spheres. As indicated, Kripke’s sceptical considerations of rule-following practices in effect face the same problems as Putnam’s early notion of understanding: the objectivity and the subjectivity of rule-following is challenged by the claim that there are neither objective nor subjective facts of the matter determining the individual’s action and that it is only by being affiliated to a community of practitioners that an individual can be taken to follow rules. As against Kripke, criticism can be raised against the notional grid that is applied in externalizing semantics while mind and understanding are left “in the head.” Such criticism is raised by McDowell (1998c) and alternatively he proposes an externalization of mind by conceiving of it neither as a physical nor as an immaterial object but rather as an evolving complex of abilities.12 McDowell dispenses with (full) scientific or third 11

PI §217. Note that Wittgenstein does not say that this is how we act, but how I act. Note also that in PI §217 Wittgenstein speaks of “architectonic demands”: the demanded explanation is something like a Scheingesims (ornamental coping) bearing no static function; therefore his remark is not meant dogmatic but to reject sceptical and unsatisfying questions leading to “rock bottom.” 12 For criticism of so-called individualist conceptions of mental states and knowledge see already Burge (1996 and 1986); even though Burge’s critique takes the right direction, it

Putnam beyond Putnam: Understanding, Pragmatism, Humanism

115

person description without falling into a Cartesian dualism or taking mind (or the subject) as an individual substance describable from a first person point of view only. An adequate description of mind requires the inclusion of a description of the practices it is or might be involved into and a description of the options it has in performance of these practices. Conceiving of mind instead as a complex of abilities engaged in interaction with the world means to reject Kripke’s trichotomy of notional spheres. Further, in rejecting the idea that naturalism is committed to a scientific description of the world, McDowell (1998d) proposes another sort of naturalism in which mind as well as human practices are conceived of as natural. In scientific naturalism “nature” is taken to stand in opposition among others to “culture,” “practice,” “subject,” or “mind.” McDowell’s idea disrupting this notional corset is that the description of this part of nature (mind and human practices) can be objective without necessarily being from a third person or scientific point of view. The failure to get a complete theoretical or third person grip on nature (including mind) is not the fault of nature but of the notional grid. The aim is to give way to a natural description of mind essentially involving a subjective or first person view without falling into the trap of the errors of philosophy of consciousness (Bewusstseinsphilosophie) or the illusions of certainty of the first person view. On the other hand, a description of objective reality also requires the inclusion of (a description of) the practices this objectivity evolves from. In this picture, practice is taken to be the fundamental part of nature which essentially involves objectivity as well as subjectivity instead of confronting them. A description of the subjective abilities and of the objective factualities can be read out of the structural traits of this practice, since the structure of the practice of an individual does not only reflect the structure of his or her physical, social and cultural environment, but also the structure of his or her subjective structure. In this, neither subjectivity nor objectivity is prior, for both result from practice. While in Kripke practice is a social notion supposed to be used in the notional realm of intersubjectivity only, the proposed alternative takes practices to encompass subjectivity, intersubjectivity and objectivity alike. All three notions are engaged in it and belong to the same notional texture.13

seems to me that his anti-individualism ultimately remains in the grip of the notional grid which Kripke and the earlier Putnam apply. 13 For point of simplification I do not extend these remarks on intersubjectivity.

116

Ulvi Doguoglu

This leads to a view McDowell (1996) puts forward, namely that the structure of the practice of a human (or another living) being reflects on the one side the structure of the world (or environment) this being lives in and on the other the subject’s modes of freedom.14 The structure of the behavioural relation of some creature to its worldly environment tells us about the world it lives in. This is what Wittgenstein is up in saying: If one sees the behaviour of a living thing, one sees its soul. (PI §357)

Essential to this view is that the description of the behaviour of an individual can only said to be normative if it is set into a certain context. Thus it is reality or objective nature that determines the objective aspects of practice while the subjective side is determined by what McDowell refers to as “second nature” (Aristotle) or Bildung (Gadamer). Second nature enables understanding and makes us sensitive and receptive to reasons of all kinds, such as moral ones. Moreover they make us able to follow rules blindly without reasoning or interpretation. Bildung or second nature is the acculturated and well-rehearsed subjective practice allows a concept of autonomously acting subjects. By these tokens, this latter concept of a subject differs vastly from the selfcertain Cartesian thinker and reflects the contrast between the two notional grids of Kripke and McDowell/late Putnam.

3. Understanding, Pragmatism, Humanism On such a base of human practices incorporating both subjectivity and objectivity, an alternative concept of subjective understanding evading the sketched problems can be developed. In addition to the levels of grammar and semantics a pragmatic level needs to be reclaimed. The relevant notions Putnam introduces for such encompassment of pragmatics are “context” and “coherence” (Putnam 2001b; Putnam 1999, esp. Part II). Understanding in this sense is not restricted to a grasp of linguistic expressions, but includes perceptions, beliefs, intentions, and the like. Whereas in Putnam’s earlier writings understanding of the sentence (5)

There is a lot of coffee on the table (Putnam 1999, p. 87),

is constituted by the meanings of the words used in the sentence and by the way they are combined in it, the main point in the view put forward 14

McDowell credits this idea to Gadamer (Gadamer 1990, p. 448), see McDowell (1996, pp. 115-119).

Putnam beyond Putnam: Understanding, Pragmatism, Humanism

117

here concerns the reference of the terms and, therefore, the constitution of truth-evaluable content: My [. . .] examples illustrate how common nouns and adjectives may have very different reference in different contexts compatibly with what they “mean.” (Putnam 1999, p. 88)

Hence, the sense of (5) is dependent on the actual context of speaking: [. . .] the truth-evaluable content of the sentence [. . .] is highly occasion sensitive: depending upon the circumstances, the sentence can be used to say that there are many cups of coffee on a contextually definite table [. . .], or that there are bags of coffee stacked on the table [. . .], or that coffee has been spilled on the table [. . .]. (Putnam 1999, pp. 87f )

Sense and intelligibility are in question if the context of use is not clear or missing. Even if I know the “literal meaning” of a sentence, I might not understand what is said on a particular occasion (Putnam 1994b, p. 256). While the meaning of words can be detached from their use (and be fixed in a dictionary), the sense of words is connected to their use. Thus, even if the meaning of a sentence, roughly determined by grammar and lexicon, remains the same, reference and truth-relevancy might vary significantly under certain conditions (see Putnam 2001b; Travis 1997). Complete understanding of a sentence, then, consists in grasping its linguistic meaning and its truth-evaluable content – that is, what it refers to and the circumstances relevant to its content. While the linguistic meaning of a sentence is independent of particular contexts (even though not of every context), the truth-evaluable content is actually determined in particular contexts and situations only. Putnam (2001b) calls what is thus understood, namely the meaning of a sentence and its truthevaluable content, its “sense.” It is important to note that the mentioned levels of grammar, semantics, and pragmatics are, though usually separable in a particular case, not strictly and theoretically separable for all times and cases, since what is part of the context-independent meaning of a term may change in view of new or developing practices. Albeit the notion of understanding something can be spelled out and made explicit in the particular case, it is never explicable in full generality and once and for all. There is no complete theoretical treatment of the matter. Any claim as to the pragmatics of speaking being fully specifiable for all times presupposes the possibility of an independent third person point of view and encounters the paradoxes mooted above. When Putnam claims that “logical truths do not have negations that we (presently) understand” (Putnam 1994, p. 256), the restriction “presently” indicates that in this

118

Ulvi Doguoglu

generality the claim can only be held as long as no alternative logic is available which stands the test of use “in the field.” It might seem that a specification or explication of the relevant features of pragmatics must in principle be possible for the particular language to be learnable at all: if the semantics of a language is radically contextual that is if the sense of an expression is wholly determined by each (token of) use, it is impossible to learn this very language. This rejoinder must be granted; it does not affect the position favoured here, though, it being emphatically contextualist, but not radically contextualist. Again, the approach pursued here holds that there are general semantic and grammatical features of linguistic expressions uttered on an occasion determining the meaning independently from particular occasions of speaking while the pragmatic context owing to the irreducible involvement of subjectivity cannot be determined in general but only for specific situations. The subjectivity of practice would collapse if there were a general theory of particular practices. Moreover, pragmatics not only affects semantics but also the grammar of a sentence. To take the above example of coffee on the table, depending on the contextual situation the individuation of the reference of the terms turns out differently. Hence, in one context the term ‘coffee’ refers to a continuum (the coffee spilled on the table); in others it refers to certain countable units (cups of coffee, coffee beans, coffee bags). Thus pragmatic coherence adds to congruence or well-formedness on the grammatical and consistency on the lexical level in understanding of language.15 This third condition is related to the irreducible aspect that was already mentioned in Putnam’s early conception of understanding, outlined in terms of paraphrasability and ability to think up discourses and situations in which a certain sentence could be used. There are, however, cases in which the grammatical or the semantic level (or both) are completely irrelevant to full understanding. Thus, we might end up in a corner of the universe in which the local population speaks a language we do not know and yet perfectly understand what is said because we might recognize the grammar of a certain practice we are familiar and conversant with. In conversation, James Conant recounted the story of him sitting in a train in Turkey and being addressed by the guard after controlling his ticket. Even though he did not speak any Turkish at that time he perfectly understood that the man was telling him 15

In conversation Hilary Putnam is very strict in using “context” exclusively in connection with pragmatic aspects of semantics. I personally would not do that and instead speak of context not only to address pragmatic aspects of language in use but of practices in full generality.

Putnam beyond Putnam: Understanding, Pragmatism, Humanism

119

he was sitting in the wrong compartment. In this example, the wellknown grammar of practice or, to use another of Wittgenstein’s expressions, the shared “form of life” enabling understanding is traveling by train. It is at this point at which understanding definitely transgresses the boundaries of (the philosophy of) language due to its fundamental boundedness to practices. The object of understanding is not merely linguistic forms and their meaning but encompasses extensive acquaintance with situations and practices. To summarize and contrast, Putnam’s early linguistic notion of understanding depends on the following features: grammatical congruence and lexical consistency (also on the sentence level). Understanding here primarily concerns grammar and semantics while a pragmatic level is not yet spelled out in detail and mistaken for being psychological. In the course of the development of semantic externalism this pragmatic level is believed to be domesticated by introducing the notion of use. Use in turn is thought to be specifiable from a third person perspective, whereby the subjective side of understanding is locked out of the account. Putnam’s later notion of understanding, however, acknowledges pragmatics as a third level of understanding. Understanding on this level is determined by pragmatic context and pragmatic coherence, in which context and coherence include the social and physical environment of the understanding individual. Hence, understanding is no longer merely linguistic, but reaches out to reality thus encompassing the objects of reference and the truth-evaluable content of the sentence. Founding on this notion of understanding, in recent years Putnam has voiced criticism of various positions and arguments in the philosophy of mind and language. For example, he argues for the pragmatic unintelligibility of theses like: (6)

There could be soulless automata (zombies) (Part II of Putnam 1999);

or the incoherence of sceptical claims like: (7)

Everything we believe might, in fact, be false and error (Putnam 2001b).

In retrospect also the argument against the possibility of us all being BiV (Putnam 1981, Ch. 1) should be seen as such an argument for the unintelligibility or incoherence. Whereas in the original text Putnam claims that the hypothesis of us being BiV be actually false, he would

120

Ulvi Doguoglu

naturally acknowledge its grammatical and lexical correctness today while pointing out its pragmatic incoherence and unintelligibility. Again, it is central to bear in mind that if understanding fails for pragmatic reasons then the linguistic expressions in question are not simply false but unintelligible. Putnam’s and Wittgenstein’s Sinnkritik, by the way, is fundamentally different from Carnap’s (see Carnap 1931 and 1968), even if one is reminded of Carnap at some places. Yet the latter’s criteria for sensefulness lack the pragmatic level and apply only on the level of “logical syntax.” Adding to these criteria of Carnap’s, the interesting critique of metaphysical doctrines in Putnam is raised on the level of pragmatics (or “pragmatic syntax,” as one might say). With the main feature of understanding being its boundedness to practices and the task of philosophy being to gain a reflective survey of human practices, their rationale and their boundaries, pragmatic understanding rises to the status of a fundamental notion of philosophy and serves as a base for a renewal of philosophy and philosophical method. Albeit natural science needs to draw boundaries in order to avoid subjectivity and to warrant repeatable and verifiable results, these boundaries are not valid for all concerns. A survey of the use of a word like ‘soul’ renders it impossible for us to see how talk of zombies or soulless automata would make sense within serious scientific or philosophical discourse although it certainly makes perfect sense in fiction and story-telling (see Putnam 1999, Part II). It is just such an extension of uses of words in fiction, i.e. in the practice of story-telling that insinuates that these words render sense within scientific discourse, too. In respect of such philosophical theses the task will plainly be this: In each case, one has to listen to the story the philosopher tells [about how a certain claim is to be understood], and show why and how it is incoherent. (Putnam 2001a, p. 23)

The challenge is to closely look at the use of the words in a context and at the story that is told, and to argue carefully whether and to what extent the extension or transfer of linguistic expressions to other practices or contexts makes sense. By this, however, in the present view narration achieves great importance in philosophy as well as in other sciences (including natural sciences). Philosophical and scientific reasoning are seen as specific forms of story-telling in each of which a specific game of giving and asking for reasons is played, in accord with a particular “grammar of narration.” This approach is in line with various contemporary strategies

Putnam beyond Putnam: Understanding, Pragmatism, Humanism

121

to recast moral reasoning via what might be called “narrative analysis.”16 Wittgenstein’s later philosophy should be seen in such a narrative light and also Putnam’s recent philosophical developments display signs of such a view in which it is the philosopher’s objective to determine whether a coherent and reasonable story about certain philosophical or practical problems can be recounted at all (Putnam 1999, Part II). Since philosophy’s aim is not to tell fictional stories for their own sake, the task of philosophical criticism is to distinguish philosophical fiction from non-fiction. It needs emphasis that the question as to the correctness of such a narrative analysis cannot be decided by strict determination but is rather a matter of ongoing argument. Such arguments are not strict arguments, even though as trained philosophers we recognize reasonable and coherent narrative explanations in many cases “in a flash” and as soon as we are confronted with them. By learning a certain practice the learner is rendered sensitive to the structure or grammar of this practice. Learning to follow rules makes sensitive to what is right and what is wrong according to certain rules. In this sense, learning to argue philosophically is to become sensitive to the structure and spirit of a good philosophical explanation or theory. In sum, then, philosophical understanding and philosophy in general are aimed at surveying human practices in the sketched Wittgensteinian sense by means of reflection on the adequacy of philosophical conceptions and survey-attempts. Methodologically, therefore, philosophy gains survey among others by scrutinizing and criticizing other philosophical conceptions. Even though it is tied to practices, in the survey and comparison of different practices understanding goes beyond the particular practices it surveys.17 The notion of understanding as bound to practice is not only central to any understanding of how language works, but rather is central to philosophical understanding and method of philosophy. The effect of this pragmatization of the notion of understanding or intelligibility in arguments thus results in a humanization of philosophy.18 With the task of philosophy being understanding of human practices, understanding 16

Cf. Giesinger (2004), MacIntyre (1981), Bieri (2001). This very idea of surveying practices is, in effect, akin to Brandom’s “making it explicit” (Brandom 1994). Applied to our discussion this would mean that understanding leaves practice as it is and making it and its boundaries explicit, marking it off from others, reflecting its rationale. 18 In Putnam this tendency is displayed in various book titles like Realism with a human face or Words and Life; further it is most recently reflected in his considerations on “Enlightenment and Pragmatism” (Putnam 2004, Part II). 17

122

Ulvi Doguoglu

lies at the heart of philosophy – at least so in Putnam’s humanistic pragmatism.

Acknowledgements For various substantial comments and discussion thanks to Holger Baumann, Michael Esfeld, Matti Eklund, Johannes Giesinger, Alessandro Maranto, Marc Neumann, Chiara Tabet, Jonna Truniger, and Peter Schulthess. I am also grateful to the organizers and participants of the Meeting in Santiago de Compostela, especially to Hilary Putnam for kind comment and criticism.

Zurich University of Applied Sciences Academic Affairs Unit P.O. Box CH-8401 Winterthur, Switzerland e-mail: [email protected]

REFERENCES Bieri, P. (2001). Das Handwerk der Freiheit. Über die Entdeckung des eigenen Willens. München: Hanser. Brandom, R.B. (1994). Making It Explicit: Reasoning, Representing, and Discursive Commitment. Cambridge, MA; London: Harvard University Press. Burge, T. (1986). Individualism and Psychology. Philosophical Review 95 (1), 3-45. Burge, T. (1996). Other Bodies. In: A. Pessin and S. Goldberg (eds.), The Twin Earth Chronicles: Twenty Years of Reflection on Hilary Putnam’s “The Meaning of ‘Meaning’”, pp. 142-160. Armonk, NY: M.E. Sharpe. Carnap, R. (1931). Überwindung der Metaphysik durch logische Analyse der Sprache. Erkenntnis 2, 219-241. Carnap, R. (1968). Logische Syntax der Sprache. Wien: Springer-Verlag. Chomsky, N. (1968). Syntactic Structures. Den Haag: Mouton. Doguoglu, U. (2004). Naturalism and Rule-Following Practices: Finding Fault with Kripke’s Implicit Notion of Objectivity. In: P. Schaber (ed.), Normativity and Naturalism, pp. 151-173. Frankfurt: Ontos. Gadamer, H.-G. (1990). Wahrheit und Methode. Grundzüge einer philosophischen Hermeneutik. Tübingen: Mohr. Giesinger, J. (2004). Der Anfang der Geschichte. Zeitschrift für Pädagogik 50 (3), 392-405. Haugeland, J. (1998). Having Thought: Essays in the Metaphysics of Mind. Cambridge, MA: Harvard University Press.

Putnam beyond Putnam: Understanding, Pragmatism, Humanism

123

Kittay, E.F. (1987). Metaphor: Its Cognitive Force and Linguistic Structure. Oxford: Clarendon Press. Kripke, S.A. (1982). Wittgenstein on Rules and Private Language: An Elementary Exposition. Oxford: Blackwell. MacIntyre, A.C. (1981). After Virtue: A Study in Moral Theory. Notre Dame, IN: University of Notre Dame Press. Malcolm, N. (1959). Dreaming. London: Routledge and Kegan. McDowell, J. (1996). Mind and World. Cambridge, MA; London: Harvard University Press. McDowell, J. (1998a). Meaning, Knowledge, and Reality. Cambridge, MA: Harvard University Press. McDowell, J. (1998b). Mind, Value, and Reality. Cambridge, MA; London: Harvard University Press. McDowell, J. (1998c). Putnam on Mind and Meaning. In: (1998a), pp. 275-291. McDowell, J. (1998d). Two Sorts of Naturalism. In: (1998b), pp. 167-197. Putnam, H. (1975a). The Analytic and the Synthetic. In: (1975d), pp. 33-69. Putnam, H. (1975b). Dreaming and “Depth Grammar.” In: (1975d), pp. 304-324. Putnam, H. (1975c). The Meaning of ‘Meaning’. In: (1975d), pp. 215-271. Putnam, H. (1975d). Mind, Language, and Reality (Philosophical Papers, vol. II). Cambridge; New York: Cambridge University Press. Putnam, H. (1979). Reference and Understanding. In: Meaning and the Moral Sciences, pp. 97-119. Boston, MA: Routledge and Kegan. Putnam, H. (1981). Reason, Truth, and History. Cambridge; New York: Cambridge University Press. Putnam, H. (1983). Models and Reality. In: Realism and Reason, pp. 1-25. Cambride; New York: Cambridge University Press. Putnam, H. (1992). Renewing Philosophy. Cambridge, MA: Harvard University Press. Putnam, H. (1994a). Model Theory and the Factuality of Semantics. In: (1994b), pp. 351-375. Putnam, H. (1994b). Words and Life. Cambridge, MA; London: Harvard University Press. Putnam, H. (1998). Kripkean Realism and Wittgenstein’s Realism. In: A. Biletzki and A. Matar (eds.), The Story of Analytic Philosophy: Plot and Heroes, pp. 241-252. London: Routledge. Putnam, H. (1999). The Threefold Cord: Mind, Body, and World. New York: Columbia University Press. Putnam, H. (2001a). Rules, Attunement, and “Applying Words to the World.” In: Ch. Mouffe and L. Nagl (eds.), The Legacy of Wittgenstein: Pragmatism or Deconstruction, pp. 9-23. New York: Peter Lang. Putnam, H. (2001b). Skepticism, Stroud and the Contextuality of Knowledge. Philosophical Explorations 4, 2-16. Putnam, H. (2004). Ethics without Ontology. Cambridge, MA; London: Harvard University Press. Quine, W.V.O. (1995). From Stimulus to Science. Cambridge, MA; London: Harvard University Press. Schulthess, P. (2001). Kontext und Signifikation: Logik und Grammatik im Mittelalter. Freiburger Zeitschrift für Philosophie und Theologie 48, 328-361. Travis, Ch. (1997). Pragmatics. In: B. Hale and C. Wright (eds.), A Companion to the Philosophy of Language, pp. 87-107. Oxford: Blackwell Publishers Ltd. Travis, Ch. (2001). The Uses of Sense: Wittgenstein’s Philosophy of Language. Oxford; New York: Oxford University Press.

124

Ulvi Doguoglu

Wittgenstein, L. (1958). Philosophische Untersuchungen [Philosophical Investigations]. Oxford: Blackwell Publishers Ltd. Wittgenstein, L. (2000). Nachlass: The Bergen Electronic Edition. Oxford; New York: Oxford University Press.

Russell B. Goodman SOME SOURCES OF PUTNAM’S PRAGMATISM

ABSTRACT. This paper considers some sources, mostly within the pragmatist tradition, for the full-fledged pragmatism that Putnam set out in the 1980s and 1990s, particularly in The Many Faces of Realism and Realism with a Human Face. In considering Putnam’s views about metaphysics, I pay particular attention to his pluralism (and hence in a way to the conference theme of diversity), which I trace back through Nelson Goodman to William James. In considering Putnam’s idea that facts and values are intertwined, I discuss both John Dewey and that neglected middle-generation pragmatist, C.I. Lewis. I briefly consider James’s essay “On a Certain Blindness in Human Beings,” where metaphysical pluralism and a pluralistic ethics of toleration come together, and I conclude with some suggestions about the relation of Putnam’s pragmatism to the very different American philosophy developed by his Harvard colleague Stanley Cavell.

Hilary Putnam’s inheritance of pragmatism can be found in his explicit discussions of pragmatist thinkers, especially William James and John Dewey, but just as much in his statements of his own position in such works as The Many Faces of Realism, Realism with a Human Face, and The Threefold Cord. In this paper I draw on all these sources in looking at Putnam’s engagement with and inheritance of pragmatism. The paper has three sections. The first and longest considers some sources of the humanistic pluralism Putnam develops in the attempt to avoid the problems of “scientific realism.” The second part of the paper considers some pragmatist strains in Putnam’s ethical philosophy, specifically his project of showing the entanglement of facts and values. In the third section I step back and briefly consider the relation of both these projects to the work of Putnam’s Harvard colleague Stanley Cavell. Putnam’s pragmatism takes its most unbridled form in The Many Faces of Realism (1987), a book that begins with a discussion of scientific realism in the form of Eddington’s argument that the table on

In: M.U. Rivas Monroy, C. Cancela Silva, and C. Martínez Vidal (eds.), Following Putnam’s Trail: On Realism and Other Issues (Pozna Studies in the Philosophy of the Sciences and the Humanities, vol. 95), pp. 125-140. Amsterdam/New York, NY: Rodopi, 2008.

126

Russell B. Goodman

which he writes is really not solid, but mostly empty space. The argument, based on contemporary physics, promises an account of reality, but it is a disconcerting account which Putnam compares to stories of “the Seducer” in nineteenth century American melodramas. The Seducer promises “various things to the Innocent Maiden,” but fails to deliver. Scientific realists propose that they alone can give us “independent” reality, and criticize many other alternatives as forms of idealism. But the reality they promise to give us turns out to be unavailable, for it is defined as, in Putnam’s words: “what “finished science” will say there is – whatever that may be” (1987, p. 4). Putnam wants to honor the Innocent Maiden’s desire for the real with the help of a pragmatism conceived as a philosophy in a broad “neo-Kantian tradition” that also includes Husserl and Wittgenstein. For all these writers, Putnam maintains, “commonsense tables and chairs and sensations and electrons are equally real [. . .]” (1987, p. 12). Putnam wants to perform a rescue of the real and achieve a return to common sense.

1. Sources In this first section, I want to consider some sources for Putnam’s position, mostly in the pragmatic tradition. If one looks back from the distance of a century and a quarter in which we now stand to those first papers of Charles Sanders Peirce we find an incredibly resilient tradition, with its detractors all along, but sustained by a set of major figures we still read (Peirce, James, Dewey), and a new set of major figures in its raucous revival in the late twentieth century, among whom Richard Rorty and Hilary Putnam are the leaders. These newest figures, like James and Dewey if not Peirce, weave a story of pragmatism that takes it back to the canonical texts of its founders. A key text for Putnam’s discussion, and for understanding why he takes James to be part of a neo-Kantian tradition despite James’s antipathy to Kant and Hegel, is James’s chapter on “Pragmatism and Humanism,” in Pragmatism (1987a). It is there that James propounds the “humanist” claim that becomes a slogan for Putnam: “the trail of the human serpent is over everything” (cf. Putnam 1987, pp. 16ff ). Putnam sees James as following Kant in understanding reality to be something available to us precisely because of its entanglement with the human. James realizes his closeness to Kant, not in his mode of argument, but in the position at which he arrives. Although he brings up the point

Some Sources of Putnam’s Pragmatism

127

himself, it is certainly an unwelcome comparison, because James thinks of himself as working in the empirical tradition of Mill and Hume, and as an opponent of rationalist and particularly Kantian philosophies that posit transcendental machinery behind or beyond experience. Yet James is too good a philosopher not to note the similarities to Kant, precisely the similarities that Putnam brings out. James writes: When we talk of reality “independent” of human thinking, then, it seems a thing very hard to find. It reduces to the notion of what is just entering into experience and yet to be named, or else to some imagined aboriginal presence in experience, before any belief about the presence had arisen, before any human conception had been applied. [. . .] Superficially this sounds like Kant’s view; but between categories fulminated before nature began, and categories gradually forming themselves in nature’s presence, the whole chasm between rationalism and empiricism yawns. (James 1978b, p. 595)

In this passage James concedes the closeness to Kant by speaking of “categories,” even if he at the same time holds that these categories evolve. Indeed, James gives an account of the evolution of these categories earlier in Pragmatism, in the chapter on “Pragmatism and Common Sense.” He speaks of “our fundamental ways of thinking,” of “conceptions” and “systems of concepts” that “rationalize” our experience. “All our conceptions,” he writes, “are what the Germans call Denkmittel, means by which we handle facts by thinking them” (James 1987a, p. 560f ). He names these concepts of “common sense,” and enumerates them as follows: “One Time, One Space, Causal influences, Minds, The same or different, the fancied, the real” (pp. 561f ). Yet James differs from Kant in two ways. First he sees these concepts as evolving: “There is not a category, among those enumerated, of which we may not imagine the use to have thus originated historically and only gradually spread” (p. 563). Second, he portrays a pluralism of categories. The One Time and One Space are useful constructions, he states, but “the great majority of the human race never use these notions, but live in plural times and spaces, interpenetrant and durcheinander.” Indeed, James finds “various types of thinking, each so splendid for certain purposes, yet all conflicting still, and neither one of them able to support a claim of absolute veracity.” In this pluralism of useful denkmittel, James finds support for the pragmatist view that “all our theories are instrumental, are mental modes of adaptation to reality, rather than revelations or gnostic answers to some divinely instituted world-enigma” (p. 571). The picture of historically evolving categories that James sets out is developed in various ways in such later pragmatist works as C.I. Lewis’s

128

Russell B. Goodman

Mind and the World Order (1929), John Dewey’s “Logic: The Theory of Inquiry” (1938), W.V.O. Quine’s “Two Dogmas of Empiricism” (1951) and The Web of Belief (1970), Richard Rorty’s Philosophy and the Mirror of Nature (1979) and Contingency, Irony, and Solidarity (1989), and Putnam’s Reason, Truth and History (1981) and Realism with a Human Face (1990). As one might infer from its title, Putnam takes an historical approach to our “categories” in Reason, Truth and History, and he states in Realism with a Human Face that “our norms and standards of warranted assertability are historical products; they evolve in time” (Putnam 1990, p. 21). If we assert that categories (Putnam’s words are ‘norms’ or ‘schemes’) evolve, there will be different categories at different times, perhaps in different places at the same time. These categories need not conflict, but if they do there may be no satisfactory scheme for uniting their disparate takes on things. For Putnam, there may be something to learn even from such a conflict of categories – more perhaps than from a reductive reconciliation. This is an outlook I shall call pluralism, and it shows up in Putnam’s already quoted statement that “commonsense tables and chairs and sensations and electrons are equally real [. . .].” The title of the work in which this quotation appears – The Many Faces of Realism (1987) – says it all in a way, for the expression “many faces” suggests that one can be a realist and a pluralist at once. Putnam puts the point by speaking of “conceptual relativity” and states that his pragmatic realism is “just the insistence that realism is not incompatible with conceptual relativity.” This “realism with a small ‘r’,” as he also calls it, is the “view that takes our familiar commonsense scheme, as well as our scientific and artistic and other schemes, at face value, without helping itself to the notion of the thing “in itself ”” (1987, p. 17). It is from the standpoints of these “schemes” that one can say that there are tables and ice cubes as well as electrons, moments of beauty and horror as well as meter readings. Putnam’s pluralism reaches back to James’s pragmatism, as we shall see in a moment, but it has closer sources in the work of two rather neglected pragmatists. The first of these is C.I. Lewis, who, like Putnam, was a pragmatist, a logician, and a Harvard professor. In his paper “Logical Positivism and Pragmatism” ([1941] 1970) Lewis wrote, in his precise, sober way: “It is one thing to say that scientific formulation is always pertinent and possible; it is a quite different thing to say that no other than scientific formulation is meaningful” (p. 100). Lewis thereby defends a plurality of significant formulations against logical positivist

Some Sources of Putnam’s Pragmatism

129

reductions to scientific discourse – against what Putnam would later call “scientism.” Another pluralist and, after his move from Penn in the mid nineteen sixties, a colleague of Putnam’s at Harvard, is Nelson Goodman. Putnam discusses Ways of Worldmaking (1978), Goodman’s most explicit pragmatist work, in Realism and Reason (1983), and he praises Goodman in Reason with a Human Face: “Goodman’s great contribution, I think, has been to urge that reconstructive reflection does not lose its value just because the dream of a total and unique reconstruction of our system of belief is hopelessly Utopian; we can learn a great deal from partial and even fragmentary reconstructions, and we can learn a great deal from reconstructing our beliefs in alternative ways” (Putnam 1990, p. 25). A full consideration of Goodman’s relation to later pragmatisms would have to take into account his work on alternative epistemologies in The Structure of Appearance, his invention or discovery of the “new riddle of induction,” his work on “languages of art,” and of course the explicit pragmatism of Ways of Worldmaking. In keeping with my current focus on pluralism, however, I want to consider a powerful little article entitled “The Way the World Is,” originally published in The Review of Metaphysics in 1960, then reprinted in Goodman’s Problems and Projects (1972). Goodman reduces to absurdity various ways of understanding “the way the world is,” advocating instead that the world is not one way but many. We may think that the one way the world is is the way it is given in experience, but this turns out to be impossible to specify. We may instead think that the way the world is is the way it is seen through the camera. But there are different ways of doing this. Shall we say that a photograph of a man with his feet towards me, so that his feet appear as large as his torso, is not the way the man is to be seen (p. 27)? Such a photograph shows something about reality. As Goodman puts it, such a photograph “reveals new facts and possibilities in visual experience” that a more conventionally composed photograph does not. Goodman makes the same point about paintings, about which he holds that there are many realistic styles, that of Picasso as much as that of Rembrandt. Even the most realistic way of depicting, Goodman argues, amounts merely to “one kind of conventionalization” (p. 30). If one asks what is the food for human beings one should hesitate: for people eat tortillas or lenguado al vino Albariño or hamburgers or poi, but there is no one food they eat. Similarly if asked what is the way the world is, Goodman answers “none.” Rather, “there are many ways the world is and every true description captures one of them” (p. 31). Nelson Goodman is

130

Russell B. Goodman

thus, what Putnam calls himself: “both a realist and a conceptual relativist” (Putnam 1987, p. 17). In a symposium with Goodman and Quine, and in a paper that repeats Goodman’s title, “The Way the World Is,” Putnam gives credit to Goodman for the “pluralistic note” (Putnam 1990, p. 265) he strikes in Ways of Worldmaking (Goodman 1978) and Languages of Art (Goodman 1968). He concludes his paper not only with an acknowledgment of the influence of Goodman and of Quine on his own work, but with the statement that “all three of us have been shaped by a continuous tradition of American thought – one that can be traced from its beginnings in the debates at Harvard between Royce and James, as well as in the work of Peirce and Dewey, through the writings of our teacher C.I. Lewis, up to and including some of the most recent developments in American professional philosophy” (Putnam 1990, p. 267). The ultimate source for the pluralism in the pragmatic tradition is, of course William James, and I would like to spend a few minutes considering how extensive a theme it is in his work, for it enters not only James’s epistemology but his metaphysics and his moral philosophy. Putnam himself mentions in this connection Pragmatism’s chapter on “The One and the Many” where James aligns himself with “the pluralistic side” (1987a, p. 556), as opposed to the side of absolute unity. As usual, James tries to stake out a position in the middle, allowing some unity or connection to the universe, but no one string or system of connection unifying everything. As a radical empiricist James believes that relations are as much experienced as the things related, so he is perfectly happy to point to real connections in experience. But these connections do not form a system and they may conflict. Putting the point metaphysically, James writes of the innumerable little hangings-together of the world’s parts within the larger hangings-together, little worlds, not only of discourse but of operation, within the wider universe. [. . .] the same part may figure in many different systems, as a man may hold several offices and belong to various clubs. (1987a, p. 545)

James also takes a pluralistic slant in the “Pragmatism and Common Sense” chapter of Pragmatism, where he speaks of “rationalizing” the “everlasting weather of our perceptions” (p. 562) through various “conceptual systems” (p. 561), of which science and common sense are two (or more). James considers the question, so important for Putnam, of which of these schemes is the true one, and he answers that they are each useful for one sphere of life or another, but that there “is no ringing conclusion possible when we compare these types of thinking, with a

Some Sources of Putnam’s Pragmatism

131

view to telling which is the more absolutely true [. . .]. Common sense is better for one sphere of life, science for another, philosophic criticism for a third; but whether either be truer absolutely, Heaven only knows” (p. 569).1 These differences among conceptual systems are memorably exemplified, and given a moral slant, in a wonderful essay entitled “On a Certain Blindness in Human Beings,” published in James’s Talks to Teachers (1899). The blindness James has in mind, he explains, “is the blindness with which we are all afflicted in regard to the feelings of creatures and people different from ourselves” (p. 229). One of James’s examples is from his own life. Traveling by wagon in the mountains of North Carolina he encounters a scene of devastation: forest ripped up, ugly scars in the land, and clumps of trees between which grew small patches of corn. James comes to see, however, that to the settlers who were living there the clearing meant a “personal victory” over nature by virtue of “honest sweat” and “persistent toil.” What to James was an environmental disaster was to the settlers a symbol of “duty, struggle, and success.” “I had been as blind to the peculiar ideality of their conditions,” James summarizes, “as they certainly would also have been to the ideality of mine, had they had a peep at my strange indoor academic ways of life at Cambridge” (pp. 233f ). Notice that these differing viewpoints on the world are to some manner accessible to each other, for James learns to some degree to see things as the mountaineers do. Another set of cases James discusses concern forms of blindness less easily overcome, those between species: Take our dogs and ourselves, connected as we are by a tie more intimate than most ties in this world; and yet, outside of that tie of friendly fondness, how insensible, each of us, to all that makes life significant for the other! – we to the rapture of bones under hedges, or smells of trees and lamp-posts, they to the delights of literature and art. As you sit reading the most moving romance you ever fell upon, what sort of a judge is your fox-terrier of your behaviour? (p. 230)

Here is an example of the partial connections and comprehensions James was to speak of in Pragmatism. Our dogs are indeed connected to us by “a tie more intimate than most ties in this world” but they are also in some ways entirely alien to us, an alienation that can not be made up as easily as James’s alienation from the North Carolina mountaineer. It’s 1

Thanks to Robert Schwartz for calling these passages to my attention in connection with James’s pluralism.

132

Russell B. Goodman

not just that the dogs miss something we apprehend but also that we miss things the dogs apprehend, not just the smells of the trees and lampposts but the “rapture” of the bone under the hedge. 2 There is both a metaphysical and a moral side to James’s pluralism in (1899), for the plurality of perspectives on the world, he holds, commands us to tolerate, respect, and indulge those whom we see harmlessly interested and happy in their own ways, however unintelligible these may be to us. Hands off: neither the whole of truth nor the whole of good is revealed to any single observer, although each observer gains a partial superiority of insight from the peculiar position in which he stands. Even prisons and sick-rooms have their special revelations. (James 1899, p. 264) 3

Notice that in James, as in Putnam, Goodman, and Lewis, there is no suggestion that these perspectives are simply ideal or subjective. James speaks of “insights” and “truth.” His pluralism is a form of realism. I have been considering some antecedents of the humanism and pluralism that Putnam includes within his pragmatic realism, and I now want to consider a third element, action. In The Many Faces of Realism Putnam writes: The heart of pragmatism [. . .] – of James’s and Dewey’s pragmatism, if not of Peirce’s – was the insistence on the supremacy of the agent point of view. If we find that we must take a certain point of view, use a certain “conceptual system,” when we are engaged in practical activity, in the widest sense of “practical activity,” then we must not simultaneously advance the claim that it is not really “the way things are in themselves.” (Putnam 1987, p. 70)

2

James returns to dogs in A Pluralistic Universe (1987a, pp. 625-781), where he writes: “We may be in the universe as dogs and cats are in our libraries, seeing the books and hearing the conversation, but having no inkling of the meaning of it all” (p. 771). Here human beings occupy the position of the dogs in James (1899). Just as the dogs have difficulty understanding our novel-loving ways, so we may have little idea of the more encompassing beings with whom we share the universe. James postulates the existence of beings greater than ourselves but still limited, whose actions we occasionally encounter but have no way of comprehending. 3 Near the end of A Pluralistic Universe James asserts that “there may ultimately never be an all-form at all, that the substance of reality may never get totally collected, that some of it may remain outside of the largest combination of it ever made, and that a distributive form of reality, the each-form, is logically as acceptable and empirically as probable as the all-form commonly acquiesced in as so obviously the self-evident thing” (James 1987a, p. 645). For another passage about dogs see James (1987a, pp. 562f ).

Some Sources of Putnam’s Pragmatism

133

So it is through the human specifically as revealed in what we do that Putnam finds reality. Putnam is quite right to say that the insistence on this point lies at the heart of classical pragmatism. Consider in the first place the word ‘pragma’, meaning an act or fact – from the Greek verb meaning “to do.” Consider secondly what is widely considered to be the originary formulation of pragmatism, Peirce’s statement “there is no distinction of meaning so fine as to consist in anything other than a possible difference of practice.” Peirce doesn’t use the word ‘pragma’ or ‘pragmatism’ here but he does employ the related word ‘practice’. Variations on these terms run through the subsequent history of pragmatism, for example in James’s idea that we “ride” on truths and that language is a program for “work,” and in Dewey’s idea that knowing is a form of doing. “We have to do a doctrine,” Dewey writes, “to know its truth” (Goodman 1995, p. 83). In fact Dewey extends pragmatism from a doctrine of significance, past its Jamesean epistemology, to a doctrine about reality that anticipates Putnam. In his important essay “Does Reality Possess Practical Character” (1908), Dewey defines pragmatism as the “doctrine that reality possesses practical character” (p. 81). Reality includes what is revealed in our interactions and practices, the “genuinely vital” and the “personal.” “Why,” Dewey asks, “should what gives tragedy, comedy, and poignancy to life, be excluded from things?” 4 Putnam speaks in the same vein when he writes in The Threefold Cord that “what has weight in our lives should also have weight in philosophy” (1999, p. 70), and when he states in The Many Faces of Realism that the world includes not only electrons and space-time regions, but tables and chairs, prime numbers, “moments of beauty and transcendence” and “people who are a menace to world peace.” Of course Putnam is not just repeating Dewey. What is new in Putnam’s neopragmatist reformulation of Dewey’s and James’s ideas is in part the result of the intervening fifty years or so of analytic philosophy (in which Putnam played his own major role). I would simply mention two figures crucial for Putnam’s neopragmatic formulations. The first is Quine, who in “Two Dogmas of Empiricism” (1951) holds that we have some flexibility in how to adjust our “scientific heritage" to our “sensory promptings,” and that when rational, the considerations guiding us in doing so are "pragmatic.". Quine therefore connects our statements about what is real – our ontology – to what works for us in our lives and

4 Cf. James’s statement that we will accept no “philosophy whose principle is so incommensurate with our most intimate powers as to deny them all relevancy in universal affairs” (1979, pp. 70f ).

134

Russell B. Goodman

in our thought (cf. Putnam 1987, p. 21). From both Quine and Wittgenstein Putnam derives an emphasis on language that is not at all present in classical pragmatism. Wittgenstein of course focuses on language as an activity, which in some broad sense – as Wittgenstein himself noticed – makes his philosophy akin to pragmatism (see my 2002). “The term ‘language-game’,” Wittgenstein writes, “is meant to bring into prominence the fact that the speaking of language is part of an activity, or of a form of life.” Wittgenstein’s picture of language is akin to pragmatism not only in its emphasis on action, but in two other ways: in his idea that language evolves historically, and in his pluralism about language games. There are “countless” kinds of sentence and uses of language, Wittgenstein writes, adding that “this multiplicity is not something fixed, given once for all; but new types of language, new language-games, as we may say, come into existence, and others become obsolete and get forgotten” (1953, p. 23).

2. Facts and Values I want to say something quite brief now concerning some pragmatist roots of Putnam’s moral and political philosophy. Here the point is that Putnam’s pluralistic realism embraces what one might call moral facts, and the truth of moral statements. Putnam states that it is “a fact, for example, that Yeats was a great poet, and a fact that the Nazis were evil” (1987, p. 63). It is true that such facts cannot be established beyond controversy, Putnam concedes, but many facts in science, e.g., the age of the earth, are not beyond controversy either (p. 65). Even if we grant that there is a theoretical end to controversy in science and not in questions of value, it does not immediately follow that science deals with truths and facts while morality does not. That inference would require the premise that the more controversial is the less objective. Echoing William James, Putnam asks: “What would be the consequences for our lives” of believing in the subjectivity of such statements as that the Nazis were evil. “What,” he continues, “is the “cash value” of believing that only what can be established beyond controversy has anything to do with “cognition,” knowledge, understanding?” (p. 65). In a later paper, written with Ruth Anna Putnam, Putnam returns to this theme, and rejects the position of such philosophers as Bernard Williams and David Wiggins, who distinguish “truth in morality” or “truth “humanly speaking”” from the “absolute truth” that we allegedly obtain in science. Pragmatism, he

Some Sources of Putnam’s Pragmatism

135

states, “urges that truth humanly speaking is all we’ve got” (Putnam 1990, p. 226). Now in refusing to draw an ultimate distinction between statements of value and statements of fact, Putnam revives a position taken by that middle generation pragmatist C.I. Lewis, who states that for “the pragmatist, there can be no final division between “normative” and “descriptive”” (Lewis [1941] 1970, p. 112). Lewis defines pragmatism as “the doctrine that all problems are at bottom problems of conduct, and all judgments are, implicitly, judgments of value, and that, as there can be ultimately no valid distinction of theoretical from practical, so there can be no final separation of questions of truth of any kind from questions of the justifiable ends of action” (p. 108). Lewis, like Putnam, draws on the philosophy of John Dewey, who is explicit in his refusal to draw a sharp line between facts and values, between “the ideal and the material” as he puts it in “The Construction of Good,” a well-known chapter in The Quest for Certainty (1929, p. 225), Dewey calls for experimental methods as applied to questions of value. “A moral law,” Dewey writes, “like a law in physics, is not something to swear by and stick to at all hazards; it is a formula of the way to respond when specified conditions present themselves. Its soundness and pertinence are tested by what happens when it is acted on” (p. 222). Dewey holds, as Putnam has stressed in a recent paper, that we debate and experiment, that to use his word, we inquire, not only about means but about ends. (Response to Brandom in Conant and egle [2002], pp. 59-65.) Putnam follows Dewey not only in refusing to draw a line between questions of fact and questions of conduct, but in his concern with democracy – one of Dewey’s main themes not only in his political writings but in such works as Democracy and Education (1916). Putnam’s wonderful paper, “A Reconsideration of Deweyan Democracy” (1995), defends the Deweyan thesis that, as he puts it, “democracy is a precondition for the full application of intelligence to solving social problems.” Drawing on both Peirce and Dewey, Putnam reminds us of the pragmatists’ allegiance to experimental methods (Goodman 1995, p. 189), and characterizes Dewey’s understanding of that method as “experimental inquiry combined with free and full discussion – which means, in the case of social problems, the maximum use of the capacities of citizens for proposing courses of action, for testing them, and for evaluating the results” (p. 190). This is to say that democracy is not just a method of expression (of, for example, an antecedently existing “will of the people”), but a method of thinking and of social transformation. The

136

Russell B. Goodman

community of scientists can only pursue objective truth if everyone in the community is free to criticize its beliefs, knowing that those criticisms will be heard and discussed. But this is just the condition of democracy as Putnam and Dewey understand it: a system of critical reflection and action in which everyone has an equal right to participate (p. 193). It is important to see that for the pragmatists, values are built right into inquiry. As Dewey learns from Peirce and James, inquiry begins with a “problem situation,” a sense that something is not in order. (Peirce had called this “the irritation of doubt” [Goodman 1995, p. 40].) In his classic discussion of “the practical character of reality,” Dewey defines a problem situation as something inherently “unsatisfactory,” in which something is “amiss.” The first stage of solving the problem is identifying it, moving from a somewhat vague state of dissatisfaction or irritation to a definition that leads to a particular experimental action. If intelligence has been properly applied the action leads to a consummation or satisfaction. Notice how, in the following statement, Dewey speaks of all this as a “situation” rather than as a private subjective experience; and that he mixes value-laden descriptions – e.g. “crisis,” “out of gear,” “menaced” – and more neutral descriptions of the situation: Awareness means attention, and attention means a crisis of some sort in an existent situation; a forking of the roads of some material [. . .]. It represents something the matter, something out of gear, or in some way menaced, insecure, problematical and strained. This state of tension, of ambiguous indications, projects and tendencies, is not merely in the “mind,” it is nothing merely emotional. It is in the facts of the situation [. . .]. (Goodman 1995, p. 87)

Notice also that Dewey is not speaking particularly about morality, but about experience generally, insofar as it involves awareness, inquiry, and growth. Dewey’s project of intensifying and enhancing the values that are present in experience from the first is central not only to his pragmatism but to his educational philosophy, and it reaches a culmination in his late great work on aesthetics, Art as Experience. In that work, Dewey writes that the aesthetic is “the clarified and intensified development of traits that belong to every normally complete experience” (1934, p. 46). He calls for a reconstruction not only of philosophy but of normal human life, which he characterizes as “stunted, aborted, slack, or heavy laden.” Dewey sees the ultimate goal of our inquiries as “conditions through which the ideal is capable of embodiment and realization” (p. 27).

Some Sources of Putnam’s Pragmatism

137

3. Putnam and Cavell It is at this point – in the idea of a reconstruction or redemption of our experience – that pragmatism as I see it appears as a form of romanticism. And it is in this context that I wish, in concluding the paper, to briefly consider Putnam’s pragmatism in relation to the philosophical romanticism developed by his Harvard colleague Stanley Cavell in the mid-nineteen eighties, at just the time when Putnam’s neopragmatism reached its full flowering. Although their approaches are in many ways dissimilar, they are comparable on the following points: (1)

(2)

(3)

they share the quest for a return to the ordinary within a Kantian problematic (giving prominence to the active role of the human mind); they are nevertheless profoundly dissatisfied with the Kantian settlement on three fronts: Kant’s banishment of the thing in itself beyond experience, his related acceptance of Newtonian physics as giving the basic forms of that experience, and his banishing of value from the world; they both owe an enormous debt to the later philosophy of Wittgenstein – on whose philosophy they taught classes together at Harvard in the 1990s.

Cavell sets out the idea of a broad, rich transcendental empiricism in his early book on Thoreau, The Senses of Walden (1971), where he writes that “what is wrong with empiricism is not its reliance on experience but its paltry idea of experience” (p. 126). Cavell’s richer conception of experience includes not only sense perception but “human forms of feeling, objects of human attraction, our reactions constituted in art.” In an explicit revision of Kant, Cavell writes that such forms of experience “are as universal and necessary, as revelatory of the world, as the forms of the laws of physics” (p. 104).5 Although Cavell had thus been considering the American form of romanticism known as Transcendentalism in the nineteen seventies, it was not until the end of the decade, in The Claim of Reason (1979, Part IV), and then in the early nineteen eighties in his Beckman Lectures, that he began to theorize romanticism as a philosophical topic. This turn to romanticism arises from Cavell’s work on skepticism about other minds, which he conceives not as the academic philosophical problem of what evidence one has for

5

See Goodman (1968) and (1978) for the idea that art contributes to understanding, and Putnam’s discussion of Goodman’s views in (1990, pp. 266ff ).

138

Russell B. Goodman

the presence of consciousness in another person, but as a problem of our lives with others. We live our skepticism concerning other minds, Cavell comes to say, and he finds examples of such lived skepticism in Shakespeare’s tragedies, specifically Othello and King Lear (see my 1990, Ch. 1). By the early nineteen eighties, Cavell had come to the thought that we live not just our skepticism about other minds but our skepticism about the world. This is the problem he comes to think of as romantic – how to recover from our lived skepticism about the world; and this is the place where his philosophy comes close to that of John Dewey (see my 1990, Ch. 4). In his 1988 book In Quest of the Ordinary, Cavell characterizes the work of romanticism as “the task of bringing the world back, as to life” (pp. 52f ). He complains that Kant negotiates away the possibility that you can “experience the world as world, things as things; face to face, as it were, call this the life of things” (p 53). The idea that we’ve lost and can recover “the life of things” is depicted, according to Cavell, in Coleridge’s “The Rime of the Ancient Mariner,” and addressed in Wordsworth’s Preface to the Lyrical Ballads. It is, he holds, given a twentieth century interpretation in Wittgenstein’s goal of bringing “words back from their metaphysical to their everyday use” (Wittgenstein 1953, p. 116). Cavell’s quest for a vital, humanly entangled ordinary, resonates with Putnam’s quest for a recovery of the ordinary from the clutches of the scientific realist, and with his sense that the world is as much valuable as it is factual. Putnam and Cavell both operate with the sense that there is something missing in philosophy, and that this has something to do with something missing in our lives. Putnam’s title Words and Life alludes to Wittgenstein and to Husserl’s idea of the “life-world,” but it also lies in the broad stream of romanticism, in which “life” is perhaps the central concept. I see Putnam, like Cavell but also like Dewey and James, and Thoreau and Emerson before them, as members of a romantic tradition in American thought that always calls for a renewal of philosophy 6 and operates with the sense that philosophical problems are, no matter how abstract, at the same time intimately connected with the way we live our lives.

6

As in the title of one of Putnam’s books, Renewing Philosophy (1992).

Some Sources of Putnam’s Pragmatism

139

University of New Mexico Department of Philosophy MSC 032140 1 University of New Mexico Albuquerque, NM 87131, USA e-mail: [email protected]

REFERENCES Cavell, S. (1971). The Senses of Walden. New York: Viking. Cavell, S. (1979). The Claim of Reason. New York: Oxford University Press. Cavell, S. (1988). In Quest of the Ordinary: Lines of Skepticism and Romanticism. Chicago: University of Chicago Press. Conant, J. and U.M. egle, eds. (2002). Hilary Putnam: Pragmatism and Realism. London: Routledge. Dewey, J. (1929). The Quest for Certainty. New York: Minton, Balch. Dewey, J. (1938). Logic: The Theory of Inquiry. New York: Henry Holt. Dewey, J. (1934). Art as Experience. New York: Minton, Balch. Dewey, J. (1916). Democracy and Education. New York: Macmillan. Dewey, J. (1908). Does Reality Possess Practical Character?. In: Goodman (1995), pp. 19-91. Goodman, N. (1951). The Structure of Appearance. Cambridge, MA: Harvard University Press. Goodman, N. (1968). Languages of Art. Indianapolis: Bobbs-Merrill Goodman, N. (1972). The Way the World Is. In Problems and Projects, pp. 24-32. New York: Bobbs-Merrill. Goodman, N. (1978). Ways of Worldmaking. Indianapolis and Cambridge: Hackett. Goodman, R.B. (1990). American Philosophy and the Romantic Tradition. Cambridge: Cambridge University Press. Goodman, R.B. (1995). Pragmatism: A Contemporary Reader. New York and London: Routledge. Goodman, R.B. (2002). Wittgenstein and William James. Cambridge: Cambridge University Press. James, W. (1987). A Pluralistic Universe. Cambridge, MA: Harvard University Press. James, W. (1979). The Will to Believe. Cambridge, MA: Harvard University Press. James, W. (1899). On a Certain Blindness in Human Beings. In: Talks to Teachers on Psychology and to Students on Some of Life’s Ideals, pp. 229-264. New York: Henry Holt. James, W. (1987a). Pragmatism. A New Name for Some Old Ways of Thinking. In: (1987b), pp. 479-624. James, W. (1987b). William James: Writings 1902-1910. New York: The Library of America. Lewis, C.I. (1929). Mind and the World Order. New York: Cambridge University Press. Lewis, C.I. ([1941] 1970). Logical Positivism and Pragmatism. In: The Collected Papers of C.I. Lewis, pp. 92-112. Edited by J.D. Goheen and J.L. Mothershead, Jr. Stanford, CA: Stanford University Press. Putnam, H. (1981). Reason, Truth and History. New York: Cambridge University Press.

140

Russell B. Goodman

Putnam, H. (1983). Realism and Reason (Philosophical Papers, vol. III). Cambridge: Cambridge University Press. Putnam, H. (1995). A Reconsideration of Deweyan Democracy. In: Goodman (1995), pp. 183-204. Putnam, H. (1987). The Many Faces of Realism. La Salle, IL: Open Court. Putnam, H. (1990). Realism with a Human Face. Cambridge, MA: Harvard University Press. Putnam, H. (1992). Renewing Philosophy. Cambridge, MA: Harvard University Press. Putnam, H. (1999). The Threefold Cord. New York: Columbia University Press. Quine, W.V.O. (1951). Two Dogmas of Empiricism. In: Philosophical Review 60, 20-43. Quine, W.V.O. (1970). The Web of Belief. New York: Random House. Rorty, R. (1979). Philosophy and the Mirror of Nature. Princeton, NJ: Princeton University Press. Rorty, R. (1989). Contingency, Irony, and Solidarity. Cambridge: Cambridge University Press. Wittgenstein, L. (1953). Philosophical Investigations. Translated by G.E.M. Anscombe. New York: Macmillan.

Chiara Tabet PUTNAM AND THE “GOD’S-EYE VIEW”: ON THE LOGICAL STRUCTURE OF ANTI-FOUNDATIONALIST PRAGMATISM

ABSTRACT: I concentrate on difficulties met by Putnam’s “internal” or “scientific” realism. They concern his attempt to reconcile pragmatism and realism. My line of argument is the following. A) By exploiting Putnam’s argument against the “God’s-eye view” (GEV) and the Brains-in-a-Vat argument (BIV), it can be shown that the realism he is defending is either a too strong metaphysical realism or a too weak “residual” position. B) If it is a metaphysical position, then it contradicts Putnam’s own views on GEV. C) If it is a “residual” position, then in the context of a Theory of meaning for formal and natural languages, it is not more explicative (and it may be even less so) than some forms of functionalism. The paper ends with an attempt to reconcile Putnam’s pragmatism with a functionalist approach to epistemic features of “meaning,” for the sake of a reconciliation between pragmatism and realism.

The aim of this paper is to investigate a fundamental question in Putnam’s arguments against both “naïve” realism (what in Reason, Truth and History is called externalism, or metaphysical reductionism) and relativistic dead-ends of (some sort of) constructivism (as well as of conventionalism). Such a question concerns what Putnam himself refers to as the paradoxes of the “God’s-eye view” (GEV), or “view from nowhere” (cf. Putnam 1990, Ch. 1; and 1981, p. 49). First of all, a few premises must be clarified. I shall take Putnam’s pragmatism as based on threemain points. These are: (1)

Meaning-holism, as a position resting on the following points: refusal of the value-fact dichotomy, acceptance of the antipositivistic principle that sentences meet the test of experience

In: M.U. Rivas Monroy, C. Cancela Silva, and C. Martínez Vidal (eds.), Following Putnam’s Trail: On Realism and Other Issues (Pozna Studies in the Philosophy of the Sciences and the Humanities, vol. 95), pp. 141-160. Amsterdam/New York, NY: Rodopi, 2008.

142

Chiara Tabet

only as a corporate whole, interdependency of belief, interests, facts and interpretations. (2)

Refusal of computational accounts of meaning, and especially refusal of mentalism (a former position of Putnam himself).

(3)

Realism about semantic content, primarily aimed at rescuing some kind of reference-determinacy from the dangers of full-blown conventionalism. It is of course a realism of quite an unusual kind, which Putnam has conceived as structurally different from metaphysical realism. His “internal realism” is thus supposed to escape the metaphysical realist’s “absolutism,” stemming from the conjunction of “Correspondence, Independence, Bivalence and Uniqueness” as, so to speak, dogmas about truth, while reconciling determinacy of reference and acceptance of “conceptual relativity.”1 On one side, we have the metaphysical realist’s position, asserting that we have (a) (b)

(c)

a fixed and unique language-independent world; a unique reference function, fixed and determined by THE world, “taking individual terms of our language onto particular real-world objects or kinds of objects” (cf. Hallett 1994, p. 67, as well as Putnam 1989, p. 214), and thus a unique true theory, or something like an ideal Theory of THE world.

At the opposite extreme we have full-blown conventionalism, resting on strong epistemic presuppositions. Its key-assumptions are: (ai) from an epistemic point of view, there is nothing like THE world, but rather we are faced with a radical semantic indeterminacy such that all we are left with is a variety of ways to conceive world(s); (bi) the reference function is radically indeterminate; and (ci) we don’t have one true Theory, but a variety of languages and theories capturing different aspects of an (itself indeterminate) reality.

1

Cf. Putnam (1988, p. 107). An interesting discussion of “internal” vs “metaphysical” realism can be found in Blackburn (1994), which in what follows I shall constantly keep in mind.

Putnam and the “God’s-Eye” View

143

In the middle, Putnam’s internal realism is, or should be, a way to accommodate (a) with (ci), by abandoning (c) and reconciling (b) and (bi) (cf. Putnam 1988, esp. p. 114). My discussion here concerns points 2 and 3. Also, inside this general framework, I shall focus on only one of the two key-notions that a theory of meaning investigates; these are given by (i)

(ii)

Truth (both as the metaphysical problem of correspondence between assertions and reality and as the epistemic problem of how such a correspondence is gained), Reference, or the question of how it is that, by speaking, we refer to an external semantic content (to a reality existing in some way independently of us), and of how it is that we have access to such a reality (an external world). This is, again, both a metaphysical question (concerning the ontological status of the outer world) and an epistemological question, concerning the structure of our linguistic access to a meaningful reality.

The notion which is going to be at stake here is the second. More specifically, I shall concentrate on its epistemological features, also as a result of criticising the relevance of metaphysical realism for a nonmodest Theory of meaning. With this in mind, we can recapitulate the general situation of Putnam’s pragmatism in the following way. Putnam’s defence of realism is complementary to his defence of the Wittgensteinian and Quinean idea that we can only make sense of the world inside some given language (or language game). No ultimate “language of the world” (no Protokolsätze) exists; we only have access to reality through different linguistic structures, or levels of description. At the same time, however, the inferential nets which we build as our world-access structures cannot “exhaust” the world: realism is then an answer to the risk of “loosing the world,” a risk which affects directly the possibility of speaking of “meaning” at all (by undermining its stability through belief creation and fixation processes, as it is argued in Representation and Reality) (Putnam 2001, Ch. 3). In this context, Putnam’s stand against functionalism rests precisely upon the necessity of keeping, so to speak, things together: stability of meaning (of external semantic content) and “open texture,” as he says, of our ways of having access to such a content.

144

Chiara Tabet

Now, the main questions that I am going to address are: (a)

Can Putnam’s pragmatism successfully preserve an antifoundational feature (implied by the very assertion that language games have an open texture: they are open to continuous change and revisions in their inferential structures) while assuming that in fact, aside this open-texture, there is a given reality that “supports” such an open-endedness?

Concerning the latter point, I shall take Putnam’s realism about semantic content as the view that operational criteria and structures of inference on one side, and internalistic objects such as mental states and representations on the other, do not entirely capture meaning, and that this is the case because there are contexts in which the referential aspects of meaning, rather than intensions, are simply “doing all the work.”2 This is, thus, a position which is prima facie much more modest than metaphysical realism, since the latter aims at accounting for a kind of ontological objectivity whose epistemological counterpart, we are being told by Putnam, is the uniqueness of the theoretical appraisal of the described reality. However, what I shall be suggesting as a key-point is that as soon as we start investigating deeper into the structure of the interaction between epistemic aspects of meaning on one side and “reference” on the other, the boundaries between realism about semantic content and metaphysical realism seem to become much less sharp. (b)

Putnam has developed his realist-pragmatist position primarily against metaphysical realism. However, a second target of his views has been given by functionalism, and especially by computational functionalism. But is Putnam’s version of pragmatism incompatible with the core assumptions of a computational-functionalist account of meaning?

I shall start by briefly recalling Putnam’s popular argument against the “God’s eye view” (the view that there is, in fact, an ultimate metalanguage, or an omniscient view, from which one has direct access to reality as well as to all levels and ways of description of reality). I shall then show that, however, the very argument is in some way contradicted by (that is: its negation is a hidden presupposition of) another popular argument by Putnam, this time one which is conceived as supporting the 2 I am thinking, of course, of the popular treatment of natural kinds terms in Representation and Reality, where the twin-Earth thought experiment shows that because of under-determination of reference two speakers can have identical mental representations while speaking about “different stuff.”

Putnam and the “God’s-Eye” View

145

view that there is an external semantic content independent from our mental representations, that we have access to this semantic content and that it constitutes the possibility of reference. This is the Brains-in-a-Vat argument, for which I refer to Putnam (1981). This reading of BIV serves the aim of showing that the status that we attribute to these (epistemic) presuppositions actually plays a major role in determining the antifoundationalist attitude underlying a pragmatic theory of meaning. I shall follow this path by considering two kinds of possible answers to the contradiction. The first consists in simply stating that the contradiction is only apparent. I shall try to show that this answer won’t do unless we take realism merely as a “useful fiction.” The second rests on a whole set of arguments explored by Putnam in (2001). To this I shall try to give a programmatic reply, concerning the scope and aim of a non-modest Theory of meaning, by suggesting that we could develop a pragmatist position which is at once less modest (in foundational terms) than Putnam’s and open to the acceptance of indeterminacy and openendedness of meaning in a way that is not incompatible with a computational level of explanation.

1. Putnam’s Argument Against GEV Putnam’s view about GEV rests on the acknowledgment of paradoxes arising from assuming semantical closure of a language (following Tarski’s classical theorems and Russell). In Putnam (1990), he defines the view underlying Tarski’s hierarchy in terms of GEV, stating that: the observer can take as large a totality as he wishes as the system – excluding totalities which include himself in the act of performing the measurement [in the case of quantum mechanics; or, in this case, the logical systems’ construction] – but [. . .] he himself [. . .] must always lie outside the system. (Putnam 1990, p. 17; but cf. also Putnam 1981, p. 49)

But the GEV (which is said to be “the view form which absolutely all languages are equally part of the totality being scrutinized”) is said to be “forever inaccessible.” This is because, taking Tarski’s case as paradigmatic, one should then presuppose a primitive “systematically unambiguous discourse” acting as an ultimate meta-language; but if this ultimate meta-language is semantically closed (that is: if it allows the use of the predicate ‘true’) it is itself paradoxical in Tarski’s sense. Thus, no level of discourse is either “ultimate” or, so to speak, “self-foundational.”

146

Chiara Tabet

2. The BIV Argument and GEV In order to frame the difficulties of Putnam’s realism, let us first briefly recall the main steps in the BIV argument. Suppose we are BIV, that is: brains placed in a vat of nutrients, connected to a computer “which causes the person whose brain it is to have the illusion that everything is perfectly normal” (Putnam 1981, p. 6). That is: brains perceive an outer reality as if it were “really there,” while in fact they are living inside a hallucination. Let us suppose that this is a collective hallucination such that all human beings are actually BIV, living in an outer reality which is in fact a purely “mental one.” The question is: is this a plausible situation, or: might we really be BIV? Now, Putnam’s move is to produce an a-priori argument, based on externalist presuppositions and exploiting implications of such presuppositions so as to produce a quasi-logical refutation of the sentence ‘we are brains in a vat’ by reductio ad absurdum. In short, the move is the following: if we were BIV, it can be shown that we would not be able to say (or think) that we were. This is because: –

–

there is no intrinsic meaning to “representations” (e.g., verbal representations of a tree, images of sky, etc.), independent of intentionality on one side and on the external reality being represented on the other. Thus “the “sense-data” produced by the automatic machinery [the computer producing the hallucination] do not represent trees (or anything external) even when they resemble our tree-image exactly. A “sense datum” might be qualitatively identical with an image of a tree without being a image of a tree” (Putnam 1981, p. 13). Therefore, because BIV do not have access to an external world, their representations cannot have the “furniture” of the external world as their objects. Thus, the reference of words in Vat-English must be different from that of words spoken by people who are not BIV. In Putnam’s terms: It follows that if (Vat) world is really the actual one, and we are really the brains in a vat, then what we now mean by ‘we are brains in a vat’ is that we are brains in a vat in the image or something of that kind (if we mean anything at all). But part of the hypothesis that we are brains in a vat is that we aren’t brains in a vat in the image (i.e., what we are hallucinating isn’t that we are brains in a vat). So, if we are brains in a vat, then the sentence ‘We are brains in vat’ says something false (if it says anything). In short, if we are

Putnam and the “God’s-Eye” View

147

brains in a vat, then ‘We are brains in a vat’ is false. So it is (necessarily) false. (Putnam 1981, p. 15, my emphasis)

Here “necessity” comes from the structure of the argument. The latter isolates reference to an outer world and shows that it is unexhausted by intension and extension acting as given conditions. Its structure is therefore similar to that of the “cat on a mat” argument (cf. Putnam 1981, pp. 33-38). Reference is taken to be a relationship (whose nature we shall temporarily leave indeterminate) between what we may call “intentional states” and the external world. Meaning attribution is thus an impure (externalist) process. Now, let us re-read the argument by isolating another variable, which we might take as co-extensive with the domain of reference, that is: instead of “outer world.” We can just assume that when we speak of semantic content we are in fact speaking of an impure semantic content. In the case of BIV, the sentence ‘we are BIV’ says something false, or does not say anything at all, because its semantic content is different from . . . from what? Precisely from the implicit semantic content of the disquoted sentence: we are BIV (we actually are BIV). We thus have, here, two distinct “we” (let us call them wefake and we real)as subject, and Putnam himself obviously acknowledges this and uses it as the central distinction of the argument. However, and here is the point, there is also a third “we,” hidden “behind” the argument: the “we” from whose point of view we can acknowledge the distinction between we fake and we real. It is precisely from the point of view of this “we,” and thus exploiting a third semantic content, that one can say that “reference” is internal in one case and external (thus “proper”) in the other. Even if this third “we” is a mere abstraction functional to the argument, this seems to cause some problem. If we did not postulate it, we would find ourselves in the following situation: when we (indeterminate we) face the alternative “we fake or we real” we face it from within our language; this language is either fake-English or actual English (because of the refusal of metaphysical realists’ GEV, tertium non datur). But again the distinction needs to be justified inside our language: if the latter is fake-English, which is the premise to be refuted, what happens is not that we have an (absolutely) fake reference, because the very meaning of the term ‘reference’ would be such that what looks externally as a “fake reference” is in fact reference tout-court. Inside vat-English, the distinction between “representation in an image” and “representation” tout-court is a linguistic distinction: BIV may call “representation in an image” something which, inside that language, is distinct from what they

148

Chiara Tabet

refer to as “representation” tout-court. But this is true also for (inside) actual English. Thus we need to postulate a third “we,” and this I shall call a we from nowhere. Can it be something different from a “we from nowhere”? This is hard to justify. We could, one might say, take the disquoted sentence WE ARE BIV as being formulated in a third language which does not aim at being in any way “fundamental” – but then we seem to find ourselves in the same position as soon as we try to show that also in this third language the sentence ‘we are BIV’ cannot be “meaningfully” asserted. The acknowledgment of this third we from nowhere as capable of making an absolute distinction between two semantic contents clashes with Putnam’s very intent, as well as his sophisticated treatment, in Putnam (1990), of the GEV view.

3. Metaphysical Realism and Realism as a Useful Fiction It seems that the metaphysical feature of Putnam’s pragmatism is far too strong to be acceptable alongside with an anti-GEV view. This is because, according to the anti-GEV argument, we cannot have a “positive” access to language limits from inside a language game. Now, there are two ways of answering this objection. The first consists in simply stating that, although there does exist an external “semantic content” out there, we only have access to it inside a linguistic framework. This is a typically pragmatic move, of a wittgensteinian sort, which would leave us with an unsolved tension between objectivity and relativity.3 Thus our “we from nowhere” would in fact be a “we from somewhere” – where this “somewhere” is identified with some specific language game. But, either this answer is inconsistent with the realist presupposition (because the individuation of semantic content is just a matter of translation – but then what about its stability?) or it is inconsistent with a completely “secularised” version of pragmatism – because saying that there is an external given semantic content but that we cannot have a direct epistemic access to it resembles a Kantian 3 In fact, it could be said that Putnam’s attempt resembles, in structure, a possible reading of later Wittgenstein’s work. Namely, in trying to locate the point of coincidence between realism and idealism, it runs the risk, quite at work in Wittgenstein’s Philosophical Investigations, of loosing the distinctive ideas of both: to quote from Blackburn again, “the difficulty is that what is promised by internalism (it is just us, just one perspective) is taken away by the absolutism (the “us” is transcendental: we cannot make sense of any other perspective)” (Blackburn 1994, p. 25).

Putnam and the “God’s-Eye” View

149

dichotomy between a Ding an Sich and levels of description (a dichotomy against which Putnam himself has convincingly argued).4 However, there is also a more sophisticated line of argument to answer the objection. This may consist in saying: fine for what concerns the refusal of a strong kind of metaphysical realism, but realism (as an hypothesis concerning the status of semantic content) may also be taken, so to speak, as a useful fiction, or, better, as a useful tool of investigation. But useful for what? Useful, perhaps, for a theory of meaning which aims at accounting for phenomena such as stability of meaning (of extensional features of meaning) through conceptual changes (through

4 here is also another way of going through the BIV argument and come to the same conclusions. That is, one could look at the BIV argument through the lens of Putnam’s model-theoretical arguments. The latter show that, against a claim attributed to the metaphysical realist, we are not in a position to hold that there is one given world, potentially described by an ideal theory, the reference relation to which is completely determined. It reaches the three following conclusions: provided that we have an ideal first-order theory , expressed in a language L(), a set N of names in L(), a function REF mapping elements of N into items of the world, and a valuation function g mapping elements of L() into the world, such that true sentences are determined by REF itself, then:

(i)

(ii)

(iii)

determining g will not enable us to determine REF if we do not already know that the “unique” world (THE world of the metaphysical realist) is the satisfying structure (which, because of the Löwenheim-Skolem Theorem, we cannot acknowledge from the beginning: in fact THE world is not a unique interpretation, since we always have unintended interpretations of the Theory); even if the REF function is determined, we cannot establish that the interpretation we are dealing with is given by THE world (that is: we cannot pick out THE world from the set of intended and unintended interpretations); and even if we are given THE world as our model, and the set of all sentences of L() made true by this model, then we cannot determine the REF function.

Thus: reference is not extensionally determined. The world does not fix the reference of (terms of) a language. (But then, what does? Putnam would say: a mixture between language and world. But this answer is of no use if one tries to determine exactly in which way such a mixture will, then, determine reference.) If reference is not extensionally determined, then how are we to know that terms in fake-English have a reference which is different from terms in actual-English? We cannot appeal to THE world, that is: THE language-independent world of the metaphysical realist (but cf. Hallett 1994, p. 75). Thus, we must appeal to the conceptual framework in which the world is experienced as THE world. In short, appeal to a third language (or to a wider Theory which acts as a conceptual framework). This is what Hallett refers to when, in a different context, he argues that “The “inside/outside explanation” is, indeed, the right one, but only if we are clear that it takes place within a Theory T” (Hallett 1994, p. 80). And this is also what is going on here: we need a third language. BUT: what about reference inside THIS third language . . . ? Thus our infinite regress.

150

Chiara Tabet

changes in the inferential nets that constitute our epistemic access to semantic content). In a way then, following this line of argument, realism would come out as a “residual” position: something that we can justify by asserting that if we were not realist about external semantic contents then we would be left in need of an explanation of phenomena related to meaning which other theories (generally what Putnam has connected with the heading “Brentano/Aristotle’s thesis”) cannot explain.5 Thus, internal realism would simply point out that there is some feature of reality which is “resistant” to being entirely captured by our inferential procedures. Its core would be given by the simple statement that information is resistant to changes in the way in which it is conceptually organized and accessed. But what about an explanation of the structure of such a resistance? Can internal realism abandon the modesty (in a Dummett-like sense of the word) of its explanatory claims and hope to say something more about that? In what follows, I shall argue that this could be done using suggestions and tools coming from a different approach to the “meaning” problem. Namely, the idea is that one should look at the work done in the framework of functionalist accounts of epistemic aspects of meaning, paying special attention to their computational tools of investigation. I can here provide just a sketch of such an idea, whose motivation is the attempt to exploit implications of the anti-GEV argument (that is: we always find ourselves as referring within a language) in antifoundationalist, but explicative, terms.6

5 Obviously the central phenomena here are: identity of meaning through time and underdetermination of reference by mental representations. 6 Before going on to it, a few things must be clarified. When I speak about functionalism, I refer to a very general direction of research, whose key-commandments may be taken to be:

(a) (b) (c)

take seriously epistemic aspects of meaning, that is: the “inferential nets” through which information is organized and understood as meaningful information; account for these by using computational models, and assume as a hypothesis that we can account for the “what” in terms of the “how” (for what it is for a piece of information to be meaningful in terms of how that information is accessed to and organized).

The framework of the argument is not given by any specific functionalist view.

Putnam and the “God’s-Eye” View

151

4. Epistemic (Computational) “Non-Transparency” and Anti-Foundationalism Let us recapitulate, for the sake of clarity, exactly which is the difficulty in Putnam’s internal realism. The main goal of internal realism is a reconciliation between conceptual relativity and determinacy of reference. It must account for the fact that, although there is no independent world (THE metaphysical realist’s world), we are able to preserve a determinate reference. It must account for stability of meaning through conceptual changes, and avoid to pay the price paid by a full-blown conventionalism in terms of conceivability of different perspectives about the same issues (cf. Blackburn 1994, p. 25). The way from metaphysical realism to conventionalism is full of dangers: if we take it that mere extensional facts do not determine reference, then all we end up with is that it is intensions that do the work, in some “mysterious” way. However, intensions do not seem to do the work at all, for the same reasons for which translating does not seem to say nothing about what we are saying in different languages. And, at the same time, we do not have an ultimate meta-language, perfectly fitting the world, to appeal to in order to determine the reference relation. Putnam’s transcendental internalism, as it has been called, obviously rests on the idea that we could locate a middle-way; but in whichever way this middle-way is reached for, it does not appear to solve the problem. That is: either it simply states the fact that what we have is an outer world which is also shaped by the lens through which we look at it, or, when it comes to the point of explaining what kind of double-folded structure we find ourselves facing, it is caught up between the two extremes. This is because, as Blackburn has emphasized, the keypositions one finds at the two extremes seem to delete each other (cf. Blackburn 1994, p. 25). An idea would be to review the whole issue from a different viewpoint. Let us start with language rather than with the world, that is: let us see if the way in which reference is “embedded” in epistemic structures must necessarily be a “mysterious” one. Now we have to look for something inside the logical structure of language games which would allow us to have some kind of “internal” reference while at the same time preserving the possibility of, so to speak, stepping outside the game. We may then take it as an hypothesis that the solution we are seeking for is to be found not in any assumption about the external world

152

Chiara Tabet

(or THE world) but in the way(s) the world is understood through a variety of languages. The first question to ask here is then: “What is going on when we understand the meaning of a term x?”, rather than “What counts as the meaning of a term x?”, since an answer to the latter question would presumably be a function of the answer to the former. We are in the kingdom of epistemic structures of access to meaning, and may start exploring it by adopting an inferentialist semantics, that is, by assuming that meaning is to be treated as “role in an epistemic net.”7 This idea is clearly strictly connected to Wittgenstein’s motto that “meaning is use,” that is, that expressions of a language only have meaning in the context of a language game, whose structure they contribute to define. We may start approaching this problem by looking at a very simple example of an inferentially-structured practice. Let us imagine we have a chess game, and suppose that “meaning” and epistemic access to meaning with respect to this game can be described in purely computational terms - in other words, let us forget anything about the game that is not directly linked to its logical structure, so that we may for a moment leave aside the question of whether the players are human beings or computers.8 We can just refer to a “program,” that is: to a set of algorithmical pieces of information describing the logical structure of the game. This will have a recursively structured internal logic, which means that, for example, if we are dealing with a computer program, the program will at each stage “think” of a move and then call itself in the position of the rival player, in order to evaluate the effectiveness of the original move. From the latter’s point of view, it will then call itself in the position of the original player who is now “seeing” the game from the rival’s position . . . and so on. Of course the recursive strategy does have an end, and this may be fixed through the choice of some fundamental parameters (e.g., concerning the number of moves to be evaluated before each decision, maximum time allowed, etc.). In a rather intuitive way, it is the very recursive structure of moves and counter-moves that we may take as constituting the meaning of any sub-set of moves in the game context, as 7 Of course the main approach that I have in mind here is a conceptual role semantics, such as the one N. Block has proposed (and Putnam himself has contrasted). What I mean by “role in an epistemic net” is thus: role in the net of believes, thoughts, and, generally, mental states, which constitute the causal structure through which we have access to extensional features of reality. 8 I borrow the example from M. Davis’ discussion of Searle’s objections to functionalism, in Davis (2000, pp. 204-205). For Searle’s objections, cf. Searle (1980, pp. 417-457), as well as (2000, pp. 16-17).

Putnam and the “God’s-Eye” View

153

well as the “meaning” of (terms referring to) the chess pieces. These are thus thought of as intensional structures, and are themselves but ways of organizing the extensional information . Otherwise said, information (e.g. what is a knight?) is here given by the very way of organizing information (the set of moves/inferences associated with the knight). The game can thus be defined in terms of an inferential context where all subcontexts are interlaced and merged one into the other so as to form a general regular structure. A chess piece is therefore defined by the computational net that describes the set of moves in which the piece itself may play a part. Thus, when we refer to “chess pieces” we refer to the computational structure they are identified with. The whole game is in fact algorithmically defined by sets of sets of moves, and by computations on these. Now, what is it to understand a chess move? It is to find its position in the general computation, locating the sub-set of inferences through which it is defined. All that we have here is algorithms, or sets of instructions: rules univocally determining what something is; we could even think of the parameters telling us where to stop in the research (so that a move is not defined in terms of the whole set of all games possible at each stage) in terms of (coded) computational instructions. If we take up the analogy between mind and machine that is at the core of computational functionalism, then the idea would be to say that the meaning of a thought content can be given in terms of the inferential nets it comes in as a constitutive part, where these nets can be specified in terms of the computational procedures which define its links with all the other thought contents that are activated in the context of the “language game” thus constituted. Thus, via such a generalization, understanding the meaning of a term is not directly linked with a referential relation to an “outer” world, and is simply taken as locating the position of the instructions which define it, determining its “use,” in the framework of a general computational process. Of course this seems to be quite a dangerous generalization, resting on a set of disputable assumptions. Some of the reasons why it seems “dangerous” to generalize in this way have been discussed especially by J. Searle. Searle tells us that we are still left with our original problem of reference unsolved, and that taking up the analogy between mind and machine actually worsens it. He develops two main related points, that is: (1)

Syntax cannot “produce” semantics: since “semantics is not intrinsic to syntax,” and “computation is defined syntactically,”

154

(2)

Chiara Tabet

then “computation is not sufficient for semantic content” (Searle 1999, pp. 34-38; cf. also Searle 2000, pp. 16-17; and 1980). Since syntax cannot produce semantics, then even in the very simple case of chess-playing you do not have, by any means, a true understanding of the meaning of “chess” (cf. ibid.).

(1) and (2) are structurally linked, and the reason why they are is of special interest for our concern. In both cases, what one seems to need in order to be able to speak of “meaning” and “meaning understanding” is access to an external semantics, which one can have by interpreting the symbols. Now, of course the key-point here is how one defines ‘interpreting’. In order to explore that, let us first re-state the two points in the following way. (1i)

In the case of a chess-playing computer, what we have is just syntactical handling of “meaningless symbols” according to certain formal rules (algorithms). These rules are obeyed rather than followed, since what seems to be required by rule-following is a certain “consciousness” on the side of the agent, and a capacity of intentionally accessing the semantic content of terms (or moves) used according to those rules.

Searle’s popular Chinese room argument re-states very clearly this point. The argument considers, as it is well known, a person who is inside a room and who receives symbols from outside the room. He has a book of instructions telling him exactly which symbols he is expected to send out in reply. The book is so written that the symbols going back and forth (according to the mechanical instructions that it contains) constitute a conversation in Chinese. It would, Searle argues, be impossible to say that the person in question knows (the meaning of symbols of) Chinese. What enables us to say that what is going on is in fact a Chinese conversation, is our external position: that is, our having access to the meaning of the symbols going back and forth. Mere instructions for handling those symbols do not do any of the work. Meaning comes from interpretation, and interpretation is essentially a mapping of symbols onto the world accessed through intentionality; the fact that the symbols are organized in a certain rule-governed regular structure does not seem to tell us anything interesting about semantics. The recursion possibly at stake in meaning-understanding (e.g., when one learns a new language with the aid of a translation manual) finds an end in the attribution of semantic content to primitive symbols, and this attribution, achieved through interpretation, is but a function of an external reference relation.

Putnam and the “God’s-Eye” View

(2i)

155

Thus, understanding, whether it is understanding of an extensional object or of an intensional structure (e.g., a computational process) means, again, interpreting.

Let us start from point (2i). The most basic example apparently showing exactly why it is that we need an interpretation in this context is probably given by a human being facing a simple Turing machine executing some operation (say, addition) on numbers. The data in the machine (input and output numbers) are coded and given as numerals. It is the human being, but not the machine (and this is the point) the one interpreting the numerals as numbers, and thus the instructions (the conditionals describing the machine’s behaviour) as saying, for example, ADD x and y, where x and y are specific tokens of an abstract type (natural numbers). It is only by recognizing that x and y are this, and not something else, that the behaviour of the machine can be described and understood, from the “outside.” Now, what does it mean to interpret x and y as natural numbers? It means to acknowledge that they belong to a certain structure, in which relative positions (and rules to establish these) are determined by certain axioms (Peano axioms, which one may take as constituting an implicit definition of the structure itself). Thus, it means to recognize their place in a rule-governed structure. Again, one can take a number x as denoting a set of instructions saying how to execute computations on it. Thus, interpreting a rule-governed symbol (e.g. a numeral) as a number amounts to establishing a relation not to an external, ultimate extensional semantics, but rather to a set of other rules. But Searle would insist that this is not the point, or, that this is only part of the problem, and perhaps the less interesting one. What must be accounted for is not just the way we determine what denoted entities are, but rather the fact that reference is a relational structure, and especially the fact that this relation is defined, here, by the inside/outside boundary. Interpretation is “external,” in the sense that it is not embedded inside computational structures, in that (in the TM example) it is the programmer, or the human being understanding the machine’s behaviour, who has access to what it is doing from the outside. But is it really necessary that interpretation should be seen as taking place only via the inside/outside boundary, where “outside” is a given external semantic content? This brings us back to point (1i). Could we not see interpretation as embedded, in some way, within rule-governed structures (thus, for example, within an ideal set of programs running on a machine)?

156

Chiara Tabet

Consider not just a TM, but a Universal TM. A UTM acts precisely as an interpreter. We have sets of programs running on it, and these are specifiable in the same coded language as input and output data of the computations it executes.9 What the UTM does is, essentially: “decoding” the data, so that it can read them as programs and apply them (the instructions they denote) to input data. Now our two main questions become: (1ii) is there any “understanding” in these coding and decoding processes? (2ii) if not, do we thereby still need an external semantic content to access in order to understand the data as, e.g., numbers, or Chinese words? Again, the answer to (1ii) would seem to be a prima facie negative one. What we have here is only “blind” rule-obeying. However, are we sure that a human being is in a better position, with respect to this, than a UTM? The way this latter question is generally answered bears on the philosophical implications of two central Theorems, that is: Turing’s halting theorem and Gödel’s first and second incompleteness theorems (cf. Rey 1997, pp. 222-224). The latter are usually cited as showing that there is something that people can do that is not possible for any machine, namely: (a)

(b)

9

producing of an incompleteness proof. What is shown to be undecidable inside a Theory of arithmetic, and thus not computable by any machine, is seen to be true by the human being producing the proof. Thus, the classical self-referential formula used in the proof is attributed truth-conditions (thus meaning) from a point of view external to the deductive Theory; and to gain knowledge of the fact that the mentioned Theory is in fact consistent, which, again, cannot be proved “inside” the Theory itself.

This is because of two central theorems in Recursion Theory:

(a) (b)

the enumeration theorem; and iteration theorem, which should be thought of as symmetrical results.

The symmetry is precisely what is presupposed by the capacity of a UTM of acting as an interpreter.

Putnam and the “God’s-Eye” View

157

Turing’s halting theorem is in turn “used” with the aim of showing that a machine cannot describe its own global behaviour, as a special case in the set of problems which are computationally unsolvable. People thus seem to be far better off when it comes to “understanding” than machines are. However, there seems to be quite a substantial confusion here, and the confusion concerns the levels at which “understanding” takes place. As a matter of fact, it boils down to the (metaphysical realist’s) assumption that all (reference determination and meaning understanding) can happen inside a language, which would thus be something like an ultimate “Theory of everything,” so powerful that it can reach for a self-foundation. But a human being burdened with the task of showing that G. self-referential formula is true inside the Theory (that is: a human being with no meta-Theory to appeal to), or asked to give a computational definition of computability, would fail exactly as a machine does.10 Perhaps an example can be of some help here.11 Imagine that we have a “message” of some sort, written in an unknown code, and for which a decoding procedure must be found. Obviously, the first step for interpreting the message is to look for a key for decoding it; such a key is given by some (set of algorithms) to be applied. Here the very condition for establishing a semantical relation (thus for recognizing the semantic content of the message) is some sort of access to an algorithmical procedure. Where can this access be found? Obviously it cannot be contained in the message itself, and it must be found in some “external” message already accessible to the interpreter. In order for the interpreter to be able to understand this second message, she (or it?) must possess 10

The results to bear in mind are

(a) (b)

the Theorem stating that no recursive function exists that enumerates at least one index of recursive functions with values in {0, 1}; and Rice’s Theorem on partial recursive functions.

It can be shown that one cannot give a self-referentially closed definition of ‘understanding meaning’ and of ‘meaning’ inside the same “program” (Theory). If we take “meaning” of a term to be the set of algorithms that determine its computational behaviour (thus, extensionally, the computations it comes in as a constituent), and “understanding meaning” of the same term as locating these (sub-) computations it comes in as a constituent inside the framework of these same computations, then what we have is the following infinite self-referential structure. Let us denote the set of computations in which our term comes in as a constituent as X. These same computations would now include the computation determined by the instruction “find the place of X ” in the computational structures which include “find the place of X ” in the computational structures which include . . . etc. 11 I borrow this example from Hofstadter (1979, Ch. 6).

158

Chiara Tabet

the algorithmic key for decoding this as well. The series of decoding procedures thus needed clearly gives rise to an infinite regress, which starts as soon as we assume that we need, each time, an external semantic content through which the needed decoding procedures are meaningfully given. However, in fact the regress must at one point have an end. We have two places where we can find such an “end.” Let us suppose that we do find it in a final “message” interrupting the infinite regress. This message obviously contains algorithmical information for decoding all messages (it contains an algorithmical formulation of some basic ability of recognizing meaningful structures and thus decoding messages). By definition, this message must also contain information about some basic ability necessary for decoding itself. Thus, for this part, the information contained by the message, which is to be decoded, is given by some procedures for decoding the message itself. The selfreferential intensional structure conveyed by the message gives rise to a paradox: what must be acknowledged as the information conveyed by the message is at the same time the intensional presupposition for having access to itself. So, let us go for the other “place.” This is the interpreter’s “hardware,” where we have some basic decoding instructions. Of course these basic instructions must be thought as obeyed rather than followed: applied rather than understood. In wittgensteinian terms, explanations must find an end somewhere, and this “somewhere” seems to be, in our case, a set of basic instructions rather than an external semantic content. It thus appears as though “stability” of meaning is provided for intensionally rather than extensionally. However, we are not yet clear about the status of our “intensional end.” Let’s have a quick look at this, by linking the problem with a possible answer to question (2ii). This would go along the following lines. If it is true that you can only make sense of a computational structure from the “outside,” however this “outside” is itself another computational structure, in which, up to a certain extent, reference (thought of in functional terms) is “embedded.” When we understand that the Goedelian self-referential formula is true, though undecidable inside any Theory of arithmetic, our understanding takes place inside a meta-Theory. What we have is a hierarchy of languages, which may be thought of as programs. Stability of meaning through conceptual changes thus seems to be a relative stability: it is stability from the viewpoint, so to speak, of the program taken to be the “final” program (the hardware) to obey in order to account for “meaning” (in the software). But then, of course, “modulo a universal Turing machine, hardware and software are

Putnam and the “God’s-Eye” View

159

interchangeable” (cf. Odifreddi 1989): one can view the “hardware” in terms of further data to compute on. It is because of the nature of this co-existence of relative stability and open-endedness (what Putnam refers to as the “open-texture” of language games) that, while one cannot find a “universal” determinant of meaning in a language game transcended “from the inside,” one should search for it in the logical (computational) structure of this transcending.

University of Pisa Philosophy Department Piazza Torricelli 3/a 56126 Pisa, Italy University of St. Andrews Department of Logic and Metaphysics Arché Research Centre St. Andrews Fife K16 9AL, UK e-mail: [email protected] [email protected]

REFERENCES Blackburn, S. (1994). Enchanting Views. In: Clark and Hale (1994), pp. 12-32. Clark, P. and B. Hale, eds. (1994). Reading Putnam. Oxford: Blackwell. Davis, M. (2000). The Universal Computer. The Road from Leibniz to Turing. New York: W.W. Norton & Company. Hallett, M. (1994). Putnam and the Skolem Paradox. In: Clark and Hale (1994), pp. 66-97. Hofstadter, D.R. (1979). Gödel, Escher, Bach. An Eternal Golden Braid. New York: Basic Books, Inc. Odifreddi, P.G. (1989). Classical Recursion Theory. Amsterdam: North-Holland Publ. Co. Putnam, H. (1981). Reason, Truth and History. Cambridge: Cambridge University Press. Putnam, H. (1983). Vagueness and Alternative Logic. In: Erkenntnis 19 (1-3), 297-314. Putnam, H. (1990a). Realism with a Human Face. In: (1990c), pp. 3-29. Putnam, H. (1990b). A Defense of Internal Realism. In: (1990c), pp. 30-42. Putnam, H. (1990c). Realism with a Human Face. Edited by J. Conant. Harvard: Harvard University Press. Putnam, H. (2000). Representation and Reality. Cambridge, MA: The MIT Press. Rey, G. (1997). Contemporary Philosophy of Mind. Oxford: Blackwell. Searle, J.R. (1980). Minds, Brains and Programs. Behavioural and Brain Sciences 3, 417-457.

160

Chiara Tabet

Searle, J.R. (1999). I Married a Computer. The New York Review of Books (April 8), 34-38. Searle, J.R. (2000). Consciousness and Language. Cambridge: Cambridge University Press.

PART III REALISMS

José Tomás Alvarado CONCEPTUAL RELATIVITY AND STRUCTURES OF EXPLANATION*

ABSTRACT. Hilary Putnam’s doctrine of conceptual relativity sustains that there are many different incompatible, yet equivalent, descriptions of what have to be considered “the same” phenomena. This fact is reason to justify the idea that metaphysical realism is wrong and that a better general view of reality should be something like a “pragmatic realism.” Putnam sustains further that the different incompatible and equivalent descriptions have to bear the same explanatory virtue. Here it is contended that there seems to be difficulties in construing this requirement in a non-question begging way in an argument designed to show that metaphysical realism should be dismissed. Different descriptions can only be deemed to have the same explanative virtue if our knowledge of, for example, structures in the natural world, seems to be simply a form to “save the phenomena,” but this is a characteristic anti-realist thesis that should not be taken for granted in the dialectical setting.

Hilary Putnam’s pragmatic realism has been thought as a philosophical position that, in a very deep sense, aims to be “realism with a human face” (cf. esp. Putnam 1987). Our contemporary culture is frequently affected by trends that, on the one hand, try to explain reductively the human mind, the human identity and, hence, all the world where our lives occur with some set of facts amenable to treatment by natural science; on the other hand, other trends try to dilute the pretensions of rationality in natural science in order to preserve the world of “soft” facts, but in doing so they destroy also the rationality in our moral and aesthetic discourses and in our moral and aesthetic concerns. The project of naturalistic explanation – as it seems to some of our contemporaries – can only be * A former version of this paper was presented in the Congress “Hilary Putnam’s Pragmatism” held in Santiago de Compostela (Spain) on May, 13-14, 2004.

In: M.U. Rivas Monroy, C. Cancela Silva, and C. Martínez Vidal (eds.), Following Putnam’s Trail: On Realism and Other Issues (Pozna Studies in the Philosophy of the Sciences and the Humanities, vol. 95), pp. 163-183. Amsterdam/New York, NY: Rodopi, 2008.

164

José Tomás Alvarado

blocked if some form of relativism is endorsed in stronger or milder forms. The problem, as professor Putnam has insistently told us, lies in the fact that the structures and patterns that shape the real lives of real human beings, their concerns and the way they see that their own lives have a sense, cannot survive either in physicalism or in relativism. The question for a responsible philosopher in our times seems to be: How can we make sense of the identity of human beings as real moral agents with a world view basically described by the natural sciences? Pragmatism, in one form or another seems to be the basic form of a correct answer for this question, at least for professor Putnam. However, the devil is in the details. It is one thing to envisage the general form that a satisfying answer to a problem may have and another to actually devise it full-blooded. This work has to do with some systematic difficulties that the project of a pragmatic realism has to face and, in particular, has to do with some difficulties that the interesting doctrine of conceptual relativity has to face. As will be pointed out soon conceptual relativity has a central importance for Putnam’s pragmatism. Here it is not contended, nonetheless, that conceptual relativity is all the content of Putnam’s pragmatism, but it should be obvious that it is, at least, one of the most characteristic ideas by which he deserves the name of a real pragmatist. Then, any difficulty in conceptual relativity should very probably be deemed as a serious difficulty for any contemporary feasible proposal of Pragmatism itself.

1. What Is Conceptual Relativity? A Rough Characterization Hilary Putnam first presented conceptual relativity in the context of a broader philosophical position he – then – dubbed “internal realism,” that in turn should be understood in opposition to metaphysical realism. In the perspective of metaphysical realism the world consists of some fixed totality of mind-independent objects. There is exactly one true and complete description of “the way the world is.” Truth involves some sort of correspondence relation between words or thought-signs and external things or set of things. I shall call this perspective the externalist perspective, because its favourite point of view is a God’s Eye point of view . (Putnam 1981, p. 49)

Internal realism, on the contrary, could not accept that there is a fixed totality of objects, or that there is a totality of properties and relations

Conceptual Relativity and Structures of Explanation

165

under which these objects fall – or fail to. Neither is there a complete unique true theory stating once and for all what exactly the states of affairs constituting the world are. In the internalist perspective, objects exist only within a theory or description and, then, the states of affairs in which these objects can appear as part of their structure can only exist also within some determinate conceptual frame given by theories and descriptions.1 This is basically the idea of conceptual relativity. Putnam also connected conceptual relativity in this period with the rejection of a conception of truth as correspondence between thought and reality (the propositions that give content to our judgements and the states of affairs in the world that make such propositions true or false). Putnam proposed instead a conception of truth as idealized rational acceptability. Whatever were the merits of that conception of truth, it was abandoned by Putnam later,2 so it will not be discussed here. Putnam has focused on a phenomenon of conceptual relativity by which several theories (or descriptions) happen to appear equivalent and also, in a sense, incompatible. The phenomenon of conceptual relativity leads to a philosophical doctrine, according to which there is no unique complete true theory of the world, but a plurality of theories and descriptions, all of them true, incompatible and yet equivalent. In the famous example in The Many Faces of Realism (1987) Putnam invites us to consider the case of a world where we can apply two different conceptual schemes, one that is called the “Carnap’s scheme” and another that is called the “Polish logician’s scheme” (after Leniewski and his mereology). Under the Carnap’s scheme one could say that the world considered has only three objects, say x i, x ii and x iii. On the other hand, under the Polish logician conceptual scheme by which mereological sums are admitted, there appear seven objects (maybe even eight if you consider a null object Ø) in the same world, that is x i, x ii, x iii,

1 What is said here of objects can also be applied to other categories of entities, like events, tropes or universals. Putnam does not mention them explicitly in these formulations of internal realism, but it should be understood that the general idea of conceptual relativity is that whatever we come to think are the categories of entities making up ontologically the world, they could only be conceived within a conceptual scheme, theory or description. 2 Cf. Putnam (1994). Nonetheless, there were some references to the notion of truth in the first formulations of internal realism those years that have persisted in time. Putnam says there that truth can be conceived as correspondence with reality, but only within the conceptual scheme by which the matching of thought and reality can be framed. It is legitimate to talk of correspondence, but only in this sense “internal” to theories, concepts and descriptions (cf. Putnam 1981, p. 52).

166

José Tomás Alvarado

x i + x ii, x ii + x iii, x i + x iii, and x i + x ii + x iii. Putnam comments about this example: Now, the classic metaphysical realist way of dealing with such problems is well-known. It is to say that there is a single world (think of this as a piece of dough) which we can slice into pieces in different ways. But this “cookie cutter” metaphor founders on the question, “What are the ‘parts’ of this dough?” If the answer is that Ø, x i , x ii , x iii , x i + x ii , x ii + x iii , x i + x iii and x i + x ii + x iii are all the different “pieces” then we have not a neutral description, but rather a partisan description – just the description of the Warsaw logician! And it is no accident that metaphysical realism cannot really recognize the phenomenon of conceptual relativity – for the phenomenon turns on the fact that the logical primitives themselves, and in particular the notions of object and existence, have a multitude of different uses rather than one absolute “meaning.” (1987, p. 19)

What appears in this example, and in others used by Putnam, is that there is no sense in trying to adjudicate which of these versions of the world is correct and which is incorrect. They are both correct, in fact, or so it seems, and we can only adjudicate between them on grounds of pragmatic relevance, considering the context where some descriptions should be made or could be made, considering the background of those to which the descriptions are addressed and those which make the descriptions, their interests, their beliefs, what they would deem relevant and informative in that precise situation, or in other situations sufficiently similar. If what objects there are is somewhat dependent on our conceptual schemes and, in turn, if our conceptual schemes are somewhat dependent on the pragmatic context where such schemes are used (or would be used, eventually), then it seems that what objects there are comes to be dependent, in a sense, on our pragmatic contexts, constituted by our beliefs, intentions and interests, both theoretical and simply practical. The doctrine of conceptual relativity, then, seems to be another form to put the old idea that in some sense what we call the “world” or “reality” has an irreducible component coming from our own mental activity. In Putnam’s metaphor, mind and world jointly made up the mind and the world.3 Putnam is not saying that reality is simply a construction

3 Cf. Putnam (1981, p. xi): “I shall advance a view in which the mind does not simply “copy” a world which admits of description by One True Theory. But my view is not a view in which the mind makes up the world, either (or makes it up subject to constraints imposed by “methodological canons” and mind-independent “sense-data”). If one must use metaphorical language, then let the metaphor be this: the mind and the world jointly

Conceptual Relativity and Structures of Explanation

167

– an arbitrary construction – of our mental activity. The idea may be better expressed saying that there is a relation of mutual dependence between our mental activity and reality in which neither of the terms can be properly made intelligible without the other. Two important points should be recalled to complement the rough idea of dependence on conceptual schemes. First, conceptual relativity is not relativism. Let “relativism” be the thesis that the truth value of propositions is dependent on what someone believes of such truth value (the one in question can be a single individual or a community of human beings selected by some special criteria, or even the entire culture, depending on the type of relativism considered). When one has two different descriptions that would be characterised as conceptually relative between them one doesn’t have two propositions in a relation of contradiction. Someone uttering them is not affirming and denying the same proposition. It is not like saying that “lying is wrong” and at the same time saying “lying is not wrong.” It is a situation in which different concepts are used so that the statements that are made with them appear to be situated at different levels of description. It is like talking, on the one hand, about geometry considering “points” as primitive objects and, on the other, talking about geometry considering “convergent sets of solid spheres.” Consider again the case of the world described by Carnap’s conceptual scheme and by the Polish logician’s. Under one of those schemes one would be inclined to say that “there are three objects in the world.” Under the other one would be inclined to say that “there are seven objects in the world.” Prima facie it seems an outright contradiction. The point of the phenomenon of conceptual relativity is that you can only state that there are some or other objects if you have a concept, or set of concepts, at your disposal by which you can select those objects in the world. In this case, as the statements in question have been done within different conceptual frames, they cannot contradict each other. They are just at different descriptive levels, so they cannot clash. At the same time, it should be considered that once one has settled what conceptual scheme is in question there are some propositions objectively true and some other objectively false. In the example considered, the point is that if you have decided to use the Carnapian conceptual frame, then you cannot but admit that there are three objects in the world and not seven, nor five. Nor any other number. Differences in the number of objects composing the world come only when you are make up the mind and the world. (Or, to make the metaphor even more Hegelian, the Universe – with minds- collectively – playing a special role in the making up.)”

168

José Tomás Alvarado

considering the alternative conceptual frames that could be used, but when any one of these is selected, the frame induces a precise answer to the question about how many objects there are. What conceptual scheme to apply may be taken as a matter of sheer decision, but it is in no way a matter of decision whether the conceptual scheme selected has or does not have a case. 4

2. What Is the Systematic Importance of Conceptual Relativity? The ideas presented above may have appeared to have a flavour reminiscent of Quine’s well-known contentions about ontological relativity, or they may have seemed a continuation of Goodman’s rhetoric of worldmaking (not to mention Rorty’s conversation of humanity). I think that, although the connection between Putnam and these other philosophers is patent, there is a deep difference between Putnam, on the one hand, and all of them, on the other, that, further, makes Putnam’s doctrine much more interesting and sophisticated philosophically. The usual contention that can be found in all these philosophers is that objects, properties, relations and all categories of entity in general are only intelligible within a conceptual scheme, just as the thesis of conceptual relativity seems to say. The point is why one should accept such a thesis. Let conceptual relativity be – at least provisionally – the following principle:

4

Putnam has also said that an important aspect of his doctrine of conceptual relativity lies in the fact that it does not affect complete languages, but it has to be understood only applying to scientific or mathematical theories. There are no interesting contrasts between different conceptual schemes describing chairs and tables. The conceptual frame within which we talk of tables and chairs is not really “optative” for us. The phenomenon of conceptual relativity, on the contrary, arises when we do have optative languages at our disposal (cf. Case 1997). I will consciously not follow this suggestion. The value of the doctrine of conceptual relativity seems to lie in the potential generalization of this phenomenon of optative languages for all our discourses. The phenomenon should be taken as indicative of a deep fact about the relationship of our thought and reality. If there are systematic reasons that allow us to conclude that, for example, in some areas of scientific discourse there are optative languages that can be indefinitely generated for expressing what, in a sense, are the same facts, then these reasons should lead us to think that our everyday “theories” about tables and chairs are affected of the same “relativity.”

Conceptual Relativity and Structures of Explanation

(1)

169

x C (x is dependent on C)5

Here x ranges over objects (or other categories of entities) and C ranges over concepts (or conceptual schemes). The idea expressed by proposition (1) is that entities can only exist if they are somehow “dependent on” some or other conceptual scheme. There is a typical strategy for arguing in favour of principle (1), that seems to be the usual type of consideration by which philosophers, like the ones cited above, put forward the thesis. Suppose, for reductio, that thesis (1) were in fact false. If it were false then it appears that: (2)

x C ¬(x is dependent on C)

Principle (2) is just the negation of (1). Now, the defender of principle (1) will invite us to consider what could be that possible entity by which the principle (1) could appear to be false. That entity satisfying principle (2) has to be such that no concept (or conceptual scheme) applies to it. The defender of conceptual relativity – taken in this very rough form – will contend that there is no such entity. Any entity we can envisage obviously should fall under one concept or another and, then, seems to be falling under principle (1). Then, the argument goes, if the critic cannot come with a counter-example it has to be because there are none. So, thesis (1) should be correct. Nonetheless, there is an obvious flaw in this line of argument. It is true that one can refer to an entity only by using our conceptual resources for doing that act of referring. No entity can be cited as counter-example for principle (1) if it is not by conceptualising that entity with one conceptual scheme or another, but then it happens that the entity referred to is immediately not eligible as putative counter-example to principle (1). The problem lies in why this fact can be taken as a reason to suppose than principle (1) becomes validated. It should be because the fact that no counter-example can be cited by the critic has to appear as a reason to think that there really are no counter-examples. But, why do we have to accept such conclusion? One reason may be that there is implicitly working in the dialectical setting a principle to the effect that: (3)

p S (p SKp)

5 I’m conscious that this formulation will not be congenial to Professor Putnam. He has explicitly stated that he is not trying to defend the idea of dependence of reality on mind (cf., for example, Putnam 1992, esp. pp. 366-368). Nevertheless, I prefer a principle like (1) because it states what contemporary philosophy has seen in Putnam’s internal realism and Putnam’s pragmatic realism.

170

José Tomás Alvarado

Principle (3) is a principle of knowability that says that for every state of affairs there is a subject S such that, if the state of affairs obtains, then it is possible that such state of affairs is known by S. Under the hypothesis of principle (3) as stated, then the fact that we cannot come with a counter-example to principle (1), that is, the fact that one cannot come with the reference to some entity such that no conceptual scheme is used to understand it, has to be taken as reason for thinking that there really is no such entity. The problem is that principle (3) is in itself a strong antirealist principle 6 that cannot be taken for granted as premise in an argument that has as its aim to show how it is that reality is, in a sense, dependent on our mental activity. Principle (3), by which everything is knowable, has to be considered as a consequence of an anti-realist contention about the dependence of reality on our conceptual schemes, and not the other way round. The defender of principle (1) can, at this point, say that his contention was not meant as a deductive argument, but simply as an inference to the best explanation. The point of his considerations, he may further add, was to make us see that the facts considered have no other reasonable and feasible explanation than the postulation of certain dependence of objects and other entities on our conceptual schemes as it is put in principle (1). But it is clear that we do have other much more feasible explanations of our cognitive contact with reality than this sort of “conceptual relativity.” The traditional realist explanation can deal perfectly well with all the facts and is much more intuitive. A traditional realist explanation could say that we in fact have to use some concepts to conceptualise reality, but not because those concepts play the role of some sort of a priori frame to which our cognitive practices are inescapable. The concepts we have framed during time come from other cognitive intercourses with reality, and should be taken as tools devised for illuminating nature at – what we reasonably take to be – its joints. Conceptual schemes have not appeared from mere arbitrariness; they are preferred normally for their fertility in the rational enterprise of understanding reality. In other words, the argumentative strategy that has been presented is simply not sufficient to render the vast theoretical reform that it purports to support. The fact that every entity we refer to has to be conceptualised in one form or another does not deductively entail that those entities are dependent on our conceptual schemes, nor is it the case that this fact is a strong reason to suppose that the thesis of relativity of entities to 6

In fact, some philosophers conceive anti-realism precisely as the endorsement of the principle of knowability. Cf. Wright (1993) and Tennant (1997).

Conceptual Relativity and Structures of Explanation

171

conceptual schemes is preferable as an overall metaphysical explanation to more traditional realist positions. At least not if some other argument or evidence is not produced for the same conclusion. Here is the point where conceptual relativity – in the form proposed by Putnam – enters. Putnam is not just thinking of the fact that any explanation one can give of what some discourse is talking about should be done using some other discourse. This is fairly obvious and inconclusive. Putnam is considering a much deeper phenomenon by which several theories appear to be in a sense equivalent and, also in a sense, incompatible. The phenomenon should be taken as indicative of a fact about our rationality and a fact about the connection between theories and the reality they purport to represent. A metaphysical realist, as Putnam understands him, could be inclined to sustain that there has to be just one true complete theory stating all the facts obtaining in the world. The world should be such that for each object and each property or relation obtaining objectively and independently there should be a term in the ideal language in which the unique complete true theory is stated. Now, the doctrine of conceptual relativity questions this realist assumption in the sense that for whatever theory presented by the realist as representing how states of affairs of the world are, there will be other theories with different ontologies and specifying what appear to be different structures over these ontologies but with the same cognitive value. Putnam comments on this situation: Talk of “theory-independent objects” is hard to keep. The problem is that such talk may retain “the world” but at the price of giving up any intelligible notion of how the world is. Any sentence that changes truth value on passing from one correct theory to another correct theory – an equivalent description – will express only a theory-relative property of the world. And the more such sentences there are, the more properties of the world will turn out to be theory relative. [. . .] All this isn’t an artefact of my simple example: actual physical theory is rife with similar examples. One can construe space-time points as objects, for example, or as events, or as properties. One can construe fields as objects or do everything with particles acting at a distance (in classical physics, at any rate). The fact is, so many properties of “the world” – starting with just the categorical ones, such as cardinality, particulars or universals, etc. – turn out to be “theory relative” that “the world” ends up as a mere “thing in itself.” If one cannot say how “the world” is, theory-independently, then talk of theories as descriptions of the “world” is empty. (1983, pp. 44-45)

172

José Tomás Alvarado

The structure of this second argument for conceptual relativity, that is, thesis (1), comes to the following. First, there is a premise connecting realism with the existence of a unique complete true theory of the world: (4)

w !T ((w is objectively structured) (w is completely represented by T ))

Here, in principle (4), w should be taken as ranging over worlds, that is, the totality of objects, properties and the states of affairs they constitute. A world is a totality of states of affairs. T is ranging over theories, that is sets of propositions closed under the relation of logical consequence. The predicate ‘[ is objectively structured’ has to be taken here as saying precisely what the realist wants to say about the world, that is, that it has a structure independent of our knowledge or of our cognitive activity, in general. The quantifier !T is saying that for any world objectively structured in a totality of states of affairs, where some totality of objects, properties and relations appear, there is one and just one theory representing it completely.7 Let us say, further, that a world is represented completely if and only if there is no state of affairs of the world s such that there is no proposition p of the theory completely representing it, and p if and only if s. The intuitive idea here is that s is the truthmaker of p, and every state of affairs of the world is the truthmaker of one proposition belonging to the theory representing the world. 8 Of course, it is not necessary that anyone actually should formulate this theory completely representing the world. It is a matter just of idealised mathematical possibility. For each element of the world there is an abstract element to which it is correlated. The abstract elements mirroring the world are structured in a theory – that is also an abstract entity, maybe here a set theoretical construction. The specific contribution given by the doctrine of conceptual relativity lies in a principle that admits of a formulation like: (5)

T T (T is conceptually relative with T)

7 Suppose that the actual world is w*, and suppose further that it is objectively structured, then principle (4) is saying that:

T{(w* is completely represented by T ) & T ((w* is completely represented by T) (T = T ))} 8

More formally, let: wT{(w is completely represented by T ) sp (((s H w) & (p H T )) (s p))}

Conceptual Relativity and Structures of Explanation

173

As theory T is also a theory, principle (5) guarantees that there is another theory T conceptually relative with T, and then T should be conceptually relative with a theory T .9 The principle stated here makes impossible an instance of principle (4) if its antecedent were true. Principle (4) can be true, nonetheless, but under the hypothesis that the antecedent is false, that is, under the hypothesis that there is no objectively structured world. Consider, in fact, that there is an objectively structured world w (the actual world), then it should follow that there is one and only one theory T representing it completely. But principle (5) says that there will be – in principle, infinite – theories different from T conceptually relative with T representing completely the world. Then, there is no unique true theory representing completely the world. If the quantified conditional is true, by hypothesis, then it follows that the world is not objectively structured. Of course, there are many things that require careful explanation in this train of thought. Maybe a metaphysical realist opposed to the conclusion that the argument purportedly renders would be inclined to reject or resist principle (4), saying that it does not follow that an objectively structured world requires a unique theory completely representing it. Maybe, also, he would be inclined to say that the class of equivalence of all the theories conceptually relative might count as just one theory. This could make principle (5) inoffensive against metaphysical realism. In any case, the relation of conceptual relativity between theories should be clarified before discussing, or settling, all those questions.

3. What Is Conceptual Relativity? (Again) As has been explained, the intuitive idea of two theories that are conceptually relative between them is the idea of two descriptions that are, prima facie, incompatible but at the same time equivalent, that is, “depicting” the “same” facts, with the same cognitive value. Putnam usually gives examples like the following about space-time: Story 1. Space-time consists of objects called points (point-events). These have no extension, and extended space-time is built up out of them just

9 Let the relation of conceptual relativity between theories be reflexive, symmetric and transitive. It follows that there results for each theory T under consideration an equivalence class of all the theories conceptually relative with T (and with each other).

174

José Tomás Alvarado

as, in classical Euclidian geometry, the extended line, plane, and solid bodies are built up out of unextended spatial points. Story 2. Space-time consists of extended space-time neighbourhoods. All parts of space-time have extension. This corresponds to the theory (advanced by Whitehead), that classical Euclidian space consists of extended spatial neighbourhoods. On Whitehead’s view, “points” are mere logical constructions and not real spatial objects: a point is (identified with) a convergent set of solid spheres (i.e., spheres together with their interiors). 10 (1983, pp. 42-43)

Another preferred example, along with the world of the Carnapian and the Polish logician, is the alternative of descriptions of the gravitational force with a field or with particles acting at a distance with retarded potential. The point of these examples is showing cases where two different theories, incompatible at face value, have the same cognitive and explicative value. There seems to be no sense in trying to discover which of them is the correct theory. It is easy to imagine, in fact, two descriptions with the same empirical value where in one of them the appearances are explained as the different states in which point-events might be, while in the other those same empirical appearances are explained as states of the regions of space-time. For any state of affairs in space-time p, there will be an explanation by which p should occur as a consequence of previous states of the point-events in the neighbourhood, while, at the same time, there will be another explanation by which p should occur as a consequence of previous states of the regions of space-time there. It is obvious from these examples that a relation of conceptual relativity between two theories or descriptions T i and Tii is not simply something that may arise due to the use of different concepts. For example, it is clear that a statement saying that: (6)

10

S believes that the cat is on the mat,

Putnam uses a very similar example in (1978). The difference is that in this text Putnam also suggest a third story: “There are only line segments with rational end points (i.e. since there aren’t “points,” in this story, except as logical constructions, (1) every line segment has rational length; (2) the piece of the line between any two line segments is a line segment, and so has rational length; (3) every line segment is divisible into n equal pieces, for every integer n; (4) there is at least one line segment; and (5) the union of two line segments is a line segment). Irrational line segments are treated as logical constructions – sets of “points,” where “points” are themselves Cauchy convergent sets of rational line segments” (p. 132). This third story comes after the other two explaining the ontological basic structure of a “lineal” world.

Conceptual Relativity and Structures of Explanation

175

cannot contradict another statement saying that: (7)

S has a physical state p in his brain.

At least, there is no contradiction here prima facie. In fact, if mental states are not reducible, nor supervenient, to physical states of the brain (or physical states of the brain-cum-environment) then one has here an example of two different descriptions that are, in a sense, talking about the “same” but at different levels. Well, if these statements cannot possibly clash between them, then the relation between them is not a relation of conceptual relativity. The reason is simple: conceptual relativity arises only when two theories have the same cognitive and explicative value. If mental states are not the same as physical states of the brain, or, if you like, mental states are not supervenient to physical states, then both descriptions cannot have the same cognitive and explicative value. In fact, proposition (6) would be stating something that no proposition or set of propositions stated in the physicalistic language of proposition (7) could possibly state. For Putnam, the conditions for conceptual relativity between two theories T i and Tii are: (i) (ii)

Ti and Tii have to be mutually relatively interpretable, and T i and Tii have to render explanations with the same cognitive value for any phenomena (Putnam 1983, pp. 38-39).

It is a requisite for these conditions to work that there be a definite set of phenomena that is taken as the domain explained and described by both theories Ti and Tii. These phenomena should be described in one theory or another, so this requisite is not trivial here. If one has no neutral description of those phenomena, then the stating of one single domain to which both T i and Tii are referred requires a hermeneutical exercise by which the mutual relative interpretation of the theories has to become a real mapping of the meanings of them. That is, it is a requisite condition that the theories falling under the relation of conceptual relativity must have the same truth conditions. The first requisite of mutual relative interpretability is purely formal. It is said that a theory Ti is relatively interpretable into another theory Tii if and only if there are possible definitions of the terms of Ti into the terms of Tii, so that all the propositions that can be meaningfully formulated in the language of T i can be expressed as propositions in the language of Tii and, further, all the theorems of Ti will appear as theorems of Tii. Two theories Ti and Tii are mutually relatively interpretable if and only if Ti is relatively interpretable into Tii and Tii is relatively

176

José Tomás Alvarado

interpretable into Ti. A theory can be seen in a purely formal way as a set of propositions in a certain language in which names and predicates are concatenated by a logical structure. A theory is relatively interpretable into another when this formal structure can be mirrored into another, so that all the theses of the theory in question can appear as theses of another theory that has, prima facie, a different conceptual repertoire. The relation of mutual relative interpretability is purely formal because the connection of the theories between which the relation is said to obtain is only concerned with the structure defined by quantifiers, logical constants and the relation of logical consequence between propositions, but it is not concerned with the meaning attributed to the names and the predicates of the theory that has been interpreted. A theory talking about cats might be relatively interpretable into another talking about dogs, if the theory about dogs has sufficient structure and concepts, but not because cats somehow come to be finally dogs, but simply because the theory talking about cats is considered under an abstract perspective where the denotation of its terms and predicates does not matter. In any case, this requisite of mutual relative interpretability, although not sufficient for conceptual relativity between theories, can be taken as a necessary condition for it. Many different discourses that, in some vague and rough sense can be considered as talking about the same will fail to meet the qualification as conceptually relative simply because they do not comply with this first formal requirement. Then, the relation of conceptual relativity between theories has to at least require of the theories falling under it that no thesis of one of them could not be translated into a thesis of the other, for lack of conceptual resources or for lack of theses from which they might deductively follow. The second requirement is more substantive and requires more attention. It concerns explanatory equivalence between the theories that are said to be conceptually relative. This is the formulation of Putnam (it is done for complete theories, that is, theories where all propositions are either provable or refutable; this condition can be overlooked here): Suppose T i and T ii are complete mutually relatively interpretable theories. (I confine attention to complete theories, because when we call descriptions “equivalent” we generally have in mind a relation between ideal completions of those descriptions, and not just between the finite fragments we actually possess.) Suppose moreover that the “translation” of each into the other preserves the relation of explanation, and that the same phenomena are explained by both (this is the informal element alluded to). Then the two theories are equivalent descriptions (we propose). (1983, p. 39)

Conceptual Relativity and Structures of Explanation

177

We do not have (at least not yet) a precise notion of explanation acceptable for all with which one could try to develop how this requirement of explanatory equivalence could actually work. But we can suppose that some requirements are necessary for any explanation, although they may not be deemed sufficient to formalise the relation of explanation. The discussion here will not be concerned with probabilistic explanation – that has many difficulties of its own – nor with singular explanation. It will be supposed here that two requirements have to be satisfied by the relation in question: (i)

(ii)

if a certain fact p is explained by another fact q, then there should be certain law L (a statement that should have the form of a universally quantified conditional and that can be deemed as capable to support counterfactual conditionals) such that p deductively follows from L (the general natural law) and q (a description of the initial state of the system considered for explaining the fact p); 11 if a certain fact p is explained by another fact q, then q should be causally connected with p. In a sense, p should be said to be caused by fact q. 12

11

This requirement introduces the ideas of the traditional nomological deductive model of explanation devised by Hempel (1993). 12 This requirement has been introduced by several philosophers like Salmon (1993) or Brody (1993). The intuitive idea behind this requirement is that the relation of logical consequence between propositions is not sufficient to discriminate the relations of ontological dependence working in nature and that natural science tries to discover and to represent in our theoretical models. The classical example of Aristotle (cf. Analytica Posteriora, I, 13) shows how two syllogisms with the same form should be distinguished for their explanatory relevance. (I)

All that twinkles is near Planets twinkle Planets are near

(II)

All that is near twinkles Planets are near Planets twinkle

Here syllogism (I) has as consequence (planets are near) what appears to be the real cause of the phenomenon, that appears as minor premise (planets twinkle). This order might be representing the epistemic process by which one could come to discover the explanatory fact that planets are near. In syllogism (II) on the contrary, the consequence is the observed phenomenon (planets twinkle) and its real cause (planets are near) appears as minor premise. Both syllogism are Barbara, first figure. They could not be distinguished

178

José Tomás Alvarado

Of course, this requirement could be of no relevance if some conception of causality, like the nomological deductive theory, were accepted. There is no single conception of causality that is introduced here in order to make precise this requirement (ii), but the relation of causality should be taken to be, at least, an ontological relation of dependence between two events not reducible to facts about the relations of logical consequence between propositions, nor reducible or supervinient to non-causal facts (say, about mere physical regularities).

4. The Requirement of Explanatory Equivalence How should the requirement of explanatory equivalence be construed, then? It has to be supposed that two theories Ti and Tii that are said to be conceptually relative have such a syntactic configuration that the class of structures that could satisfy them is, at least prima facie, different. Suppose, say, that the structure of the models of Ti is characterised by the triple A = where A is the domain of entities over which the theory quantifies. Let Ti be the description of space-time by points, like it appears in the “story 1” given above by Putnam, then the domain A is a set of points out of which space-time is constructed. P i is the set of all n-adic properties (that is, monadic properties and n-adic relations) that obtain, or fail to, between objects of the domain. c i is the set of selected objects to which the theory makes reference by proper names or other devices apt for individual reference. On the other hand, suppose further that the structure of the models of Tii is characterised by the triple B = . In the same way as in the case of triple A, here B is the domain of entities over which the theory quantifies. It could be the set of neighbourhoods of “story 2” above. Pii is the set of properties and relations obtaining or failing to between neighbourhoods and c ii the set of selected individuals to which Tii makes reference. Now, the property of mutual relative interpretability that has to obtain between Ti and Tii, if they are really conceptually relative, means here that all the entities that in theory Ti are basic, that is the set of points A, have to be definable as certain states of affairs obtaining between the entities basic for T ii, that is, the neighbourhoods composing B. In the same way, neighbourhoods of B have to be definable as set of points of A. Any state of affairs making true a proposition T i, that is any fact about points falling under using only formal criteria. By criterion (i) above each of them is as good as the other. But certainly, they are not.

Conceptual Relativity and Structures of Explanation

179

one or other property or relation, has to be represented as a state of affairs making true a proposition actually pertaining to Tii, that is, a state of affairs about neighbourhoods falling under one or other property or relation, or maybe a set of states of affairs about neighbourhoods falling under properties or relations in Pii connected between them in some way. Consider now a certain fact p that should be explained by both T i and Tii, say two balls falling freely, at the same time, from the top of the Tower of Pisa to the surface of the earth (we may suppose further that Galileo himself drops the balls of iron and all other professors and students of the University of Pisa are seeing the event from below expecting how the theories of young Galileo will finally result put to the test). It is going to be supposed further that fact p – both balls falling at the same time, or almost, to the surface of the earth, after obtaining certain acceleration starting from rest at the top of Tower of Pisa – can be described from a neutral language. Fact p can be described using common language as simply both balls falling at the same time using the phenomenological point of view of the observers and what they normally take to be the entities that compose the world. How will both explanations work here? First, from the point of view of theory Ti, fact p has to be expressed as a fact about a set of points of space-time (its preferred ontology) composing some states of affairs under properties and relations also of the theory. Only in this way could fact p be deduced from some antecedent state of the system in question q and a law of T i. Then, it has to find a certain initial state of the physical system q, that here should include the altitude of the Tower of Pisa from the surface of the Earth in that place and the masses of each of the balls and of the Earth. This fact q, described only in terms of states of affairs affecting points of space-time, has to appear as causally relevant for the obtaining of p. It is supposed that a natural law L, or a set of natural laws, pertaining as theses to Ti will allow the deduction of p from q as premise. From the point of view of theory T ii, on the other hand, fact p has to be expressed as a fact about a certain region of space-time. The initial state of the system in question, i.e. q, is also described as a certain region of space-time from which p could follow deductively, granted some theses of T ii guaranteeing the transition. And here p should be caused by q. As can be seen, it is supposed that both theories T i and Tii are perfectly empirically adequate. By hypothesis there is no way of deciding between them as a matter of predictive precision. In so far, then, as empirical adequacy is concerned, they are, indeed, explanatory equivalent, just as Putnam wanted them to be. There is a fair question that remains to be answered, nonetheless, concerning this supposed explanatory

180

José Tomás Alvarado

equivalence. It is true, indeed, that empirical adequacy is a very important epistemic criterion when coming to decide between alternative theories, but it is not the only criterion that could or should be relevant. If the decision between alternative theories were just a matter of empirical adequacy, then for the class of all the empirical appearances there are countless alternative theories satisfying the requirement, and countless completely outrageous with no plausibility at all. Theories about devils and witches interfering with physical systems by magic can surely be empirically adequate. Probably they will be hopelessly ad hoc, and not simple, nor very coherent with all we think and believe about the large scale structure of reality, but these other considerations need not enter here. We are now just focusing on empirical adequacy. The point is that if criteria other than empirical adequacy are admitted, then there seem to be reasons to think that one might be able to establish preferences between empirically adequate theories according to considerations of simplicity, conservativeness, beauty, internal coherence and coherence with other bodies of well entrenched beliefs. It is just not obvious that the structure of physical space-time is equally feasibly built from points or built from regions, if what constitutes a good explanation is not only empirical adequacy but other epistemic criteria must enter into the balance of considerations. It should be recalled here that the business of natural science, at least as many philosophers think, is to discover and describe the objective structures of natural processes, that is, what types of events are dependent on others, and among them, what types of event are ontologically more fundamental. The question is to discover and describe the objective order of the causal dependences in nature. From this perspective, the structure of explanation posited by a theory is simply not indifferent. Those structures of explanation should be representing the objective order of ontological dependence and should not be taken as mere predictive devices to save the phenomena. Consider again the case here. In principle, explanations within theory Ti and within theory Tii are describing the causal process by which the balls fall from the top of the Tower of Pisa dependent on the distance from the top of the Tower to the surface and the mass of the Earth. In principle, also, there is no difference between them as to how they describe the law followed by the accelerated movement of the balls to the surface. But there is a difference as to what those explanations take to be the fundamental facts in question concerned here. For one explanation it is a fact about point-events, while for the other it is a question about regions of space-time. Now, surely one may say here that the mathematical descriptions are equivalent in so far as the description and

Conceptual Relativity and Structures of Explanation

181

prediction of observable phenomena is concerned. But surely also the fact that two mathematical theories are equivalent from a formal viewpoint is not the only relevant factor here either. The theories we are considering are not just mathematical theories with some formal structure, for they are taken to be physical theories describing the physical structure of nature. There is, of course, a perspective within which these differences between “points” and “regions” appear as irrelevant, but it is an antirealist perspective of science. A constructive empiricist would say that theories are equivalent indeed if they are empirically adequate. The point here is that Putnam cannot just take for granted this perspective in an argument designed to show that reality is somewhat dependent on conceptual schemes because every theory one can think of as explaining the structure and workings of reality has countless conceptually relative counterparts. In fact, two theories are conceptually relative if and only if they are mutually relatively interpretable and explanatorily equivalent. And those theories are explanatorily equivalent if criteria other than empirical adequacy do not matter. It follows, then, that theories will have countless conceptually relative counterparts only if the explanatory structure of those theories does not matter, which comes to be blatant anti-realism. Consider again the whole dialectical situation. Putnam is trying to argue that there are neither objects nor properties if not within conceptual schemes. The argument for this conclusion consists in showing that any theory one may purport to put forward as representing the true objective structure of reality will have conceptually relative counterparts, so that the structure represented seems to be no more objective than any other of those that other theories with which conceptual relativity obtains represent. But now, at this point, one must ask what is conceptual relativity and how we know that any theory has conceptually relative counterparts. The answer is that conceptual relativity is mutual relative interpretability and explanatory equivalence. Then, when one comes to understand how it is that any theory will have explanatorily equivalent conceptually relative counterparts, one finds that this will happen only supposing that the objective structures represented by theories in their explanatory devices do not matter, in so far as empirical adequacy is not hampered. Then, the argument can only show that there is no true objective unique structure in reality if it is assumed that what our theories tell us about the objective structure of reality should not be taken seriously. Hence, the entire train of thought seems to be begging the question.

182

José Tomás Alvarado

5. Conclusions I do not pretend to put forward these considerations as an objection to the doctrine of conceptual relativity, nor against pragmatism in general as a philosophical stance. But it seems to me that these considerations should be taken seriously. There is a systematic question here that should be addressed by anyone willing to say that objects and entities in general are possible only within a conceptual frame. The crucial premise (5) seems to get its strength from the implicit assumption that only empirical adequacy matters when coming to assess explanations and the theories under which those explanations are framed, and this assumption seems to be begging the question. I do not know exactly how could a reparation of the argument for conceptual relativity be framed. Maybe conceptual relativity is a real phenomenon (considering a true explanatory equivalence between theories) but much more difficult to find and more abstruse than the usual examples. Maybe, the defender of conceptual relativity could try to rely more explicitly on constructive empiricism in philosophy of science in order to defend premise (5). Maybe, finally, he could simply accept that his argument is weaker than he expected initially, but endorse it anyway as an overall image of the nature of the human endeavour to understand reality that taken globally makes more sense than its competitors. In all these cases, there is work ahead to do and I will find myself satisfied if I have just convinced you that such is the case.

Acknowledgements I thank specially Hilary Putnam for his comments and suggestions on this work.

Pontificia Universidad Católica de Valparaíso Instituto de Filosofía Av. El Bosque 1290 Viña del Mar, Chile e-mail: [email protected]

Conceptual Relativity and Structures of Explanation

183

REFERENCES Aristotle. Analytica Posteriora. Brody B. (1993). Towards an Aristotelian Theory of Scientific Explanation. In: Ruben (1993), pp. 113-127. Case, J. (1997). On the Right Idea of a Conceptual Scheme. Southern Journal of Philosophy 35, 1-18. Hempel, C.G. (1993). Explanation in Science and in History. In: Ruben (1993), pp. 17-41. Putnam, H. (1978). Realism and Reason. In: Meaning and the Moral Sciences, pp. 123-140. London: Routledge. Putnam, H. (1981). Reason, Truth, and History. Cambridge: Cambridge University Press. Putnam, H. (1983). Equivalence. In: Realism and Reason (Philosophical Papers, vol. III), pp. 26-45. Cambridge: Cambridge University Press. Putnam, H. (1987). The Many Faces of Realism. La Salle, IL: Open Court. Putnam, H. (1992). Reply to David Anderson. Philosophical Topics 20, 361-369. Putnam, H. (1994). Sense, Nonsense and the Senses. An Inquiry into the Powers of the Human Mind (The Dewey Lectures). The Journal of Philosophy 91, 445-517. Reprinted in: The Threefold Cord. Mind, Body and World, pp. 1-70. New York: Columbia University Press, 1999. Ruben, D.-H., ed. (1993). Explanation. Oxford: Oxford University Press. Salmon, W.C. (1993). Scientific Explanation and the Causal Structure of the World. In: Ruben (1993), pp. 78-112. Tennant, N. (1997). The Taming of the True. Oxford: Clarendon Press. Wright, C. (1993). Realism, Meaning and Truth. Oxford: Blackwell.

Celeste Cancela Silva CONCEPTUAL RELATIVITY AND SPEAKING-SENSITIVE SEMANTICS*

ABSTRACT. The main goal of this paper will be to analyze the relationship between the phenomenon of conceptual relativity and the speaking-sensitive semantics. My paper will be divided into three parts. In the first part I will present Putnam’s the phenomenon of conceptual relativity, in the second part I will exhibit the main ideas that characterize the speaking-sensitive semantics defended by Putnam and, finally, in the last part I will connect Putnam’s semantics with the doctrine of conceptual relativity.

1. What Does the Phenomenon of the Conceptual Relativity Consist In? This part of my paper will present the phenomenon of conceptual relativity, therefore it will be divided into brief sections whose purpose will be to make conceptual and linguistic clarifications related to this phenomenon. The thesis of conceptual relativism is one of the core theses of Putnam’s internal realism (1976-1982). Although Putnam’s natural realism implies the abandon of some ideas defended by Putnam himself in his internalist period, the thesis of conceptual relativity is not rejected and it is compatible with the kind of epistemology assumed by Putnam in his most recent period (natural realism). According to Putnam “one great philosopher who described at least a special case of the phenomenon of conceptual relativity was Kant” (Putnam 2004, p. 33). Putnam adds:

*

This paper has been carried out with a Postgraduate Grant from the Fundación Caja Madrid.

In: M.U. Rivas Monroy, C. Cancela Silva, and C. Martínez Vidal (eds.), Following Putnam’s Trail: On Realism and Other Issues (Pozna Studies in the Philosophy of the Sciences and the Humanities, vol. 95), pp. 185-201. Amsterdam/New York, NY: Rodopi, 2008.

186

Celeste Cancela Silva

I am thinking of the Second Antinomy in the Dialectic of the Critique of Pure Reason [. . .]. That “antinomy,” or rather the part that will concern us, has to do with the question whether points are genuine individuals, of which space consists (in which case Kant said they would be “simples”), or whether they are “mere limits.” (Putnam 2004, pp. 33-34)

Even though the phenomenon of conceptual relativity is central to The Many Faces of Realism, Hilary Putnam had already defined it in his article “Equivalence” (Putnam 1983a) calling it “cognitive equivalence” instead of “conceptual relativity.” Actually, Putnam’s best defense of conceptual relativity is in a lecture entitled “A Defense of Conceptual Relativity” that he gave at Hermes Lectures (University of Perugia) and published in his latest book Ethics without Ontology (Putnam 2004, pp. 33-51). 1.1. Two Different Uses of the Term ‘Conceptual Relativity’ in Putnam’s Writings Jennifer Case in her paper “On the Right Idea of a Conceptual Scheme” (Case 1997, pp. 1-19) points out that Putnam uses the term ‘conceptual relativity’ to refer to two different things.1 We can see this if we read the next texts of Putnam’s, where he describes or explains what he understands by this term: (1)

(2)

The doctrine of conceptual relativity, in brief, is that while there is an aspect of conventionality and an aspect of fact in everything we say that is true, we fall into hopeless philosophical error if we commit a “fallacy of division” and conclude that there must be a part of the truth that is the “conventional part” and a part that is the “factual part” (Putnam 1990, p. x). A corollary of my conceptual relativity – and a controversial one – is the doctrine that two statements which are incompatible at face value can sometimes both be true (and the incompatibility cannot be explained away by saying that the statements have “a different meaning” in the schemes to which they respectively belong) (Putnam 1990, p. x).

1 On one hand, the term ‘conceptual relativity’ is sometimes use to refer to the interpenetration of fact and convention while, on other occasions, it refers to what Putnam calls the phenomenon of conceptual relativity. Although the phenomenon of conceptual relativity has some connection with the interpenetration of fact and convention, they are not the same thing.

Conceptual Relativity and Speaking-Sensitive Semantics

(3)

187

There are ways of describing what are (in some way) the “same facts” which are (in some way) “equivalent” but also (in some way) “incompatible” (Putnam 1987, p. 29).

1.1.1. The Interpenetration of Fact and Convention In text (1) Putnam uses the term ‘conceptual relativity’ to refer to the doctrine of the interpenetration of fact and convention. Putnam presents this doctrine in his article “Convention: A Theme in Philosophy” (Putnam 1983b, pp. 170-183). The following quotes in which this doctrine is described belong to that article: Kant already taught us that the whole idea of comparing our conceptual system with a world of things-in-themselves (which Kant did accept, as a sort of postulate of reason) to see if the conceptual system “copies” the unconceptualized reality is incoherent. The idea of comparing our conceptual system with a conceptual system which is “built in” to the world, a unique “right way” in which the world can be seen, is also incoherent. The pluralists are surely right: there are many right ways of representing the world, an even more wrong ways, as every artist knows. (p. 177) We don’t wish to suggest that knowledge is merely made up by the human mind; that would be an idealism beyond all idealisms. So we say that there is a “factual component” to knowledge. We don’t, or at least we shouldn’t, want to say that knowledge is simply a “copy” of mindindependent reality or an approximation to the “one true theory.” So we say that knowledge has a “conventional component.” But this talk of “components” lands us in the soup. For to speak of two components is to suggest that we should be able, in principle at least, to factor our knowledge into parts; that we should be able to say which part is factual and which part is conventional. But to do this would precisely be to recover the idea that there is a “world in itself”; for that is what a statable “factual component,” freed of all “conventional components,” would amount to. A better way to describe the situation would be to say that our knowledge, or any piece of it, is conventional relative to certain alternatives and factual relative to certain others. (p. 178) Everything we say is conventional in the sense that we might have said something else, perhaps something verbally incompatible; and everything we say is factual in the sense that we could not have said just anything else. (p. 179)

The doctrine of the interpenetration of fact and convention may be summarized in Putnam’s slogan: “the mind and the world jointly make up

188

Celeste Cancela Silva

the mind and the world” (Putnam 1981, p. xi). Since my interest in this section is clarifying the phenomenon of conceptual relativity, I will not stop any longer in the doctrine of the interpenetration of fact and convention. 1.1.2. The Phenomenon of Conceptual Relativity In texts (2) and (3), Putnam uses the term ‘conceptual relativity’ to refer to the phenomenon of conceptual relativity. Text (3) belongs to The Many Faces of Realism, while, surprisingly, text (2) is the continuation of text (1). Therefore, we can observe that Putnam himself mixes both uses, i.e., in the same paragraph the expression ‘conceptual relativity’ refers to the doctrine of interpenetration of fact and convention and to the phenomenon of conceptual relativity. The result of that can be a certain degree of confusion, such as the interpretation that both theses are equivalent, which is not correct. According to Putnam the phenomenon of conceptual relativity understood as “cognitive equivalence” means that: Two formalized or at least mathematically formulated theories or scientific descriptions are instances of conceptual relativity in this sense if (1) they at least look incompatible; indeed, if we simply conjoin them without taking account of the different ways the terms are used, we get a contradiction at once. (E.g., if we conjoin a theory in which points are sets of convergent spheres, or equivalence classes of such sets, with a theory in which spheres are sets of point or mereological sums of points, and we assume the Axiom of Foundation of set theory we get an immediate contradiction [. . .] (2) There is a way of translating each statement in the “optional language” of the one theory or description into a statement in the optional language of the other which “translates” scientific explanations into equally acceptable scientific explanations. (Putnam 2001, p. 433)

In several of his works Putnam proposes the following example to illustrate what the phenomenon of conceptual relativity consists in. Let us imagine a world made up of the following individuals x 1 , x 2 , and x 3 . How many objects are there in this world? Putnam asks to imagine that for a Carnapian there will be three objects, whilst for a Polish logician there will be seven objects, 2 as the table below illustrates:

2 The Polish logician assumes the calculation of parts and wholes, invented by Leniewski; according to this calculation, for each two given particles there is an object, which is the sum. Putnam ignores the “null object,” so that for a “mereological” logic there will be seven objects.

Conceptual Relativity and Speaking-Sensitive Semantics

189

World 1

World 2

x1, x2, x3

x1, x2, x3, x1 + x2, x1 + x3, x2 + x3, x1 + x2 + x3

(A world according to Carnap)

(The same world according to Leniewski)

Putnam observes: Given a version, the question, “How many objects are there?” has an answer, namely “three” in the case of the first version (“Carnap’s World”) and “seven” (or “eight”) in the case of the second version (“The Polish Logician’s World”). Once we make clear how we are using ‘object’ (or ‘exist’), the question “How many objects exist?” has an answer that is not at all a matter of “convention” [. . .]. Our concepts may be culturally relative, but it does not follow that the truth of falsity of everything we say using those concepts is simply “decided” by the culture. But the idea that there is an Archimedean point, or an use of ‘exist’ inherent in the world itself, from which the question “How many objects really exist?” makes sense, is an illusion. If this is right, then it may be possible to see how it can be that what is in one sense the “same” world (the two versions are deeply related) can be described as colored, possessing dispositional properties, etc.) in one version and as consisting of space-time regions, particles and fields, etc., in other versions. To require that all of these versions must be reduced to a single version is to make the mistake of supposing that “Which are the real objects?” is a question that makes sense independently of our choice of concepts. (Putnam 1987, p. 20)

Before entering the explanation of the phenomenon of conceptual relativity, I am going to clarify the term ‘conceptual scheme’, since it is crucial to know how he uses that term in order to understand his thesis of conceptual relativity. Jennifer Case in her paper (1997) points out that the term ‘conceptual scheme’ in Putnam should not be identified with the same term used by Donald Davidson in his article “On the Very Idea of a Conceptual Scheme” (1984). The latter identifies conceptual schemes with natural languages. However, if we consider Putnam’s example, the difference between Carnap’s and Leniewski’s conceptual schemes cannot be attributed to the natural language used by each of them. Putnam’s example does not suggest that the world as described by Carnap or by Leniewski implies that it is necessary to make use of two different natural languages. It should be remembered that in Davidson

190

Celeste Cancela Silva

conceptual schemes are associated to natural languages, so the fact that we have different conceptual schemes implies a plurality of natural languages. Nevertheless, in the example given by Putnam, the fact that we have two conceptual schemes that may be employed to describe the world, Carnap’s and Leniewski’s, is not determined by the fact that the former logician uses a different natural language from that of the latter. As I have said this difference in the notion of conceptual scheme in Davidson and Putnam is dealt with by Jennifer Case in the mentioned article, where she introduces the notion of “optional language,” which enables her to draw a clear and fundamental distinction between the way in which Davidson and Putnam use the notion of “conceptual scheme.” Carrying on with the same example from Putnam, we can see how in the same natural language we can use two or more different conceptual schemes. Paraphrasing Davidson, to have a conceptual scheme is to have an optional language, therefore the schemes are different, so are the optional languages. Of course, as Case points out, “To speak the language of the Polish Logician is to employ the conceptual scheme of mereological sums, but it is not to speak Polish. A speaker of Polish may employ the Polish Logician’s conceptual scheme on one occasion and Carnap’s on another, all the while speaking Polish” (Case 1997, p. 11). The main ideas of the phenomenon of conceptual relativity can be summarized in two points: 1.

2.

Some words such as ‘object’ and ‘exist’, have more than one possible use and none of the uses is metaphysically privileged. They are open words in the sense that the way that we understand them is not fix forever and these uses always admit of extension in different ways. This does not mean, however, that we have different meanings for the expressions ‘object’ and ‘exist’, which would otherwise imply as many dictionary entries as the senses we can associate to them. The truth or falseness of some sentences depends on the conceptual scheme. A scheme in which the word ‘object’ is used in a sense different from the one in which this notion can be used in another conceptual scheme will produce one sentence containing the word to be true in relation to a scheme and false in relation to another scheme. Therefore, if we conjoin the sentences formulated in both schemes we get a contradiction.

If we were to examine the cases in which Putnam uses examples of conceptual schemes to explain his theory of conceptual relativism, we could point out (like Jennifer Case) that these conceptual schemes are

Conceptual Relativity and Speaking-Sensitive Semantics

191

theories that can be formalized into a formal language. Putnam’s theory of conceptual relativity can be expressed as follows: the truth or falseness of certain sentences is dependent on the theory (i.e., on the conceptual scheme). Thus, if two cognitively equivalent descriptions use the word ‘object’ with different senses, there will be the possibility that one sentence is true according to a description and false according to the other one, in such a manner that if we conjoin the cognitively equivalent descriptions we get a contradiction. In summary, what Putnam postulates in his theory of conceptual relativity is that there are multiple ways of describing reality that are mutually equivalent and incompatible between themselves. In the case of two equivalent descriptions being incompatible, in the sense that if we amalgamate the sentences of both we obtain a contradiction, we would be faced with a case of conceptual relativity. What renders these two equivalent descriptions incompatible is the different use that is made of certain terms in each one of them. 1.2. Putnam’s Dilemma Putnam presents in his latest book the attack directed by his critics against his phenomenon of conceptual relativity as a dilemma. The dilemma is as follows: 1. 2.

Either Putnam is talking about a mere change of meaning, or what he is saying is unintelligible, since: 2.1. If the word ‘exist’ means the same when Leniewski says “There is an object which the mereological sum of x 1 , x 2 , x 3 ” and when Carnap says “There does not exist such a thing as a mereological sum of x1 , x 2 , x 3 ,” then both statements contradict each other. 2.2. If Leniewski’s and Carnap’s previously mention statements do not contradict each other, that means that the word ‘exist’ has a different meaning in both statements.

Putnam argues that such a dilemma does not exist (Putnam 2004, pp. 40-47). With that aim he begins by clarifying the word ‘meaning’. That very word can be used in several different ways, two of which are pointed at by Putnam. One such use corresponds to Linguistics, where the meaning of a word is its dictionary definition. When we ask about the meaning of a word in this sense, we expect either its definition or a synonym.

192

Celeste Cancela Silva

But there is at least another way in which the word ‘meaning’, the one we can find in Wittgenstein. In this case asking about the meaning of a word is not any longer asking for a synonym or a definition, but asking about its use. In Putnam’s own words: [T]here is another, perhaps looser, notion of meaning made famous by Wittgenstein, in which to ask for the meaning of a word is to ask how it is used, and explanations of how a word is used may often involve technical knowledge of a kind ordinary speakers do not posses, and may be of a kind that would never apper in a lexicon or be offered as translation. (Putnam 2004, p. 41)

According to Putnam’s example of a universe formed by three individuals, the word ‘exist’, whether it is used by Carnap or by Leniewski, has the same meaning from a narrow linguistic sense, but it has a different meaning in Wittgenstein’s sense. Putnam says: In the wide sense of the term ‘meaning’, meaning as use, there is a difference in “meaning” here. But it is not trivial, because it is not the case that the person who gives this description of the Polish Logician’s language has to agree that what the Polish Logician says is true, or that the disagreement between the Carnapian logician and the Polish Logician is “only apparent.” (Putnam 2004, p. 42)

And he adds: Conceptual relativity [. . .] holds that the question as to which of these ways of using ‘exist’ (and ‘individual’, ‘object’, etc.) is right is one that the meanings of the words in the natural languages, that is, the language that we all speak and cannot avoid speaking every day, simply leaves open.. Both the set theory that developed in the nineteenth (and early twentieth) century and the mereology that Leniewski invented are what I will call optional languages (a term suggested by Jennifer Case), in the sense that one may count as a master of the (English or German or Polish . . . ) language without learning these particular sublanguages. The optional language of set theory and the optional language of mereology represent possible extensions of our ordinary ways of speaking. (Putnam 2004, p. 43)

Therefore, point 1 in the dilemma, the one which refers to the fact that Putnam in his thesis of conceptual relativity is simply pointing at a simple change in meaning, is solved when the word ‘meaning’ is clarified. If we understand ‘meaning’ in Davidson’s sense, then the word ‘exist’ used by Leniewski and the same word used by Carnap mean the same. However, the difference comes from the different use that each of these two logicians make of it. It is just in this case that it is correct to

Conceptual Relativity and Speaking-Sensitive Semantics

193

speak of a change in meaning, i.e., there is a change in meaning in the word ‘exist’ or ‘object’, or whatever other word when we use it in a different sense. What happens with point 2 in the dilemma then? Do the sentences ‘There is a mereological sum of x 1 , x 2 and x 3 ’ and ‘There is no such thing as the mereologican sum of x 1 , x 2 and x 3 ’ when the first sentence is used by Leniewski and the second one is in Carnap’s language? Actually, if we conjoin the previous sentences overlooking the different uses of the word ‘exists’ in each of them, we will obtain trivially a contradiction. But if we do not ignore that Carnap and Leniewski are using different optional languages, then according to Putnam [the statements] are not in fact contradictory, if we understand each of them as belonging to a different optional language, and recognize that the two optional languages involve the choices of incompatible conventions. What are “incompatible” are not the statements themselves, which cannot simply be conjoined, but the conventions. (Putnam 2004, p. 46)

2. Speaking-Sensitive Semantics In this part I will present the main theories of a speaking-sensitive Semantics like the one that Hilary Putnam defends in his most recent writings.3 Charles Travis in The Uses of Sense (1989) makes a difference between two conceptions of the semantics of a statement uttered in natural language. In that book Travis makes a detailed study of the thought of the second Wittgenstein centered in one of conception of Semantics that he calls “speaking-sensitive Semantics,” whose main representatives would be Wittgenstein and Austin. From Travis’ point of view, this perspective of Semantics is alternative to the one he calls the “classic conception” which he attributes to Grice. As it is well-known to those familiar with the philosophy of language of the last century, in natural language we have some expressions whose reference is context-sensitive known as “deictics” or “reference expressions” (personal pronouns, demonstrative pronouns, adverbs, etc.) What the presence of one of those expressions provokes in a given sentence is its truth-value to be context-sensitive (since deictics and 3 In Cancela (2001) I make a more in depth presentation of speaking-sensitive Semantics as defended by Putnam. Here I repeat some ideas already expressed in that paper.

194

Celeste Cancela Silva

reference expressions are characterized by a kind of reference that enjoys this kind of sensitivity), so that the truth-value can change from one context to another. For example, the sentence ‘I’m not an actress’ is true if uttered by Celeste Cancela, but it is false if the utterer is Nicole Kidman. Whilst in the classic semantic approaches the context-sensitivity of the truth-value of one sentence is an exceptional phenomenon, from a semantic perspective, such as the one defended by Putnam, this phenomenon, far from being exceptional becomes something common or typical of natural language. Thus, if we already knew that the truth-value of sentences such as ‘She went to Paris one month ago’, ‘Nadia is eating at the moment’, ‘This cat is wild’, etc., is context sensitive, now we will defend the truth-value of sentences such as ‘There is a lot of coffee on the table’, ‘There is a desk in the office’ or ‘The walls are blue’ is also context sensitive. What the defenders of a speaking-sensitive semantics highlight is that the truth-value of the sentences above is not only determined by the meaning of the words that those sentences are made of, but also by the context in which those sentences are uttered. We can advance in a summarized way the main ideas that characterize context- sensitive semantics in the following points: (1)

(2)

Knowing the meaning of the words that are joined to form a sentence is a necessary condition to know the content of that sentence, but this knowledge is not enough. The reason why this is so is that we may know the meaning of a sentence and still not know its content if the context in which that sentence is used is unknown to us, i.e., we cannot know the content of a sentence independently from its context. The different ways of understanding some words should not be identified with their different meanings. The meanings of words appear in dictionaries, but the fact that a given word may be understood differently depending on the context does not imply that it changes its meaning (or that it has a different meaning) form one context to another. 2.1. Putnam and Speaking-Sensitive Semantics

In order to begin the explanation of what this semantic image basically means we can concentrate for a few moments on one of the examples that Putnam uses to explain the main idea of his contextual perspective of language, the one in which he invites us to consider this sentence: ‘There is a lot of coffee on the table’ (Putnam 1999, pp. 87-88; and 2001b, p. 3).

Conceptual Relativity and Speaking-Sensitive Semantics

195

Some of the many different ways in which that sentence can be understood, according to Putnam’s text, are: (1)

(2)

(3)

There is a lot of (brewed) coffee on the table (e.g., in cups or mugs). Sample context: “There is a lot of coffee on the table. Help yourself to a cup!” There are dozens of bags full of coffee beans on a table standing near a place where a truck comes to get them and take them to a warehouse. Sample context: “There is a lot of coffee on the table. Load them in the truck.” A lot of coffee has been spilled on a table. Sample context: “There is a lot of coffee on the table. Please wipe it up” (Putnam 2001b, p. 4).

All the different ways in which the previous sentence can be understood are consistent with the meaning of the words that form it, i.e., any speaker who knows the meaning of each and every word in the sentence ‘There is a lot of coffee on the table’ will be able to associate to it one of the aforementioned senses4 without violating the meaning of the words used in the utterance. However, the way in which the sentence in the example is understood will depend not only on the meaning itself of the words, but also on the particular context in which the sentence has been used. The central idea is that the sentence ‘There is a lot of coffee on the table’ does not have a specific content if considered in isolation from its use any context of use. Therefore, it is not possible to know what is meant with this sentence if it is not considered within a specific speech act. If the truth-evaluable content of a sentence is determined not only by what its words mean, but also by the specific context in which it is being used, then its truth or falsity is also dependent on the context. We can summarize this very schematically as follows: (A)

The truth-evaluable content of a sentence depends to a great extent on the specific context in which it is used. This does not mean we can ignore the importance of the meanings of words in the truthevaluable content of the sentence, but rather that those meanings

4 From now on I will use the term ‘sense’ just as it is used in this paragraph, i.e., I will understand as the sense of a sentence or a word the different ways in which a sentence or word can be understood.

196

(B)

Celeste Cancela Silva

are not enough in themselves to determine that content. As we have seen, in depending on the context the sentence ‘There is a lot of coffee on the table’ can be used to refer to different things, so that the truth-evaluable content of this sentence will vary with the circumstances in which it is uttered. To determine the truth-value of a sentence we need to know in the first place which content is being evaluated. If the content of a sentence is partly dependent on the context, as we have seen in (A), the truth-evaluable content of a sentence will also depend on the context. The sentence ‘There is a lot of coffee on the table’, like my previous example ‘I’m not an actress’, is true under certain circumstances and false under different circumstances.

From a context-sensitive semantics, not only common nouns and adjectives are context-dependent, but verbs are also speaking-sensitive. In consequence this kind of semantics defends the contextual-sensitivity of the verb ‘to know’, meaning basically that we cannot interpret the sense of a sentence such as ‘Peter knows what gold is’ without its context. The main idea is that sentences such as that one do not possess “a unique truth-evaluable content that automatically accompanies it (like a shadow) every time it is uttered” (Putnam 2001b, pp. 8-9). In short, what the defenders of a context sensitive semantics show is that the truth-evaluable content of a sentence depends not only on the meaning of the words in that sentence, but also on its context of use, in such a way that its truth-evaluable content is context dependent. In Austin’s words: “The truth or falsity of a statement depends not merely on the meanings of words but on what act you were performing in what circumstances” (Austin 1962, p. 144). There is a very important aspect which Austin introduces on which Putnam puts special emphasis in his writings: the fact that the action of understanding the words uttered and the action of understanding the context in which those words are being uttered are closely connected, that they are interdependent. According to Putnam, it would be a mistake to think that “the relevant notion of the “context” in which a statement is made is one according to which it can be specified prior to and independently of grasping the content of the statement” (Putnam 2001b, p. 4). A few lines below he says that “We can only say that the “context” is (in the relevant sense of ‘context’) when we can say what the content of the statement is (and vice-versa)” (Putnam 2001b, p. 4).

Conceptual Relativity and Speaking-Sensitive Semantics

197

2.2. Leaving the Shadows Behind Charles Travis in Unshadowed Thought (2001) analyses thoroughly some of the most well-established ideas in analytic philosophy. In his own words such ideas are: One finds expression in the idea that such a thing as an English sentence has a truth condition – a condition under which assertive uses of it would be true; and that such a thing as an English predicate (‘is red’, say) has a satisfaction condition – a condition of an object under which the predicate would be true of that object. The other is the idea that where someone thinks something – takes something to be so – there is a particular representation, structured in a particular way out of particular component ways of representing, which represents precisely that as so, and to which the person thus relates in a particular way. That representation, the idea is, represents that person’s way of representing the thing in question to himself as so. The way he exploits that representation is the means by which he qualifies as thinking what he thus does. (Travis 2001, p. xi)

While he analyses these two ideas, Travis shows that it is easy to find behind each of them theories which are defending “shadows.”5 The main aim of Travis’ book is, precisely, the rejection of any theory proposing idea of shadow. We can see his definition of shadow in the following quote: A shadow is (or is fully identified by) a specifiable form for a representation to have – a specifiable way there is for a representation to represent – which does not admit of understandings. In the case of representations of the world as being thus and so – such a thing as stating that pigs grunt – this means that, for any way the world might be, a shadow fixes an unique answer to the question whether that is how things are according to a representation that has that shadow’s form (a possible answer being, perhaps, that there simply is no answer, fertig). (Travis 2001, p. xii)

In the previous section we have seen that the fact that some words may be understood in many different ways does not imply a change in their meaning. A crucial aspect in the semantic approach defended by Putnam and Travis is that the different ways in which a word (and therefore also a sentence) may be understood are not fixed once and forever. One of the many examples of this that Travis presents is the following one. Let us imagine that a customer asks the shop assistant in a stationery store for some blue ink. The assistant then gives her a bottle containing 5

Travis takes the expression ‘shadow’ from Wittgenstein

198

Celeste Cancela Silva

ink which looks green, but which (as she demonstrates to the customer) produces blue text when one writes with it. If, in this case, what the customer wants is a bottle of ink that produces blue text, then what the shop assistant has given the customer is what the latter had requested. We could imagine, however, different circumstances in which what counts as blue ink is ink that not only produces blue text, but that also looks blue in the bottle. In this case the shop assistant has not given the customer what she had asked for. But assuming that what counts for a thing to be considered as such is established previously and regardless of any context implies that there is a unique truth-evaluable content of a sentence which accompanies it like a shadow, with independence of its context of use. In this respect, Travis says that: “When you understand words, there is something you understand.” There certainly is. You understand what would make things as the words say. You understand how to take them. But what you understand need not be that they represented in such-and-such way, where that is a fact which entails all the facts as to of what they would be true, or false. That we understand words as we do does not require that our statements be shadowed – that they have specifiable forms which would engage with the world as a shadow would. (Travis 2001, p. 211)

As a summary we may say that Putnam’s semantic picture distances itself from the classic image mainly in his treatment of common nouns, adjectives and verbs, in the way he considers the extension of this kind of context dependent expressions. According to Travis: On one conception of that, to say how things are from such a perspective would be to describe things in terms that would say precisely the same as to how things are, no matter what the circumstances of their use – terms which depend in no way on those circumstances for their saying what they do. It would be to use ways of representing the way things are which would represent just the same as so, no matter when, or how, used [. . .] Occasion-sensivity poses a significant challenge to the idea that for a sentence of a language there are the conditions (Umstände) of which it would be true – a much more fundamental challenge than does mere indexical reference. (Travis, 2001, pp. 251-252)

The main idea that I have intended to explicit in this section has been that the different ways in which words can be used are not determined once and forever. Although it has been especially in recent years when Hilary Putnam has been involved in the defense o a semantic image like the one I have just presented, we can already find in previous works some of the main ideas of speaking-sensitive semantics. Actually, a semantic

Conceptual Relativity and Speaking-Sensitive Semantics

199

image of this kind was already implicit in his thesis of conceptual relativity. However, there exist obvious differences between a speakingsensitive semantics and his thesis of conceptual relativity, which will be dealt with in the next part.

3. Conceptual Relativity and Speaking-Sensitive Semantics In this part I will show the relationship between Hilary Putnam’s thesis of conceptual relativity and the semantic image that he has been defending recently. According to Putnam, what conceptual relativity claims is that for some concepts, but not for all, e.g., “object,” “point,” “line,” it may be the case that two conceptual schemes (which could be formalized as “theories,” with a well defined notation and axioms, in the case of his examples) contain sentences which contradict each other, if we try to simply conjoin the two schemes (e.g., “Points are sets of regions,” “Regions are sets of points”), but it is still right to treat the schemes as wholes, as cognitively equivalent (in the sense of his paper “Equivalence”). We have also seen in the first part of this paper that Putnam does not accept as correct the explanation that those words (‘point’, ‘object’, ‘exist’) have different meanings in both conceptual schemes, understanding as “meaning” the linguistic (dictionary) meanings of words. What a conceptual scheme says about the world is not simply determined by the linguistic meanings of individual words, but also by the use those words are put to. What has just been said is quite similar to context-sensitive semantics developed by Travis, and in fact it is compatible with that semantic image. However, we should not think that any statement is an example of conceptual relativity. One of the many examples that Travis uses to show speaking-sensitive semantics is the following. Let us imagine that I have in my office a door held horizontally by two blocks that I use as a desk. If I had to describe my office, the sentence ‘I have a desk in my office’ would be part of my description. However, depending on what we understand as a desk, my sentence ‘I have a desk in my office’ may be true or false. If the way in which we understand what a desk is excludes a door held by two blocks, then my sentence is false. What both Travis and Putnam insist on is that there are many different ways of understanding the same word without altering the meaning of the word itself. This is actually the only point

200

Celeste Cancela Silva

that a speaking-sensitive semantics and conceptual relativity have in common: the many ways in which some words can be understood are not fixed once and for all. Putnam’s and Travis’s semantic picture is intends to account for how the natural language works, whereas conceptual relativity is a phenomenon which takes place sometimes in scientific descriptions and mathematically formulated theories. Even if the truth-evaluable content of some sentences is context dependent, these sentences will not represent an example of conceptual relativity. Putnam says about this: It was never part of my doctrine of conceptual relativity that every statement is an example of it, and while the relativity of the content of our utterances to the context of their use is indeed something that interests me very much, it is not an example of what I called conceptual relativity. (Putnam 2001, p. 431)

Finally I would like to finish saying that the thesis of conceptual relativity should not be mistaken with conceptual pluralism. What is understood as conceptual pluralism is that reality may be described in many different ways, all of which are correct. Conceptual pluralism is a phenomenon wider than conceptual relativism, and, although it is true that the latter implies the former, the converse is not the case. Putnam explains this in the following way: The fact that the contents of a room may be partly described in two very different vocabularies cannot be an instance of conceptual relativity [. . .], because conceptual relativity always involves descriptions which are cognitively equivalent (in the sense that any phenomenon whose explanation can be given in one of the optional languages involved has a corresponding explanation in the other), but which are incompatible if taken at face value (the descriptions cannot be simply conjoined). But the fact that the contents of a room may be partly described in the terminology of fields and particles and the fact that it may be partly described by saying that there is a chair in front of a desk are not in any way “incompatible,” not even “at face value”: the statements “the room may be partly described by saying there is a chair in front of a desk” and “the room may be partly described as consisting of fields and particles” don’t even sound “incompatible.” And they are not cognitively equivalent. (Putnam 2004, p. 48)

Conceptual Relativity and Speaking-Sensitive Semantics

201

Acknowledgements I am very grateful to Hilary Putnam for comments on a previous version and help.

Universidad de Santiago de Compostela Dept.de Lógica y Filosofía Moral Praza de Mazarelos s/n 15.782 Santiago de Compostela, Spain e-mail: [email protected]

REFERENCES Austin, J. (1962). How to Do Things with Words. London: Oxford University Press. Cancela, C. (2001). Semántica sensible al habla después de Wittgenstein. In: J.M. Sagüillo, J.L. Falguera and C. Martínez, (eds.), Formal Theories and Empirical Theories (Proceedings), pp. 557-569. Santiago de Compostela: Universidad de Santiago de Compostela. Case, J. (1997). On the Right Idea of a Conceptual Scheme. The Southern Journal of Philosophy 35, 1-19. Davidson, D. (1984). On the Very Idea of a Conceptual Scheme. In: Inquiries into Truth and Interpretation, pp. 183-198. Oxford: Clarendon Press. Putnam, H. (1981). Reason, Truth and History. Cambridge: Cambridge University Press. Putnam, H. (1983a). Equivalence. In: (1983c), pp. 26-45. Putnam, H. (1983b). Convention: A Theme in Philosophy. In: (1983c). pp. 170-183. Putnam, H. (1983c). Realism and Reason (Philosophical Papers, Vol. III). Cambridge: Cambridge University Press. Putnam, H. (1987). The Many Faces of Realism. LaSalle, IL: Open Court. Putnam, H. (1990). Realism with a Human Face. Edited by James Conant. Cambridge, MA: Harvard University Press. Putnam, H. (1999). The Threefold Cord: Mind, Body and World. New York: Columbia University Press. Putnam, H. (2001). Reply to Jennifer Case. Revue Internationale de Philosophie 55, 431-438. Putnam, H. (2001b). Skepticism, Stroud and the Contextuality of Knowledge. Philosophical Explorations 4 (1), 2-16. Putnam, H. (2004). Ethics without Ontology. Cambridge, MA: Harvard University Press. Travis, Ch. (1989). The Uses of Sense. Oxford: Clarendon Press. Travis, Ch. (2001). Unshadowed Thought. Cambridge, MA: Harvard University Press.

Matti Eklund PUTNAM ON ONTOLOGY

ABSTRACT. I here critically discuss Hilary Putnam’s views on ontology, especially as recently expounded in his Ethics without Ontology. In particular, I discuss Putnam’s thesis of conceptual relativity and his criticism of the thesis that objectivity requires objects. Although I think that much of what Putnam says is important, and that there are important elements of truth to it, my points will largely be negative. Along the way I will discuss Putnam’s and Wittgenstein’s views in the philosophy of mathematics, and I will compare Putnam’s metaontological views with those of William James, Michael Dummett, Crispin Wright and Simon Blackburn.

1. Introduction In this paper, I will discuss, and to some extent criticize, Hilary Putnam’s views on ontology, recently summarized in his Ethics without Ontology (2004). I will start out with a critical discussion of the thesis of conceptual relativity. Then I will turn to what is the main issue in the book: Putnam’s criticism of the focus on ontological matters in recent philosophical discussions of mathematics and ethics.

2. Conceptual Relativity In many writings from the 1980s and onwards, Putnam has defended a view he calls the thesis of conceptual relativity. Here is a statement of the view from his most recent book (2004). Putnam considers the specific case of the existence of mereological sums, but what he says about this is supposed to generalize. (How far is an issue that I will return to later.) [. . .] to ask whether mereological sums really exist would be stupid. It is, in my view, a matter of convention whether we say mereological sums

In: M.U. Rivas Monroy, C. Cancela Silva, and C. Martínez Vidal (eds.), Following Putnam’s Trail: On Realism and Other Issues (Pozna Studies in the Philosophy of the Sciences and the Humanities, vol. 95), pp. 203-222. Amsterdam/New York, NY: Rodopi, 2008.

204

Matti Eklund

exist or not [. . .]. How can the question whether something exists be a matter of convention? The answer, I suggest, is this: what logicians call “the existential quantifier,” the symbol ‘(x)’, and its ordinary language counterparts, the expressions ‘there are’, ‘there exist’ and ‘there exists a’, ‘some’, etc., do not have a single absolutely precise use but a whole family of uses. (Putnam 2004, p. 37)

To my mind, the thesis is somewhat obscure. I will later turn to what exactly the thesis comes to. But let me first give an argument to the effect that, however exactly we should conceive of the thesis, it cannot be true while saying something significant. The idea is that ‘exists’ and what we may call other ontological expressions of English (‘there are’, ‘object’, ‘some’, . . . ) are somehow indeterminate in meaning. This would appear to mean that there are two possible languages English 1 and English 2 we could speak, with the ontological expressions of English being relatively precisified in one way in English1 and in another way in English2 .1 To stick with the example Putnam employs, we can imagine that in English 1 ‘there are mereological sums’ comes out true, and in English 2 ‘there are mereological sums’ comes out false. I think we can reduce this claim to absurdity. Here it goes. In English1 there can be a singular term t, purporting to refer to some mereological sum, such that there are some true atomic sentences of the form F(t) of English 1 . Now, what should be said about the sentence F(t) of English1 in English2 (or in English, for that matter)? It seems clear that the correct thing to say is that it is true. (Indeed, I just said in English that it is true.) But for an atomic sentence, of any language, to be true, the singular terms in that sentence must refer.2 This is a fact we can surely give expression to in English and in English2. So we can conclude in English and in English2 that t refers. But for t to refer, there must be a referent for t. In English 2 we can conclude that t refers. This, in conjunction with the fact that the referent of t, if any, is a mereological

1 Where the ‘relatively’ is included to ward off the objection that it may be in principle impossible completely to precisify the ontological expressions. Expression e relatively precisifies expression e only if e is more precise than e is. (E.g., every precisification of e is a precisification of e but not vice versa.) 2 This may be argued to be an overgeneralization. Some theorists would hold that although Vulcan does not exist, ‘Vulcan=Vulcan’ and ‘Vulcan is a planet’ are true sentences. Roughly, on these views, when the referent of t does not exist, t can still occur in some true atomic sentences: namely those that ascribe certain essential or core properties to the referent of t. Even if this is correct, the argument in the main text is not impugned: just focus on sentences not belonging to this special class.

Putnam on Ontology

205

sum contradicts the supposition that in English2 ‘there are mereological sums’ is not true. Hence the thesis of conceptual relativity is false. In his recent (2004), Putnam himself brings up objections similar in spirit to one I have just stated. Specifically, he considers an objection against his position that takes the form of a dilemma: either (say) the seemingly contradictory statements of someone (Putnam’s “Polish Logician”) who says “there are mereological sums” and someone (Putnam’s “Carnap”) who says “there are no mereological sums” really contradict each other or they do not. If the statements are genuinely contradictory there is no way in which their respective statements can be jointly true, and if the statements are not genuinely contradictory then the case can simply be described as one where one of them is using the quantifier less inclusively. (The second horn can be seen as the one I have been focusing on.) Putnam’s response to the dilemma confuses me. First, he outlines the different senses in which Carnap and the Polish logician mean the same or different things by the words in their respective ontological vocabularies (Putnam 2004, pp. 40-45). There may well be different ways of talking about sameness of meaning, and disentangling what different things such talk can amount to might well be useful. But here it is not immediately relevant: for the potential problem is that whatever we say about whether they mean the same, the example fails to support the thesis of conceptual relativity. Later Putnam does turn explicitly to the question of whether Carnap and the Polish logician contradict each other, and says that they do not – but fails to ddress the question of why this does not just impale him on the second horn of the dilemma (Putnam 2004, pp. 45-47). In several places in (2004), Putnam simply makes fun of opposition to his thesis of conceptual relativity. For instance, he says, once we assume that there is, somehow fixed in advance, a single “real,” and single “literal” sense of ‘exist’ – and, by the way, a single “literal” sense of ‘identity’ – one which is cast in marble, and cannot be either contracted or expanded without defiling the statue of the god, we are already wandering in Cloud Cuckoo Land. (Putnam 2004, pp. 84f )

One aspect of the fun-making is quite obviously misconceived. Putnam talks as if his opponent assumed that the symbol ‘exist’ independently of its use had a single literal sense, and of course that is not right. What the opponent of the thesis of conceptual relativity (when this thesis is formulated as a semantic indeterminacy thesis – and we will later see that this way of conceiving of the thesis may not be all that happy) wants to claim is rather that given its use, the sense of ‘exist’ is uniquely determined. A less polemical point of Putnam’s would be that all, or

206

Matti Eklund

virtually all, words of our language are somehow semantically indeterminate, as he claims that ontological expressions like ‘exist’ are, and so the opponent of conceptual relativity unmotivatedly claims that ontological expressions are special. What my above argument shows is that ontological expressions really are special. Here is a general reflection relating to this. In another part of his critical discussion of ontology, Putnam defends what he calls pragmatic pluralism, the recognition that it is no accident that in everyday language we employ many different kinds of discourses, discourses subject to different standards and possessing different sorts of applications, with different logical and grammatical features – different “language games” in Wittgenstein’s sense – no accident because it is an illusion that there could be just one sort of language game which could be sufficient for the description of all of reality! (Putnam 2004, p. 22)

I do not think this is entirely clear. But I think we can anyway see that even if the claim is correct, there can be expressions which occur in different discourses and whose meanings are invariant across discourses: the logical expressions. Getting back to the main topic, even if the vast majority of our expressions are somehow semantically indeterminate, the logical expressions very arguably constitute an exception to the rule. And, and this is the point, those who oppose Putnam’s thesis of conceptual relativity often already for independent reasons want to assimilate ontological expressions to logical expressions. Incidentally, there are independent reasons for concern with how Putnam in (2004) conceives of the thesis of conceptual relativity. As stated, it is clearly a kind of semantic indeterminacy thesis (that ontological expressions “do not have a single absolutely precise use”).3 But there are reasons to be skeptical of the idea that this is a happy characterization of what Putnam is after. First, Putnam’s thesis of conceptual relativity is quite clearly meant as a metaphysical thesis: but how can the truth of a metaphysical claim turn on (something as shallow as) the semantic indeterminacy of some words we actually employ? Second, in order for the extension of any predicate to include some thing a under any acceptable assignment, a must exist. So, under the 3

There are several slightly different things Putnam’s formulation might come to: he might be saying that ontological expressions are ambiguous, or that they are polysemic, or that they are context-sensitive, or that they are semantically indeterminate, or he might be suggest a hybrid of some or all of these claims. For my purposes, the differences between these possibilities do not really matter. For simplicity, I will lump all these possibilities together under the heading “semantic indeterminacy.”

Putnam on Ontology

207

supposition, for reductio, that ontological expressions have different semantic values under different acceptable assignments, it turns out that the most generous assignment is the correct one. Reductio. It is perhaps instructive to compare Putnam’s earlier, more metaphysical sounding characterizations of the view. For example, in (1981) he says, ““Objects” do not exist independently of conceptual schemes.” Not that there is not a trivializing reading also of this claim. The just quoted claim occurs just after the passage In an internalist view also, signs do not intrinsically correspond to objects, independently of how those signs are employed and by whom. But a sign that is actually employed in a particular way by a particular community of users can correspond to particular objects within the conceptual scheme of those users. (Putnam 1981, p. 52, Putnam’s emphasis)

Putnam’s characterization of his view certainly sounds radical. But in light of the clarificatory remarks, does it amount to anything more than a truism? On all views, the meanings of signs depend on how these signs are employed. The notion of a conceptual scheme is notoriously slippery, but absent a characterization of the notion that determines a different reading, it is natural to hold that it is only given a particular use, within a particular “conceptual scheme,” that signs refer to the objects they refer to, and in particular that it is only within a particular conceptual scheme that ‘exist’ and ‘object’ mean what they do. In his discussion of conceptual relativity in his (2004), and also earlier in his (2001b), Putnam makes remarks on what he means by conceptual relativity that I think serve partly to obscure matters. First, he distinguishes conceptual relativity from what he calls conceptual pluralism. Conceptual pluralism, as Putnam explains it, is what is at issue when “‘the contents’ of a room” may be described “very differently by using first the vocabulary of fundamental physical theory and then again the vocabulary of tables and chairs and lamps and so on” (Putnam 2004, p. 48). This is not a matter of conceptual relativity, Putnam rightly remarks, for the two descriptions can rather unproblematically be true together. This distinction strikes me as useful. (Indeed, the upshot of my criticism of the thesis of conceptual relativity can be said to be that every case that Putnam takes to be a case of conceptual relativity is really just an instance of conceptual pluralism.). But another supposed clarification that Putnam offers is more problematic. He says, The most common misunderstandings [of the notion of conceptual relativity] are (1) that by a ‘conceptual scheme’ I meant any body of thought and talk at all, including our ordinary talk of tables and chairs;

208

Matti Eklund

and (2) that by ‘conceptual relativity’ I meant a doctrine which implies that every conceptual scheme in this sense, every body of thought and talk, has an alternative which is incompatible with it (sometimes my critics miss the qualifier – “at face value”) but equally true. According to these misunderstandings of the notion, ‘there is a computer on this desk’ is true or false depending on what “conceptual scheme” one happens to pick. But the problem with this is that, as things stand right now (I happen to be typing these words into a computer on the desk in front of me), if anyone were to use those wordt [sic] to make a false statement, then either (s)he would be talking about a different desk or a different time and place, or (s)he would be giving at least one of those words a different meaning or reference in some other way. (Putnam 2001b, p. 431)

If we understand the above quoted remarks from Putnam’s (1981) to constitute a statement of the thesis of conceptual relativity – and I think I am not alone in finding it natural to regard them as such even though Putnam does not there employ the label – then what Putnam here describes as a misunderstanding cannot justly be regarded as such. For there the view is explicitly that objects, quite generally, do not exist independently of conceptual schemes. However, given Putnam’s more recent statements of the view, for example in (2004), the remarks make more sense. For if Putnam is only talking about the semantic indeterminacy in our ontological expressions, then there is a distinction worth marking between entities that fall under “there exists” under every relative precisification of the existential quantifier and (would-be) entities that do so under some but not all ways of relatively precisifying it. The latter entities are perhaps, in an intuitive sense, optional. So if, like Putnam, we focus on the actual semantic indeterminacy of the ontological expressions, there is a distinction to be drawn, of the kind Putnam here calls attention to. But there is still an oddity. This oddity serves to bring out further what is wrong about focusing on semantic indeterminacy. I can best bring out the point by contrasting the way Putnam now appears to conceive of conceptual relativity with how Eli Hirsch views his related doctrine of quantifier variance. Hirsch’s statement of his view is that there are “many possible perspectives on “the existence of objects,” which all are adequate for describing the same facts, the same “way the world is”” (Hirsch 2004, p. 231). He is not focused on any actual semantic indeterminacy in our ontological expressions, but on the supposed possibility of there being significantly different sets of ontological expressions, all adequate for describing the way the world is. For Hirsch, the question is not whether ‘there is a computer on this desk’ is true on all ways of making our actual

Putnam on Ontology

209

ontological expressions more determinate, but rather whether there are alternatives to our actual ontological expressions, all adequate for describing the way the world is, such that if we understand ‘there exists’ one of these ways the statement comes out true and if we understand ‘there exists’ another way it comes out false. This alternative way of conceiving of conceptual relativity (or maybe: this alternative thesis) faces the same problem as that I first outlined for the thesis of conceptual relativity. But it appears preferable – and should be preferable by Putnam’s own lights – to the way that Putnam conceives of the thesis. For the way Putnam appears to conceive it, there is a significant ontological difference between the existence of tables and chairs and the existence of what we only talk about in what Putnam calls our “optional languages.” But this difference is grounded only in how we happen to employ ontological expressions.

3. “Objectivity Requires Objects” In fact, I am not sure about the significance of these critical remarks on conceptual relativity for Putnam’s overall argument concerning ontology in Putnam’s (2004). The main issue there is the following. Consider ethics and mathematics. Traditionally, analytic philosophers have regarded ethics and mathematics as somehow problematic, and have been more drawn toward antirealism about these areas than about science. Putnam thinks that the discussion of the realist credentials of ethics and mathematics has been overly focused on ontology. In particular, he thinks that antirealism about ethics is often motivated as follows. First, it is assumed that “if a claim is objectively true, then there have to be objects to which the claim “corresponds”” (2004, p. 52). (In slogan form we may say: objectivity requires objects.) Second, it is held that natural objects cannot be truth-makers for ethical claims. So ethics is objective only if there are “non-natural” objects. But then, third, we should be doubtful of the objective truth of ethical claims. Putnam’s main aim in (the first part of) the book is to criticize this line of reasoning. I think there is something importantly right about Putnam’s criticism, but also something importantly wrong about it. I want first to stress that whatever is right or wrong about what Putnam aims to show, it is not clear that there is any need to bring in the thesis of conceptual relativity. For suppose that, as I have argued, the thesis of conceptual relativity is false, or worse, unintelligible. Still one can hold that the above argument against the objective truth of ethical

210

Matti Eklund

claims just gets the cart before the horse. One can hold that the question of whether there are ethical properties – whether natural or not – is to be determined by investigating the nature of ethical discourse, our practice of making ethical claims: and not vice versa. This type of claim is for instance made by theorists in the neo-Fregean tradition, most prominently in the case of mathematics (see Dummett 1956, 1973, and 1981; Wright 1983; and Hale and Wright 2001). According to neo-Fregeans, we arrive at the conclusion that there are mathematical objects through arriving at the view that some mathematical claims are true and noting, through an investigation of the logical form of these claims, that for their truth some mathematical terms must refer. This line of argument by no means requires conceptual relativity. There are other ways too that the reasoning which Putnam criticizes can be criticized. Here is a straightforward point. One can take it to be a Moorean fact that (say) it is objectively true that torturing babies for fun is wrong. In other words one can take this to be a belief of such a standing that whenever we have a philosophical argument whose conclusion contradicts it, it is more reasonable to doubt the conclusion of this philosophical argument than it is to doubt this. This is compatible with adherence to the view that “objectivity requires objects.” All it involves is the view that sometimes beliefs about what is and is not objectively true should be given more credence than beliefs about what types of objects there can and cannot be. Putnam argues against the “objectivity requires objects” view by considering the case of logic. Logic is held to constitute a counterexample to the view. Putnam asks us to consider statements like, If all platypuses are egg-laying mammals, then it follows that anything that is not an egg-laying mammal is not a platypus. This statement is objectively true, and yet there are no objects which make it true. That is the claim. But someone who thinks that the objective truth of mathematical claims requires mathematical objects will presumably rely on the fact that some true mathematical claims – say, “there is an even prime number” – require for their truth that there are mathematical objects, on grounds of their being quantified claims. This contrasts with logic. To argue that ethical claims, if true, are ontologically committing, one will have to assume either that claims of the form “a is good” or “a is right” are ontologically committing to ontological correlates for the predicates, or focus on what is required for the truth of claims like “goodness is a virtue,” or focus on quantification over things like rights. The issues here are perhaps not as straightforward

Putnam on Ontology

211

as in the case of mathematics, but still it appears reasonable to hold that ethics is on the side of mathematics rather than logic here. So one reason for dissatisfaction with Putnam’s appeal to the case of logic is that apparently relevant features of the case do not unproblematically carry over to the cases of mathematics and ethics. Another reason for dissatisfaction is this. Someone generally attracted to the “objectivity requires objects” view might well be attracted to an ontology of facts, and to the view that for every true claim there is a fact that makes it true, even when the claim is not otherwise ontologically committing. This may be an objectionable view in all sorts of ways: but surely it is not off the map, and Putnam presents no argument against it. There must on this view be facts which make logically true claims true. But then the case of logic does not in any way suggest that there must not be ethical facts for the truth of ethical claims. There is another puzzling feature of Putnam’s discussion. Putnam starts off the relevant chapter by discussing, in the context of Wittgenstein interpretation, whether there is a distinction between those discourses that do and those that do not describe reality. The examples, from Wittgenstein, are ethical discourse and mathematical discourse: do these discourses describe reality? The discussion of this issue leads Putnam to consider the question “how there can be such a thing as a truth which is not a description of some object or objects?” (Putnam 2004, p. 55). This in turn brings him to logic: logical truths seem to be of this kind. But somewhere in this reasoning, ‘describe’ starts being used with a different meaning. The initial question about ethical and mathematical discourse is whether the apparent assertions of these discourses are genuine assertions rather than, say, mere avowals. The question is here framed in terms of whether the discourses are “descriptive.” Very arguably logical discourse is in this sense descriptive: at any rate nothing Putnam says in any way casts doubt on this claim. But when Putnam says that statements of logic are not descriptions, the issue is rather whether these statements are about some objects. I have already criticized Putnam’s reasoning about logical and ethical discourse. But if the reasoning was cogent it would primarily show something about the “objectivity requires objects” view. Putnam’s discussion suggests that it also poses difficulties for the way that realism/antirealism disputes are normally conceived, but this is misleading.

212

Matti Eklund

4. Putnam and Wittgenstein As already mentioned, Putnam appeals to Wittgenstein in his discussion (see Putnam 2004, esp. Ch. 3). To explain the relevant points, let me provide some background. There is a discussion in the literature of what Wittgenstein really holds about discourses like ethical discourse and mathematical discourse. Is he some sort of realist? Some sort of antirealist? Or a quietist? Rather than engaging directly in Wittgensteininterpretation let me briefly outline some different possible views. Realism across the board Wittgenstein holds that superficial features – declarative form, etc. – suffice to determine that the use of a particular class of sentences is descriptive, and this is all that’s required by a realist view. Selective antirealism Wittgenstein is an antirealist about (e.g.) ethics and mathematics. Quietism All we can say about declarative sentences of various discourses is that they are used in such-and-such a way. Questions about realism/antirealism cannot be raised. Motley The realism/antirealism distinction – the distinction between genuinely assertoric or genuinely descriptive discourses and those that discourses that somehow fall short of this ideal – is too coarse-grained. Instead there is a motley of distinctions to be drawn between different discourses.4 On Putnam’s interpretation – which seems reasonable although I will not get into Wittgenstein interpretation – the “motley” interpretation is the right one. On this interpretation, the right question to ask about ethical discourse or mathematical discourse is not whether this discourse really is descriptive, but what are the salient similarities and differences between this and paradigmatically descriptive discourses. For Putnam, this is not only how Wittgenstein is best interpreted: it is also the most reasonable view to take. What, then, are the claims of Putnam and Wittgenstein with respect to mathematical discourse? What are some salient similarities and differences between this discourse and, say, ordinary object discourse? 4

For discussion of these interpretations of Wittgenstein, see, e.g., Blackburn (1990) and Conant (1997).

Putnam on Ontology

213

Both in the relevant parts of Wittgenstein’s writings on mathematics – especially his lectures XXV and XXVI in (1975) – and in Putnam’s discussions I find a significant unclarity. Neither Putnam nor Wittgenstein properly distinguishes between the following two claims: (a)

(b)

Non-existence. Mathematical terms like ‘0’, ‘1’, ‘2’ do not refer to objects; there is no special class of objects that mathematical assertions are about; mathematical assertions cannot be said to correspond to reality. Non-explanation. The appearance that we somehow explain the nature of mathematical discourse by appeal to the fact that mathematical terms refer, that mathematical assertions are about mathematical objects, and that true mathematical assertions correspond to reality is illusory.

For the most part, Wittgenstein is apparently concerned to establish (b). Consider, e.g., Suppose we said first, “Mathematical propositions can be true or false.” The only clear thing about this would be that we affirm some mathematical propositions and deny others. If we then translate the words “It is true . . . ” by “A reality corresponds to . . . ” – then to say a reality corresponds to hem would say only that we affirm some mathematical propositions and deny others. We also affirm and deny propositions about physical objects. – But this is plainly not Hardy’s point. If this is all that is meant by saying that a reality corresponds to mathematical propositions, it would come to saying nothing at all, a mere truism: if we leave out the question of how corresponds, or in what sense it corresponds. (Wittgenstein 1975, p. 239)

But there are also passages where Wittgenstein appears to be concerned to establish (a): To say “A reality corresponds to “2 + 2 = 4”” is like saying “A reality corresponds to “two.”” It is like saying a reality corresponds to a rule, which would come to saying: “It is a useful rule, most useful – we couldn’t do without it for a thousand reasons, not just one.” (Wittgenstein 1975, p. 249) If you have a mathematical proposition about  0 , and you imagine you are talking about a realm of numbers, – I would reply that you aren’t as yet talking about a realm of anything, in the most important sense of ‘about’. You are only giving rules for the use of ‘ 0 ’. (Wittgenstein 1975, p. 251)

Compare too a passage from Wittgenstein (1964):

214

Matti Eklund

The comparison with alchemy suggests itself. We might speak of a kind of alchemy in mathematics. Is it the earmark of this mathematical alchemy that mathematical propositions are regarded as statements about mathematical objects, – and so mathematics as the exploration of these objects? (IV, §16)

There is a similar problem with Putnam’s discussion of Wittgenstein. Putnam says, Wittgenstein was scornful of the idea that talk of numbers or sets is analogous to talk of objects. We should not think that set theory has discovered an unimaginably large “universe” of intangible objects; this is all confusion. The difference between the use of the quantifiers here and the use of the quantifiers when we are literally generalizing about objects is too great. (Putnam 1996, p. 247) 5 [. . .] [B]ut this supposed “real question” [of what makes mathematical claims true] rests on the following picture: truth is what results when a statement “is made true by” (or “corresponds to”) something. But, while that picture fits some statements, for example, the statement that a sofa is blue may “correspond” to a particular blue sofa on a particular occasion – it doesn’t fit other familiar statements without strain. (Putnam 1996, p. 251)

Here the focus seems clearly to be on (a). But what should be most important for Putnam’s purposes is (b). After all it is the thesis that objectivity requires objects that is Putnam’s main focus: and this thesis appears only to require (b). Consider for instance Putnam’s characterization of his main point on ontology in (2004): I see the attempt to provide an Ontological explanation of the objectivity of mathematics as, in effect, an attempt to provide reasons which are not part of mathematics for the truth of mathematical statements and the attempt to provide an Ontological explanation for the objectivity of ethics as a similar attempt to provide reasons which are not part of ethics for the truth of ethical statements; and I see both attempts as deeply misguided. (p. 3)

Here what is at issue is, explicitly, the relevance of an ontological explanation. And if anything, the stance that Putnam here gives voice to is in conflict with (a). It is in the spirit of these remarks to take the question of whether there are numbers to be one settled by criteria internal to mathematics: but clearly by these criteria there are numbers.

5

Putnam goes on to support his interpretation of Wittgenstein by quoting the latter part of the just quoted passage from Wittgenstein (1964).

Putnam on Ontology

215

Numbers, moreover, are objects in the sense that is at issue: they are in the range of the first-order variables. Furthermore, recall that Putnam defends a thesis of conceptual relativity. This conceptual relativity thesis meshes well with (b). If different things can truly be said to exist under different conceptual schemes, then it seems natural further to say that facts about what entities there really are cannot serve to explain the nature of our practices. (I am saying this somewhat hesitantly as I have earlier raised doubts about what exactly the thesis of conceptual relativity comes to. But surely (b) is in the spirit of conceptual relativity.) It is harder to reconcile (a) with the thesis of conceptual relativity. Suppose the thesis of conceptual relativity is true. Then suppose further that it is common practice to take there to be numbers: in this sense our conceptual scheme licenses us to take numbers to exist. I would have thought that statements to the effect that numbers exist, made “within our conceptual scheme,” are true. Benacerraf ’s problem, and other problems that potentially confront someone who is a metaphysical realist about mathematics, would seem beside the point. For numbers to exist, it is, roughly, sufficient that the hypothesis that they do so is coherent, that it is not in conflict with the empirical facts (in the way the hypothesis that there are yetis is), and that numbers exist “according to our conceptual scheme.” Naturally the metaphysical question of what is must be the case for numbers to exist is different from the epistemic question of what knowledge of the existence of numbers requires. But it is clear that the relevant metaphysical issues are directly relevant also to the epistemology of mathematics. In light of this, it is odd that Putnam in (2001a) appeals to Benacerraf in what appears to be an attempt to establish (a).6 It may be suggested on Putnam’s behalf that he when seeking to establish (a) he is only that arguing against the view that there are numbers when this is understood in the metaphysical realist’s preferred way. But if this is Putnam’s target, the considerations on behalf of conceptual relativity, if successful, ought by themselves to be sufficient. No considerations specific to mathematics are necessary against it.

6

See Putnam (2001a, pp. 150 and 160). The discussion is confusing, for later (p. 182) Putnam makes what seems to me much more sensible remark, also from his point of his view, on Benacerraf ’s problem: that the problem arises only on a mistaken conception of what reference to mathematical objects is.

216

Matti Eklund

5. Allies? Throughout the book, Putnam writes as if he is pretty much alone in criticizing the emphasis on ontology in mainstream analytic philosophy. This is for several reasons unfortunate. First, it is far from accurate. Second, a discussion of how Putnam’s conception of the issues here relates to those conceptions that are broadly in the same camp as Putnam’s would have helped clarify just what the view amounts to. I will focus on three types of conceptions that are broadly in the same camp as that of Putnam: the neo-Fregean view of Michael Dummett, Crispin Wright and Bob Hale; the conception of realism/antirealism issues defended by Crispin Wright in (1992) and (2003) (which although related to Wright’s neo-Fregeanism is a somewhat separate idea); and Simon Blackburn’s quasi-realism. Of these theorists, neither Wright nor Dummett is ever mentioned in Putnam’s (2004); and Blackburn is brought up only to be criticized, and an unsuspecting reader might get the impression from the criticism that Blackburn’s views on the relevant issues are radically different from Putnam’s. (While Putnam has elsewhere compared his view on the nature of truth – specifically the possibility of unknowable truths – with those of Wright and Dummett, he has never directly addressed the relation between his metaontological views and the aspects of Wright’s and Dummett’s metaontological views that I will focus on.) Consider first the neo-Fregean view of Michael Dummett (see especially the early 1956, but also 1973 and 1981) and of Crispin Wright and Bob Hale (see Wright 1983, and Hale and Wright 2001). Here is Dummett justifying the claim that mathematical objects exist: If a word functions as a proper name, then it is a proper name. If we have fixed the sense of sentences in which it occurs, then we have done all there is to be done towards fixing the sense of the word. If its syntactic function is that of a proper name, then we have fixed the sense, and with it the reference, of a proper name. If we can find a true statement of identity in which the identity sign stands between the name and a phrase of the form “the x such that Fx,” then we can determine whether the name has a reference by finding out, in the ordinary way, the truth-value of the corresponding sentence of the form “There is one and only one x such that Fx.” There is no further philosophical question whether the name – i.e., every name of that kind – really stands for something or not. (1956, pp. 40f ; compare Wright 1983, pp. 13f )

Dummett is concerned to defend the existence of mathematical objects. Here there is a clear difference between Dummett and Putnam. But there

Putnam on Ontology

217

is a deeper similarity. Dummett argues that mathematical objects exist because some mathematical terms refer; and the referring mathematical terms refer because they occur in true mathematical sentences, and we arrive at whether these sentences are true “in the ordinary way.” This is in the same spirit as thesis (b) above. But it is the absolute opposite of (a). In (2004), Putnam emphasizes Kreisel’s point that “the question is not the existence of mathematical objects but the objectivity of mathematics” (Putnam 2004, p. 67). Putnam does not acknowledge Dummett in this connection – but it is arguably very much via Dummett that this remark of Kreisel’s is as well-known as it is. Dummett famously calls it “Kreisel’s dictum” (see esp. Dummett 1973, p. 228). Although Dummett in the passages considered is concerned with mathematical discourse, the stated reason for taking mathematical terms to refer to objects rely on perfectly general considerations. Dummett’s defense of thesis (b) in the case of mathematics is so general that it appears that, on pain of incoherence, he must defend something similar across the board. Truth is always in the way hinted at prior to reference. What Dummett is relying on is Frege’s context principle: only in the context of a sentence does a word have a meaning. He takes this principle to imply that, even if words have references, somehow the semantic values of sentences are logically prior. Putnam and Wittgenstein, by contrast, appear to base their defense of thesis (b) on considerations meant to be peculiar to the case of mathematics; contrasting mathematics with discourse where talk of correspondence to reality appears to have a more natural home. (Although Wittgenstein elsewhere gives voice to views similar in spirit to Frege’s context principle.) Turn now to a second type of view which is similar in spirit to that of Putnam. In (1992) – and in later works like the articles collected in (2003) – Crispin Wright outlines and defends the following conception of realism/antirealism disputes. Standard alternatives for the would-be antirealist about ethics is to opt either for non-factualism (which denies that ethical statements are truth-evaluable) and a Mackie-style error theory (which holds that no ethical properties are exemplified and hence that no – atomic – ethical statements are true). But Wright thinks there are devastating problems for both these antirealist paradigms. There is no denying that ethical statements are truth-evaluable, and that some atomic ethical statements are true. But this does not mean that the realist wins. Nor does it mean that the quietist wins. Ethical discourse, and even discourses like comic discourse, are like, say, ordinary scientific discourse in that the statements made are truth-evaluable and some are

218

Matti Eklund

true. But there are differences – differences relevant to the issue of realism – with respect to what the truth of statements belonging to these different discourses consists in, or involves. We can ask the following questions about the statements of a particular discourse: (1) Can their truth be evidence-transcendent? (2) Is their truth response-dependent? (3) Is it a priori that there is a guarantee that all differences of opinion about the truth-values of these statements can be traced to cognitive shortcoming (like that at least one of the disputants believes what she does on the basis of erroneous information, or that the conditions are unsuitable, or one of the disputants’ cognitive systems is malfunctioning)? (4) Is there something else that can be explained by the facts to which true statements of the discourse can be said to correspond, besides the truth of the statements? These questions all correspond to different “marks of realism,” to use Wright’s label. A discourse can possess some of these marks of realism without possessing them all.7 This is very close to the “motley” interpretation of Wittgenstein. There is a sense in which statements of all discourses can be said to have a descriptive and assertoric function. But within the realm of such statements there are many potentially cross-cutting realism-relevant distinctions that can be made. And on Wright’s view, the realism-relevant questions do not turn on ontology in the way that Putnam finds objectionable. The “motley” view that theorists like Putnam ascribe to Wittgenstein is often held to be in tension with the view that discourses can be usefully distinguished into the ones that realism is true of and the ones that antirealism is true of (and Putnam clearly takes there to be a tension there). But Wright’s view suggests that reconciliation may be possible. It may be that although it is too simplified to employ a black and white realism/antirealism distinction, the motley of ways in which discourses may differ from each other may include a number of realism-relevant ways. A discourse may possess all or only some marks of realism. This does not mean that questions about realism and antirealism are somehow moot. Next, Blackburn. In Putnam’s discussion, Blackburn is mentioned as one of the all too scientistic enemies of the kind of views Putnam espouses (Putnam 2004, p. 83f ). But it is far from clear that Blackburn’s quasi-realism is all that different from the kind of view favored by Putnam. Here is how Blackburn explains his quasi-realism as applied to the case of modal discourse: 7

This is a very quick summary of some very delicate issues. See Wright (1996) for Wright’s own summary of his position.

Putnam on Ontology

219

The aim is to see these propositions as constructions that stand at a needed point in our cognitive lives – they are the objects to be discussed, rejected, or improved upon when the habits, dispositions or attitudes need discussion, rejection, or improvement. Their truth corresponds to correctness in these mental states, by whichever standards they have to meet. (Blackburn [1986] 1993, p. 55)

Blackburn does not deny that the propositions belonging to modal discourse can be true and false. He is only opposing a realist construal of modal discourse in that he thinks that, in a certain way, the truth-values of modal propositions are tied to our responses. I think there are some deep unclarities with respect to just how to understand Blackburn’s position. But given as quasi-realism about a particular discourse does not involve denying that statements made within the discourse are true and false, the position appears compatible with much of what Putnam says. Specifically, note that on Blackburn’s view on modal discourse, we do not explain the truth of modal propositions by appeal to the entities which they somehow describe or correspond to: rather we turn to their role in our cognitive lives. Blackburn accordingly agrees with Putnam that the truth of statement of a particular discourse is not always to be explained by appeal to objects to which the statements correspond, or which the statements are about.8

6. Pragmatism In this section, let me briefly relate the preceding discussion to pragmatism. Putnam likes to relate his stand on the ontological issues discussed here to his affinities with that of the classic American pragmatists. How accurate is this? Through his defense of “pluralism” and “humanism,” William James is sometimes taken to have defended something similar to Putnam’s thesis of conceptual relativity, and Putnam himself stresses his debt to James, especially in (1987).9 But consider what exactly James says in the passages that underlie the attribution of this metaontological view to him. 8 Although Blackburn often focuses on special cases like moral discourse – and hence gives the impression that he holds there to be a sharp divide between such cases and “ordinary” descriptive discourse, a view Putnam would surely reject – he does also have global aspirations for his quasi-realist view. See e.g. Blackburn (1993, introduction and essay 1). 9 See e.g. Pihlström (1996, pp. 64-88) and Thayer (1968, pp. 352-357). Putnam stresses his debt to James repeatedly in the sections leading up to the argument for ontological pluralism in Putnam (1987).

220

Matti Eklund

First, reflect on the well-known passage “Common sense is better for one sphere of life, science for another, philosophic criticism for a third; but whether either be truer absolutely, Heaven only knows” (James [1907] 1946, p. 190). This does not justify ascribing to James anything like a thesis of conceptual relativity. Rather, he only expresses a kind of (principled) agnosticism about what is “truer,” together with a lack of concern regarding this issue. Or consider another well-known passage from James: We carve out groups of stars in the heavens, and call them constellations, and the stars patiently suffer us to do so, – though if they knew what we were doing, some of them might feel much surprised at the partners we had given them. We name the same constellation diversely, as Charles’s Wain, the Great Bear, or the Dipper. None of the names will be false, and one will be as true as another, for all are applicable. In all these cases we humanly make an addition to some sensible reality, and that reality tolerates the addition. All the additions “agree” with reality; they fit in, while they build it out. No one of them is false. Which may be treated as the more true, depends altogether on the human use of it [. . .]. What shall we call a thing anyhow? It seems quite arbitrary, for we carve out everything, just as we carve out constellations, to suit our human purposes. (James [1907] 1946, pp. 252-253)

There are two ways of reading this passage. On one way of reading it, James here espouses conceptual relativity. Different, and incompatible, ways of conceptually carving up reality into objects are all equally acceptable. On another way of reading the passage, however, James holds that all names are applicable – that they all refer – without saying anything about this being relative to conceptual schemes, in any interesting sense. I can let my second remark about pragmatism be briefer. Consider theses (a) and (b), about non-existence and non-explanation, respectively. I have earlier criticized Putnam for running the theses together. Now let me just add that the one of these theses with the better pragmatist credentials is (b), concerning the (supposed) futility of ontological explanations.

7. Concluding Remarks Let me conclude by briefly stressing some of the main points. I started out by criticizing Putnam’s thesis of conceptual relativity. However, as I

Putnam on Ontology

221

went on to note, even if this thesis is rejected, it is not clear that this impugns Putnam’s main aim in the discussion of ontology in (2004): to, so to speak, dethrone ontology. About this my verdict is more equivocal. I criticized Putnam for not properly distinguishing between, for example, the view that there are mathematical objects and the view that we can usefully explain the nature of mathematical discourse by appealing to what mathematical objects there are. Putnam criticizes both views. I am more sympathetic to the criticism of the latter view (although I have not here provided any substantive arguments for why we ought to be sympathetic to this criticism).

Cornell University Sage School of Philosophy 218 Goldwin Smith Hall Ithaca NY 14853, USA e-mail: [email protected]

REFERENCES Benacerraf, P. (1973). Mathematical Truth. Journal of Philosophy 70, 661-679. Blackburn, S. ([1986] 1993). Morals and Modals. In: (1993), pp. 52-74. Blackburn, S. (1990). Wittgenstein’s Irrealism. In: J. Brandt and R. Haller (eds.), Wittgenstein: Towards a Re-Evaluation, pp. 13-26. Vienna: Holder-Pickler-Tempsky. Blackburn, S. (1993). Essays in Quasi-Realism. Oxford: Oxford University Press. Conant, J. (1997). On Wittgenstein’s Philosophy of Mathematics. Proceedings of the Aristotelian Society 97, 195-222. Dummett, M. (1956). Nominalism. Philosophical Review 65, 491-505. Reprinted in: (1978), pp. 38-49. Dummett, M. (1973). The Philosophical Basis of Intuitionistic Logic. In: (1978), pp. 215-247. Dummett, M. (1978). Truth and Other Enigmas. Cambridge, MA: Harvard University Press. Hale, B. and C. Wright (2001). The Reason’s Proper Study. Oxford: Clarendon Press. Hirsch, E. (2004). Sosa’s Existential Relativism. In: J. Greco (ed.), Sosa and His Critics, pp. 124-132. Oxford: Blackwell. James, W. ([1907] 1946). Pragmatism: A New Name for Some Old Ways of Thinking. New York: Longmans, Green & Co. Putnam, H. (1981). Reason, Truth and History. Cambridge: Cambridge University Press. Putnam, H. (1996). On Wittgenstein’s Philosophy of Mathematics. Proceedings of the Aristotelian Society (Supplementary Volume) 70, 243-264. Putnam, H. (2001a). Was Wittgenstein Really an Anti-Realist about Mathematics? In: T. McCarthy and S.C. Stidd (eds.), Wittgenstein in America, pp. 140-194. Oxford: Oxford University Press.

222

Matti Eklund

Putnam, H. (2001b). Reply to Jennifer Case. Revue Internationale de Philosophie 55, 431-438. Putnam, H. (2004). Ethics without Ontology. Cambridge, MA: Harvard University Press. Wittgenstein, L. (1964). Remarks on the Foundations of Mathematics. Oxford: Basil Blackwell. Wittgenstein, L. (1975). Lectures on the Foundations of Mathematics: Cambridge, 1939. Edited by C. Diamond. Chicago and London: The University of Chicago Press. Wright, C. (1983). Frege’s Conception of Numbers as Objects. Aberdeen: Aberdeen University Press. Wright, C. (1992). Truth and Objectivity. Cambridge, MA: Harvard University Press. Wright, C. (2003). Saving the Differences: Essays on Themes from Truth and Objectivity. Cambridge, MA: Harvard University Press.

Fernando González García María Uxía Rivas Monroy THE PRAGMATIC REALISM OF HILARY PUTNAM

ABSTRACT. This paper will concentrate mainly on the analysis of some features present in Putnam’s internal realism and in his “natural realism” that he shares with the pragmatist thinkers Peirce and James. Following the middle way which Putnam tries to reach between “reactionary metaphysics” and “irresponsible relativism,” the first part of the paper deals with what is the positive insight of traditional realism, i.e., the reality of external things as independent from our mind, as it is emphasized by Putnam. It also analyzes how this idea of the mind-independent character of reality has been very well worked out in Peirce’s thought through the category of Secondness. The second part focuses on the right insight of relativism that Putnam wants to maintain: How our notion of reality changes and evolves together with our language, our life and our handling the world. The attention that H. Putnam pays to W. James is, also, especially important in relation to his way of understanding reality.

1. Introduction For H. Putnam the issue of realism is one to which he has dedicated major attention, as we can see from a close study of the developing of his work. Although Putnam has labeled his mature philosophical position about realism with quite a number of different tags according to the perspectives he wanted to emphasize in each case (‘internal realism’, ‘pragmatic realism’, ‘natural realism’), we will speak about his “pragmatic realism” as a term that summarizes and unifies the main stream of his thoughts about realism. We want to draw attention to some close connections that can be seen between Putnam’s conception of

The present work forms part of the research project HUM2006-04955 funded by the Spanish Ministry of Education and FEDER.

In: M.U. Rivas Monroy, C. Cancela Silva, and C. Martínez Vidal (eds.), Following Putnam’s Trail: On Realism and Other Issues (Pozna Studies in the Philosophy of the Sciences and the Humanities, vol. 95), pp. 223-242. Amsterdam/New York, NY: Rodopi, 2008.

224

Fernando González García and María Uxía Rivas Monroy

reality and that of the classic pragmatists, especially C.S. Peirce and W. James, such as: (a) (b)

(c)

From a general perspective: a kind of humanism, which is related with the common sense point of view. From a metaphysical point of view: the role played by externalism, that is, the contribution of the world or the idea that the world is as it is independently of the particular interests of the person describing it. From an epistemological point of view: (1)

(2)

The way in which externalism is balanced with the idea that the independence of the world does not need to be supported both by a kind of objectivity that sees description as mere copy, and truth as correspondence. The insight that there are not intermediaries between the world and us.

The two parts into which this paper is divided explore these more typically pragmatic aspects of Putnam’s thought. The first part deals with the positive perspective brought by the traditional realism, which Putnam tries to preserve and which was also elaborated by Peirce through the category of Secondness: The feature of reality as “independent of the mind.” The second part of the paper pays attention to the positive perspective brought in by relativism, and which Putnam also wants to preserve: the sense in which the notion of reality is intertwined with our language, our life and our interactive practices with everything around us. Putnam rescues very often the reasons and the images presented by James to insist on the importance of these factors. In Putnam’s conception of “the real” the balance between its independence (realism) and its dependence on the knower (relativism) is a very complex process, where he tries to escape from the menace, always present, of relativism and at the same time he is eager to preserve the more obvious insights of common sense and scientific knowledge. The result is that Putnam prevents himself from any compromise with essentialisms, substantialisms or Platonisms, linked with a unique view of the real (Uniqueness: The “God’s-Eye View”) and denying the possibility of pluralism. The new pragmatic realism of Putnam, his “natural realism,” is faithful to the ontological point of view that characterized his internal realism, for his acceptance of the independence of the world (externalism), typical of the traditional realistic position; however at the same time he distances himself from it since he does not assume the

The Pragmatic Realism of Hilary Putnam

225

epistemology usually linked with this realism: the existence of intermediaries between the knower and the world.

2. The Pragmatic “Temperament” of H. Putnam One of the clearest features that undoubtedly gives Putnam’s thoughts something more than a “family resemblance” with classical pragmatism is his “temperament” (see James 1978, pp. 11-18) when he is doing his philosophical reflection. Like James, who stated that “pragmatism may be a happy harmonizer” (1978, p. 39) of empiricism and rationalism, Putnam shows us as well this reconciliation mood between two extreme positions, both equally unsatisfactory, in his search for “a middle way between reactionary metaphysics and irresponsible relativism” (Putnam 1999b, p. 5), especially when the subject at stage is realism. Putnam’s thought is characterized by this conciliatory attempt, which seeks to preserve what is right in both perspectives, namely, traditional realism and cultural relativism, without ever assuming their radicalism. Thus, Putnam rescues from the metaphysical realism the idea that the world is independent of the mind, i.e., it is independent of the interest, beliefs, desires or rationalizations of human beings.1 The proximity to relativism is established through what Putnam sees with perfect clarity in the same way as the classical pragmatists did: That the world is always a world-for-us; we are obliged to it due to our own condition of knowers and agents; it is not possible to understand what the world is, what the world means if it is not through our epistemic and active access to it. In the dilemma of what comes first, Being or Knowledge, Putnam does not want to do without any of the two poles, he does not even want to establish what comes first and what comes later. In the background of Realism what comes first is Being,2 in that of Relativism what comes first is Knowledge. The option of either horn of this dilemma twists the answer to the question “what is reality?” to one side. But, would it not be possible to conciliate both points of view, bearing in mind that they are the two sides of the same phenomenon and of the same process which 1 “The traditional metaphysician is perfectly right to insist on the independence of reality” (Putnam 1999b, p. 8). 2 In (1988), Putnam says that the metaphysical realism wasn’t able to seduce him because of the inadequacy of its description which accepts Uniqueness, Bivalence and Correspondence; but what was having that effect on him was the idea that “the way to solve philosophical problems is to construct a better scientific picture of the world. That idea retains the ancient principle that Being is prior to Knowledge, while giving it a distinctively modern twist” (p. 107).

226

Fernando González García and María Uxía Rivas Monroy

implies, inexorably and at the same time, both the world and the subject? On the one hand, without the mediation of the subject (a reflective, acting, social, historical one) it is not possible to determine what and how the world is; but this mediation does not have to leave us with a radical skepticism that keeps us from stating that what we know is real. On the other hand, and in parallel with the twist imposed by the mediation of the subject, we have the “bedrock,” the place “where my spade is turned” as Putnam says following Wittgenstein (Putnam 1987, p. 85), the “brutal facts” (Peirce 1931-1958 [CP], 1.419), which imposes a corset over what we can say about reality, balancing the possible excesses of mediation. Kant’s lesson is clear in this respect, and it is natural for Putnam to consider Kant as the first internal realist in the history of thought.

3. Externalism as the Rejection of Objectivity as Copy As we have already said, externalism is equivalent to recognizing that the world plays a relevant role in our knowledge of it. Externalism has to do with the idea that the world is as it is independent of the particular interests of whoever is describing it. Putnam calls this role “the contribution of environment.” 3 The “natural realism” Putnam has recently defended in (1999b) seeks to make justice of what James expresses as the “natural realism of the common man,” and to accept that we perceive “external” things, or rather, “aspects of “external” reality” (Putnam 1999b, p. 10). Therefore, we can see that Putnam is carrying on with one of the features that he wants now to insist on even in a stronger way as he did before with his internal realism, namely, to take into account the point of view of common-sense. This insight takes for granted that the entities around us and with which we have active dealings are real.4 What is going on here is the acceptance, without big compromises, of the existence of a naïve 3 This idea of the contribution of the world or of the environment is still a permanent one in Putnam’s thoughts, as we can see through the developing of his work. It has started in (1975), and was going on in (1982; 1987), or even in one of his latest books (1999b), where he states: “I agree with the critic that the world is as it is independently of the interest of describers” (p. 6), but as he also points out next: “It is true that a knowledge claim is responsible to reality, and, in most cases, that means a reality independent of the speaker. But reflection on human experience suggests that neither the form of all knowledge claims nor the ways in which they are responsible to reality is fixed once and for all in advance, contrary to the assumptions of the traditional realist” (p. 7). 4 “If there is any appeal of Realism which is wholly legitimate it is the appeal to the commonsense feeling that of course there are tables and chairs” (Putnam 1987, p. 3).

The Pragmatic Realism of Hilary Putnam

227

and pre-theoretical reality that does not question the authenticity, genuineness or the very “nature” of things, but, on the contrary, assumes that there are tables and chairs, and combs and toothbrushes, and so on, which are daily experienced as such, and not, following Russell, as “theoretical constructs” or as sense data. This defense of the reality of things, proper of common-sense, is linked not only with the natural realism of James, but also with the common-sense philosophy of The Scotch School, after its rediscovering from Peirce. Peirce says that one of the main insights of Scotch’s common-sense philosophy was the statement that there are “acritical” propositions, meaning that they are indubitable, unquestionable and original beliefs (see CP 5.440-5.446). The following text from C. Hookway’s (2000) explains clearly what being acritical means and why these beliefs are taken for granted: Justification comes to a halt upon propositions which function as the “bedrock of truth” which is accepted without grounds or justification. We regard them as “the very truth” (CP 5.505): “if you absolutely cannot doubt a proposition . . . it is plain that there is no room to desire anything more” (CP 6.498). Although these claims are, in fact, fallible, they are “acritical.” [. . .]: an acritical belief is one of which we are certain, which does not issue from the kind of process of deliberation or reasoning which can be subjected to critical monitoring. We do not know why we believe these things; we cannot imagine being able to doubt them; and they have a foundational role for our practices of inquiry and justification. (p. 150) [. . .] these beliefs [those of common-sense] do concern the nature of empirical reality. They result from experience, but they are not the products of deliberate self-controlled inquiry. When asked what supports them we can only say: “Everything counts for them and nothing counts against them.” It does not occur to us that we can question them: their extreme vagueness renders them immune from straightforward falsification. (p. 190)

There are many ways of giving an account of externalism, i.e., of the role that what is independent of us plays in the cognitive process. Traditionally externalism was linked with the scientific and positive conception of philosophy, which has come to our days since the time of classical Greece and which considers the world as being there, preexisting, and where the knower discovers or takes off the veil, which covers its deepest secrets. Science or the scientific activity, as the major guarantor of knowledge, had, as its primordial function, to discover the “very” and “real” nature of things, their essence, so that any kind of relativism was not allowed. From this point of view, objective knowledge

228

Fernando González García and María Uxía Rivas Monroy

is that which “copies” or “corresponds” with the world; that which “progresses” under the accumulative discovering of new facts, and that which inexorably gets closer and closer to the Truth, ideal limit of knowledge. Therefore, in this traditional view, externalism was identified with objectivity understood as a particular correspondence with the objects and their semantic counterparts, in such a way that the description of the world would be no more than a mere copy of it. Nevertheless there is another way of understanding externalism which does not lead necessarily to understanding objectivity as a mere activity of copying and, what is more important, which does not have as a consequence the chiasmic separation of cognitive subject and cognitive object. In this alternative picture, what is prevalent is the idea that human beings and the world are not completely separated. This picture reinforces the view that human beings are beings who interact with the world, developing simultaneously a cognitive or intellectual activity as well as a practical or instrumental one. This activity has not only effects upon the subject who knows, but also upon the object to be known. Putnam’s statements that “the mind and the world jointly make up the mind and the world” (Putnam 1982, p. xi), and that “the trail of the human serpent is over all” (Putnam 1987, pp. 16, 77) not only refer directly to James, but also and more importantly support the Jamesian insistence on the fact that “you can’t weed out the human contribution” (James 1978, p. 122). We can say that the world is a world-for-us, human beings, and that it is utterly impossible to get out of the world-for-us to enter in the world-in-itself. For that reason objectivity as a copy is not the appropriate way of understanding what the world is like, because this view seeks for the knowledge of the world-in-itself, a pure unattainable fantasy, which does not care about the human being who interacts with it, even by knowing it. The world we know is the world we continuously deal with, and these dealings have obvious consequences over the knowledge we build upon it.5 These non-scientific but ordinary dealings are also part of the reality of things. They are based as much on the malleability and resistance of things – as James expressed it (see James 1978, p. 124) – as on our own interests and aims to force this malleability to the extent it can be forced. This reality proper of common-sense, reality with a small ‘r’, humanized reality in the end, is the one which pragmatism seeks to recover in order to neutralize the excesses of centuries of philosophical 5 “[T]he real insight in James’s pragmatism, the insight that “description” is never a mere copying and that we constantly add to the ways in which language can be responsible to reality” (Putnam 1999b, p. 9).

The Pragmatic Realism of Hilary Putnam

229

reflections “bewitched,” as Wittgenstein would say, for a picture of a Unique, Eternal, Objective Reality. In this picture, the main dimension is the cognitive and theoretical one, which rules over the practical and interactive one, at the same time that it encourages all kinds of dualisms and dichotomies, instead of favoring the mediations by the persons who know and manipulate this reality.6

4. Externalism and Secondness: The Independence of Reality Externalism is also related to the importance of recognizing that there is something which is independent of our thoughts and us, and that means a guarantee in front of the possible excesses of reflection and reasoning. Putnam is of that opinion when he says: “The notion that our words and life are constrained by a reality not of our own invention plays a deep role in our lives and is to be respected” (Putnam 1999b, p. 9). Peirce also paid attention to this role of something external to thought, especially in order to warrant as far as possible the scientific activity and the value of our beliefs. Thus, in “The Fixation of Belief” he stated that the fundamental hypothesis of the scientific method was the following one: There are Real things, whose characters are entirely independent of our opinions about them; those Reals affect our senses according to regular laws [. . .] we can ascertain by reasoning how things really and truly are [. . .]. The new conception here involved is that of Reality. (CP 5.384)

Even though the element of mediation is very important to understand pragmatism and it is well developed in Peirce’s thought with his category of Thirdness – the category linked to representation, thought, reason, law (the category which could open the door to relativism, or even skepticism) – and also one can find it in the internal realism of Putnam as “conceptual relativism,” we will focus here on this feature of realism which allows a certain kind of objectivity, which gives a peculiar sense to the notion of truth and which prevents us from extreme relativism. Peirce uses the category of Secondness to do this job, and Putnam the idea of externalism.7 6 “We endlessly renegotiate – and are forced to renegotiate – our notion of reality as our language and our life develop” (Putnam 1999b, p. 9). 7 Putnam lays great stress on the reality of external things as independent of our mind – a feature he always was well disposed to maintain through his different approaches to realism: from internal realism to natural realism. Thus he speaks about the contribution of the environment (1975); or about the “objective” factor of experience (1982); or when he

230

Fernando González García and María Uxía Rivas Monroy

Externalism emphasises that feature of reality, which enables us to understand reality as independent of the mind. The categorical scheme proposed by Peirce of Firstness, Secondness, and Thirdness – or quality, reaction and mediation, respectively – obtained from phenomenology and mathematics, articulates in a logical, clear, coherent and simple way the basic elements involved in the conception of reality. In our opinion, Peirce’s thought about this subject could help to get a better understanding of the role and the position played by externalism in Putnam’s philosophy,8 for Secondness prevents us from falling into the kind of “irresponsible relativism” from which Putnam tries to escape. Let us see some of the features that Peirce used to characterize Secondness mainly in “The Logic of Mathematics; An Attempt to Develop my Categories from Within” (CP 1.417-1.520). The phenomenological analysis carried out by him shows that Secondness entails real facts, which are individual events that happen here and now.9 According to Peirce “we feel that facts resist” the individual will, and for that reason facts are called “brutal” (CP 1.419). Peirce also states that “[w]henever we come to know a fact, it is by its resisting us” (CP 1.431) and this resistance is what reveals that something independent is there. For Peirce what gives reality to a fact is opposition, i.e., its being opposite to another one. Peirce attributes to facts the feature of “having brute fighting force, or self-assertion. The individual fact insists on being here irrespective of any reason” (CP 1.434). That is why Peirce calls facts “brutal,” because the thought does not play any role in Secondness, it is only the reaction of things with each other, without any mediation to make them intelligible, which should be taken into account. Peirce points out that fact, as individual or singular, is determined with respect to any states that “[t]here are ‘external facts’, an we can say what they are” (1987, p. 33); or when he points out that “we can and should insist that some facts are there to be discovered and not legislated by us” or he suggests “the hypothesis that truth does not transcend use” (1988, pp. 114, 115), and so on. He also explains this idea in an even more explicit way through statements such as: “the world isn’t a product. It’s just the world” (1990); or the more stronger one “I have, especially in recent years, tried not to state my own doctrine as a doctrine of the dependence of the way things are on the way we talk [. . .]. In any sense of ‘independent’ I can understand, whether the sky is blue is independent of the way we talk” (1994, p. 301). 8 Putnam declared in several writings that he does not share the Peircean conception of truth (see, for example, Putnam 1990, pp. vii-viii), namely, the conception of a destined or fated truth, got in the long run and convergent in the final opinion of the scientific community: nevertheless, in relation with externalism, Putnam is much more Peircean that Jamesian. 9 “Secondness is, strictly speaking, when and where it happens, and it does not have any other being” (CP 1.532).

The Pragmatic Realism of Hilary Putnam

231

quality or possibility, differing from generality or law, which is undetermined. The mathematical perspective involved in Secondness is that of the Two, the Dyad. For Peirce “the nature of fact is linked in some way with number two” (CP 1.430). On the one hand, a fact does not need of a Third for being; in the same way as the number two does not need of three in order to be. On the other hand, a fact gets its reality when it is confronted with another reality. Peirce expresses this idea as follows: The fact fights its way into existence; for it exists by virtue of the oppositions which it involves [. . .]. For just as we can only know facts by their acting upon us, and resisting our brute will [. . .], so we can only conceive a fact as gaining reality by actions against other realities. (CP 1.432)

Peirce gives us some examples that illustrates these features of the fact: A man may walk down Wall Street debating within himself the existence of an external world; but if in his brown study he jostles up against somebody who angrily draws off and knocks him down, the sceptic is unlikely to carry his scepticism so far as to doubt whether anything beside the ego was concerned in that phenomenon. (CP 1.431) One ball “acts” on the other; i.e., it makes an exertion minus the element of representation. (CP 1.532)

As we have seen Peirce explains the main characters linked with Secondness (or externalism) focusing on the lack of intelligibility. From this point of view the reality of things is completely isolated from our mind, our thoughts, or our reason; and as such they are “brute” and our relation to them is through reaction, opposition and resistance. But because the nature of facts is represented by the dyad, the apprehension of a fact, in its simple being here and now takes place for our reaction to it; we wouldn’t know that anything is there if it did not act upon us, if it did not resist “our brute will,” namely, our action without reasoning. In the “semiosis,” the process in which something functions as a sign (CP 5.484), when a relation between a “representamen,” an “object” and an “interpretant” (CP 1.541) is established, the category of Secondness corresponds to the object. Peirce distinguishes between two types of objects: the immediate object, that which is represented by the sign; and the dynamical object, that which is external to the semiosis, but which determines that the sign represents it in a certain way. Secondness corresponds to the dynamical object, and it is that way of being which is independent of the mind and of its representation, in that sense it would

232

Fernando González García and María Uxía Rivas Monroy

not need Thirdness in order to be.10 Thirdness, on the contrary, corresponds to the immediate object and it is that way of being which is not independent of representation. Representation establishes a mediation that enables us to understand or to apprehend how objects are like. It seems apparent that reality as Secondness resists arbitrary conceptualizations and exerts an “external force” to lead research not where it wants but where it will irretrievably end sooner or later. But Thirdness takes also part in reality, giving the rational element, the thought, to the Secondness. That is why reality is the “represented” opinion. Thus, at least, the guarantee of knowledge is Secondness: It points at the necessary external item needed to stop knowledge from being a mere fantasy or imagination built by the knower; and it is needed to put limits and to constrain the possible relativistic chimeras.

5. Humanism Putnam has always had a belligerent position with respect to relativism, which, oddly, does not prevent him from going back to W. James – one of the founders of Pragmatism and very often accused of relativism – in order to build his epistemology. Certainly, Putnam does not accept all of James’s philosophy, but the Jamesian attempt of epistemologically reassessing commonsense as a formula to humanizing philosophy – or rather, re-humanizing it – was utterly assumed by him. Thus Putnam has always explicitly admitted James’s influence both in his internal realism and in his natural realism. In his first presentations of functionalism, Putnam described the mental activity in computational terms. From this point of view there are not big behavioral differences between a computer and our mind, what counts in both cases is the organization, internal organization in the former, mental organization in the latter. What follows from that is that knowing is reduced to an internal process, which does not make it possible to determine the objective reference of the semantic terms. As recognized by Putnam, it was the seeking for a solution to this special relation between language and reality as it was established by functionalism, and the conviction of maintaining realism,11 what, little by 10

“Different minds may set out with the most antagonistic views, but the progress of investigation carries them by a force outside of themselves to one and the same conclusion” (CP 5.407). 11 “The problem for [. . .] myself is to show the possibility of a return to what James called natural realism [. . .]. Before I wrote “Realism and Reason” and “Models and

The Pragmatic Realism of Hilary Putnam

233

little, drove him to the pragmatists, and, specifically to James. This can be seen in Putnam’s attempt to elaborate an epistemological theory that, trying to be close to common sense, eliminates intermediaries between the knower and the object to be known. There is a keystone in the way followed by Putnam since that time: his internal realism, which drew the general lines of his epistemological thought and that allows him to reach his present position, called, following James here once more, natural realism or common sense realism (see Putnam 1999b, p. 15). In James’s attempt – with which Putnam is in agreement – to liberate our thought from all the metaphysical excesses introduced by centuries of philosophical tradition (see James 1996, pp. 194-195), he aspired to build what in his opinion could be a kind of realism, in spite of the fact that our descriptions of reality would depend to a big extent on our interests, which certainly put James dangerously on the verge of relativism (see James 1978, p. 98). Nevertheless, we can call this view realism, because, even if our descriptions of reality were always charged with a strong subjective character, and truth – and in a certain sense reality itself – could be considered as a human construction, these descriptions would be true in the whole sense of the word, i.e., these descriptions would adjust to reality and not the other way round where reality would adjust to these descriptions. James struggled on many occasions to remove that relativist impression, which emerged from his conception of knowledge.12 What is true however is that his humanism (a term which was coined to refer, among others, to the conception of truth as a construction) was displayed through the expressions of a certainly very excessive subjectivism – as a reaction to the traditional conception of truth as something fixed beforehand, once and for ever. This humanism, than, situated him in a place that for many authors, among them Putnam, is not compatible with an authentic realism, which nevertheless is necessary, even when traditional realism must be rejected. In Putnam’s words:

Reality,” in particular, I did not see either how to defend realism or how there could be any other way of understanding the relation of language to reality. It seemed to me that I had become entangled in hopeless antinomies!” (Putnam 1999b, pp. 18-19). 12 “That reality is “independent” means that there is something in every experience that escapes our arbitrary control. If it be a sensible experience it coerces our attention; if a sequence, we cannot invert it; if we compare two terms we can come to only one result. There is a push, an urgency, within our very experience, against which we are on the whole powerless, and which drives us in a direction that is the destiny of our belief ” (James 1978, p. [45]211).

234

Fernando González García and María Uxía Rivas Monroy

As will become clear, I do not, myself, side completely with James, nor do I side completely with his traditional realist critic. I agree with the critic that the world is as it is independently of the interests of describers. James’s suggestion that the world we know is to an indeterminate extent the product of our own mind is one I deplore. (Putnam 1999b, p. 6)

Therefore, if it is true that Putnam does not accept the subjectivism of the Jamesian philosophy, far from rejecting it as a whole, he recognizes the importance of many of its insights and, to a great extent, he allows himself to be influenced by those aspects of James’s humanism that, as a vindication of common sense, he considers useful to overcome the antagonism between the most radical skepticism and the most dogmatic essentialism. As a matter of fact, Putnam considers that many of the criticisms directed to James were mistaken, because he was influenced by an approach to reality and truth too rooted in the essentialist tradition (see Putnam 1999b, p. 6). The ontological compromise of classic realism is very arguable, as Putnam says, because to admit a fixed totality of objects implies to admit the very concept of object, which is in itself a very problematic one, and, besides, the developing itself of knowledge produces variations in its own conditions in such a way that a situation in a specific moment has little to do with precedent or later ones. The traditional realist, instead, is willing to hold that, without taking into account concrete situations, knowledge conditions remain always unalterable (see Putnam 1999b, pp. 7-11). That James’s humanism has deeply penetrated Putnam’s thought is shown by the fact that Putnam admits the variability of knowing conditions as well as the possibility of discussing the existence of objects with a permanent and prefixed meaning, but, at the same time, Putnam has introduced some variety into humanism in such a way that the realist compromise has become more credible than that of James. It is very probable that the key difference between James’s pragmatism and Putnam’s neo-pragmatism – as we have already said, Putnam tries to make clearer the independence of reality – had to do with the peculiar application by James of the pragmatic maxim, first stated by Peirce (see his “How to Make Our Ideas Clear,” CP 5.402), and afterwards by himself. In his version of the maxim, James manifests explicitly that the expected consequences of an object are the whole meaning of it, in such a way that if there were some practical differences

The Pragmatic Realism of Hilary Putnam

235

in these consequences we would speak actually not only of different concepts, but also of different objects.13 For James, supposing the possibility of distinguishing between objects that have similar practical consequences is absurd, because the concept of an object is no more than the sum of all the consequences to be expected from it. Even though it is true that Putnam is not speaking of concepts, but of meanings, when he argues about the actual difference between H2 O and what in a twin world of ours is called “water” and it behaves as such but is not H2 O (see Putnam 1975), he is not being pragmatically consequent with the sense in which James understands pragmatism. James says the following about those cases, about which it is possible to discuss if two notions or conceptions refer to the same thing or to a different one: The pragmatic method in such cases is to try to interpret each notion by tracing its respective practical consequences. What difference would it practically make to any one if this notion rather than that notion were true? If no practical difference whatever can be traced, then the alternatives mean practically the same thing, and all dispute is idle. Whenever a dispute is serious, we ought to be able to show some practical difference that must follow from one side or the other’s being right. (James 1978, p. 28)

If someone admits that water is H2 O, and not XYZ, it is because there are practical consequences that follow from that belief, and those consequences are different from the ones derived from the belief that water is XYZ. What is not credible from a strictly pragmatic point of view of a Jamesian type is that two objects, which have different microstructures, using a term of Putnam, behave in exactly the same way. The fact that James did not consider anything underneath the mere expected consequences of an object, as part of its conception, constitutes, according to Putnam, the gist of the well-worn Jamesian subjectivism. Nevertheless, in our opinion, when we speak of the expected effects of an object, in reference to the pragmatic maxim, we are not only speaking of its individually expected effects, but also of all practical effects we all, inevitably, expect from it. The knowledge of an object has to do, without doubt, with action – and, for that reason, it would not make any sense to speak of a knowledge which did not reflect it; but this 13

“To attain perfect clearness in our thoughts of an object, then, we need only consider what conceivable effects of a practical kind the object may involve  what sensations we are to expect from it, and what reactions we must prepare. Our conception of these effects, whether immediate or remote, is then for us the whole of our conception of the object, so far as that conception has positive significance at all” (James 1978, p. 29).

236

Fernando González García and María Uxía Rivas Monroy

action is not only concerned with an individual, but, on the contrary, it is included in the objective content, shared by everyone. That means that when we speak about the differences among objects, those cannot simply be theoretical ones. If one conceives the solar system with nine or ten planets, the difference is not only that there is one more planet, but also that there are conceivable effects to be expected from that conception – for instance, the spaceships which intended to go out of the solar system should consider the modification to be introduced in their trajectory in case they went past the tenth planet, etc. According to Putnam, it is true that reality does not dictate to us all of its possible descriptions in a closed and definitive way, but that there is always place for a linguistic interpretation, because we are the ones who add to reality the several forms in which language takes charge of it; but, in spite of this, reality is always independent and resists its being arbitrarily interpreted (see Putnam 1987, p. 36). This, as we have pointed out above, possibly approaches Putnam more to Peircean pragmatism than to James’s pragmatism, for in Peirce the realistic compromise is much more evident. It is not to be denied that the focus James is putting on in the constructive feature of the cognitive process, which is, without doubt, basic, and could incline us to a subjectivist reading of his thought; but, it is also true that to constrain everything James is saying about reality to constructivism reduces the philosophy of James to a caricature of itself. In many places James assures that reality exists independently. Why, then, should we not see behind the Jamesian statements, as the author himself declares, an independent reality which we cannot disregard because that would precisely imply invalidating any hope of knowledge? According to James: To say that our thought does not “make” this reality means pragmatically that if our own particular thought were annihilated the reality would still be there in some shape, though possibly it might be a shape that would lack something that our thought supplies. That reality is “independent” means that there is something in every experience that escapes our arbitrary control. If it be a sensible experience it coerces our attention; if a sequence, we cannot invert it; if we compare two terms we can come to only one result. There is a push, an urgency, within our very experience, against which drives us in a direction that is the destiny of our belief. (1978, p. [45]211).

Why should we not believe James when he states, as Putnam, that it is necessary to maintain a view about an independent reality as a warranty to truth?

The Pragmatic Realism of Hilary Putnam

237

6. Occam’s Razor: Why Intermediaries If We Can Do without Them? The way to the Common Sense Realism or Natural Realism that characterizes Putnam’s latest thoughts, and that James had foreshadowed in its basic points, takes place already in the internal realism of the seventies, and this journey is related by Putnam in the first part of his (1999b, pp. 3-20). According to Putnam, underneath the question of how language hooks on to the world is hidden the problem of a majority and acritical acceptation of a perception theory, according to which the immediate objects of perception are mental representations and not physical objects, and which in its main outlines is unaltered since the XVII century. This model is basically the same one for both the rationalistic view of modern philosophy and the empiricist one, which, given the general acceptance of it, has propitiated its applicability until our days, so that it is not even questioned by the present cognitive science. But now the question is: what basic, metaphysical and epistemological assumption is behind this view about perception? The answer given by Putnam leads us again to James and the humanism defended by him in many of his writings (especially in 1996). Behind the traditional models of knowledge lies the statement that there are intermediate elements between us and the world, elements which not only make cognition understandable, but also make it possible. This assumption constitutes, in Putnam’s words, a truly metaphysical and epistemological disaster, and it was James who overcame it with his humanism. The disaster lies in that the epistemic intermediaries take us away from an authentic and direct cognitive touch with the world; but what humanism proposes is just to join the object and the consciousness in a unique experience, which eliminates the artificial gap that the traditional approach to knowledge had introduced, removing, therefore, once and for ever, the problems generated by dualism (James 1996, p. 233). Putnam does not adhere in any way to the Jamesian view according to which the objective and subjective poles of knowledge constitute parts of one and the same experience and that they are made of the same substance.14 But James’s rejection of the traditionally attributed roles to 14

“[. . .] in James’s metaphysics there is, if not exactly a totality of all “objects,” still an etwas – “pure experience” – that all conceptualizations are constrained by but that cannot itself be captured by concepts. This element in James’s thinking is not one with which we should have any sympathy” (Putnam 1999b, n. 10, p. 178).

238

Fernando González García and María Uxía Rivas Monroy

object and subject and, above all, his refusal to admit the intermediary elements between them, made of James a kind of Hume who awakens Putnam from his dogmatic realistic sleep, which was heir of a philosophical tradition with respect to knowledge shared both by rationalism and empiricism. Putnam himself implicitly admits this debt when in (1999b) recognizes that the internal realism of (1982) was still dependent on the epistemological tradition of Modernity, which took for granted the epistemic intermediaries; and that was so because this tradition had still not been able to get rid of the “Cartesian cum materialist” picture of the world, according to which cognitive access to something different from our qualia was completely unconceivable (see Putnam 1999b, pp. 18-19). This is, basically, the problem considered by the functionalist theory that internal realism has not been able to solve because, anchored in this tradition, it was impossible for it to determine how knowledge, which is an internal process, can reach an objective, external reference. Putnam cannot say yet of the internal realism that it is a natural realism, because he maintains the intermediation between the two extremes that, from any of the traditions of modern philosophy, are irreconcilable with one another. This dualism is the endemic evil of Western philosophy, and that was what James tried to overcome with his Radical Empiricism, according to which both the object and the subject are part of one and the same experience in which, however, we cannot speak of reduction of one extreme to the other, as the British philosopher B. Russell pointed out when he said: [. . .] William James and the American realists have urged that there is no specific character of “mental” things, but that the things which are called mental are identical with the things which are called physical, the difference being merely one of context and arrangement [. . .]. There are [. . .] those who deny that there is a character called “mental” which is revealed in introspection. These men may be called “neutral monists,” because, while rejecting the division of the world into mind and matter, they do not say “all reality is mind,” nor yet “all reality is matter.” (Russell 1983, p. 7)

It is precisely in that point where, in our opinion, Putnam begins to disagree with James, because of his interpretation of radical empiricism, and of the constructivism inherent in it, and, in short, of his interpretation of the more radical statements of Jamesian humanism in strictly subjectivist terms. What Putnam does not accept of James is his radical proposal of the elimination of the two traditional poles of knowledge, namely, the consciousness and the object, as it is proposed in (1996).

The Pragmatic Realism of Hilary Putnam

239

Putnam wants us to consider the objects of perception as external things, and perception itself as something completely far away from what could be taken as a subjective interpretation of the data offered by reality. Putnam attempts to get a more pragmatic realism, withdrawn from traditional realism but without falling into the struggle that, according to him, confronts James with reality, because, in Putnam’s understanding of James, reality is capable of being subjectively conformed. Therefore this implies, in our opinion, that Putnam is giving an excessively conservative interpretation of the philosophy of James, which – despite the defense Putnam is given of James in some places (see Putnam 1999a, pp. 22-23), as opposed to the criticism coming from the more traditional realistic ranks, such as Russell’s 15 – awards to it some statements that, we believe, James would not dare to state. On this matter and even though we do not share the relativism of R. Rorty, we accept his argumentation (see Rorty 1982, pp. xiii-xlvii) that a defense of pragmatism does not have to fall in the temptation of seeking to answer the question whether in order to define what truth is, it has to be appealed to utilitarian criteria, and, therefore, to subjective ones. Criticism against James for being anti-objectivist could only be done if from the beginning an objectivist or essentialist approach to reality is assumed, and, in that case, the conclusion is obvious: we are asked to assume a petitio principii when supposing the original evilness of James’s doctrine. Putnam’s reading persuades us that a more generous approach and a non-relativistic interpretation of James from Putnam’s own assumptions are possible, and that many of Putnam’s statements are not so different from the ones actually made by James. Finally, there is still a question to be asked, one which we leave for Putnam’s readers: Is Putnam perhaps struggling between the choice of withdrawing definitively from a traditional understanding of reality, assuming James’s humanism and radical empiricism, and the choice of sticking, after all, to a “weak” metaphysical realism, which prevents him from giving the definitive leap to held a consequently pragmatic view of a Jamesian kind?

15

Russell acussed James of saying that the true is simply what has convenient effects. He made these critics in two of his papers collected in his (1966). These papers are: “Pragmatism” (pp. 79-111) and “William James’s Conception of Truth” (pp. 112-130). James answered to them in his paper “Two English Critics,” collected in (1996, pp. [146]312-[153]319). James’s answer is specifically directed against the second of the mentioned papers.

240

Fernando González García and María Uxía Rivas Monroy

7. Conclusion Even when the humanization of philosophy proposed by James is in general accepted by Putnam, in our opinion, the weight that externalism keeps having in his views slants the scale to Peirce. Putnam agrees with James and Dewey in their fight against the phantom of Metaphysical Realism; the pictures and arguments offered by James are appropriate to put to an end this image so condemned by Putnam, in which the knower is restricted to a mere spectator, who apprehends the nature of things as they are, but scorns the different uses they have in relation to our interest, desires or beliefs as not relevant to the very nature and being of those same things. The picture of metaphysical realism is an incomplete one to understand the reality of things, because it seems to be only interested in its scientific research, but not in our ordinary dealings with them. Philosophy has to be humanizing in order to reconcile it with common-sense and life. James was, in our opinion, the thinker who brought pragmatism to its further consequences with his doctrine of “radical empiricism,” leaving aside the frame of dilemmas and dichotomies, such as realism/idealism, rationalism/empiricism, fact/value still present in his pragmatism but trying to harmonize them. In the end, and in spite of focusing on mediation, pragmatism is still moving itself around the framework of dualisms. Dualisms are the point of reference, which are taken to establish mediation, so that, in some ways, they continue to offer the coordinates, or the background, or the general frame of pragmatism. James gave the definitive leap to solve the problem of the epistemological dualism and traditional epistemology when proposing “radical empiricism.” Putnam’s pragmatism, call it internal or natural, does not give such a definite leap as that of James. Putnam, in his internal realism, has dualisms as framework, and, even when the declared aim of his present naturalism is to put an end to them, until now he has not achieved it. Although it is true that the division fallacy must be avoided,16 Putnam has not yet elaborated a real alternative to metaphysical realism. Internal realism or natural realism are pictures that still do not offer, as Putnam 16

“The doctrine of conceptual relativity, in brief, is that while there is an aspect of conventionality and an aspect of fact in everything we say that is true, we fall into hopeless philosophical error if we commit a “fallacy of division” an conclude that there must be a part of the truth that is the “conventional part” and a part that is the “factual part”” (Putnam 1990, pp. 117-118). This fallacy reminds greatly Quine’s arguments against the empiricist dogma of the radical and sharp distinction between analytical and synthetic statements (see Quine 1961).

The Pragmatic Realism of Hilary Putnam

241

himself recognizes, a well-elaborated ontology and epistemology; they do not give a fully explanatory alternative, which will provide another theoretical framework, and which lead us to think away from dichotomies. Peirce overcame metaphysical realism through the weight given to mediation by means of the category of Thirdness and the theory of signs; but even so, the tendency to a metaphysical realism is very strong in his work, in spite of the reference to a scientific community. This similar tendency is to be seen in Putnam throughout all his attempts to safeguard the external reality. James tried also to overcome the metaphysical realist framework through a genetic conception of truth, where truth goes beyond the narrow frame of logic, and is extended to ideas, beliefs, and behaviors. Furthermore, he elaborated a monist epistemology, based on the notion of a pure experience. In spite of the fact that Putnam offers a recovery of the valuable perspectives of Peirce, James or Dewey in their criticism of the metaphysical realism and their alternatives to it, Putnam’s internal and natural realism continue to be pictures, which enable us to see what has to be taken into account in order to get a much more human view of the world (or of reality); but which still do not have by now enough explanatory power to overcome all the dualities and dichotomies. Perhaps the definitive and satisfactory explanation of realism in Putnam is still to come.

Universidad de Santiago de Compostela Facultad de Filosofía Dpto. de Lógica y Filosofía Moral Praza de Mazarelos, s/n 15782 Santiago de Compostela, Spain e-mail: [email protected] [email protected]

REFERENCES Hookway, C. (2000). Truth, Rationality and Pragmatism. Oxford: Clarendon Press. James, W. (1978). Pragmatism and the Meaning of Truth. Cambridge, MA: Harvard University Press. James, W. (1996). Essays in Radical Empiricism. Lincoln: University of Nebraska Press. Peirce, C.S. (1931-1958). Collected Papers of Charles Sanders Peirce. 8 vols. Edited by C. Hartshorne, P. Weiss, A.W. Burks. Cambridge, MA: Harvard University Press.

242

Fernando González García and María Uxía Rivas Monroy

Putnam, H. (1975). The Meaning of ‘Meaning’. In: Mind, Language, and Reality (Philosophical Papers, vol. II), pp. 215-271. Cambridge: Cambridge University Press. Putnam, H. (1982). Reason, Truth, and History. Cambridge: Cambridge University Press. Putnam, H. (1987). The Many Faces of Realism. La Salle, IL: Open Court. Putnam, H. (1988). Representation and Reality. Cambridge, MA: The MIT Press. Putnam, H. (1990). Realism with a Human Face. Cambridge, MA: Harvard University Press. Putnam, H. (1994). Words and Life. Cambridge, MA: Harvard University Press. Putnam, H. (1999a). La vigencia de William James. In: El pragmatismo. Un debate abierto, pp. 17-44. Barcelona: Gedisa. Putnam, H. (1999b). The Threefold Cord. Mind, Body, and World. New York: Columbia University Press. Quine, W.V.O. (1961). Two Dogmas of Empiricism. From a Logical Point of View. Cambridge, MA: Harvard University Press. Rorty, R. (1982). Introduction: Pragmatism and Philosophy. In: Consequences of Pragmatism: Essays, 1972-1980, pp. xiii-xvii. Minneapolis: University of Minnesota Press. Russell, B. (1966). Philosophical Essays. Londres: George Allen & Unwin Ltd. Russell, B. (1983). The Collected Papers of Bertrand Russell. Theory of Knowledge. The 1913 Manuscript, vol. 7. London and New York: Routledge.

PART IV FACT/VALUE DICHOTOMY

Ángel Manuel Faerna MORAL DISAGREEMENT AND THE “FACT/VALUE ENTANGLEMENT”

ABSTRACT. In his recent work, “The Collapse of the Fact-Value Dichotomy,” Hilary Putnam traces the history of the fact-value dichotomy from Hume to Stevenson and Logical Positivism. The aim of this historical reconstruction is to undermine the foundations of the dichotomy, showing that it is of a piece with the dichotomy – untenable, as we know now – of “analytic” and “synthetic” judgments. Putnam’s own thesis is that facts and values are “entangled” in a way that precludes any attempt to draw a sharp distinction between “value judgments” and “matters of fact.” The idea of an “entanglement of facts and values” Putnam rightly attributes to John Dewey, who – we should add – made of it the main issue in his controversy with Logical Positivism. Nevertheless, a closer inspection of the problem whose history Putnam summarizes could bring into light important aspects of it that have been neglected. It is worth reading in this connection the intercourse between Dewey and Stevenson. Secondly, it is striking that Putnam’s version of the history of the fact-value dichotomy hardly mentions the problem that caused this very dichotomy to arise in the first place: i.e., the problem of (apparently) insoluble moral disagreements. By contrast, Dewey’s attack on the dualism of fact and value can be read as an attempt to redescribe this kind of disagreements in a way that makes room for intelligent inquiry, and consequently for rational expectations of agreement. The “entanglement thesis” surely is a part of this re-description. But then it must have implications – particularly for the analysis of value-justification – which are overlooked, or by-passed, by Putnam. This leaves the question open whether Putnam and Dewey subscribe to different versions of pragmatism with regard to norms and values.

In: M.U. Rivas Monroy, C. Cancela Silva, and C. Martínez Vidal (eds.), Following Putnam’s Trail: On Realism and Other Issues (Pozna Studies in the Philosophy of the Sciences and the Humanities, vol. 95), pp. 245-264. Amsterdam/New York, NY: Rodopi, 2008.

246

Ángel Manuel Faerna

I. In his recent essay “The Collapse of the Fact-Value Dichotomy” (2002),1 Hilary Putnam traces the philosophical history of the fact/value dichotomy from Hume to Logical Positivism. The aim of this historical reconstruction is to show that the dualism of facts and values was of a piece with central theoretical commitments endorsed by the empiricist tradition: namely, the dualism between “relations of ideas” and “matters of fact” in the case of Hume, and the dualism between “analytic” and “synthetic” propositions in the case of logical positivists. Putnam argues that these commitments are no longer tenable. Hume’s epistemological analysis depended upon a sensualistic, representational interpretation of knowledge which could not survive the science’s revolution of the first half of the twentieth century (Putnam 2002, p. 21). In an effort to place empiricism on better grounds, Logical Positivism turned from epistemology to semantics. This shift leaded to the various versions of the Verifiability Theory of Meaning, none of which ever reached a satisfactory formulation; finally, the philosophical credit of the theory could not survive Quine’s energetic attack on the analytic/synthetic distinction. It seems that the collapse of these fundamental dualisms, epistemological and semantic, should have provoked the abandonment of the fact/value dichotomy, since the very meaning of its first term, ‘fact’, obtained its philosophical clarification from such dualisms (as far as ‘fact’ is a term of the standard vocabulary of empiricism). However, this has not been the case. To quote Putnam’s own words: What it has led to is a change in the nature of the arguments offered for the dichotomy. Today it is defended more and more on metaphysical grounds. At the same time, even the defenders of the dichotomy concede that the old arguments for the dichotomy were bad arguments. (Putnam 2002, p. 40, his emphasis)

The new arguments are arranged to prove that in a world described as it is in itself there would be no room for the kind of properties that ethical judgments predicate, which leaves noncognitivism and relativism as the only way to cope with the realm of value – in sharp contrast with the (ideal) possibility to achieve objective knowledge and absolute truth in the realm of fact. Not to say, these “metaphysical grounds” presuppose notions as questionable as “the absolute description of the world,” “the

1

My comments will concentrate on sections 1-2 of the essay (pp. 1-45).

Moral Disagreement and the “Fact/Value Entanglement”

247

world in itself,” or “completed physics,” which Putnam and other neopragmatists have successfully undermined in the last decades. Anyway, there must be additional reasons, apart from metaphysical puzzles, that make the fact/value dichotomy so appealing; otherwise, it would be difficult to account for its entrenchment among nonphilosophers. Putnam mentions two of them (2002, pp. 44-45): first, it is a swift expedient for evading the difficult task of trying to solve our ethical disagreements. As this may not sound very respectable as a reason, it would normally remain unspoken. The second reason seems more noble, but not necessarily sounder from a logical point of view: relativism can be (wrongly) seen by many people as the only alternative to ethnocentrism and cultural imperialism. In short, the alleged reasons, old and new, for drawing a sharp line between facts and values are, at best, non-conclusive, and, at worst, totally misconceived. This does not mean that we would be better off without the distinction, it only means that we cannot expect to reach a better understanding of the problems involved in it, problems concerning how our descriptions relate to our valuations, by keeping them apart and putting valuations out of the domain of objectivity, rational argument, and empirical testability, a domain that we take for granted in descriptions. This, I think, is the gist of the “entanglement thesis” that Putnam borrows from John Dewey: you cannot explain the activities that conform the task labeled as “describing what the facts are” without introducing a good deal of values in the picture. When the facts of the matter have to do with how the world is, those values are epistemic values; when they have to do with human action and human products, we are then confronted with ethical, political, or aesthetic values. Of course, it would be fatally misleading to conclude that there is no difference between these two kinds of values, but the difference has nothing to do with “objective” versus “subjective” concerns. The concerns are different simply because the interests displayed in the activity of investigating how the world is are different from those connected with the aims of our action (although, as far as both types of interests are human interests, in this case “difference” does not amount to “independence”). Each concern, however, requires its own standard of objectivity if it is to be pursued rationally. The entanglement of facts with values does not preclude the possibility of making objective judgments about facts; it simply underscores the pervasiveness of values whenever human interests are at stake. Consequently, unless you are ready to admit that human beings have an interest, let’s say, in ethics but not in science, you cannot use these two activities (as logical positivists did) to instantiate

248

Ángel Manuel Faerna

the fact/value dichotomy. And, in my book, nobody has ever advanced better candidates.

II. Up to this point, I have been summarizing what I think to be Putnam’s main line of argument in the essay mentioned above. If I have not misread him, I will then feel happy to say that I find no objections to oppose to it. Therefore, the comments that follow are not to be interpreted as aiming to question the central point of the argument, that is, the “entanglement thesis.” But a closer inspection of the historical narrative – or, more precisely, of some threads of it – used by Putnam to make his point 2 will bring into light important aspects of the problem that have been neglected. On these grounds, some obscurities in Putnam’s position will be revealed, as well as some important differences between his view and that advanced by Dewey. In his exposition of the metaethical theses associated with Logical Positivism, Putnam concentrates mainly on Carnap. Stevenson is mentioned several times also, but his analysis of the nature of ethical arguments is not carefully discussed. I think this is a mistake, for Stevenson, who was sympathetic with the positivist school, exhibited a genuine concern for the language of ethics, whereas Carnap and others approached the topic only incidentally. Reading Stevenson’s Ethics and Language (1960) one understands the positive role played by the fact/value dichotomy in making moral controversies more manageable and in reducing them to rational terms. For the dichotomy was intended to explain why it is so difficult to reach rational agreements when “values” are at stake, in contrast with the comparatively swift procedures we display when the disagreement involves “bare facts.” Hence Stevenson’s well-known distinction between “disagreement in belief” and “disagreement in attitude,” which could be paraphrased in terms of our dichotomy as “differences in description of facts that are relevant to a particular moral judgment” and “differences in the values attached to such facts,” respectively. It is all important to realize that Stevenson does not mean to say that moral disagreements are disagreements in attitude. 2 As Putnam himself recognizes in his Introduction to the book, “although I have criticized the fact/value dichotomy in chapters of previous books, this is the first time I have tried to examine the history of the dichotomy from David Hume to the present day” (2002, p. 2). Therefore, we will omit references to previous works of Putnam and concentrate on the new materials provided by this one.

Moral Disagreement and the “Fact/Value Entanglement”

249

Far from it, he insists that ethical controversies are characteristically dual in nature (see, for instance, Stevenson 1960, pp. 11 and 19), for they involve both beliefs and attitudes. Moreover, Stevenson claims that beliefs and attitudes do not subsist in mutual isolation (Stevenson 1960, p. 5), though we need to think of them separately in order to clarify in each particular case what the disagreement is about and what would be the best strategy in trying to reduce it. To this extent, Stevenson’s position could be phrased as follows: every moral conflict is likely to have a descriptive side, and this gives experience and reasoning a chance. Then, if we succeeded in identifying and assessing relevant facts, we could expect some progress in our efforts to reach rational agreements in ethics.3 As I have declared my sympathy for the “entanglement thesis,” I cannot say that Stevenson is right. Nevertheless, I find his intention perfectly respectable because, far from evading the difficult task of trying to solve our ethical disagreements, he presents a method to achieve this goal. Certainly, the method does not always work: agreement in belief does not entail agreement in attitude (Stevenson 1960, p. 10) (this would be Stevenson’s version of the non-derivability thesis between “is” and “ought” advanced by Hume). Thus, it is conceivable that a moral disagreement persists even when all the facts of the matter have been agreed on by both parts; in such case, experience and reasoning are no longer useful, and persuasion takes their place. In conclusion, reason has a place in ethics (and a place not necessarily small or subsidiary); but it has a limit too, as attitudes can express ultimate, incompatible differences in the “moral sense” by which different persons perceive the same facts. Now let us take a look at the way the “entanglement thesis,” as entertained by Putnam, faces the same problem. In order to illustrate how facts and values are entangled, we do not need to restrict ourselves to ethical contexts. As I said earlier, Putnam is clear in that we cannot answer even the purely descriptive question “Is this a fact of the world?” without presupposing a whole set of values which form our standards of epistemic justification, exactly in the same sense that we cannot answer the prescriptive question “Is this action right?” without presupposing the 3

It is true that Stevenson’s attention is preferentially devoted to disagreements in attitude, but this is due to the fact that historically it is that aspect of moral disagreement that has been normally overviewed (see Stevenson, p. 16). This, of course, has influenced the usual interpretation of his position as purely noncognitivist, which Putnam seems to subscribe, since he does not find it necessary for his argument to distinguish between Stevenson’s position and Carnap’s.

250

Ángel Manuel Faerna

set of values which form our standards of moral justification (Putnam 2002, pp. 30-31). But if we ask “How do we know that our standards are the right ones?,” we find out that there is no second set of meta-values to rely on for assessing the reliability of the first set (in any case, this move would lead us to an infinite regress). We know that our epistemic values are right because by using them we reach beliefs that are acceptable according to the standard those very values prescribe. Putnam’s commentary on this cannot be more explicit: To say this is not to express any sort of skepticism about the superiority of these criteria [. . .]. If this is circular justification, it is still justification enough for most of us. But it is to say that if these epistemic values do enable us to correctly describe the world (or to describe it more correctly than any alternative set of epistemic values would lead us to do), that is something we see through the lenses of those very values. It does not mean that those values admit an “external” justification. (2002, pp. 32-32, his emphasis)

It must be assumed that the same holds for justification of ethical judgments: we know that our moral values are right because by using them we arrive at courses of action that are commendable, or attain ends that are worthy, or avoid undesirable results, etc., according to the standard these very values prescribe. What Putnam seems to have in mind is that kind of situation in which members of the same community try to fix their beliefs according to a normative framework that they all share (say, the scientific community, or the community of liberal, democratic citizens). Here the practice of dealing with justifications preserves its meaning in spite of its selfconfessed circularity, for, to repeat Putnam’s words, it is still justification enough for most of us (i.e., for the persons already partaking in that common set of values). Then, saying that values do not admit an “external” justification, is the same as to say that there is no normative framework outside the community which could justify those values when they are challenged – normally by a second community with its own set of epistemic and/or moral values. Any appeal to “experience and reasoning” in order to settle the question, as if these formed a neutral jury, would mean an appeal to facts as disconnected from values; the kind of appeal that the “entanglement thesis” prevent us from doing. Regarding the question of moral disagreement, we seem to be led to the same conclusion we arrived at by Stevensonian means: reason has a place in ethics (maybe a large and fundamental one), but it also has limits. And those limits are essentially the same in both analyses: the limits imposed by a sort of “normative incommensurability” that could

Moral Disagreement and the “Fact/Value Entanglement”

251

manifest itself when persons belonging to different communities (with specific cultural, social, political, or religious traditions) adhere to different value judgments.4 However, this conclusion has obvious relativistic implications that Putnam is not willing to admit. So he invites us to “give up the very idea of a ‘rationally irresolvable’ ethical dispute” and replace it by “the idea that there is always the possibility of further discussion and further examination of any disputed issue” (Putnam 2002, p. 44). But the question is: how is this to be accomplished when each chain of justifications is moving into its own circle? How, if every justification is “internal,” is the discussion supposed to reach a point placed beyond the limits? To this problem the solution is neither to give up on the very possibility of rational discussion nor to seek an Archimedean point, an “absolute conception” outside of all contexts and problematic situations, but – as Dewey taught his whole life long – to investigate and discuss and try things out cooperatively, democratically, and above all fallibilistically. (Putnam 2002, p. 45, his emphasis)

But, again, do all these commendations mean something more than a solemn statement of our shared values as a democratic community? In my opinion, this line of discussion leads to a dead-end. And, which is more important, it is the kind of discussion that Dewey was trying to avoid when he claimed that facts and values are entangled. Indeed, Dewey’s attack on the dualism of fact and value can be read as an attempt to re-describe this kind of situation in a way that makes room for intelligent inquiry, and consequently for rational expectations of agreement. But then the “entanglement thesis” must have further implications which are overlooked, or by-passed, by Putnam. It was William James who said that, in order to be real, a conceptual difference must make a difference elsewhere. According to Stevenson, we must separate attitude from belief – and, to that extent, values from facts – if we are to keep moral disputes under rational control; according to Putnam, we simply cannot do that. But what is the difference this difference makes regarding the possibility of arriving at rational agreements when a shared framework is lacking? As we have seen, that difference is, at the very least, difficult to ascertain. Does this mean that the “entanglement thesis” does not make a real, clear-cut difference? I do not think so. But maybe it is high time we took a glance at the intercourse between Dewey and Stevenson to seek for some light. 4

Stevenson is perfectly aware of the social origin of “attitudes” and its dependence on community-patterns; see his (1960, p. 13).

252

Ángel Manuel Faerna

III. Putnam’s historical narrative may suggest that there is a gulf between Dewey’s philosophical stance and the thinking trend represented by Stevenson and Logical Positivism. The collapse of this thinking trend, that finds its origin rooted in Hume, would put us now in a position that enables us to appreciate the potentialities of the alternative approach that Dewey was trying to enforce – an approach dismissed or sidelined by those who adhered to that mainstream of philosophy. Some readers can get the impression that, in turning their eyes to Dewey’s contribution, recent philosophers like Putnam are behaving as the instrument of some kind of historical justice. But, despite the undeniable merits that Putnam and others deserve for bringing Dewey’s work back in, I would like to express a warning about the wrong image that narrative conveys. A full assessment of pragmatism’s place in the empiricist tradition is, needless to say, out of the scope of this paper. I will just state the fact that the image I was referring to is not tenable as far as Stevenson is concerned.5 It is not insignificant that Stevenson’s Ethics and Language starts with a long quotation from Dewey; nor that the quoted lines concern the desirability of keeping moral theory (that is, values) in contact with human nature (that is, facts) in order to “put an end to the impossible attempt to live in two unrelated worlds” (Stevenson 1960, p. iv, emphasis added). Besides, Chapter VIII of the book – entitled “Intrinsic and Extrinsic Value” (pp. 174-205)6 – makes an intensive use of Dewey’s ideas that are, in the first paragraph of Chapter XII – entitled “Central Aspects of John Dewey’s Analysis” (pp. 253-264) – the purpose of a separate assessment. It is also pertinent to remember that Stevenson is the author of the Introduction to Volume 5 of Dewey’s Middle Works, which contains the first edition (1908) of Dewey’s and Tufts’s Ethics (Stevenson 1978). Dewey, in return, reviewed Ethics and Language in an article entitled “Ethical Subject-Matter and Language” (Dewey 1945). At this point it is worth quoting a comment Dewey made on Stevenson’s book the same year in a letter to Horace S. Fries: “In some respects it is better than most writing on ethical method – but its ‘psychological’ so called-foundations are terrible” (1981-1991, Vol. 16, p. 470, editor’s 5 Of course, I do not mean to charge Putnam of ignoring this fact or designedly omitting it. My point is rather that his historical reconstruction induces the wrong impression that Dewey and Stevenson (or even Dewey and Logical Positivism as a whole) represented antithetic positions. 6 In the first footnote (p. 175), Stevenson acknowledges that he is making use of Dewey’s ideas throughout the chapter.

Moral Disagreement and the “Fact/Value Entanglement”

253

note). And we can take Dewey’s contribution to the International Encyclopedia of Unified Science, his (1939), as an attempt to present his views as akin to the spirit of Logical Positivism. These references do not pretend to prove that the differences were scarce or non-important, but they do prove that a dialog was ongoing; it is only natural to presume that the terms of that dialog should throw some light on our problem. To this effect, Stevenson’s commentary on Dewey in his (1978) gives important clues. (Let me mention by the way that in these pages Stevenson places Dewey’s work on ethics in that mainstream of British empiricism that arises from Hobbes, Hume and Mill [pp. xiv-xv].) There Stevenson offers a tentative reconstruction of Dewey’s position on the problem of moral disagreement (Section 5 is totally devoted to this subject), using his own categories of analysis: “Dewey used neither the term ‘disagreement in attitude’ nor any equivalent term; but like a multitude of other men he must certainly have been aware, intuitively, of the sort of disagreement to which the term refers” (p. xxv). This reconstruction is summarized by Stevenson as follows: I venture to ascribe to Dewey the following view: he recognized, implicitly, the logical possibility of scientifically unresolvable disagreements in attitude, along with the logical possibility that men, even if wholly rational, would continue to judge and act in opposition to one another rather than in accord with one another; but he went on, in the interest of making the most of scientific reasoning, to assume that such disagreements will not in fact arise. In other words, he believed (or ”willed to believe”) that men who have opposing attitudes would no longer have them, but instead would let their predominating attitudes stamp the same things with the same values, if they could complete their dramatic rehearsals. [. . .] Relative to that assumption he ascribed to ethical methodology an intersubjectivity of its own, not far distant from that of the sciences. (Stevenson 1978, p. xxvi, all emphases added, except that on ‘if ’)

This reconstruction affords the answer to another question Stevenson is willing to establish: that is, the extent to which Dewey “was prepared to compromise with a straightforward naturalism” (Stevenson 1978, p. xx) (which, in this metaethical context, means a purely descriptivist analysis of the meaning of ethical terms). In Stevenson’s reconstruction of Dewey’s view, descriptivism is avoided once the logical possibility of scientifically irresolvable disagreements is acknowledged. But, at the same time, these disagreements are discarded in fact on the basis of an unwarranted assumption, or by virtue of an act of faith: that is, that rational individuals will converge progressively toward the same

254

Ángel Manuel Faerna

attitudes.7 It is only “relative to that assumption” that Dewey can ascribe the “method of intelligence” – or the scientific method – to ethics. As far as I can see, the view that Stevenson is ascribing to Dewey does not differ, in any essential feature, from Putnam’s own contention with respect to the problem of moral disagreement: in both cases, at most we can trust that discussion, if sustained long enough, will lead to convergent value judgments. Stevenson does not object to this, for it does not contradict his own analysis. All that he needs to point out is that we are trespassing here the limits of pure analysis to enter the field of moral influence.8 Nonetheless, the “entanglement thesis” is supposed to be a contribution to the analysis of normative statements, not a normative statement in itself. So we must ask again, what is the difference that thesis makes? The situation is this: if Stevenson’s reconstruction of Dewey’s view is right, then this view is not able to question Stevenson’s analysis of moral disagreements, and, consequently, Putnam’s contention that the “entanglement thesis” represents an alternative to positivists views is wrong; but, if Stevenson’s reconstruction is wrong, then Putnam (or Putnam’s reading of Dewey) cannot be right either, for he is arriving at essentially the same conclusion regarding the problem of persistent disagreements – remember that there is no possible “external” justification for the values of a given community, and that leaves us only with the hope that communities converge in the long run. I lean towards the second option of this dilemma. I think both Stevenson and Putnam are missing something in their interpretation of Dewey. And this “something” can be disclosed by examining Dewey’s side in the dialog.

IV. In his article “Ethical Subject-Matter and Language” (1945) – a review of Ethics and Language, which had appeared the year before – Dewey charges Stevenson with having confused the function and use of ethical 7 In support of this interpretation, Stevenson mentions (1978, p. xxvii) a “postulate” that Dewey introduced into his (1891, p. 332), by virtue of which the realization of the individual and of the community are satisfied by the same conducts. Although Dewey did not mention such postulate ever since that early date, “he may have taken [it] for granted throughout his subsequent writings” (Stevenson’s emphasis). 8 This explains an earlier remark made by Stevenson in (1960), concerning an important difference between Dewey’s method and his: “Dewey, unlike the present writer, does not wish to isolate the special tasks of analysis” (p. 261), that is to say, Dewey does not separate philosophical analysis in ethics from the promotion of a certain moral attitude. On this point, Stevenson is undoubtedly right.

Moral Disagreement and the “Fact/Value Entanglement”

255

sentences with their structure and contents, thus taking “the cases in which “emotional” factors accompany the giving of reasons as if this accompaniment factor were an inherent part of the judgment” (Dewey 1945, pp. 127 and 129, his emphasis). Then the argument focuses on the notions of “sign” and “meaning,” in order to refute Stevenson’s claim that an utterance can have emotive meaning “in a sense which excludes descriptive reference” (Dewey 1945, p. 130). I will not go into these semantic topics, but concentrate instead on a second, connected criticism which has been already anticipated: that is, the psychological assumptions Stevenson makes, which Dewey described to Fries as “terrible.”9 For an expression to have emotive meaning but no reference at all, there must be something meant which is not an object “in the world.” That “something” are emotions and feelings, not as they have “a specified position in a complex situation in which occur also the things the “emotions” are at, about, or for,” but – here Dewey quotes Stevenson’s own words – as “designating an affective state that reveals its full nature to immediate introspection, without use of induction” (Dewey 1945, pp. 134-135, his emphasis). Of course, this appeal to an introspective self-revelation is totally incompatible with the psychological views Dewey had been developing his whole life long; but, what is more pregnant in the context of the dialog we are examining, it is a part of a mythology of “the mental” wholly devoid of scientific basis. Therefore, Dewey’s rejection of emotivism springs from deep scientific, psychological sources, and this fact is of paramount importance in order to understand his attack on the fact/value dichotomy, as well as to assess his positioning with respect to the empiricist-positivist tradition. Now we can see that Stevenson’s tentative reconstruction of what Dewey would have to say as to the problem of moral disagreement is hopelessly wrong. For he is wrong in assuming that Dewey would accept the “intuitive” notion of “disagreement in attitude,” if ‘attitude’ means an “emotive” response not integrally connected with the things that cause it to go off. On the other hand, if attitudes are taken, as Dewey certainly does, as a non-separable element of the “complex situation in which occur also the things” those attitudes are about, then the question whether “men, even if wholly rational, would continue to judge and act in opposition to one another rather than in accord with one another,” is not well formulated. For ‘rationality’ is here a name for the method of choosing the response that fits better with the requirements of the situation, that is, a method for reflectively establishing the right 9

See p. 252 above.

256

Ángel Manuel Faerna

judgment. Were men wholly rational, they would judge and act in accord with one another, because they would set aside their emotional preferences – that is, their unreflective impulses – and try to react as experience and reasoning advise them to do. As Dewey clearly puts it: The theoretical view about ethical sentences which is an alternative to that put forward by Stevenson is, that as far as non-cognitive, extracognitive, factors enter into the subject-matter or content of sentences purporting to be legitimately ethical, those sentences are by just that much deprived of the properties sentences should have in order to be genuinely ethical. (1945, p. 137, his emphasis) 10

As we can see, the “assumption” that Stevenson was attributing to Dewey is totally absent: only if ‘attitude’ is interpreted in terms of “unreflective impulse,” can we contemplate “the logical possibility of scientifically unresolvable disagreements in attitude”; but, of course, this is no more than a truism. As to the question of “naturalism,” I think the following statement is pretty clear also: If moral theory has any distinctive province and any important function it is, I would say, to criticize the language of the mores prevalent at a given time, or in given groups, so as to eliminate if possible this [noncognitive] factor as a component of their subject-matter; to provide in its place sound matter-of-fact or “descriptive” grounds drawn from any relevant part of the whole knowledge possessed at the time. (Dewey 1945, p. 138, his emphasis)

In Stevenson’s use of the term, Dewey is undoubtedly a naturalist, because he does not recognize other meaning for ethical terms than the descriptive one. He do recognizes that ethical language possesses nondescriptive, “quasi-imperative” functions,11 but “the point at issue is whether the facts of use and function render ethical terms and sentences not fully comparable with scientific ones as respects their subject-matter and content” (Dewey 1945, p. 137, his emphasis). Nevertheless, Dewey’s naturalism, as the statement above shows, does not deprive ethical theory of its critical role. Far from limiting itself to validate whatever the current usage of moral language may sanction, moral theory – as any other part of philosophy, according to Dewey – must provide the methods for correcting and improving the judgments that usage conveys. 10

The term ‘genuinely ethical’ is to be interpreted in connection with the narrow sense of ‘ethical judgment’ as a reflective activity. 11 “Of ethical sentences as ordinarily used, it may be said, I believe, that their entire use and function [. . .] is directive or “practical”” (Dewey 1945, p. 137, his emphasis).

Moral Disagreement and the “Fact/Value Entanglement”

257

I find it difficult to conciliate this naturalism with Putnam’s idea that justifications are “internal.” What does it mean for a value to have an internal justification? Does it mean that we cannot know that our response to the situation is the right one unless we see things “through the lenses of those very values”? In that case, we would be taking values as a kind of a priori pattern, by virtue of which the situation will look – or, maybe, will be – one way or another. That is exactly what lenses do. But it is also what emotions and feelings are supposed to do, according to Stevenson. Therefore, I dare to say that Dewey’s reply to Putnam would be similar to the one he gave to Stevenson: if that pattern is defined as extra-cognitive, it must be eliminated from the subject-matter of genuine ethical sentences. If, on the contrary, the pattern is thought of as cognitive, then it is itself subjected to evaluation and criticism by means of any relevant part of the available knowledge. I do not see any preestablished framework here which could make sense of the internal/external dichotomy.

V. Indeed, Dewey never used a dichotomy of this sort to account for the logic of moral (or any other kind of) justification. He was just interested in showing that “ethical terms and sentences are fully comparable with scientific ones as respects their subject-matter and content”; this content was for him genuinely descriptive, and hence objective. For this reason, when Putnam declares, “what I am saying is that it is time we stopped equating objectivity with description” (Putnam 2002, p. 33), I am not quite sure that these words correspond to the same spirit that inspired Dewey’s “entanglement thesis.” As I stated earlier, the “entanglement thesis” means that we cannot expect to reach a better understanding of how our descriptions relate to our valuations by keeping them apart and putting valuations out of the domain of objectivity and empirical testability. But, when Dewey comes to grips with the question as to how descriptions and valuations relate to each other, he never tries to intimate that there is something wrong with the idea of “fact.” Rather, he sets to prove that deliberation with respect to values is essentially a matter of factual description and prediction. This is the point I think Dewey was eager to make in his discussion with Logical Positivism. In other words, Dewey put the emphasis not so much on the role played by values in scientific knowledge, than on the possibility of treating values

258

Ángel Manuel Faerna

scientifically. And this was so because positivists had expelled value judgments from the field of scientifically warranted propositions. This is fairly evident when we look at the way Dewey introduces the topics of his Theory of Valuation: “Since the propositions of the natural sciences concern matters-of-fact and the relations between them, and since such propositions constitute the subject matter acknowledged to possess preeminent scientific standing, the question inevitably arises whether scientific propositions about the direction of human conduct, about any situation into which the idea of should enters, are possible” (1939, p. 192). Here Dewey develops the same argument that he will later repeat in his review of Stevenson: emotivism rely on “an alleged psychological theory which is couched in mentalistic terms, or in terms of alleged states of an inner consciousness or something of that sort” (p. 199). Such inner states, in case they existed, are irrelevant and unnecessary in an analysis of value-expressions, for “value-expressions have to do with or are involved in the behavioral relations of persons to one another” (p. 200); that is, they have to do with interpersonal public phenomena. Since any kind of valuation refers “directly to an existing situation and indirectly to a future situation which it is intended and desired to produce,” the phenomena involved are: “(i) aversion to an existing situation and attraction toward a prospective possible situation and (ii) a specifiable and testable relation between the latter as an end and certain activities as means for accomplishing it” (p. 202, his emphasis). On these grounds, Dewey’s argument proceeds to show that such phenomena can be phrased in terms of “propositions about observable events – propositions subject to empirical test and verification or refutation” (p. 201). I do not think it necessary to pursue Dewey’s argument further in order to support my claim that it does not follow Putnamian lines. While Putnam seems most interested in showing how values intervene in the description of matters-of-fact, Dewey’s emphasis is on the “matter-offactness” of values: “propositions about valuations have, indeed, been shown to be possible. But they are valuation-propositions only in the sense in which propositions about potatoes are potato-propositions. They are propositions about matters-of-fact” (p. 208, his emphasis). Of course, the “entanglement thesis,” if true, must work in both directions, but I think that the insistence in one direction or in the other reveals different attitudes toward those to whom the thesis was initially addressed to (i.e., philosophers of the empiricist-positivist tradition). To say it more explicitly: when Putnam asserts that to think without the fact/value dichotomy is to depart from “classical empiricism as well as [from] its

Moral Disagreement and the “Fact/Value Entanglement”

259

twentieth-century daughter, logical positivism” and “to enter upon a genuine “post-modernism”” (Putnam 2002, p. 9, his emphasis), in my opinion he is parting from Dewey in fundamental respects. In brief, Dewey’s position in his dialog with empiricists and positivists amounts to the following. The difference between “factualpropositions” and “value-propositions” consists only in that “valuepropositions” make reference to the direction of human interactions and human relationships in definite existential situations. Such interactions and relationships require value-terms to be adequately described. Then, as far as value-propositions express in what relation desire and purpose are supposed to be with respect to the existential situation and its future development, they must be treated as any other scientific, empirically ascertainable propositions. This means that moral disagreements, inasmuch as they emerge from competing value-judgments, should be solved by the very same methods that we run when the issue point is not “moral” or evaluative. In other words, propositions about valuations are themselves capable of being evaluated12 – note that there is here no trace of the kind of restriction Putnam introduces with his notion of “internal” justification. To carry out this second appraisal, we need to examine things “as they sustain to each other the relation of means to ends or consequences” (Dewey 1939, p. 211, his emphasis). But, whereas the possibility of an empirical evaluation of something as means is generally recognized, ends tend to be thought of as endowed with “intrinsic” value, and therefore as final. For this reason, Dewey’s well-known theory of the “continuum of endsmeans” becomes central in his conception of ethics as a form of scientific inquiry.13 For the sake of brevity, I will omit here the details of the theory; instead, I will conclude by illustrating the difference it makes with respect to the problem of moral disagreement as addressed by Putnam in his analysis of “thick” ethical concepts (see Putnam 2002, pp. 34ff ).

12

“Are propositions about existent valuations themselves capable of being appraised, and can the appraisal when made enter into the constitution of further valuations? [. . .] If this condition is satisfied, then propositions about valuations that actually take place become the subject matter of valuations in a distinctive sense, that is, a sense that marks them off both from propositions of physics and from historical propositions about what human beings have in fact done” (Dewey 1939, pp. 208-209). This, of course, illustrates the critical role of Dewey’s ethical theory that was mentioned above. 13 The relation ends-means is considered at length in Sections V (“Ends and Values”) and VI (“The Continuum of Ends-Means”) of (Dewey 1939, pp. 220-236).

260

Ángel Manuel Faerna

Words like ‘rude’, ‘brave’, ‘generous’, ‘cruel’, have the peculiarity of admitting descriptive as well as normative uses; therefore, the concepts they carry cannot be considered as purely evaluative (like ‘good’, ‘bad’, ‘virtue’, ‘vice’, or ‘duty’) nor as referring only to facts. The ethical “thickness” of such concepts would speak for the “entanglement thesis,” since it defies the idea that there is an absolute dichotomy between facts and values. It is only natural, then, that Putnam takes the trouble in defending the claim that they are genuinely “thick,” that is: (i) (ii)

that they cannot be replaced by a purely descriptive equivalent term; and that they are not “factorable” in two different components, descriptive and “attitudinal.”

Now, if such concepts are genuinely thick, then we cannot use their descriptive function without incorporating in the same speech-act, so to say, their evaluative content. But, to learn a concept is to learn how to use it, then we cannot understand a thick ethical term either without adopting the corresponding evaluative standpoint. We cannot, for instance, know what facts are involved in the description of someone as “cruel” unless we endorse, at least hypothetically, certain value judgments. Accordingly, “proponents of the entanglement” would maintain “that if one did not at any point share the relevant ethical point of view one would never be able to acquire a thick ethical concept, and that sophisticated use of such a concept requires a continuing ability to identify (at least in imagination) with that point of view” (Putnam 2002, pp. 37-38). Or else: “What is characteristic of “negative” descriptions like ‘cruel’, as well as of “positive” descriptions like ‘brave’, ‘temperate’, and ‘just’ [. . .] is that to use them with any discrimination one has to be able to identify imaginatively with an evaluative point of view” (p. 39, his emphasis). Although Putnam has invited us to “give up the very idea of a “rationally irresolvable” ethical dispute,” his analysis of thick ethical terms supplies an excellent basis for describing what such a dispute would consist in. If A says: “Ariel Sharon is a cruel statesman,” and B replies: “I concede that Sharon’s policy concerning Palestinians is harsh and painful, but I do not think he is being cruel at all,” their disagreement is not likely to be solved by a patient examination of every single military action ordered by Sharon in recent years to repress Palestinian people; for A and B will differ again in referring to these actions by means of different thick terms. As long as B refuses to call those actions “cruelty,” he is resisting to identify with A’s evaluative point of view;

Moral Disagreement and the “Fact/Value Entanglement”

261

and, since such point of view is not rigidly associated with any definite set of non-evaluative statements whatsoever (for there is no “descriptive component” that we can isolate as corresponding to the meaning of ‘cruel’), there is no hint about what sort of rational argument A could use here. Putnam’s suggestion is that A and B should continue discussing and trying things out just the same. I think this is wise, considering the alternatives. But if, as a result of these sustained efforts, B finally acknowledges that Sharon’s policy is cruel, all that we could say is that a “switch” has taken place in B’s evaluative point of view. As this switch is contingent, unpredictable, and non-compulsory, it hardly could be described as the effect of rational force, nor is it easy to distinguish from the “change in attitude” that Stevenson associated with sheer persuasion.14 To say that B’s final opinion is “internally” justified according to A’s evaluative framework does not alter the fact that the disagreement has been, so to say, solved accidentally in a vacuum of rules; for, in case that B would have convinced A that Sharon’s policy is morally impeccable, the opposite judgment would be “internally” justified as well according to B’s evaluative framework. Per contra, I presume that Dewey’s analysis of the case we are considering would have hinged on the means/ends relation. After all, what A is claiming is that Sharon’s policy is unnecessarily tough, whereas for B it is tough but necessary, and this is a typical problem of assessing the adjustment between ends and means. Then, what we have to do is to evaluate those conflicting value-judgments, which means to examine things rationally “as they sustain to each other the relation of means to ends or consequences.” This is not “to seek an Archimedean point, an “absolute conception” outside of all contexts and problematic situations,” a mistake against which Putnam rightly warns us. On the contrary, it is to put under discussion the whole situation in order to examine what conditions have to be brought into existence to serve as means, what ends are actually reached, how these ends turn out to be “means to future ends as well as a test of valuations previously made. Since the end attained is a condition of further existential occurrences, it must be appraised as a potential obstacle and potential resource” (Dewey 1939, p. 229). According to Dewey, such a thorough examination of the

14

Moreover, it could be legitimately asked whether, in using the term “to identify imaginatively with an evaluative point of view,” Putnam is committing to that “terrible” mentalistic psychology that Dewey deplored in Stevenson, for such term echoes the idea of “empathy,” a notion too obscure indeed to play such a central role in elucidating how rational agreement can be achieved in ethics.

262

Ángel Manuel Faerna

situation should be enough to empirically settle the question whether Ariel Sharon is being “cruel” or not. On this perspective, the term ‘cruel’ is but a linguistic resource to indicate that a particular ill-adjustment between means and ends is taking place. As this constitutes an empirically ascertainable matter, the term is, to that extent, purely descriptive. But the fact it describes is undesirable in itself, for it means that the responses actually elicited do not fit with what the situation requires. For this reason the term has also an evaluative import that is inseparable from its descriptive content.15 Let me ask again the question with which this discussion begun. Why is it so difficult to reach rational agreements when “values” are at stake? Stevenson’s explicit answer is that there is no strict implication between beliefs and attitudes. Putnam’s implicit answer is that it is hard to move from one evaluative point of view to another. Dewey’s answer is that we persist in the untenable idea of “ends-in-themselves,” the view that, as compared with the importance of the selected and uniquely prized end, other consequences may be completely ignored and brushed aside no matter how intrinsically obnoxious they are. This arbitrary selection of some one part of the attained consequences as the end and hence as the warrant of means used (no matter how objectionable are their other consequences) is the fruit of holding that it, as the end, is an end-in-itself, and hence possessed of “value” irrespective of all its existential relations. (Dewey 1939, p. 228, his emphasis)

Dewey’s doctrine of “the continuum ends-means” has been favorably commented by Putnam in different previous works. Thus, it is rather striking that he did not mention it in his historical reconstruction of the collapse of the fact/value dichotomy. For, as this paper has been suggesting, that doctrine is essential to understand Dewey’s contention that values are subjected to the same methodology of inquiry that applies to physical science – a contention that Dewey badly wanted to urge on positivists. In omitting this point, Putnam’s version of the “fact/value entanglement” falls short in evading the non-cognitivist conclusions arrived at by the proponents of the dichotomy. No matter whether we frame values in terms of “subjective attitudes” or in terms of “evaluative points of view,” they will acquire that static “final” character of “endsin-themselves” which, according to Dewey, blocks the road of moral inquiry. Dewey’s pragmatism has been defined by a Spanish scholar as 15

I venture to say that Dewey would have disliked the expression ‘thick ethical concept’, for it puts the emphasis on the “ethical” side; ‘value-laden descriptive concept’ is perhaps closer to Dewey’s mind.

Moral Disagreement and the “Fact/Value Entanglement”

263

“the critic of culture” (see Esteban 2001), which means the critic of any “community” that presents itself as identified with a given set of values, or as endowed with a set of “lenses” of its own. Such “values” are the mores that moral theory must criticize in order to abolish them, or to correct them, or to improve them, according to the verdict of inquiry. As opposed to those traditional communities, Dewey’s democratic community must be experimentalist, that is, it must enter an endless process of forming and reforming its ends and values: fallibilism, tolerance, cooperation, are but the milieu of such process; they are means, not ends. If one looks at the three answers given above, one realizes that only the third one makes a real difference with respect to the dominant view that value judgments are subjective and that moral disagreements escape rational control. I have argued that the claim that facts and values are entangled is not strong enough to remove this dominant view, unless it is complemented, as Dewey did, with a more full-blooded naturalism. The question whether such naturalism is or is not objectionable in its own terms is, of course, another story.

Acknowledgements I am indebted to my colleague Ramón del Castillo for some profitable conversations on the topics of this paper, and to my student Lorrina Niclas Barrientos for her help with the English manuscript.

Universidad de Castilla-La Mancha Facultad de Humanidades de Toledo Plaza de Padilla, 4 45071 Toledo, Spain e-mail: [email protected]

REFERENCES Dewey, J. (1891). Outlines of a Critical Theory of Ethics. In: (1967-1972), vol. 3, pp. 237-388. Dewey, J. (1939). Theory of Valuation. In: (1981-1991), vol. 13, pp. 191-251. Dewey, J. (1945). Ethical Subject-Matter and Language. In: (1981-1991), vol. 15, pp. 127-140.

264

Ángel Manuel Faerna

Dewey, J. (1967-1972). The Early Works of John Dewey, 1882-1898. 5 vols. Edited by J.A. Boydston. Carbondale: Southern Illinois University Press. Dewey, J. (1976-1983). The Middle Works of John Dewey, 1899-1924. 15 vols. Edited by J.A. Boydston. Carbondale: Southern Illinois University Press. Dewey, J. (1981-1991). The Later Works of John Dewey, 1925-1953. 17 vols. Edited by J.A. Boydston. Carbondale: Southern Illinois University Press. Esteban, J.M. (2001). La crítica pragmatista de la cultura. Ensayos sobre el pensamiento de John Dewey. Heredia (Costa Rica): Universidad Nacional de Costa Rica. Putnam, H. (2002). The Collapse of the Fact/Value Dichotomy. In: The Collapse of the Fact/Value Dichotomy, and Other Essays, pp. 1-64. Cambridge, MA & London: Harvard University Press. Stevenson, C.L. (1960). Ethics and Language. New Haven: Yale University Press. Stevenson, C.L. (1978). Introduction. In: Dewey (1976-1983), vol. 5, pp. ix-xxxiii.

Francisco Javier Gil Martín Jesús Vega Encabo TRUTH AND MORAL OBJECTIVITY: PROCEDURAL REALISM IN PUTNAM’S PRAGMATISM

ABSTRACT. Drawing upon several questions raised in a recent exchange between Putnam and Habermas, this paper exposes Putnam’s pragmatist arguments against both moral skepticism and substantive moral realism and examines his realist account of moral objectivity in connection with the notion of truth. We defend that Putnam’s conception of moral objectivity can be better understood as a new version of a procedural realism. Furthermore, we discuss an internal tension and some problems that threaten the defense of this kind of moral realism.

1. Introduction Hilary Putnam and Jürgen Habermas have met on various occasions and their philosophies have been mutually enriched in such topics as the conception of truth, the notion of equality, the defense of democracy, and the rejection of sceptical and relativist trends. 1 The last episodes of this dialogue have revolved around some divergences on the explanation of moral objectivity. In a lecture delivered in Frankfurt in July, 1999, Putnam unmasked the dangers of holding a sharp distinction between norms and values (2001; 2002a, Ch. 7). In a few words, Putnam presents a sort of indispensability argument against the threat of losing content, if all we need to ground moral objectivity is to obey the norms of “discourse ethics.” The argument can be summarized like this: the objectivity of norms presupposes the objectivity of some values. 1 See the last two lectures of Putnam (1987), as well as Putnam (1990a, pp. 217-231; 1990b; 1995; and 1994, Part III). On the other hand, see Habermas (1988, pp. 153-186; 1991, pp. 120-125; 1992, pp. 53-57; and 1999, especially the introduction and Ch. 5).

In: M.U. Rivas Monroy, C. Cancela Silva, and C. Martínez Vidal (eds.), Following Putnam’s Trail: On Realism and Other Issues (Pozna Studies in the Philosophy of the Sciences and the Humanities, vol. 95), pp. 265-285. Amsterdam/New York, NY: Rodopi, 2008.

266

Francisco Javier Gil Martín and Jesús Vega Encabo

Therefore, some value judgments must be correct, even “true,” for the norms to have content and be valid at the same time. Habermas defends himself, claiming that his differentiation between norms and values doesn’t implicate a non-cognitive status of the latter, and rejecting the frequent charges of empty formalism against Kantian ethics. His response insists on the idea that our moral point of view grows from within the ethical conditions of a community and its shared form of life. Values are not completely separate from norms; they are authorized by rational procedures and can also be subject to universality constraints. Even the formal procedures to solve moral problems embody a certain normative substance, such as the demand of full inclusion of the participants and their mutual perspective-taking. Furthermore, Habermas suspects that Putnam’s view about the objectivity of value judgments tends to assimilate them to descriptive (even scientific) statements. The issue about truth has played a role in this clear misunderstanding (Habermas 2002; Putnam 2002b). But more significant is Habermas’ contention that Putnam should elucidate the priority of questions regarding justice over questions of good life, such as they are established in modern democracies.2 Among all the points raised in this open dialogue, we are going to focus on the understanding of moral objectivity in connection with the notion of truth. Both philosophers would be willing to refute a version of the so-called substantive moral realism. Habermas tries to secure his moral cognitivism with a constructive account of practical discourse, and renounces the application of the predicate ‘. . . is true’ to moral judgements. Putnam, on the contrary, likes speaking of the truth and falsity for evaluative and normative judgments and commits himself to a kind of moral realism. And so Habermas thought of Putnam as a “thoroughgoing realist.” But Habermas himself ought to take seriously both claims of Putnam. Given that norms and values interpenetrate, the objectivity of norms falls or stands with the objectivity of values; and given that issues of objectivity and truth are intimately connected, the defense of a certain kind of moral realism must be possible. In this paper, we argue that Putnam and Habermas should agree on a neutral kind of procedural realism. Naturally we do not claim that appealing to procedures exhausts their understanding of our moral point of view, at least in Putnam’s work. Nevertheless, certain pragmatist undertakings, together with a plain dismissal of a metaphysical realism of values and

2

That was one of the contentious questions in the debate between Putnam and Habermas at Northwestern University, Evanston; November 13 th , 2002.

Truth and Moral Objectivity

267

norms, lead them to emphasize the role of our standards and procedures of moral justification in order to account for the objectivity of moral statements. Our argument will run as follows. First, we will see that indispensability arguments are not designed to explain what the objectivity of a certain discourse domain consists in. Second, we will add that the appeal to the notion of truth does not even help in this task. Third, we will note that the rejection of the analysis of truth in epistemic terms still leaves open the possibility of a close link between justification and truth in explaining moral knowledge. We will conclude by arguing that moral procedural realism exposed in pragmatist terms only resists constructivist temptations by accepting primitive and underived obligations.

2. Indispensability Arguments (IA) in Ethics IA go very deep in Putnam’s philosophy. We are before an antiskeptical argument, with robust roots in Kantian thought (even with a transcendental flavour) and with a prominent pragmatist inspiration. It is antiskeptical, because it points up a cognitivist and objectivist account in different domains; Kantian, because it raises the question about the possibility of knowledge; and Pragmatist, because the indispensability is always seen from the priority of practice. To appreciate the scope of IA, it is necessary to answer (a) (b)

questions about the conclusions of the argument and questions about what grounds indispensability in every domain.

First, IA in Putnam’s version are designed to support cognitivism and objectivity; they are thus opposed to certain skeptical threats. Second, Putnam uses two distinct strategies to motivate the indispensability in the normative case: the “companions in the guilt” and the inseparability arguments. Let us examine these questions closer. (a) How could we derive cognitivism from indispensability? Putnam thinks that indispensability justifies our treatment of some kind of statements as true or false (1994, p. 154; 2002a, p. 111). Nevertheless, we could suspect that this superficial fact is not enough to establish that they could be really true or false. What prevents us from thinking that we are under a mere illusion of cognitive content and an appearance of objectivity? But, from where could we ground the suspicion that we entertain an illusion? By the very premise of the argument, we cannot but

268

Francisco Javier Gil Martín and Jesús Vega Encabo

place ourselves within the game of language we are considering, because only within it are the statements viewed as indispensable. Dispensability could only be established by legitimizing a jump to a perspective external to the game. (When Quine presented the argument in philosophy of mathematics, he claimed that no first philosophy could ever transcend science to put science itself into doubt.) The IA excludes an Archimedean point of view, from which we could discover the dispensability of concepts and forms of discourse that are deeply integrated in our scientific and daily practices. Such a detached point of view is adopted by non-cognitivist approaches in ethics that lead to revisionist descriptions of (the presumed cognitive content of) our ordinary moral convictions. Indispensability means that adopting an Archimedean perspective to criticize and revise moral language has consequences for the understanding that agents have of their own activities of giving reasons and asking for them. In other words, treating moral statements as true or false, as cognitively significant because they are indispensable in describing the language game of morals, is equivalent to eliminating any possibility for any revision that would make illusory our practices. So, from the internal point of view of our normative practices, it would be incoherent to consider moral judgments as lacking cognitive status. The IA defends cognitivism from the threats of an external, Archimedean skeptic. Putnam’s argument resembles a Kantian mode of thought; it can be seen as the result of a transformation in the way we ask certain questions. Instead of arguing about how we could cognitively have access to Platonic entities in order to secure the objectivity of mathematical knowledge, we ask about the function that mathematical statements play in science. Practical considerations on the indispensability of some statements suffice to establish the possibility of knowledge and the cognitive status of them. Otherwise, more entrenched areas of discourse would be affected. Indispensability concerns the intelligibility of our practices and gives reasons for legitimating a certain conceptual scheme without following the full path of a transcendental foundation. So, arguing for the indispensability amounts to a way of justifying the applicability of some concepts in practice. For example, it is indispensable in giving sense to the statements of physical science to accept the conceptual scheme of mathematics. We could also talk about a necessity derived from the impossibility to make sense of our practices without supposing the cognitive status of a certain domain of language. Although it is not about transcendental necessity committed to the unrevisability of those presuppositions, the Kantian roots of the argument

Truth and Moral Objectivity

269

are evident: we could not actually think without applying those concepts (Putnam 1994, pp. 245-263). (b) But we have not yet established the real indispensability of any conceptual scheme. The indispensability must be motivated and grounded in each particular case. For mathematical statements, the argument seems clear enough. Given the deep integration of mathematics in our physical objective knowledge of the world, to consider mere fiction the conceptual scheme of mathematics would also alter such knowledge and affect the very cognitive status of physical laws.3 The same argument does not work for normative knowledge and moral knowledge. One difference concerns the scope of the agreement reached in mathematics and in morals. But it is not the only difference: normative predicates are not integrated in science or daily life in the same way as mathematical concepts are integrated in fundamental laws of physics. To overcome these difficulties, Putnam proposed two further arguments for the indispensability of normative discourse and against the subjectivist conception of normative concepts. (i) The “companions in the guilt” argument stressed the incoherence of being non-cognitivist in ethics and cognitivist in epistemology. “All values are in the same boat,” as Putnam often says. All values are predicates that guide actions; so their descriptive use is not detachable from an evaluative or normative one. If all predicates that are actionguiding would be “queer,” as Mackie contends, those predicates integrated in scientific practices (such as coherence, simplicity or explanatory power) would be also queer. The practice of science would be affected by the distortion of evaluative and normative language encouraged by expressivists and subjectivists. The notion of “fact” would be unintelligible unless the criteria for justifying the facts are subject to rational scrutiny and have cognitive status. We cannot build a scientific image of the world without essentially involving the reasons that guide the selection of hypotheses. Describing a scientific theory as simple or coherent is equivalent to saying that we are justified in accepting the theory and claiming that we ought to accept the theory. Therefore, either it has sense to talk about “facts involving duties” or we relinquish our claim that science provides us with objective facts. If “values” seem a bit suspect from a narrowly scientific point of view, they have, at the very least, a lot of “companions in the guilt”:

3 “Mathematics and Physics are integrated in such a way that it is not possible to be a realist with respect to physical theory and a nominalist with respect to mathematical theory” (Putnam 1979, p. 74). See also Putnam (2004, p. 66).

270

Francisco Javier Gil Martín and Jesús Vega Encabo

justification, coherence, simplicity, reference, truth and so on, all exhibit the same problems that goodness and kindness, from an epistemological point of view. (Putnam 1990a, p. 141)

(ii) This first argument is combined with a second one. The doubts casted on the objectivity of the normative are consequences of a mistaken and strict dichotomy between facts and values. The entanglement and interweaving of facts and values, facts and theories, facts and interpretations has become a recurrent theme in Putnam’s work during the last two decades. Thinking of facts in an empiricist or positivist vein unables us to see how value statements could be rationally justified. Ethics cannot be about matters of fact insofar as “facts” are observationally verifiable or reducible to observation predicates that would allow us to confirm factual statements by a simple test procedure. Obviously, evaluative and normative terms are not descriptive in this sense. The second part of Putnam’s argument is positive: if “value and normativity permeate all of experience” (2002a, p. 30) and if it is not possible to factorize in our epistemic or moral experience between the descriptive and evaluative (normative) components, then it is better to think about facts and values as entangled. The entanglement is shown by words as cruel, “used sometimes for a normative purpose and sometimes as a descriptive term” (2002a, p. 35), and it points to the fact that we are not actually able to describe such kind of terms without using the very same terms in a normative context. The normative point of view is unavoidable. This is how ethical concepts bear on the world. This way of seeing the fact/value entanglement leads us to the core of IA in the context of normativity. As we have said, indispensability disallows an external perspective from which to judge our moral life or to reject our normative practice as illusion. The primacy of practice is the expression of the indispensability. In fact, IA capture the idea that reasons have authority for us. Our life could be unintelligible without viewing ourselves as being subject to the authority of duties, norms and values. Normative language is indispensable because it is constitutive of the point of view of the participants. We cannot understand what it could mean to deny that there are normative truths, in the same way in which we cannot make sense of our thought without assuming that there are logical truths. It would be foolish to talk about what thinking signifies for us without presupposing the conceptual scheme of normativity. We cannot get rid of evaluative and normative vocabulary without risking the loss of the normativity as a whole. The normative problem does not rest solely on explaining the function of normative concepts, but on understanding how we are justified in our rational requirements. The

Truth and Moral Objectivity

271

main idea of the argument would be the following: all the efforts to explain normative concepts in terms of projections performed by the agents are condemned to lose the specificity of their normative authority, which manifests itself only from the first-person point of view. This fact also offers an answer to the possibility of normative knowledge in general. The Kantian roots reappear at this point: reflection on our practices clears the path to recognize the conditions that make possible our justified moral agency. Of course, instead of being transcendentally shaped, reflection arises from the conditions of the practice of inquiry and thereby, values and norms will not be infallibly justified. Revisability is not prevented, but we will not be able to know the conditions for it from the outside of the very normative stance.

3. Truth and Objectivity If the IA pervades the realm of normativity, then we have “reasons for rejecting the widespread belief that ethical judgments lack objectivity” (Putnam 2004, p. 3). But, what do we understand by the objectivity of normative language? Putnam will claim that moral statements can be taken as correct, incorrect, true or false, in the sense that they are “under rational control, governed by standards appropriate to their particular functions and contexts” (2002a, p. 33). However, Habermas contends that talking about “truth” in ethics compels us to embrace a substantive evaluative realism. Let us see briefly how Putnam’s notion of “objectivity without objects” uncouples the moral truths from this supposed bond with a substantive realism. There are logical reasons to claim the truth-aptness of moral statements and to state that the same notion of truth is applicable to any kind of statements. The difficulties found by the expressivist to account for conditionalization and for arguments using descriptive and evaluative statements are at the same time good reasons to uphold that moral statements are truth-apt. For instance, if the government succeeds in convincing the rebels to lay down the arms, the situation of the civilian population will improve; the government has succeeded in convincing the rebels to lay down the arms; therefore, the situation of the civilian population has improved. These facts can be accommodated within a minimal account of truth. The Equivalence Principle explains our use of the predicate ‘. . . is true’ within our languages. Accordingly, the judgment that it is true that murder is wrong is equivalent to the judgment that murder is wrong.

272

Francisco Javier Gil Martín and Jesús Vega Encabo

Someone could argue that this scheme accounts for superficial facts about the use of moral statements and that those superficial facts would disappear after providing a correct interpretation of moral statements. In other words, that moral language presents some features of the assertive mode of speech does not let us draw any conclusion about the realism/antirealism issue and the objectivity of moral judgments. What does it explain that we can assert, deny and ascribe truth or falsity in moral language? And what is meant by correctness and validity in ethics? The problem of truth for moral statements becomes interesting when it involves other theses about the value of truth and about how moral statements bear on reality. In the Dewey Lectures, Putnam displays an account of truth that provides some responses to these questions. He tries to block the deflationist’s use of the Equivalence Principle together with a verificationist theory of meaning. But he does not support this rejection by positing a substantive property that would make our assertions really true or false. Once the Equivalence Principle is accepted by logical reasons, the philosophical task is to give an account of what understanding is. And for this task, we need to escape the alternative between the verificationist and the metaphysical realist: on the one hand, the statements we use to assert something can transcend verification and, on the other, this does not require postulating a queer substantive property, a “substantive sort of rightness” (1999, p. 54) with a special metaphysical burden. Putnam then puts forward the following theses. First, we understand the predicate ‘is true’ by understanding what is asserted by the proposition from which we predicate truth; second, it is possible to formulate meaningful statements which are recognition-transcendent, because in asserting them we are just claiming that what is asserted is beyond our recognitional powers; and third, understanding does amount to seeing something in the marks and noises without positing something that goes beyond them. These theses articulate a new IA that points out that verificationists and metaphysical realists cannot do justice to “who we are and the sense that our practices have for us” (1999, p. 64). It is our self-understanding that would be affected by both positions. Suppose that understanding consists in knowing the assertability conditions of sentences. Then every change in the context of use would cause a change in meaning. But if this is so, we would distort the way in which understanding takes place for us. On the other hand, if we posit something beyond our practices to explain how understanding is possible, then we are trying to reflect a necessity external to our practices “and the

Truth and Moral Objectivity

273

ways of thinking internal to them” (1999, p. 63). We could even claim that learning a new set of concepts entails the ability “to see the face” of the activities that ground our own practices in other activities. If we cannot do that, then we are reaching the limits of the intelligibility with respect not only to the practice we are learning and evaluating, but also to our own practice. Correctness denotes the way a sentence in use “bears on reality.” Not to violate the priority of the normative viewpoint is a formal constraint on every notion of correctness. As such, it does not say anything about the way we understand how normative (and moral) concepts bear on reality. But we know that, given the intimate link between our language, our actions and the world, we are testing the sense and normative understanding of our own point of view when we inquire about the extension of our set of evaluative and moral concepts. Since not every statement “bears on reality” in the same way, we no longer need to explain the truth and objectivity of moral statements as necessarily performing a descriptive function. Moral statements are not made true or false by (non-natural) objects described by these statements. Habermas was therefore misguided when he objected that Putnam, in talking about the truth and falsity of value statements, was treating them as descriptions. Identifying the way moral language bears on reality is crucial to account for the objectivity of moral statements. We find valuable insights in this regard in Putnam’s recent book Ethics without Ontology. There he defends that logical and mathematical truths “bear on reality” without describing a set of special entities and properties. Something similar is argued for evaluative and normative statements; it’s possible to talk about objectivity without objects in logic, mathematics and normativity. We will not repeat the arguments against the descriptive use of logical and mathematical truths. We are interested in Putnam’s positive proposal about the objectivity of some statements considered as conceptual truths. The latter are not conceived as analytical, necessary and unrevisable, but as truths for which “it is impossible to make (relevant) sense of the assertion of its negation” (2004, p. 61; 1994, pp. 255-256). These truths are identified by way of interpretation into a body of beliefs; they are interwoven with other empirical beliefs. They are revisable as a result of deep changes in opinion about what we can “conceive”; certain intuitions can then disappear after taking into account alterations in the justification of those conceptual truths. It is our familiarity with what the justification is, for example, in the domain of logic that explains how one learns what logical truth is. Analogous considerations apply to mathematical truths and to methodological value judgments. Nothing prevents us from applying the same strategy for

274

Francisco Javier Gil Martín and Jesús Vega Encabo

evaluative and normative truths. Under the conditions of the ethical life in which we are involved as participants, it would make no sense to affirm the negation of certain evaluative and normative assertions. Of course, this kind of truths are again revisable as part of a whole of empirical and conceptual beliefs. And such a revision becomes possible because we are in fact familiar with moral justification. We have learned the standards and procedures of morals. Learning what moral truths are is internally connected to learning the procedures that help us to establish them. And among these procedures, those ones used to apply the truths in our individual and social life play a significant role. It is the familiarity with these procedures that it would explain that one cannot recognize certain truths but from a certain (normative) perspective. The comparison with logical and mathematical truths highlights why moral knowledge is subject to the same constraints of fallibilistic inquiry. Practical reasoning is guided by norms of truth and correctness. Evaluative and normative truths are revisable through learning processes, but they are not made true or false by objects and properties such as intrinsically non-natural values. Therefore, explaining how evaluative and normative statements bear on reality does not entail the defense of a substantive moral realism. We claim that it is closer to a version of a procedural realism, in which the objectivity of moral knowledge is explained in terms of the procedures we use to recognize evaluative and normative truths. It does not mean that all evaluative and moral truths are “conceptual” truths. But, regarding the normative point of view from within which these truths are available, some of them have to be conceptually true if we want to preserve the whole world of normativity.

4. Truth and Justification Nevertheless, such a line of reasoning engenders obvious questions. Some of them are also prompted by the debate between Putnam and Habermas. Are both authors implicitly committed to an epistemic concept of truth? Are normative and moral truths identified with statements whose validity is established by procedures learned through inquiry? And are those procedures, in a sense, an essential aspect of normative and moral validation? The concepts of truth and justification are not linked by definition. In the last decade, Putnam has argued against the temptation to assimilate truth with warranted assertability even in ideal conditions. Truth is not definitionally the outcome of an ideal conduct of inquiry, because there

Truth and Moral Objectivity

275

can be truths that we cannot ever verify. Putnam’s objections to Habermas emphasized this point in the case of ethics. It was not only that we find difficulties making sense of an ideal speech situation without admitting “thick” ethical terms. It was also that “there is no reason to believe that the outcome of an ideal and sufficiently prolonged discussion of ethical questions would inevitably be correct” (2002a, p. 126). It must be noted, however, that Habermas has distanced himself from his previous Peircean account of truth. Criticisms by A. Wellmer and C. Lafont convinced him that “truth” is a property that statements cannot lose, and that the unavoidable epistemological link between truth and justification “must not be turned into a conceptually inseparable connection in the form of an epistemic concept of truth” (Habermas 1999, p. 52). The very notion of an ideal speech situation is also abandoned because it falls into the fallacy of misplaced concreteness in suggesting an state of definitive agreement which would make all further communication superfluous. The procedural conditions of discourse permit only to anticipate the argumentative force of those reasons that would convince us. Instead of evoking an end state as a regulative ideal, they favour an ongoing process of critical discussion by which objections are weakened and the acceptable becomes what survives criticism. Nevertheless, Habermas does trace a clear-cut distinction between the concept of validity at work in scientific and descriptive discourse and the validity of moral statements as ideal warranted assertability. Whereas truth claims in the case of empirical statements pragmatically presuppose a world of objects existing apart from our descriptions, moral validity claims lack such a realist commitment and in themselves hinge on the recognition by all the participants in the discourse. Moral correctness is conceptualized as a validity claim in a sense analogous to the truth, albeit without the same presuppositional commitments, and defined by the deontology of the just and the right, in terms of the norms and actions being equally in the interest of all possibly affected (Habermas 1999, pp. 271-318). 4 Moral objectivity is thereby explained in a cognitivist and constructivist way. Value judgments do not respond to the same validity conditions. But they are candidates for embodiment in norms. And when values are discussed from an impartial point of view and are rationally accepted by all the participants in a moral discourse, then the content of the judgment can be authorized as obligating under universality conditions. In that

4

Obviously, the terms of the analogy were more convincing before dropping the epistemic notion of truth (Habermas 1991, pp. 11, 126, 130-131).

276

Francisco Javier Gil Martín and Jesús Vega Encabo

sense, they get entangled with moral norms. At the same time, Habermas accepts a weak notion of cognitive content for evaluative statements and insists that the meaning and significance of values can be rationally debatable in ethical discourses concerning the self-understanding and life-projects of persons and groups. Values are not objective just in the sense of being accepted within a community as a set of cultural conventions or customs. They deserve acceptance when subject to rational scrutiny and supported by good reasons. They are objective insofar as they are recognized intersubjectively by means of compelling reasons. Values have to be authorized in a practical context of rational discussion and inquiry. So Habermas is reluctant to consider them in naturalist terms, as sociological facts. Some convergences may not be less important than the differences between Putnam’s realist cognitivism about evaluative and normative judgments and the Habermasian account of ethical and moral discourses. We know that value judgments cannot be authorized or made true by reality itself. Habermas is willing to differentiate the assertoric mode of speech from the deontological mode as regards the validity claims. But his only reason is to secure that validity claims in the moral domain are unconditional without admitting the existence of substantive “moral truths.” Moreover, rather than underestimating the cognitive status of value judgments, Habermas tries to hold a distinction internal to our normative discussion between the questions regarding justice and the questions regarding goodness, a distinction that he finds neglected in Putnam’s philosophy. But what is more important is that for both of them what is true (or valid) in ethics does not diverge from what is endorsed by the best reasons, even with regard to value judgments. As Putnam puts it: “In the case of ethics (unlike science), the true view cannot differ from the view for which there are the best reasons” (2002a, p. 175). He occasionally expounds that intimate relation by referring to how we judge objectivity in an ordinary sense, roughly equivalent to what Kant named the “enlarged mind” in the “maxims of common understanding”: Statements are called “objective” if their claim to truth is not dependent on idiosyncratic standpoints or on disregarding the standpoints and interests of others. A sufficient condition that an ethical claim be objective in this “ordinary” sense is that it be reasonable from the standpoint of an interest in the common welfare, where the common welfare is [. . .] something that is itself to be determined by intelligent discussion among persons who share this commitment. I wish to emphasize that this is something that I suggest only as a sufficient condition, and by no means a necessary one. “Value judgments” are not a homogeneous class, and different sorts of judgments possess different

Truth and Moral Objectivity

277

sorts of objectivity. If it is reasonable for those affected to accept an ethical claim of this sort after experimentation and discussion, it will be – so long as reasons to question it do not arise – what John Dewey called a “warrantedly assertable” claim. (Putnam 2002c, pp. 17-18)

Such a minimal condition is essentially procedural insofar as it links the objectivity of moral truths to their justification. In these terms, it is akin to what Habermas says about the deliberative process of the moral point of view. He emphasizes the role of the procedures in filtering the reasons given in support of the validity of moral norms. This filtering works by rationally grounding common interests in shared values and by examining the norms that incorporate these values: universal norms are worthy of recognition because they are in the common interest of all participants in the discussion and because they are equally good for all of them. That minimal condition actually applies to the ethical discussion about values, as we the affected deliberate to select our best values and to identify the common good from the point of view of what is the best “for us.” In other words, Putnam and Habermas agree on some minimal procedural conditions that would link the justification of our judgments with the consideration of what is an ethical truth (Putnam) or a valid norm (Habermas). Putnam would object that the procedural conditions do not exhaust the necessary conditions for the objectivity of ethical statements, because not every evaluative judgment work in the same way. But once we have given up postulating a realm of intrinsic moral facts or values, it is not easy to see how objectivity can be ascertained otherwise. A brief reflection about the reasons to disavow a substantive moral realism would clarify the inevitability of the so-called procedural realism.

5. Procedural Realism In The Sources of Normativity, Christine Korsgaard introduced a distinction between two kinds of moral realism, substantive and procedural. Any theory of normativity, also in ethics, has to explain how reasons have normative force without distorting the role they play in our life, seen from the first-person perspective. Traditionally, it has been thought that only some version of moral realism might reply to subjectivist and relativist arguments and in general to any sceptical position. But it is possible to identify different realist commitments. The

278

Francisco Javier Gil Martín and Jesús Vega Encabo

question about moral “reality” is one about how objective moral judgments are possible.5 Korsgaard (1996, p. 35) characterizes substantive moral realism as “the view that there are answers to moral questions because there are moral facts or truths, which those questions ask about.” According to Korsgaard, what distinguishes such realism is the way it explains the objectivity of moral statements. In a certain sense, it adopts some dubious metaphysical commitments. Putnam consistently rejects its distortion of our understanding of the objectivity of reasons: “I see the attempt [. . .] to provide an Ontological explanation of the objectivity of ethics as a[n] [. . .] attempt to provide reasons which are not part of ethics for the truth of ethical statements” (2004, p. 3). If the realist accounts for our moral “reality” in terms of the discovery of intrinsically normative entities, then the link between the standards and procedures prevailing in our moral language and the truths to be known would be severed. But our moral life involves per se what agents think from the inner perspective of their moral activities; there is no such a thing a point of view from nowhere. The substantive realist strategy is pseudo-explanatory: on the one hand, it posits objects whose explanatory necessity is not obvious; on the other, it has to appeal to the self-evidence of normative truths, and then refuses to answer the normative question. The procedural realist proposes a different account for the objectivity of ethical judgments. He would defend three theses. The first one is very neutral and could accommodate a wide variety of alternatives, while the other two are more restrictive. (1)

(2) (3)

There are correct and incorrect answers to normative (moral) problems; there are better or worse reasons; there are correct or incorrect modes of applying normative (moral) concepts. Moral statements can be treated as true or false, valid or invalid. Moral truths are not independent of the procedures to establish them.

The procedural realist notion of moral objectivity cannot be understood without invoking the way agents determine the existence of reasons for moral truths. It is not enough to claim that there are correct or incorrect answers; it is also required that what explains the objectivity be connected with our means to reach those answers. 5

In Putnam’s words: “As a being who makes value judgments every day, I am of course committed to the idea that there are true value judgments: what must be the case if there are to be true value judgments? In what kind of a world can there be true value judgments?” (1995, p. 43).

Truth and Moral Objectivity

279

Putnam staunchly upholds that there are no recognition transcendent facts in ethics. He also pays attention to the role played by procedures in explaining moral objectivity. In “William James’s Ideas,” written together with Ruth A. Putnam, he characterizes James’s ethics as an “imperfect procedural conception” (1990a, p. 225). The core of the position is that we cannot “discover” what truth or goodness is independent of our procedures. Rightness amounts to being the outcome of a self-correcting procedure. The success of a procedural rule cannot be established irrespective of the very same procedure and, given that our procedural rationality is imperfect, changes and revisions are possible. In short, Putnam’s explanation of moral objectivity has to do, at least in part, with the application and correction of procedural rules. The only constraints are to avoid conceiving procedures as algorithms and as not subject to fallibility conditions. Now, the thesis (3) is ambiguous and admits two different readings. (3a)

Normative truths are essentially dependent on rational procedures.

What it is to be a normative truth is to be the outcome of a procedure of justification. This thesis needn’t be interpreted as implicating a definitional link between truth and rationality. But in talking about conceptual truths Putnam has insisted on the inseparability of our conceptions about what a normative truth is and our judgements about what is reasonable or unreasonable. And in some places he has argued that the notion of truth in ethics is internal to the practical reasoning itself. If we ask about what it really makes sense from the moral point of view, we identify some truths that are not understandable without considering exercises in reasoning. But what kind of dependence are we talking about? If there is some necessity in the way moral truths connect with the reasons moral agents may articulate through the use of procedural rules, such a necessity derives from the fact that we could not make sense of our moral language without invoking the standards and procedures we use to justify our moral demands towards others and towards ourselves. Are we then faced with a weak transcendental necessity? (3b)

Normative truths are not completely independent of rational procedures.

Truth does not always transcend warranted assertibility. Truth is not radically non-epistemic. In Putnam’s terms, “truth does entail warranted assertibility if conditions are sufficiently good” (2002a, p. 108). But this thesis could be just the claim that there must be a connection between the

280

Francisco Javier Gil Martín and Jesús Vega Encabo

truth and the justification of our moral claims if we are not to be condemned to skepticism. Some truths must be justifiable if we have any moral knowledge at all. Even if one is not very worried about skeptical challenges, to maintain that the very nature of moral truths entails that they are justifiable needs more argument, surely a kind of dependence as derived from (3a). Is it possible that normative truths transcend the best reasons we have for our moral claims? If we take seriously Putnam’s criticisms of the idea that the “outcome of an ideal and sufficiently prolonged discourse of ethical questions would inevitably be correct,” we would have to ask how to understand correctness if goodness and truth are nothing but “regulative ideals” independently of our imperfect procedures. There are two options here: either the procedure is not good enough or some standards of correctness are in fact separate from the procedures. This last option seems to commit us to some version of substantive realism. We should thus assume that our procedures are incomplete and revisable in such a way that we might identify those factors that would improve our methods. And then, what are the reasons to prefer some procedures? Are they independent of the procedure itself? In any case, it could be argued that this line of reasoning moves away from the real intentions of both Putnam’s insistence on procedures and his criticisms of Habermas’s discourse ethics. The real concern is that of emptying moral claims of any content while preserving an extreme minimalist conception of ethics. The question is then how to fill ethical reasons with content while avoiding cautiously the postulation of intrinsically moral facts. (A) Our maxims frequently incorporate thick ethical terms. Habermas would accept that values play a role in formulating those maxims that arise from contingent forms of life and are subject to the universalization test. Again, this cannot be the issue at stake. Rather it is a matter of how to understand the validity of ethical maxims with thick terms regardless of the procedures that would lead us to judge them from a moral point of view. Putnam needs to claim that these kinds of evaluative judgments are in a sense objective before entering the discourse. An IA is here available. Although conditions in moral discussion, the rules of discourse ethics would be empty without incorporating thick ethical concepts; then, we have to presuppose that these concepts have some kind of cognitive content. Thick ethical vocabulary is inextricably linked to the very description of the procedures we use to discuss and evaluate our normative statements. If we would not assume that they are objectively correct or incorrect, then it would be in danger the validity and

Truth and Moral Objectivity

281

objectivity of the outcomes of our moral procedures. “Relativism of any kind with respect to value-terms cannot leave the objectivity of “norms” unaffected” (Putnam 2002a, p. 120; 2002b, pp. 308-309). But as we have seen, the IA does not say anything about what we understand by objectivity of evaluative judgements. It only shows that they must be treated as objective if we do not want to renounce the objectivity in the whole domain of ethics. Following Dewey at this point, Putnam would claim that it is to be subject to criticism and rational discussion what converts the valued within a community into the valuable.6 But this was precisely the point of the procedural realist. (B) Another possibility would be to recognize that our procedures embody certain substantive assumptions as, in Habermas’s case, the nonrestrictive inclusion of all the affected and the mutual exchange of perspectives in examining the moral correctness of maxims. Putnam conceives of these sorts of norms as non-formal conditions in our procedures (1990a, p. 224; 1994, pp. 154-155). Even if we concede that our conflicts of value and the discrepancies between the life-worlds of different communities are hard to solve, we may yet admit that we have in fact arrived at some agreements about these substantive conditions of our procedures. The question is then how to explain the emergence of such agreements. How do the most general truths about normativity appear as having normative authority for us? Putnam claims that it is not possible to derive principles from reason itself; normative truths are not a priori principles of reason. Nonetheless, he does defend the Kantian intuition that norms are “discovered not by mere trial and error, but by normative reflection on our practice” (1994, p. 168). Reflection engenders corrigible knowledge of those conceptual truths that are presupposed in our practices. These truths bring to light the presuppositions guiding the learning and experience of the participants in the ethical inquiry. Reflection takes place within the process of an open and cooperative inquiry. Conceptual truths about what is rational are not extracted by a transcendental reflection from the reason itself, but from the very process

6

In his reply at the conference at Santiago de Compostela, Putnam insisted that he is more interested in Dewey’s recovery of the Aristotelian idea that “the judge of the good life is the good man.” However, such an idea of a judicious or prudent man as a norm of moral judgments leaves the problem unresolved. According to Aristotle himself, it is the adequacy of the process of practical deliberation that offers genuine moral reasons. What is at stake here is how to properly build the procedures of moral reasoning. We find valuable what a good man does because it is the virtuous outcome of the forms of moral reasoning.

282

Francisco Javier Gil Martín and Jesús Vega Encabo

of research. So, the validity of a procedure cannot be unfolded in terms of being coherent with a priori norms of reason. Our standards and procedures are developed through criticism from inside the practices of moral agents. This pragmatist trend is crucial in understanding how we can improve substantively our procedures. Again, they are valuable insofar as they are accepted after a continuous process of inquiry and criticism. But that still does not answer our question about the agreement regarding the procedures and the set of abstract values embodied in very general procedures. If the agreement must grow from within the activities performed by moral agents, there must be some questions and solutions that are not in dispute. Putnam always argues from the impossibility of a global skeptical doubt, but in no place he expounds how it is possible that we really develop a moral point of view as an impartial means of solving ethical conflicts. It is not enough to charge contextualism with incoherence; it is also needed an explanation about why we ought to accept the conditions of a moral point of view. The IA should not assume that this could not be put into doubt. In fact, our recent past and our present places us before numerous cases in which an impartial moral point of view was not or is not accepted as the rational way to discuss moral conflicts. And we are yet confronted with difficulties to explain the improvement of procedures. Habermas would answer that we already presuppose, as a kind of anticipation facilitated by the very moral point of view, the inclusion in the discourse of people whose interests could be violated in the resolution of a conflict. He believes moreover that the mechanism for moral learning processes is internal to the pragmatic structure of rational discourse. We know that we cannot think of the success of the procedure irrespective of the procedure itself, and also that it is difficult to secure that reflection would lead us to the best procedures. To ask about the best conditions for the improvement of our procedures is not to adopt an external and transcendental point of view. The difficulties arise from the very rational acceptability of our practices of ethical valuation. Even if we suppose that there are some values and criteria that could be shared by different communities or ways of life, we still do not know what the correct resolutions of their conflicts are. And in any case, we could not guarantee that the deliberation would lead, in the domain of the procedures, to revise our standards of rationality toward the presuppositions of general inclusiveness and mutual perspective-taking.

Truth and Moral Objectivity

283

Or is it empirically true that societies which do not organize the inquiry in a democratic form are irrational? Or is it rather a moral truth, insofar as certain norms of inquiry can be considered as obligations and, even, moral obligations? What does it mean that to organize our inquiry by some principles, like those of discourse ethics, has normative authority? We tend to think that either we read the IA as a transcendental argument 7 or we need to posit some primitive obligations regarding the search of truth and the search of goodness. And it is not clear whether Putnam is willing to make such a radical move.

6. Conclusion Cognitivism in ethics does not require a moral substantive realism that postulates intrinsically normative properties. Putnam and Habermas agree on this point. Having concentrated selectively on only one of the “legs of the table,” according to Putnam’s image of ethics as a system of interrelated concerns (2004, p. 28), we have maintained that their attempts to explain moral objectivity by recourse to procedures, viewed from a pragmatist perspective, are not far apart. The way we carry out the inquiry is in itself open, cooperative, fallibilistic. The objectivity of moral statements cannot be understood independently of the best reasons we have as moral agents. Nonetheless, the internal skeptic could bring up the difficulties to solve some ethical disagreements, especially when the discussants do not agree on the minimally good (substantive) procedures to solve practical problems in general. We cannot discover any necessity in the idea of rationality as the result of our involvement in an incessant and inclusive discussion. How do we accept the (rational, and even moral) obligations to conduct our inquiry under the principles of a democratic and unblocked scientific order? This is not a question about why we are rational or ethical beings at all. It is a question internal to some conflict, a question that has been answered either by appealing to our transcendental intimacy with obligations of rationality – as Habermas has suggested – or by being confident in the results of our historically shaped human experience. Undoubtedly, we have left aside one crucial point in Putnam’s thoughts about ethics. Maybe there are other ways of “bearing on reality”

7 That is the case of Habermas’ repeated arguments concerning the lack of “functional equivalents” and the “normative” content of the inescapable conditions of communicative praxis.

284

Francisco Javier Gil Martín and Jesús Vega Encabo

and responding the demands made by others. Moral life is shaped by the experience of certain demands made on us by other sentient life. There are even facts that correct our mistakes in reasoning, facts regarding “the cries of the wounded,” facts that generate also primitive moral obligations. Putnam (2004, p. 24) writes that, “for Levinas, the irreducible foundation of ethics is my immediate recognition, when confronted with a suffering fellow human being, that I have an obligation to do something.” The human being, an actual existing person, is then immediately valued. If we don’t have reasons to arise skeptical worries, this appearance of normativity seems enough to obligate me. But this is the philosophical position that authors like Nagel has identified as a modern version of substantive moral realism. Only stupidity of mind, as Samuel Clarke said, or corruption of customs can lead us to doubt about things so naturally plain and self-evident.

Acknowledgements We appreciate the comments and criticism of Hilary Putnam in response to our paper, presented at the Second Meeting on Pragmatism. Hilary Putnam’s Pragmatism (Santiago de Compostela, Spain, 13th -14 th May, 2004). We are indebted to Fernando Broncano and Carlos Thiebaut for helpful discussions. Cristina Lafont and Axel Müller provided us with insightful remarks and inestimable information. We are also grateful to David Jamieson for improvements to the English text. Javier Gil benefited from a postdoctoral grant at Northwestern University, funded by the Spanish Government (MECD). Javier Gil benefited from a postdoctoral grant at Northwestern University, funded by the Spanish Government (MECD), and from a grant provided by the Caja Madrid Foundation, Spain.

Universidad de Oviedo Departamento de Filosofía Facultad de Filosofía Campus de Humanidades 33011 Oviedo, Spain e-mail: [email protected]

Truth and Moral Objectivity

285

Universidad Autónoma de Madrid Departamento de Lingüística, Lógica y Filosofía de la Ciencia Facultad de Filosofía y Letras Campus de Cantoblanco 28049 Madrid, Spain e-mail: [email protected]

REFERENCES Habermas, J. (1988). Nachmetaphysisches Denken. Frankfurt: Suhrkamp. Habermas, J. (1991). Erläuterungen zur Diskursethik. Frankfurt: Suhrkamp. Habermas, J. (1992). Faktizität und Geltung. Frankfurt: Suhrkamp. Habermas, J. (1999). Wahrheit und Rechtfertigung. Frankfurt: Suhrkamp. Habermas, J. (2002). Werte und Normen. Ein Kommentar zu Hilary Putnams Kantischem Pragmatismus. In: Raters and Willaschek (2002), pp. 280-305. Korsgaard, C. (1996). The Sources of Normativity. Cambridge: Cambridge University Press. Putnam, H. (1987). The Many Faces of Realism. La Salle, IL: Open Court. Putnam, H. (1990a). Realism with a Human Face. Edited by J. Conant. Cambridge, MA: Harvard University Press. Putnam, H. (1990b). A Reconsideration of Deweyan Democracy. Southern California Law Review 63 (6), 1671-1698. Putnam, H. (1994). Words and Life. Cambridge, MA: Harvard University Press. Putnam, H. (1995). Pragmatism: An Open Question. Oxford: Blackwell. Putnam, H. (1999). The Threefold Cord: Mind, Body, and World. New York: Columbia University Press. Putnam, H. (2001). Werte und Normen. In: L. Wingert and K. Günther (eds.), Die Öffentlichkeit der Vernunft und die Vernunft der Öffentlichkeit, pp. 280-313. Frankfurt: Suhrkamp. Putnam, H. (2002a). The Collapse of the Fact/Value Dichotomy and Other Essays. Cambridge, MA: Harvard University Press. Putnam, H. (2002b). Antwort auf Jürgen Habermas. In: Raters and Willaschek (2002), pp. 306-321. Putnam, H. (2002c). Pragmatism and Nonscientific Knowledge. In: J. Conant and U.M. egle (eds.), Hilary Putnam. Pragmatism and Realism, pp. 14-24. London & New York: Routledge. Putnam, H. (2004). Ethics without Ontology. Cambridge, MA: Harvard University Press. Raters, M.-L. and M. Willaschek, eds. (2002). Hilary Putnam und die Tradition des Pragmatismus. Frankfurt: Suhrkamp.

Óscar L. González-Castán THE FACT/VALUE ENTANGLEMENT AS A LINGUISTIC ILLUSION

ABSTRACT. This paper addresses four different, albeit related, issues concerning Hilary Putnam’s fact/value entanglement theory. It is organized around a series of questions. First, what set of problems does the theory attempt to resolve? The revision of the empiricists and the positivist traditions will be the main concern of this first section. Secondly, how does the notion of “fact/value entanglement” solves these problems? I argue that while Putnam has persuasively defended the possibility of an entanglement between descriptive functions and evaluative functions at the level of language use, he nevertheless provides us with only indirect indications about the existence and properties of the bearers of this kind of entanglement; namely the assumed existence of thick ethical concepts and their necessity for some central human practices. These indirect signs are not sufficient to buttress Putnam’s entanglement thesis between facts and values. Thirdly, what are the most appropriate candidates to play the role of bearers of the description/evaluation entanglement? I challenge the claim that the ontological notion of entanglement between facts and values, itself obscure, is indispensable. Instead, I argue that we can substitute the idea of fact/value entanglement for the simpler, and less problematic, notion of human action. Finally, does Putnam commit a petitio pincipii when he attempts to lay down the criteria by which to judge if a thick ethical concept is being appropriately used in a given situation? I argue that Putnam faces what I shall call “Nietzsche’s problem” – the problem of not prejudging, in giving the relevant examples for the discussion, how the situation should be morally valued.

This paper was written thanks to the Plan Nacional I+D+i of the Spanish Ministry of Education within the Research Project HUM 2007-60464.

In: M.U. Rivas Monroy, C. Cancela Silva, and C. Martínez Vidal (eds.), Following Putnam’s Trail: On Realism and Other Issues (Pozna Studies in the Philosophy of the Sciences and the Humanities, vol. 95), pp. 287-305. Amsterdam/New York, NY: Rodopi, 2008.

288

Óscar L. González-Castán

I. Putnam develops the fact/value entanglement theory out of a profound dissatisfaction with the previous state of the art concerning the relationship between facts and values as it had been settled by the empiricist tradition. According to this tradition, at least in its most popular forms, there is a dichotomy between facts and values. Such dichotomy boils down to the idea, many times attributed to Hume, that you cannot derive “oughts” from “ises.”1 A set of non-moral premises cannot entail a moral conclusion. Therefore, a sharp undisputed distinction exists between factual assertions and moral judgments or, more generally, between factual judgments and evaluative judgments. Putnam’s criticism of the fact/value dichotomy is partly based on the idea that the concepts of fact, and of cognitive meaning, that the empiricist tradition has used to draw this dichotomy are too poor to cope with certain intuitions of ordinary language and with some general human practices. Words that express the so-called thick ethical concepts such as ‘cruel’, ‘brave’ or ‘compassionate’ have meaning, one that the empiricist and positivist traditions cannot recognize given that they refer neither to sense impressions nor to logical or mathematical entities nor to theoretical entities. Words of this kind, they assert, merely convey the emotional reactions that certain circumstances cause, for whatever reasons, in a subject. Only these objective circumstances – and not the feelings or passions that they invoke – can be described in purely objective terms, and, hence, be subject to rational discussion and justification. Hume, for example, claims that the relationship between the facts that constitute the objective world and the passions that they elicit in the emotional screen of the subject is a special case of the relationship between primary and secondary qualities ([1739-1740] 1973, III, i, 1). Cruelty is not the property of an objective fact of the world and, consequently, a property that exists independently of us. Rather, ‘cruelty’ is the word that we use to express characteristic feelings of dislike that objective facts arouse, given the way we are contingently made by nature. However, we tend to project these feelings of dislike onto the object, and invest it with them as if the object had the property of cruelty. And if cruelty is a negative value, a “vice,” then values are also subjective, since the passions on which they are based are subjective.

1

Some authors have challenged very convincingly the attribution of these ideas to Hume. See, for example, MacIntyre (1971) and Hampshire (1949).

The Fact/Value Entanglement as a Linguistic Illusion

289

Furthermore, passions, Hume claimed, are “original existences” that are neither rational nor irrational, at least prima facie ([1739-1740] 1973, part III, i, 1). Properly speaking, only the judgments that affirm whether the object of the passion exists, or whether there are the means to achieve such an object, considered as the end of action, are rational or irrational. Since, for empiricism, ethics is related to our passions and subjective values, ethics cannot be a domain for rational discourse. Only judgments about matters of fact can be rationally justified. Judgments about passions cannot. An important consequence of the empiricist and positivist position is that feelings, sentiments, and a fortiori, values have no possible correspondence to facts, and, hence, statements in which we express our emotions and value judgments can be neither true nor false, neither justified nor unjustified. The idea of a correspondence between our values and the facts, or the circumstances, that we value should be wrenched from any respectable philosophical theory. Facts do not ontologically support our valuations. Human subjects simply project value and meaning on an uncaring and senseless objective world of (natural) facts. This leaves the door open to non-cognitivist theories of value. And once we eliminate Hume’s metaphysical statement that a universal human nature exists which guarantees that normal, sane individuals will experience similar emotional reactions to the ones that Hume himself, and a specific eighteenth-century English culture had, the argument immediately gives way to relativism. Different cultures and individuals hold different values that can be neither rationally justified nor criticized because there is no fact of the matter about them. And there is nothing we can do to avoid it. Accordingly, there is no neutral, objective set of values.

II. Putnam rejects this position as well as logical positivism’s variations on the theme. Following Iris Murdoch and Bernard Williams, among others, he asserts that terms expressing thick ethical concepts do not denote merely contingent, subjective values in the sense of expressing unjustified attitudes, with no possible grounding in a non-subjective reality, towards a state of affairs or a person. They also have a characteristic descriptive function as it occurs, for example, when we say that “Nero was cruel.” This sentence would contain, for example, a description of Nero’s character.

290

Óscar L. González-Castán

This supposedly descriptive use is attached many times to an evaluative and normative one. For example, to say that Nero was cruel is to both describe and evaluate the kind of person Nero was at the same time. In this sense, we can say that Putnam is defending the existence of a description/evaluation entanglement at the pragmatic level of discourse (what we intend to communicate through words like ‘cruel’). Moreover, to say that Nero was cruel is to say that Nero should not have been cruel, because being cruel is wrong from the relevant moral point of view. In this point Williams, Putnam, and McDowell agree about this thesis. Picking up the content of a thick ethical word requires to share – or at least to understand – its evaluative point (Williams 1985, pp. 141-142; McDowell 1998, pp. 70-71). It requires, as Putnam puts it, to learn “to imaginatively identify with a particular evaluative outlook” (Putnam 2004, p. 69). Hence, justification in the realm of values is internal to a set of values, although the set may be subject to change, enhancement, correction, or even elimination in the long run. In Putnam’s view, this application of his internal realism should not be incompatible with the idea that thick ethical concepts are “world-guided” in that their right application is also determined by how the world is like (Scheffler 1987, p. 414). If terms that express thick ethical concepts appear in descriptive sentences, then there is some sort of fact or property, although not necessarily a supernatural or non-natural one, that is being described by these concepts (Putnam 2004, p. 69). For the motto dictates that sentences are only truly descriptive and, correspondingly, descriptively true, if there is a fact that is being described. Hence, thick ethical concepts must refer to a certain kind of fact, though these facts are neither things of which we can have a sense impression, nor mathematical nor theoretical entities. Rather these facts are intermingled or encrusted with values instead of being simply facts that we think or feel as having value for various reasons that have to do more with the peculiar constitution of our subjectivity than with the objective constitution of the world. They are not just facts that happen to have value for us as if we were adding value to them, but are “value/facts,” as it were, whose value we must recognize if we do not want to miss an objective feature of the world. That Nero is cruel is one of these value/facts. Putnam’s position represents a Copernican revolution in relation to the empiricist and positivist traditions. Such a radical change obliges us to liberate the notion of fact from the straightjacket within which this tradition had held it bound, for we must honor the intuition that terms

The Fact/Value Entanglement as a Linguistic Illusion

291

such as ‘cruel’ form part of sentences that express value judgments and descriptive judgments at the same time, and in an indivisible way. The recognition that descriptions and evaluations often go hand in hand suggests that rational discussion should also be possible in the arena of the new and broader conception of fact. We can only discuss rationally when we discuss about facts. And when we argue about facts, we have very good prospects of saying either true or false things about them given that, in principle, there is no reason to assume that we cannot know what the facts are. Thus, according to Putnam, ethical sentences, and evaluative sentences generally, are either true or false, closing the door to non-cognitivist theories of value that state that ethical sentences establish no factual claims. As we can see, Putnam’s revision of the concept of fact would lead us to further revise, in domino fashion, the concept of value, the relationship between facts and values, and the role that rationality plays in the world of values. Putnam’s central idea concerns the inherent entanglement of facts and values. I take it that the slash between the words ‘fact’ and ‘value’ in the complex expression ‘fact/value entanglement’ is meant to express the idea that we want to talk about the kind of “value/facts” that I have mentioned above. If such facts exist, they would be ones that do not acquire value from the outside, from the subject’s perspective, for example, but facts that, under certain conditions, have a recognizable and objective value. The main defense of the entanglement thesis involves a peculiar variation of the Kantian transcendental strategy: namely that we cannot get rid of this entanglement because we cannot get rid of thick ethical concepts, and of the linguistic terms that we use to express them, and that have both evaluative and descriptive functions. Putnam bases his claims on the more general indispensability argument that he advances in order to redefine the notions of “objectivity” and “existence,” and to take distance from what he calls “metaphysical realism” (Putnam 1992, pp. 138-139). Something is objective when our best scientific theories and our common practices would be completely altered to the point of becoming unrecognizable and unfeasible, if we did not accept it as an objective fact. Since normative and evaluative discourse is indispensable to social and personal life, as well as to science, values must be objective. In other words, values are objective because our practices require them. And if they are objective, then they have to have some kind of existence as facts. And, since Putnam is neither a radical skeptic about our capacities to know the world (although he accepts that there are facts that human beings will never know), nor an idealist who posits an

292

Óscar L. González-Castán

unknowable noumenal world, then, in principle, we can know these facts, much in the same way that we can know everything else – through the democratization of inquiry. Fact/value entanglement? Can we penetrate this notion and understand what it means? Thus far, Putnam has only given us indirect indications – a few external signs – for the existence of this kind of entanglement. They include the putative existence of thick ethical concepts, and their indispensability for certain significant human practices. In what follows, I do not challenge the idea that there are words, such as ‘cruel’, that we use with both descriptive and evaluative purposes. Nor shall I criticize the idea that these words play an important role in language and in certain social practices in which language is essential. Nevertheless, what remains problematic is the inference from the recognition of how people at different times and in different cultures use thick ethical concepts, (i.e., as if the concepts really describe some traits of reality that are thus invested with value), to the idea that a fact/value entanglement therefore actually exists without any further ado. In other words, you cannot travel for free from the useful reconstruction of the logic and pragmatics of thick ethical concepts to the full and unproblematic ontological recognition of an entanglement between facts and values that would buttress the logical and pragmatic properties of these kinds of words. This difficulty leads us to the third problem, namely, the identification of the most appropriate candidates to be the bearers of the description/evaluation entanglement.

III. Although common language attributes both descriptive and evaluative dimensions to expressions like ‘Nero was cruel’, Putnam does not say much about what kind of “things” thick ethical concepts both describe and value, and why it is that these “things” cannot be described and valued in two separate acts – a cognitive act related to the description of facts, and a non-cognitive one related to the evaluation of them. Thus, the relevant question is this: What is there in the world that cannot be described without also being valued, and valued without also being described? Do fact/value entanglements attach to natural states of affairs, persons, traits of character, actions, acts of will, consequences of actions, mental states, or a mixture of these? Drawing from Putnam’s examples, and the fact that we are dealing with ethical concepts, it is not surprising that Putnam’s most adequate candidates spring from our ethical life, and

The Fact/Value Entanglement as a Linguistic Illusion

293

includes motives for action, acts of will, situations, and persons (Putnam 1992, p. 166). But Putnam devotes little thought as to why they are the most appropriate candidates for bearers of some descriptions that are inseparable from evaluations, and vice versa, i.e., for bearers of a description/evaluation entanglement. Saying that these bearers are fact/value entanglements appear to be begging the question, and does not throw light on the problem. In other words, Putnam’s pragmatist thesis concerning the existence of fact/value entanglements does not tell us what ontologically supports a description/evaluation entanglement. And I take it that the striking title of Putnam’s most recent book, Ethics without Ontology, should not lead us to conclude that we are no longer entitled to talk about bearers of the description/evaluation entanglement. Nothing in the book suggests that we should not. Again, the main point of Putnam’s argument is that in talking about cruelty or compassion we are describing properties, although they are neither supernatural nor non-natural ones (Putnam 2004, p. 69). Since my argument does not hinge on the acceptance of supernatural nor of non-natural properties, it is not affected at all by the idea of an ethics without ontology. Putnam refuses to divide thick ethical concepts into two parts, such as into a descriptive component that would state the matter of fact, and an attitudinal component that would express an emotion or volition as in non-cognitivist accounts of “good” and “ought.” But the main problem is not that thick ethical concepts cannot be factorized; it is not a conceptual problem. Neither is the issue one of whether there are linguistic expressions that are both evaluative and descriptive at one and the same time; it is not a linguistic problem. The real problem is ontological – the existence of something in the world whose very constitution does not allow the factorization of the concepts through which we think about it, nor does it allow to express these concepts with terms that are not descriptive and evaluative at the same time and in an indivisible way. I propose a tentative solution to this problem. I begin with the premise that what ontologically supports a description/evaluation entanglement cannot be a fact in the empiricist sense. Indeed it need not be conceived of as a fact at all, not even in the non-reductive sense. I construct a notion of value, and of moral value, in which the dichotomy between descriptions and evaluations does not make sense, but in which we do not need to appeal to a fact/value entanglement. I argue that human acts are the most appropriate candidates to play the role of bearers of the description/evaluation entanglement in ethics. In plain, ordinary language, as well as in more sophisticated philosophical language, we say that actions are cruel or compassionate. What we need is an account

294

Óscar L. González-Castán

of actions that can accommodate the sound intuitions of Putnam’s theory without its attendant problems. Human acts go beyond the limits of the empiricist notion of fact. Certainly, acts are neither mathematical nor theoretical entities. And although some aspects of human actions can be the object of sense impressions, what we can call their “physical execution” (seeing, for example, the physical movements of the body of someone who is hitting or torturing someone else), nevertheless perception of the physical execution is not the same thing as perceiving a human action. It is to perceive a physical movement that can be correctly described in purely naturalistic terms, (i.e., with words that do not understand the act in terms of its meaning, purpose, and moral standing). I think Putnam would agree (Putnam 1981, p. 145). But apart from the physical execution of the act, we need at a minimum to distinguish between the external act, (not to be confused with the physical execution), and the internal act. The external act and the internal act form a unity, a single human act, and it is this unity that is the subject matter of both our moral judgments and of the efforts of philosophers when they are interested in providing theories of action and of morality. The external act is morally what it is because of the object of the act. The internal act is morally what it is because of the end of the act. Take, for example, the giving of alms out of motives of vainglory. Giving alms would be the object of the external act. We morally evaluate this external act from its object. We can also think of the execution of the external act as the realization of the intention of the agent. And we identify the intention of the agent when we correctly answer what the agent wants: giving alms. The end of the internal act would be, in the example, our personal vainglory. And we morally judge this internal act in virtue of its end. We can conceive of the execution of the internal act as the realization of the motive of the agent. And, correspondingly, we identify the motive of the agent when we answer why the agent wants what he wants: because of his vainglory. The external and the internal act form a unity because an agent cannot act so as to fulfill his intentions without coming into play his motives as well, whatever these are. According to some classical theories of human action, both the internal and the external acts have to be morally good for the act, as a whole, to be called “good.” We might say that giving alms in pursuit of personal glory is a morally bad act. I shall not develop this, except to say that it involves a mixture of Aristotle, Aquinas, and, perhaps, our

The Fact/Value Entanglement as a Linguistic Illusion

295

common, everyday judgments, as well.2 For my purposes it is enough to say that this sort of theory requires two kinds of concepts. The first kind of concepts is internal to the theory such as “physical execution,” “external act,” “internal act,” “object,” “end,” “intentions,” “motives,” etc. The theory uses these concepts with the aim of attempting to describe human action in an objective way, (i.e., in both its main and abstract constituents, so that we do not forget anything that is important for the step by step description of an act). We might want to force the philosophical language a bit and say that these types of words are part of the thin vocabulary of the moral theory. At this thin, albeit very important, level of the theory we have not yet decided if a particular external act, or a particular internal act, or both are good or bad, cruel or compassionate. We only say that there are human acts with all the parts we have distinguished for purposes of a reflective and analytical philosophical discussion. The problem is whether we can fully describe any human act without also saying something about its morality, amorality or immorality as well. Two possible kinds of answers exist. Hume, for example, thought that you could give a full account of human action without talking about the morality of the action. This is revealed in part by the fact that Hume devoted separate books of his Treatise to the problem of human action ([1739-1740] 1973, Book II) and to the problem of the morality of human action ([1739-1740] 1973, Book III). Nevertheless, as MacIntyre and Nussbaum have argued, this strategy would have made no sense in the classical theories of Plato, Aristotle or Aquinas who defend that the description of a human action automatically entails an evaluation of its morality. For them, a theory of morality is an essential part of a general theory of human action. If this position is adopted, then concepts of appraisal, many of which are thick ethical concepts, become necessary. And this is the second kind of concepts that the theory needs. Descriptions of human actions should be satisfactory from a human, moral perspective. Our thick ethical vocabulary serve this purpose. Thus, we have good, complex, and interesting examples in the history of philosophy of concepts that intend both to describe and evaluate actions at the same time, and in an indivisible way. Another question is whether or not the moral evaluation of an action is a genuine description of its moral value. Evaluations have to be warranted in order to be descriptive.

2

An account of Aquinas’ general theory of human action and of the distinction between internal and external acts can be found in MacInnerny (1992).

296

Óscar L. González-Castán

Nevertheless, the point here is that, for these types of theory, the notion of fact/value entanglement is superfluous. This is so because they do not take positivism as the starting point of discussion and, hence, they do not pose the problem only in terms of facts and values. They also need the notion of human action, which is a different category from that of both fact and value. Yet actions are not mysterious entities like the platonic entities that Putnam wants so eagerly to disqualify, which “somehow guarantees or stand behind correct judgments of the reasonable and unreasonable” (Putnam 2004, p. 70). If there is something mysterious, it is the entanglement between facts and values. And if this entanglement is not a platonic entity, which certainly it is not, it is at least a very strange philosophical notion. This clarification is important if we want to develop a position similar to the fact/value entanglement but without the troubles it involves. First of all, the classical theorists avoid the factorization of acts into a descriptive side and an evaluative side. We have seen that merely describing the objective physical movements that someone makes when he or she acts, is not sufficient to fully describe it. At best, it would only be a description of the most external aspect of the act; its facade, so to speak. This is not even to describe what I have called the “external act,” although the external act is an observable performance, at least in many cases, as in the alms example. If we want to provide a fully-fledged description of an act, we need to look “behind” what can be seen, and say something about the intentions (the objects of the act) and the motives (the ends of the act) of the agent who performs the action. These intentions and motives cannot be seen in the physical sense, but they remain an essential and inseparable part of the act, and we have to account for them if we want to describe the act as a whole. The only language we have to talk about intentions and motives is an evaluative one. This means that actions cannot be described with the same physicalistic language that we might use to talk about the physical execution of the act. We talk about intentions and motives (and thus think about actions) as being good or bad, cruel or compassionate. Hence, if human acts cannot be described without appealing to the intentions and motives of the agent, then they cannot be described without using the same kind of evaluative language that we use to talk about intentions and motives, no matter how independent they may be with regard to the physical execution of the act. Hence, we cannot describe the action as a whole without valuing it, and we cannot value it without saying something about its descriptive properties, quite independently of whether or not we are successful. Thus, human actions

The Fact/Value Entanglement as a Linguistic Illusion

297

are kinds of things that “attract” both descriptions and evaluations in an inseparable way. Yet, this does not necessarily implies a fact/value entanglement as the bearer of the description/evaluation entanglement. It only means that there might be descriptions of “things” that need not be thought of in terms of facts, as is the case of human actions.

IV. This leads us to the ticklish problem of what is involved in the examples Putnam provides to illustrate and assess the moral value of certain situations. I shall contend that he commits a petitio pincipii. He presupposes that the situations he describes have an objective moral value. This strategy allows him to judge from the realistic point of view of the “God’s Eye” anyone who does not accept the moral value that he has already assigned to them. In order to have an adequate description of a situation, one needs to be able to acquire and use the relevant set of concepts. In Reason, Truth, and History Putnam argues that if someone from a completely different culture who had never seen the chairs and tables that decorate one of our living rooms, made a description of the room, he would give a description based on to his own conceptual tools. Yet, his description “would hardly convey the information that a member of this [our] culture would wish to have about the room. His description might consists only of true statements but it would not be adequate” (Putnam 1981, p. 138). In this sense he would describe some facts, but not other facts (for example, the fact that there are tables and chairs). This is one of the arguments that Putnam uses to defend that our perception of facts is conceptually laden. In the same way, if we asked someone who has witnessed some cruel deed that he described such an act, he would made an imperceptive, superficial and inadequate description of the act had he lacked both the relevant concept of “cruelty” and the right intellectual and emotional abilities to apply it to the circumstances. Again, in order to make the “right” kind of description, one needs to learn to “imaginatively identify” with a given moral perspective (Putnam 2004, p. 69). This is not a very convincing argument. Moreover, it relies, in an important respect, on the kind of metaphysical realism that Putnam denies. In these examples we are tempted to occupy the place of the Eye of God, having from the outset all the relevant and true descriptions of what is going on in the world. Putnam’s second example introduces an unrecognized meta-ethical judgment in that the deed has been pre-

298

Óscar L. González-Castán

determined to be cruel – for this is what is perceptive, deep and adequate – and we are expected to share this evaluation in order for the example to work. If we do not, we are blind to reality, for the reality of the situation has already been defined as cruel. And anyone who participates in the interaction is judged according to the previously meta-judgment. This is, I think, tricky philosophy. Both in the examples of the cruel deed as well as that of the living room, Putnam employs, in a metaphysically realistic way, the same concepts that he then considers as relevant to make an appropriate description. This strategy looks very much like a vicious circle, and a petitio principii. It establishes from the outset what needs to be established, (i.e., what concepts are relevant for the description of the act). The examples appear to be metaphysically and epistemologically innocent because, for example, it seems obvious that there are chairs and tables in the room, and nobody would dare to challenge this judgment. But as soon as we move from familiar rooms to value judgments about human actions, the situation might not be so obvious. Can one say whether or not an action is cruel, cowardly, or humble without prejudging the situation in the provision of the example? The problem arises not only because uncertain or controversial cases exist, in which a range of moral words, such as ‘torture’, ‘humiliation’ and ‘abuse’ (i.e., from least to most offensive), might compete.3 Problems arise as well due to the fact that we often have to choose between different, but opposed, “families” of value words; such as between ‘cowardly’ (with a negative connotation) and ‘humble’ (with a positive one). Thus, we inevitably face what I shall call Nietzsche’s problem – how to handle the existence of radically opposed moral evaluations of, apparently, the same action. Is one and the same external movement of turning the other cheek a cowardly action or a humble action? Nietzsche posits this and similar questions in his book Morgenröte (Nietzsche [1887] 1999, §38) and he claims that the answer depends on the censorship that the customs and habits of a given culture impose on the moral consciousness of the individual that it has formed. According to Nietzsche, in a Christian culture people are trained to feel a pleasant

3 For example, when photographs from the infamous prison of Abu Ghraib became public, the Bush Administration avoided use of the word ‘torture’ to describe the American soldiers’ behavior. Instead, the most it admitted was that there were “abuses” and that some prisoners had been humiliated. Secretary of Defense Donald Rumsfeld stated that “my impression is that what has been charged thus far is abuse, which I believe technically is different from torture. And therefore I’m not going to address the “torture” word” (Sontag 2004).

The Fact/Value Entanglement as a Linguistic Illusion

299

spiritual emotion, despite the possible physical pain, when they turn the other cheek, because doing so is considered morally good, whereas in a heroic society, such as the ancient Greece of Homer’s epics, at least in Nietzsche’s interpretation, such an act might simply be considered cowardice. In itself the action, or in Nietzsche’s words, the instinct on which the valuation is based, is neither good nor bad. In Nietzsche’s philosophy these problems give rise to lengthy and sophisticated psychological and moral reflections on human actions and moral motives, and contribute to his notion of the Will to Power. Putnam lacks the tools to deal with Nietzsche’s problem. Putnam states with assurance, for example, that terms such as ‘kindness’ or ‘compassion’ are terms of moral praise. This being the case, thick ethical concepts are “action-guiding” because they convey a normative stance (Scheffler 1987, p. 414). As we have seen, these kinds of meta-ethical judgments often provide the basis upon which the rest of the philosophical discussion is based. Certainly a given moral community might, as a matter of fact, use these terms in this way. But if Nietzsche is right, then he has unmasked these accepted and customary judgments. For it may be the case that ‘kindness’ and ‘compassion’ are not unconditionally terms of moral praise in all cases. For Nietzsche, they might simply mask less praiseworthy interests and ideals that are secretly at work behind the scene of our moral consciousness. What we may find behind the backdrop of morality is a “weak” human being and a strong hierarchical structure that distorts, alienates and subdues the “true” life of the individual. This was Nietzsche’s final verdict. Of course, with this short reminder of Nietzsche’s philosophy I am not saying that we should accept Nietzsche’s ideas without any further discussion. In fact I think we should not do it. Nor am I saying that Nietzsche is the one who is necessarily in the right track against Putnam. What I am suggesting is that Putnam must first deal with Nietzsche’s problem before constructing his argument, instead of starting with the unjustified assertion that there are accurate and objective moral evaluations of human life based only on the existence of thick ethical concepts with supposed descriptive functions, with which we cannot dispense. Failing to do so, his argument is vulnerable to claims of metaphysical realism and a petitio principii. Putnam is a pre-Nietzschean moral philosopher. Though this is not a philosophical sin, one must at least address the issue. Otherwise, to claim that ‘kindness’ and ‘compassion’ are terms of praise would be sheer philosophical dogmatism or, to say the least, unjustified normativity. The most you can say is that they are words of praise for a certain group of people. And the

300

Óscar L. González-Castán

question remains as to whether they employ these words as they do because they stem from a larger, inherited set of values that are never questioned. What Nietzsche wants to subvert with his genealogy of morals, is this larger set of values that work behind the scenes of certain western moral and metaphysical traditions. And he is persuaded that this critical enterprise makes philosophical sense. Putnam has some powerful resources available to deal with these criticisms. He defends a bunch of related propositions that could counteract the impact of Nietzsche’s problem on his pragmatism. The meaning given to the paradigmatic value words that we use in our moral judgments is historically determined. Their meaning is internal to a particular group of people whose members have been socialized into a characteristic understanding of their application. And this meaning is neither fixed nor eternal. Moreover no algorithm exists by which to determine in an exact manner whether an action is cruel or not. Lacking an exact and objective means to settle these issues, controversies about value judgments remain unavoidable. Moral disagreement seems to be an intrinsic feature of human life. To put it in another way, the right application of value terms to actions is a matter of interpretation. Despite this, Putnam argues that values are objective, and that their application is not an arbitrary matter. As stated above, the strength of the indispensability argument is one reason why Putnam thinks that values are objective. But is their application not simply arbitrary? Putnam offers no explicit argument in his defense on this score. Instead, he argues by means of a defense of the parallel thesis in the realm of epistemic values, and the implications it has for ethics. Given that both realms have general isomorphic characteristics, important conclusions can be drawn for ethical values from the study of epistemic values. According to Putnam, careful observation of the practice of science reveals that it, too, lacks a neat algorithmic procedure to determine whether or not a theory is simple, coherent, has good empirical consequences, is self-sufficient, and so on. This fact notwithstanding, scientists have to make decisions to determine these questions so as to make a good rational selection among competing theories, which is always a pragmatic one. Following Dewey, Putnam asserts that, fortunately, scientists collectively possess enough experience to endowed them, in general, with a sense that allows scientific communities to determine in an objective way, (i.e., through consensus), whether a theory has the desired epistemic virtues, and whether it exhibits these virtues to a greater degree than a rival. Of course, scientists are not free from making mistakes in formulating these

The Fact/Value Entanglement as a Linguistic Illusion

301

“methodological value judgments” (Putnam 2004, p. 67). Success is never taken for granted. This is why we have to be fallibilists in epistemology, and why the democratization of inquiry is necessary. Nevertheless, one of the outcomes of the scientific enterprise is that we can sometimes say that the theories scientists have sanctioned as simple, coherent and so on, objectively possess these properties, at least prima facie. Value terms in science “stand for properties of the things that they are applied to, and not just for feelings of the person who uses the terms” (Putnam 1981, p. 135). The fact that there is no algorithm to determine these issues means that the properties of the theories about which we judge are fuzzy or imperfect, but it does not mean that they are not objective. Can this argument be applied to ethics? Without having first solved Nietzsche’s problem, I think not. First, we need to clarify what is the relationship between the acquisition of a thick ethical concept, the acceptance by the well-trained members of a moral community of the most widely spread moral judgments of their community, and the claims of objectivity within this community. An illuminating aspect of this problem can be found in the following quotation from Putnam’s book The Collapse of the Fact/Value Entanglement: “They [the proponents of the entanglement theory] maintain that if one did not at any point share the relevant ethical point of view one would never be able to acquire a thick ethical concept, and that sophisticated use of such a concept requires a continuing ability to identify with that point of view” (Putnam 2002, p. 38). This is a very tricky thesis indeed. We might agree, at least in principle, that, in order to acquire a thick ethical concept and apply it correctly, or in a sophisticated way, one must share the worldview of the community. Unfortunately this principle of agreement is undermined by the almost imperceptible word ‘relevant’, but also by the word ‘sophisticated’. I take it that the expression ‘sophisticated use’ means in this context something similar to ‘correct use’, given that the general framework of Putnam’s discussion of values has to do more with the objectivity of values and the correct way of dealing with them, (i.e., the way that allows us to describe them), rather than with our virtuosity in using and applying them, although they are not unrelated. Thus, in what follows the adjective ‘sophisticated’ can be substituted by ‘correct’. Both adjectives function here as terms of praise or, rather, as terms of “metapraise” for they describe how someone uses first-order terms of praise, such as ‘cruel’. When I employ the term ‘correct’ it is to make the discussion clearer.

302

Óscar L. González-Castán

Thus, let us start with the idea that we can legitimately say that there is a correct or a sophisticated use of first-order moral predicates. What does it mean to apply correctly a thick ethical word in a given situation? It might mean either of two different things. First, ‘correct’ might mean that the action or the situation is such that, if we do not apply the concept, then we are not rightly describing a morally objective feature of a given action or situation. In this case, the use of the appropriate thick ethical word in a moral judgment about the action or situation is a truly descriptive one that generates truth and knowledge. This is a strong sense of ‘correctness’ that conveys an objectivist model of ethics, because it portrays people as seeking the truth about values (Gibbard 1992, p. 142). Secondly, it might mean that people who teach us how and when and why to use it, do agree, as a matter of fact, with us about how, and when and why we use it. This is a weak sense of ‘correctness’. In this second case, saying that the thick ethical word has a descriptive use might mean that it indirectly describes, for the attentive interpreter, the idiosyncratic way in which a particular society or culture morally judges the world. For in saying that a person uses a thick moral word correctly we are describing how a given group of people tend to use this word in connection with the action that is being morally judged. Thus, correctness here means that there is a coherent match between how the group and the individual judge the action. They share a sensibility, and, in so doing, accept a roughly similar system of norms. In this sense, thick moral concepts “are encrusted with the thickest of cultural deposits” (Blackburn 1992, p. 285). They are even “too thick” because they carry “such a complex drapery of religious, cultural, and ritual history that thinking of [them] as tied to just one attitude is impossible” (Blackburn 1992, p. 296). All this suggests a non-objectivist or, at least, an alternative model of thick ethical concepts, given that they are just “cultural artifacts” that people have come to inhabit without thinking about how and when and why they did so (Gibbard 1992, p. 142). Of course, the strong and the weak senses of ‘correctness’ are partially related. Agreement can be a good sign of world-guidedness. But agreement per se is not always sufficient for world-guidedness. And disagreement with the community of reference does not necessarily mean being out of touch with the world and its characteristics. If Putnam has the first interpretation in mind, then the problem of metaphysical realism, once again in the form of the “Eye of God,” arises. He takes for granted what remains contested; namely, that a given situation has certain objective moral characteristics not subject to interpretation or dispute. What Putnam suggest, then, is that given this

The Fact/Value Entanglement as a Linguistic Illusion

303

objectivity in the world, the moral community can confidently educate the rest in the proper use of ethical terms, as if our tutors and parents did not want that we failed to describe a fact of the moral world. Yet, alternative interpretations, possibly valid, may exist, although the community of reference does not share them. This may be the reason why it holds tightly to the ruins of education and socialization. In this scenario, it is not longer clear what the correct use of the term is, what the correct or sophisticated description is (Scheffler 1987, p. 414). It is not even clear whether or not it makes sense to talk of ‘correctness’ in the strong sense. In this case, the term ‘correct’ can only be understood according to the second, weak interpretation which, by itself, is not a very interesting philosophical claim. However, it seems clear that the interpretation that Putnam supports is the first, strong one. He buttresses this interpretation when he says that “the fact that an act is cruel is available through the lenses of value concepts” (Putnam 2002, p. 119). This is true, but only if the act is cruel. The same problem exists with the word ‘relevant’. It is possible to say that the community’s moral point of view is relevant because it allow us to describe objective characteristics of the world. Alternatively, we could say, following the second interpretation above, that it is relevant simply because the community upholds it. The problems posed by the first interpretation are the same as the ones we have seen with the first interpretation of the words ‘correct’ or ‘sophisticated’. And the second interpretation fails for the same reasons. In sum, to say that there is a relevant ethical point of view that is embodied in a community such that we need to share this viewpoint if we want to make correct or sophisticated moral descriptions amounts to the idea that we need to accept metaphysical realism and reject Nietzsche’s problem. Putnam tries to walk a middle course by saying that “recognizing that our judgments claim objective validity and recognizing that they are shaped by a particular culture and by a particular problematic situation are not incompatible” (Putnam 2002, p. 45). It is true that they are not incompatible a priori but they might in fact be incompatible if we are not convinced in advance that a particular culture uses thick ethical concepts that are relevant, correct, and sophisticated to describe objective characteristics of the action, or of the situation at hand.

304

Óscar L. González-Castán

Acknowledgements I should like to thank the participants in the II Meeting on Pragmatism. Hilary Putnam’s Pragmatism that was held in Santiago de Compostela (May 12 th -14 th , 2004), and specially Prof. Hilary Putnam and Matti Eklund, for their valuable comments on an earlier draft of this paper. I would also like to thank the Organizing Committee (María Uxía Rivas, Concepción Martínez, and Celeste Cancela) for their friendly hospitality.

Universidad Complutense de Madrid Facultad de Filosofía Departamento de Filosofía I (Metafísica y Teoría del Conocimiento) Despacho A-205/2 28040 Madrid, Spain e-mail: [email protected]

REFERENCES Blackburn, S. (1992). Morality and Thick Concepts. Through Thick and Thin. Proceedings of the Aristotelian Society (Supplementary Volume) 66, 285-299. Gibbard, A. (1992). Morality and Thick Concepts. Thick Concepts and Warrant for Feelings. Proceedings of the Aristotelian Society (Supplementary Volume) 66, 265-283. Hampshire, S. (1949). Some Fallacies in Moral Philosophy. Mind 58, 466-482. Hume, D. ([1739-1740] 1973). Treatise of Human Nature. Edited by L.A. Selby-Bigge. Oxford: Clerendon Press MacInnerny, R. (1992). Aquinas on Human Action. A Theory of Practice. Washington D.C.: The Catholic University of America Press. MacIntyre, A. (1971). Hume on “Is” and “Ought.” In: Against the Self-Images of the Age, pp. 109-124. London: Duckworth. McDowell, J. (1998). Mind, Value, and Reality. Cambridge, MA: Harvard University Press. Nietzsche, F. ([1887] 1999). Morgenröte. Sämtliche Werke. Kritische Studienausgabe, vol. III. Edited by G. Colli, M. Montinari. München/Berlin: dtv/de Gruyter. Putnam, H. (1981). Reason, Truth and History. Cambridge: Cambridge University Press. Putnam, H. (1992). Realism with a Human Face. Cambridge, MA: Harvard University Press. Putnam, H. (2002). The Collapse of the Fact/Value Dichotomy and Other Essays. Cambridge, MA: Harvard University Press. Putnam, H. (2004). Ethics without Ontology. Cambridge, MA: Harvard University Press. Scheffler, S. (1987). Morality Through Thick and Thin. A Critical Notice of “Ethics and the Limits of Philosophy.” Philosophical Review 96 (3), 411-434. Sontag, S. (2004). Regarding the Torture of Others. The New York Times Magazine, May 23.

The Fact/Value Entanglement as a Linguistic Illusion

305

Williams, B. (1985). Ethics and the Limits of Philosophy. Cambridge, MA: Harvard University Press.

ABOUT THE CONTRIBUTORS

Juan José Acero is Professor in Logic at Granada University. Philosophy of language and philosophy of mind are his main fields of academic interest, but has also published on 20 th Century analytic philosophers (Wittgenstein, Russell, Carnap, Quine, Austin). He has edited the 16th volume of Enciclopedia Iberoamericana de Filosofía: Filosofía del lenguaje, I. Semántica (Trotta, 1998), and co-edited the monograph issue on Gareth Evans published in 2004 by Agora. He has contributed to F.P. Ramsey: Critical Reassessments, edited by Mª J. Frápolli (Continuum, 2005). José Tomás Alvarado (Santiago de Chile, 1969) is Lecturer in Philosophy in the Institute of Philosophy of the Pontificia Universidad Católica de Valparaíso (Chile). He has worked in Putnam’s internal realism, especially in Putnam’s model-theoretic argument against realism, as well in the general contemporary realism/anti-realism debate. He is author of Hilary Putnam: el argumento de teoría de modelos contra el realismo (Eunsa, 2002) and of several articles about Putnam’s thought. Maria Baghramian is a Associate Professor in the School of Philosophy, UCD, Dublin. She is the author of Relativism (Routledge, 2004) and Hilary Putnam: Language, Mind and the World (Polity Press, forthcoming). She also edits the International Journal of Philosophical Studies. Celeste Cancela Silva studied Philosophy at the University of Santiago de Compostela (Spain) where she is realizing her Ph.D. on Putnam’s philosophy. She has published articles on philosophy of language, philosophy of science and epistemology, all of them connected with Putnam’s thought. Ulvi Doguoglu works at the University of Zurich, Switzerland. He is the author of “Understanding at Limits” (forthcoming) in which he develops a philosophy of what it means to learn and understand practices along lines drawn by Wittgenstein and Putnam. He currently is engaged in a research project on the role emotions play in epistemological processes.

308

Contributors

His fields of specialization include Wittgensteinian pragmatism and the philosophy of “Bildung” or education. Matti Eklund is Associate Professor, Sage School of Philosophy, Cornell University. Speccializes in metaphysics, philosophy of language, and philosophy of logic. Has published articles on, among others, the liar and sorites paradoxes, personal identity, and various aspects of metaontology. Ángel Manuel Faerna (Madrid, 1962), Ph.D. (Universidad Complutense, Spain) with a doctoral thesis on “El pragmatismo conceptualista de C.I. Lewis” (1993). Professor of Philosophy at the Universidad de Castilla-La Mancha. Author of Introducción a la teoría pragmatista del conocimiento (Singlo XXI, 1996), and of a critical edition of selected writings by John Dewey (La miseria de la epistemología [Biblioteca Nueva, 2000]). Co-editor of Identidad, individuo e historia (with Mercedes Torrevejano) (Pre-Textos, 2003), and of Caminos de la hermenéutica (with Jacobo Muñoz) (Biblioteca Nueva, 2006). Francisco Javier Gil Martín teaches courses in ethics and political philosophy at the University of Oviedo, Spain. He received his Ph.D. degree from the University of Salamanca, Spain, and was postdoctoral fellow at Northwestern University, in Evanston, IL, USA. He is co-author (with Jesús Vega Encabo) of the Spanish edition of the debate between Habermas and Putnam on norms and values (forthcoming). Fernando González García studied philosophy at the University of Santiago de Compostela, where he did a postgraduate course and a dissertation on “Truth in C.S. Peirce and W. James.” Since 1990 he teaches philosophy at secondary school. At the moment he is working on his doctoral thesis entitled “Truth, Pluralism and Relativism in William James’s Thought.” He has written several papers on James’s philosophy, among others: “La concepción de la verdad en William James” (in: Formal Theories and Empirical Theories [Universidad de Santiago de Compostela, 2001]), “La verdad como construcción. La propuesta humanista de William James” (in: La ciencia ante el público: Cultura humanista y desarrollo científico-tecnológico [Universidad de Salamanca, 2003]), “El monismo epistemológico no reduccionista de W. James: la noción de ‘experiencia pura’” (Ágora, vol. 21-II, 2003).

Contributors

309

Óscar L. González-Castán is Professor of Philosophy, Department of Philosophy I (metaphysics and theory of knowledge), Universidad Complutense de Madrid, Spain. He is author of two books: La conciencia errante. Introducción crítica a la filosofía de la psicología (Tecnos, 1999) and Las direcciones de la mirada moral (Universidad de Alcalá, 2000). He is also author of numerous articles and chapters on metaphysics, theory of knowledge, and ethics published in Europe and Spain. Russell B. Goodman is Regents’ Professor of Philosophy at the University of New Mexico, studied philosophy at the University of Pennsylvania, Oxford, and Johns Hopkins. He has written American Philosophy and the Romantic Tradition (Cambridge University Press, 1990) and Wittgenstein and William James (Cambridge University Press, 2002), and is the editor of Pragmatism: A Contemporary Reader (Routledge, 1995), Contending with Stanley Cavell (Oxford University Press, 2005), and Pragmatism: Critical Concepts in Philosophy (Routledge, 2005). Christopher Hookway is Professor of Philosophy at the University of Sheffield in the UK. His research is in the philosophy of Peirce and in epistemology, and his publications include Peirce (Routledge and Kegan Paul, 1985), Quine: Language, Experience and Reality (Polity Press, 1988), and Truth, Rationality, and Pragmatism: Themes from Peirce (Oxford University Press, 2000). Jaime Nubiola (Barcelona, 1953) is Professor of Philosophy at the University of Navarra, Spain. He is the author of El compromiso esencialista de la lógica modal (Enusa, 1984), La renovación pragmatista de la filosofía analítica (Enusa, 1994), El taller de la filosofía. (Enusa, 1999), as well as a good amount of papers on philosophy of language, history of analytic philosophy, American philosophy, Peirce and pragmatism. He is the editor of Anuario Filosófico and head of the Grupo de Estudios Peirceanos (http://www.unav.es/gep/). Hilary Putnam is Cogan University Professor Emeritus at Harvard University. His books include Reason, Truth and History, Renewing Philosophy, Pragmatism an Open Question, The Threefold Cord: Mind, Body, and World, The Collapse of the Fact/Value Dichotomy, and Ethics without Ontology.

310

Contributors

María Uxía Rivas Monroy (Vigo, Spain, 1958) is Senior Lecturer at the Department of Logic and Moral Philosophy at the University of Santiago de Compostela. Her research is focused on Philosophy of Language and Semiotics (Fregean semantics, Analytic Philosophy, American Pragmatism). She is author of the book El significado disgregado. La interpretación de M. Dummett de la semántica fregeana (Novo Século, 1994) and co-editor of the book Truth in Perspective (Ashgate, 1998). Chiara Tabet is currently undertaking postgraduate research in analytic philosophy as a Ph.D. student in Pisa (Italy) and St. Andrews (UK), where she is a member of the research Centre Arché. Her main research topic is Constructivism as an ontological and a semantical position in the Philosophy of Mathematics and of Logics, but she is also interested in various issues in the Philosophy of Language (especially contextualism and relativism about meaning) and in the Philosophy of Mind (especially functionalism and computationalism). Jesús Vega Encabo is Associate Professor in the Autonomous University of Madrid (Spain), where he teaches courses in epistemology and philosophy of science. His main interests are in the epistemology and the philosophy of mind. He has published papers about the nature of knowhow and is now working on topics related to the non-conceptual content of experience. Recent publications are also devoted to different problems in the philosophy of science, such as the nature of the scientific representations, the conditions of technological knowledge and the relations between science and democracy.

r o d o p i [email protected]–www.rodopi.nl

The Self-Correcting Enterprise Essays on Wilfrid Sellars Edited by Michael P. Wolf and Mark Norris Lance

This volume presents ten new papers on the work of Wilfrid Sellars and its implications for contemporary philosophy. Contributors run the gamut from established voices in the Sellarsian literature to the newest voices in the field. It addresses topics ranging from cognitive science and philosophy of mind to epistemology and the philosophy of language. This volume is of interest to those studying cognitive development, perception, justification and semantics. It will also be of great interest to anyone following the recent work of John McDowell or Robert Brandom. Michael P. Wolf is an Assistant Professor of Philosophy and currently Director of the Cognitive Science Program at California State University, Fresno.

Amsterdam/New York, NY, 2006 274 pp. (Pozna n´ Studies in the Philosophy of the Sciences and the Humanities 92) Bound € 60 / US$ 81 ISBN-10: 9042021446 ISBN-13: 9789042021440

Mark Norris Lance is a Professor at Georgetown University in both the philosophy department and the Program on Justice and Peace.

USA/Canada: 295 North Michigan Avenue - Suite 1B, Kenilworth, NJ 07033, USA. Call Toll-free (US only): 1-800-225-3998 All other countries: Tijnmuiden 7, 1046 AK Amsterdam, The Netherlands Tel. +31-20-611 48 21 Fax +31-20-447 29 79 Please note that the exchange rate is subject to fluctuations

r o d o p i [email protected]–www.rodopi.nl

Interpretation and Transformation Explorations in Art and the Self Michael Krausz

In this book, Michael Krausz addresses the concept of interpretation in the visual arts, the emotions, and the self. He examines competing ideals of interpretation, their ontological entanglements, reference frames, and the relation between elucidation and selftransformation. “This book marks a decisive moment in the philosophical scholarship on interpretation. Krausz is a unique figure in the current philosophical climate, equally capable of theoretical sophistication, eloquence, and compelling argumentation. Widely acclaimed for his major contributions to interpretation theory, he has now added a crucial dimension to his work, and to the field itself.” Andreea Deciu Ritivoi, Carnegie-Mellon University (Author: Yesterday’s Self; Editor: Interpretation and Its Objects: Studies in the Philosophy of Michael Krausz)

Amsterdam/New York, NY, 2007 XIII-154 pp. (Value Inquiry Book Series 187) Paper € 34 / US$ 46 ISBN-13: 9789042021808

USA/Canada: 295 North Michigan Avenue - Suite 1B, Kenilworth, NJ 07033, USA. Call Toll-free (US only): 1-800-225-3998 All other countries: Tijnmuiden 7, 1046 AK Amsterdam, The Netherlands Tel. +31-20-611 48 21 Fax +31-20-447 29 79 Please note that the exchange rate is subject to fluctuations

r o d o p i [email protected]–www.rodopi.nl

Propositions Semantic and Ontological Issues

Amsterdam/New York, NY, 2006 V-252 pp. (Grazer Philosophische Studien 72) Paper € 52 / US$ 70 ISBN: 9789042021945

Edited by Massimiliano Carrara and Elisabetta Sacchi

This special issue of GPS collects 11 papers (and a long introduction), by leading philosophers and young researchers, which tackle more or less from close the topic of propositions by trying to provide the reader with a cross-section of the ongoing debate in this area. The raised issues range over the semantics, the ontology, the epistemology, and the philosophy of mathematics and stimulate the reader to reflect on crucial problems such as the following: are propositions objects? In the positive case, what kind of objects are they? Can they be grasped by cognitive creatures such as we are? When can we say that two people entertain the same proposition? Have propositions any role to play in speech act theory? Even though the notion of proposition has received considerable attention in the past philosophical debate, it is still of great interest, in particular in connection with the attacks which have recently been launched against it in the theory of language. The volume, which is equipped with a long and detailed introduction that supplies the young reader with useful background information on the different stances in the debate, could prove useful also for didactic purposes.

USA/Canada: 295 North Michigan Avenue - Suite 1B, Kenilworth, NJ 07033, USA. Call Toll-free (US only): 1-800-225-3998 All other countries: Tijnmuiden 7, 1046 AK Amsterdam, The Netherlands Tel. +31-20-611 48 21 Fax +31-20-447 29 79 Please note that the exchange rate is subject to fluctuations

r o d o p i [email protected]–www.rodopi.nl

Grazer Philosophische Studien Vol. 73 – 2006

Amsterdam/New York, NY, 2007 269 pp. (Grazer Philosophische Studien 73) Paper € 54 / US$ 76 ISBN: 9789042022324

Herausgegeben von Johannes L. Brandl, Marian David und Leopold Stubenberg

Inhaltsverzeichnis/ Table of Contents Abhandlungen/ Artic les Werner SAUER: Die Einheit der Intentionalitätskonzeption bei Brentano Tanja PIHLAR: Zur Th eorie der Vorstellungsproduktion (“Grazer” Gestalttheorie I: France Weber) Thane Martin NABERHAUS: Does Husserl Have an Argument against Representationalism? Torsten WILHOLT: Lost on the Way from Frege to Carnap: How the Philosophy of Science Forgot the Applicability Problem John PRESTON: Janik on Hertz and the Early Wittgenstein Friedrich Christoph DOERGE: Re-Denition and Alston’s ‘Illocutionary Acts’ Michael VEBER: Not too Proud to Beg (the Question): Why Inferentialism Cannot Account for the A Priori Anthony Robert BOOTH: Can there Be Epistemic Reasons for Action? Achim LOHMAR: Why Content Relativism Does Not Imply Fact Relativism Stephen HETHERINGTON: So-Far Incompatibilism and the So-Far Consequence Argument Christoph JÄGER & Anne BARTSCH: Meta-Emotions Diskussionen/ Discussions William H. HANSON: The Paradox of Nonbeing Dan LÓPEZ DE SA: Flexible Property Designators Buchnotizen/ Critical Notes

USA/Canada: 295 North Michigan Avenue - Suite 1B, Kenilworth, NJ 07033, USA. Call Toll-free (US only): 1-800-225-3998 All other countries: Tijnmuiden 7, 1046 AK Amsterdam, The Netherlands Tel. +31-20-611 48 21 Fax +31-20-447 29 79 Please note that the exchange rate is subject to uctuations

r o d o p i [email protected]–www.rodopi.nl

Naturalized Epistemology and Philosophy of Science Edited by Chienkuo Michael Mi and Ruey-lin Chen

Much has happened in the eld of contemporary epistemology since Quine’s “Epistemology Naturalized” was published in 1969. Even before Ronald Giere published his article “Philosophy of Science Naturalized,” naturalized philosophy of science had been inuenced by the so-called historical approach. Kuhm, Lakatos, Feyerabend and Laudan all contributed importantly to this trend. In this light it has emerged, without a doubt, that philosophy of science is closely related to epistemology. This volume explores some of the relevant relations and will be of interest to epistemologist and philosophers of science.

Amsterdam/New York, NY, 2007 196 pp. (Rodopi Philosophical Studies 7) Paper € 40 / US$ 54 ISBN: 9789042021983

USA/Canada: 295 North Michigan Avenue - Suite 1B, Kenilworth, NJ 07033, USA. Call Toll-free (US only): 1-800-225-3998 All other countries: Tijnmuiden 7, 1046 AK Amsterdam, The Netherlands Tel. +31-20-611 48 21 Fax +31-20-447 29 79 Please note that the exchange rate is subject to uctuations

r o d o p i [email protected]–www.rodopi.nl

Contemporary Pragmatism Volume 4, Issue 1 Edited by John R. Shook and Paulo Ghiraldelli, Jr.

Contents Special Issue on Cornel West Eddie S. GLAUDE, Jr.: On Cornel West and Pragmatism Thomas MCCARTHY: The Natural Order of Things: Social Darwinism and White Supremacy Eduardo MENDIETA: Translating Democracy or Democratic Acts of Translation: On Cornel West’s Democracy Matters Jeffrey STOUT: A Prophetic Church in a Post-Constantinian Age: The Implicit Theology of Cornel West Jason A. SPRINGS: The Priority of Democracy to Social Theory Marc LOMBARDO: On the Frank Speech of Cornel West’s Prophetic Witness Paul C. TAYLOR: Making Niagara a Cataract: Cornel West, Greatness, and the Music of Ideas

Amsterdam/New York, NY, 2007 IV-115 pp. (Contemporary Pragmatism 4:1) Paper € 24 / US$ 34 ISBN: 9789042022461

USA/Canada: 295 North Michigan Avenue - Suite 1B, Kenilworth, NJ 07033, USA. Call Toll-free (US only): 1-800-225-3998 All other countries: Tijnmuiden 7, 1046 AK Amsterdam, The Netherlands Tel. +31-20-611 48 21 Fax +31-20-447 29 79 Please note that the exchange rate is subject to uctuations

r o d o p i [email protected]–www.rodopi.nl

Essays on Frege’s Conception of Truth

Amsterdam/New York, NY, 2007 VIII-236 pp. (Grazer Philosophische Studien 75) Paper € 50 / US$ 70 ISBN: 9789042021563

Edited by Dirk Greimann

In his writings on the foundations of logic, Gottlob Frege, the father of modern logic, sketched a conception of truth that focuses on the following questions: What is the sense of the word “true”? Is truth a denable concept or a primitive one? What are the kinds of things of which truth is predicated? What is the role of the concept of truth in judgment, assertion and recognition? What is the logical category of truth? What is the signicance of the concept of truth for science in general and for logic in particular? The present volume is dedicated to the interpretation, reconstruction and critical assessment of Frege’s conception of truth. It is of interest to all those working on Frege, the history of logic and semantics, or theories of truth. The volume brings together nine original papers whose authors are all widely known to Frege scholars. The main topics are: the role of the concept of truth in Frege’s system, the nature of the truth-values, the logical category of truth, the relationship between truth and judgment, and the conception of the truth-bearers.

USA/Canada: 295 North Michigan Avenue - Suite 1B, Kenilworth, NJ 07033, USA. Call Toll-free (US only): 1-800-225-3998 All other countries: Tijnmuiden 7, 1046 AK Amsterdam, The Netherlands Tel. +31-20-611 48 21 Fax +31-20-447 29 79 Please note that the exchange rate is subject to uctuations