TERMINOLOGY
TERMINOLOGY AND LEXICOGRAPHY RESEARCH AND PRACTICE Terminology and Lexicography Research and Practice aim...
321 downloads
3068 Views
2MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
TERMINOLOGY
TERMINOLOGY AND LEXICOGRAPHY RESEARCH AND PRACTICE Terminology and Lexicography Research and Practice aims to provide indepth studies and background information pertaining to Lexicography and Terminology. General works will include philosophical, historical, theoretical, computational and cognitive approaches. Other works will focus on structures for purpose- and domain-specific compilation (LSP), dictionary design, and training. The series will include monographs, state-of-the-art volumes and course books in the English language.
Series Editors Helmi Sonneveld Sue Ellen Wright
Volume 1 M. Teresa Cabré Terminology Theory, methods and applications
TERMINOLOGY THEORY, METHODS AND APPLICATIONS
M. TERESA CABRÉ Universitat Pompeu Fabra Edited by Juan C Sager Translated by Janet Ann DeCesaris
JOHN BENJAMINS PUBLISHING COMPANY AMSTERDAM/PHILADELPHIA
8
TM
The paper used in this publication meets the minimum requirements of American National Standard for Information Sciences — Permanence of Paper for Printed Library Materials, ANSI Z39.48-1984.
Original title: M. Teresa Cabré. La Terminolgia. La teoria, els mètodes, les aplicacions. Barcelona, Emúries 1992
Library of Congress Cataloging-in-Publication Data Cabré, M. Teresa (Maria Teresa) [Terminologia. English] Terminology : theory, methods, and applications / M. Teresa Cabré; edited by Juan C. Sager ; translated by Janet Ann DeCesaris. p. cm. Includes bibliographical references (p. ) and index. 1. Terms and phrases. I. Sager, Juan C. II. Title. III. Series. P305.C2713 1998 418--dc21 98-44718 ISBN 90 272 1633 9 (Eur.) / 1 55619 787 X (US) (Hb; alk. paper) CIP ISBN 90 272 1634 7 (Eur.) / 1 55619 788 8 (US) (Pb; alk. paper) © 1999 – John Benjamins B.V. No part of this book may be reproduced in any form, by print, photoprint, microfilm, or any other means, without written permission from the publisher. John Benjamins Publishing Co. • P.O.Box 75577 • 1070 AN Amsterdam • The Netherlands John Benjamins North America • P.O.Box 27519 • Philadelphia, PA 19118 • USA
To Antoni And our children Toni, Albert, and Eva
Contents Acknowledgements
xi
Chapter One an overview of terminology 1. Social and political aspects 1.1 Origins 1.2 Development of the field 1.3 The evolution of modern terminology 2. Scientific and functional aspects 2.1 The theory of terminology 2.2 Terminology, a new practice 2.3 The functions of terminology 2.4 Schools and working methods in terminology 3. Organizational aspects 3.1 Different orientations according to spheres of influence 3.2 Organization of terminology 3.3 International cooperation
1 1 2 5 6 7 9 10 12 14 15 20 22
Chapter Two terminology, an interdisciplinary field 1. Terminology and linguistics 1.1 Theoretical linguistics and applied linguistics 1.2 Variety of the language system 1.3 Lexicology 1.4 Lexicography 1.5 Terminology 1.6 The specificity of terminology 1.6.1 Terminology and lexicology 1.6.2 Lexicography and terminology
25 27 28 29 30 32 33 34 37
viii
contents
2. Terminology and cognitive science 2.1 The analysis of terms 2.2 Concept formation 2.3 The relationships between concepts 2.4 Subject classification 3. Terminology and communication 3.1 Specialized communication 3.2 Terminology and specialized communication 3.3 Terminology and translation 3.4 Terminology and language planning 4. Terminology and documentation 4.1 The relationship between terminology and documentation 5. Terminology, computer science and knowledge engineering 5.1 Computer science at the service of terminology 5.2 The usefulness of terminology for computer science
39 39 42 44 44 45 45 47 47 48 50 51 52 53 55
Chapter Three the foundations of terminology 1. Special languages 1.1 Types of discourse 1.2 General language and special languages 1.2.1 The concept and scope of special languages 1.2.2 General language and special languages 1.2.3 Variation in special languages 1.3 Special language documents 1.4 The role of terminology in special language texts and documents 2. The terminological unit 2.1 Terms as systematic units 2.1.1 The designation 2.1.2 Concepts 2.1.3 The term-concept relationship 2.1.4 Function 2.2 Terms as pragmatic units
56 57 58 59 71 76 78 80 80 81 82 95 107 111 112
Chapter Four terminology in practice: terminography 1. The foundations of terminological practice 1.1 Theoretical principles 1.2 International standards
115 115 116
contents 2. Materials used in terminography 2.1 Reference materials 2.1.1 Documentation on documentation 2.1.2 Documentation on the special subject field 2.1.3 Documentation on terms 2.1.4 Documentation on the research method and presentation of work 2.2 Specific materials for terminographic work 2.3 Support materials 2.3.1 Extraction records 2.3.2 Terminological records 2.3.3 Correspondence records 2.3.4 Query records 3. Working methods 3.1 Systematic searches 3.1.1 Systematic monolingual searches 3.1.2 Systematic multilingual searches 3.2 Ad-hoc searches 3.2.1 The query 3.2.2 The search 3.2.3 General process of ad-hoc searches
ix 116 117 117 117 118 119 121 121 121 123 127 127 129 129 130 151 152 153 155 157
Chapter F ive computerized terminology 1. The concept and scope of computerized terminology 2. Contribution of computer science to terminology 2.1 Computer science and terminological methodology 2.2 Computer science and terminological practice 2.3 Artificial intelligence and terminology 3. Terminology and data banks 3.1 The creation of a data bank 3.2 The organization of the data 3.3 Data banks of interest to terminology 3.4 The evolution and limitations of data banks 4. Terminological data banks 4.1 Definition 4.2 Classification
161 162 162 164 166 168 169 171 172 173 175 176 178
contents
x
4.3 The design of a terminological data bank 4.3.1 Compilation 4.3.2 Storage 4.3.3 Retrieval 4.4 Limitations and problems 4.5 The future of terminological data banks
181 181 184 186 190 191
Chapter Six terminology and standardization 1. General standardization 1.1 Basic features 1.2 Standardization bodies 2. Terminological standardization 2.1 Standardization of terms 2.2 Standardization of principles and methods of terminology 2.3 ISO Technical Committee 37 3. Terminology and neology 3.1 An overview of neology 3.2 Neologisms 3.3 Linguistic characteristics of neologisms 3.4 Pragmatic and sociolinguistic aspects of neologisms 3.5 Standardization of neologisms 3.6 International criteria for creating terms
194 195 197 199 199 201 201 203 204 205 207 207 209 210
Chapter Seven professional terminology: the role of terminologists in a language service 1. 2. 3. 4. 5.
Linguistic needs and language planning Planning and language services The technical tasks of language services Language services and terminology The training of terminologists 5.1 Background assumptions 5.2 Training in terminology versus training terminologists
notes references index
215 216 217 219 220 221 222 225 233 245
Acknowledgements This book was first published in Catalan in 1992. I devoted three years to writing it after having being the head of the Catalan Centre of Terminology (termcat). At that time, the theoretical understanding that I had assumed of terminology did not question the universality of the principles that had been established by the gtt, although in its applied side it introduced social elements. The latter are essential to account for the needs that terminology has when standardising the use of a minority language, such as Catalan, which is spoken within a technologically-developed society. A few years later, I realised that the theory that I had assumed until then and the practice I had developed were somehow contradictory and that these contradictions would not be overcome solely with a social-based terminology. The major contradictions had to do necessarily with the very understanding of the discipline, its delimitation, the description of its subject of study and its general aims. The universal validity of the premises established by the classical theory, which are indebted to the historical and scientific context when they were formulated, was thus questioned; but questioning them did not mean that their appropriateness for particular applications, subject areas and goals be denied. This book, however, does not cover the new formulation which was very vaguely insinuated in 1992 and has been progressively taking shape in the works I have published since then. Fortunately, several specialists from all over the world have agreed on this position, although we still cannot count on a totally integrated theory, which accounts for terms conceived as units that represent and transfer specialised knowledge and also for their integration into the specialists' competence and their use in different registers of specialised communication and different scientific and cultural contexts. The English version of this book would not have been possible without the contribution made by several people who I would like to name explicitly. I would like to thank J. C. Sager for the friendship he has shown me in his revision of the English version of the book but, above all, I want to thank him for everything I have learned from him. Without this knowledge, I would not have thought about terms in the same way. I want to give special thanks to Janet DeCesaris, a colleague and a friend, for her generosity in undertaking the translation of the book into English, which undoubtedly has been a difficult and hard task.
xii
acknowledgements
And many thanks to Judit Feliu and Cristina Corcoll, iula colleagues, for having so efficiently contributed to the preparation of this manuscript. Finally, I would not want to finish these acknowledgements without thanking Bertie Kaal for the interest she has put in the publication of this text. Without the generosity of all these people, it would have been extremely difficult for me to publish this book in English. M. Teresa Cabré
Chapter 1 An overview of terminology
1.–Social and political aspects The initial motivation for the study of terminology was both spontaneous, like the motivation for technology, and theoretical, like the motivation behind the birth of science. During the simultaneous expansion of knowledge and the growth of technology and communications in the eighteenth century, terminology was seen as a necessary tool for overcoming some of the difficulties associated with these multiple developments. Only in the twentieth century has terminology acquired a scientific orientation while at the same time being recognised as a socially important activity.—Rey (1995)
Terminology, the discipline concerned with the study and compilation of specialized terms is not a new field of study, but only in recent decades has it been systematically developed, with full consideration of its principles, bases and methodology. Its social and political importance has now also been recognized on both the national and the international scale. Terminology, as we understand it today, first began to take shape in the 1930s and has only recently moved from amateurism to a truly scientific approach. 1.1–Origins Although the systematization of terminology and its scientific status are recent developments, activities in the field date from much earlier. In the 18th century research in chemistry by Lavoisier and Berthollet or in botany and zoology by Linné exemplify the interest that the naming of scientific concepts has always had for the real protagonists—the specialists. Due to the growing internationalization of science in the 19th century the need for scientists to have at their disposal a set of rules for formulating terms for their respective disciplines became apparent. Botanists (in 1867), zoologists (in 1889) and chemists (in 1892) expressed this need at their respective international meetings. In the 18th and 19th centuries scientists were the leaders in terminology; in the 20th century engineers and technicians have become involved. The rapid progress
2
terminology
and development of technology required not only the naming of new concepts, but also agreement on the terms used. The Austrian E. Wüster (1898–1977), considered the founder of modern terminology and the main representative of what is known as the Vienna School,1 came from the field of engineering, as did the Russian D. S. Lotte (1889–1950), founder of the Soviet School of Terminology.2 The first international association of standardization, the International Electrotechnical Commission (IEC), was founded in Missouri in 1904. During the first half of the 20th century neither linguists nor social scientists paid special attention to terminology; only from the 1950s onwards did they begin to show any interest and even then it was just in passing. It is curious that linguists have generally shown little interest in terminological studies; instead they have been concerned with developing a theory to account for the principles governing all possible human languages but have been less concerned with the multiple aspects of language seen as a tool for communication. Only within this latter approach is terminology afforded a place in linguistic analysis. 1.2–Development of the field It is no coincidence that the development of both theoretical and applied terminology in the second third of the 20th century occurred thanks to the interest of scientists and technicians. Subject matter and methodology develop when there is a need, and are pursued to the extent that they are the result of clear social needs. If we analyze the organization of societies and their representative ideologies, we can establish the causes behind the appearance of organized, systematic terminology and explain the importance it has acquired in most advanced countries. Analysts of contemporary society and culture believe that we are living in a transitional period and that our society is characterized by a series of changes that make it notably different from earlier periods. We are entering a new civilization that is marked by overwhelming technological control, to the point that we can even alter the most basic laws of nature—genetic engineering and biotechnology are good examples of this. From the rural society of the first two-thirds of the 19th century, and the industrial culture of the last thirty years of the 19th century and the first half of the 20th century, we have leapt to the so-called post-industrial culture of the second half of the 20th century. This succession of civilizations, each with new cultural parameters, does not imply the elimination of those that existed previously; rather, the cultural forms of successive civilizations coexist for long periods of time, and only gradually do certain cultural modes displace others without eliminating them altogether. The change from rural to industrial society brought with it a change in civilization that became evident in new economic patterns, in different geographic
overview
3
distribution of populations, and in different concepts of family, religion and work. Rural society was characterized by people working the land. The population lived spread out in small settlements; the economy was based on subsistence; very few people went to school and most were illiterate. In addition, there was a hierarchical and authoritarian sense of intra-group relationships; organized religion was quite important. On the other hand, the highly developed societies of the industrialized world concentrate their populations into large cities, which usually contain major industrial complexes. Market economies base their development on fostering production and consumption, which favour the loss of the idea of the family as an economic and protective unit. The concentration of the population with the resulting appearance of class awareness and the general spread of education are factors that determine the democratization of interpersonal relationships and political regimes. Ethnologists refer to this new civilization, derived but distinct from the previous one, as a post-industrial civilization. The loss of traditional ideologies has led to a society based on the search for material possessions and on increased individualism and competition, all in the glorification of power and success. In this new stage there are two major areas in which cultural changes are obvious: the technologization of society, and the value attached to information. These two new cultural pillars have replaced other elements from earlier times which characterized industrial society. Some of these changes have had major effects on language and interpersonal communication and have created the need for new linguistic products, new professions related to language, and new ways of organizing communication. In the industrial period, the spread of education and the growing importance of written communication (as opposed to the prevalence of oral communication in rural culture) made it necessary to codify languages and establish operative standard registers. As a result, hierarchical rules for good usage were established in many languages of culture, and the concept of a ‘‘standard language’’ was born. Thanks to this unification the dominant languages were consolidated. At present, however, we are witnessing a paradoxical phenomenon in terms of the status of languages: on the one hand there is a trend towards monolingualism across cultures which is justified by the need for direct and efficient communication; on the other, national languages are being recognized as the natural tools of communication at all levels of usage, whether general or specialized. The need to assert cultural identity justifies the defensive attitude of non-dominant languages opposed to the cultural and linguistic colonialism of dominant languages. In this context, major efforts are being made to rationalize the diversification of languages by means of government-sponsored language planning projects. These projects should respect cultural identity and encourage international relations at the same time. A process of language standardization must respect the cultural and formal idiosyncrasies of
4
terminology
each language, but at the same time allow a language community to become part of larger groups and not force it to shut itself off in sterile isolation. Terminology is also affected by social changes which have had a major effect on linguistic needs: a. The accelerated development of science and technology in recent times has been accompanied by the appearance of a large number of new concepts and even new conceptual fields which require new names. b. Technology is growing rapidly and pervades all spheres of society. Technological developments in the fields of information and communication create the need for new ways of communication that did not previously exist; and the vocabularies of these languages require constant updating. This has brought about the appearance of new fields of activity, such as the so-called language industries. c. Mass production is both the result of and the driving force behind the overriding importance of standardized products. The idea of ‘‘hand-crafted’’ is becoming outdated. d. The transfer of knowledge and products, one of the most significant features of modern society, brings about, on the one hand, the appearance of new markets for scientific, technical, cultural and commercial exchange; on the other, the need to deal with the multilingualism of the new arenas for exchange. It also results in a need to standardize the elements that convey the exchange—the systems and basic units of transfer. e. Information has become of the utmost importance and the amount of information has increased exponentially. This great mass of data requires powerful and effective support. Databases of all sorts are being created and require continuous updating. They must be easy to access and multidimensional. As a result, there arises a new need for information storage and retrieval, as well as for standardized systems for the automatic transfer of the contents of the increasingly sophisticated large stores of data. f. The development of mass communication allows the widespread dissemination of terminology, with the resulting interaction between the general and specialized lexicons. Specific terms become part of popular culture through their use in the mass media. g. Government intervention in language subjects terminology to standardization processes and makes it necessary to create official organizations to manage this work. The fact that scientific and technological creation occurs almost exclusively in the dominant economic powers means that there is a one-way transfer of knowledge and new products, entailing large-scale borrowings of technical and scientific vocabulary in other countries. To combat this situation language
overview
5
policies, often protectionist ones, are implemented and thereby favour the emergence of new professions in these fields. Small countries, whether politically independent nations or not, with unstable languages also engage in language planning and require many language professionals. Special languages and their prime component, terminology, are one of the most important areas for language standardization. 1.3–The evolution of modern terminology As stated earlier, modern terminology emerged in the 1930s with the work of E. Wüster in Vienna. In his doctoral dissertation, Wüster presented arguments for systematizing working methods in terminology, established a number of principles for working with terms and outlined the main points of a methodology for processing terminological data. As Rondeau (1983) notes, at this time Wüster was particularly concerned with methodology and standards as opposed to theory, since he considered terminology a tool that should be used as effectively as possible to eliminate ambiguity from scientific and technical communication. His interest in theory would come later. At the opening session of the Infoterm symposium in 1975, Wüster himself named four scholars as the intellectual fathers of terminological theory: A. Schloman from Germany, who was the first to consider the systematic nature of special terms; the Swiss linguist F. de Saussure, who was the first to draw attention to the systematic nature of language; E. Dresen, the Russian who was a pioneer in underscoring the importance of standardization and the principal force behind the isa,3 and, J. E. Holmstrom, the English scholar who was instrumental in disseminating terminologies on an international scale from Unesco and who was the first to call for an international organization to deal with the issue.4 Following Auger (1988) we identify four basic periods in the development of modern terminology: a. the origins (1930–1960) b. the structuring of the field (1960–1975) c. the boom (1975–1985) d. the expansion (1985–present) The initial period of development of the study of terminology (1930–1960) was characterized by the design of methods for the systematic formation of terms. The first theoretical texts by Wüster and Lotte appeared at this time. When he wrote the dictionary The Machine Tool, published in 1968, Wüster had the opportunity to check the rationale and suitability of the methods he had presented in his doctoral thesis.
6
terminology
In the second stage of development (1960–1975), the most important innovations in terminology came from the development of mainframe computers and documentation techniques. At this time the first databanks appeared, and the international coordination of principles of terminology processing was initiated. During this period the first approaches were made to standardize terminology within a language. The third stage—the boom of terminology between 1975 and 1985—is marked by the proliferation of language planning and terminology projects; some countries like the former USSR and Israel had begun their language policies earlier. The significance of the role of terminology in the modernization of a language became apparent in this period. The spread of personal computers brought a major change in the conditions for processing terminological data. In the most recent period of development (1985–present) some new issues are worth mentioning. Computer science is one of the most important forces behind changes in terminology. Terminologists now have at their disposal tools and resources that are better adapted to their needs, more user-friendly and more effective. At the same time there emerges a new market—that of the language industries—in which terminology occupies a privileged position. International cooperation is broadened and consolidated, as international networks are created to link agencies and countries which share characteristics or are interested in cooperation. Examples of this are the exchange of information and the international cooperation in training terminologists. Finally, the model of terminology linked to language planning, which is so necessary for developing countries, is being consolidated at this time.
2.–Scientific and functional aspects Throughout the last fifteen years, and in spite of often differing ideologies concerning a substantial number of the guiding principles of terminology, agreement has been reached. The fact remains, however, that in every country, the objectives and the working methods of terminology are subject to a wide range of factors, such as the country's political, socioeconomic and linguistic situation.—Rondeau & Sager (1986)
Not all experts agree that terminology constitutes a separate discipline, nor do all consider it a theoretical subject. For some, terminology is a practice dealing with social needs that are often related to political and/or commercial ends. In the opinion of others, terminology is a true scientific discipline that owes much to the other subject fields from which it borrows fundamental concepts; but it is, nevertheless, considered a separate discipline in the sense that it has reformulated and synthesized the original foundations so that it could build its own field. There are many interme-
overview
7
diate positions which, although recognizing that terminology contains some original theoretical aspects, only conceive of it within the framework of other, more consolidated disciplines. 2.1–The theory of terminology In the 18th- and 19th-century, scholars were alarmed by the proliferation of terms and were most worried about the diversity of forms and the relationships between forms and concepts. They were not concerned with the nature of concepts nor the foundations for creating new terms. Theoretical concerns about the nature of terms arose later when terminological work began to be organized in some special fields as a result of practice. Wüster’s work is a good example. He was initially interested in methods of compilation and standardization of terms, and once they had been applied in The Machine Tool (1968), he concentrated on aspects of the theory of terms. Almost thirty years separate his doctoral thesis (1930), which was practice-oriented, from his publications on theory.5 We can thus say that terminological theory arose and is even today developed through practical experience that is, in its turn, motivated by the need to provide solutions to language-based problems in communication. The work carried out in the 1930s, simultaneously but independently by Austrian, Soviet, and Czech scholars, is the basis for the beginning of what the Austrians would call terminology science. The three classical schools of terminology—the Austrian, the Soviet, and the Czech schools—all emerge from this work. Looking at the development of these three schools we can identify three different approaches that are not mutually exclusive: • A first approach that considers terminology to be an interdisciplinary but autonomous subject at the service of scientific and technical disciplines • A second approach focusing on philosophy, which is primarily interested in the logical classification of concept systems and the organization of knowledge • A third approach focusing on linguistics, which considers terminology a subcomponent of a language’s lexicon and special languages as subsystems of general language. A general theory of terminology is based upon the first approach in which the nature of concepts, conceptual relations, the relationships between terms and concepts and assigning terms to concepts are of prime importance. This focus on moving from concepts to terms distinguishes the methods used in terminology from those used in lexicography. The aim of terminographers is to assign names to con-
8
terminology
cepts; i.e. they move from the concept to the term (an onomasiological process). By contrast, lexicographers start with the word—the dictionary entry—and characterize it functionally and semantically; i.e., they move from the word to the concept, precisely in the opposite direction (a semasiological process). This view, which today is considered the most systematic, coherent theoretical approach to terms, differs from lexicological theory in three ways: in the priority of the concept over the designation; in being exclusively concerned with the level of the terminological unit and not with the other levels of linguistic description; in excluding any diachronic approach or information. Wüster considered terminology an independent subject which he defined as being concerned with the relationship between the sciences such as physics, chemistry, medicine, etc. and a combination of other disciplines such as linguistics, logic, ontology, and computer science. The autonomy of terminology in relation to linguistics or, more directly, in relation to lexicology is fully justified. Terminology and lexicology differ in the way they conceive and deal with their approach to the object of study, in the object of study itself, in their methodology, in the way terms are presented and in the conditions that must be taken into account when proposing new terms. Terminology shares with logic a basic interest in concepts. As opposed to semantics, which is interested in the name-meaning relationship, terminology is primarily concerned with the relationship between objects in the real world and the concepts that represent them. Logicians use a process of abstraction to generalize from various objects that exist in the real world to arrive at the concept or class of objects. To accomplish this, they eliminate the contingent and irrelevant characteristics from the individual objects and only retain those features that are pertinent for characterizing the class that represents the diversity. Terminology and logic also share an interest in the way concepts relate to one another. Indeed, the type of relationships and the system of symbols terminologists use to represent these relationships come from logic. Terminology shares with ontology an interest in the nature of ‘things’ in the real world and the relationships established in this world. The concern of how to classify referents is not new for semanticists and philosophers, and ontology deals with the relationships that are not based on logic. Unlike logical relationships, these relationships do not start from the similarity between concepts but rather from their situation in the real world. As far as the ties between terminology and computer science are concerned, Wüster claims that computer science is one of the keys to terminology because of the enormous possibilities it offers to store and retrieve information and to order conceptual systems. Information science uses terminology to order concept fields that subsequently provide access to information about the documents. In Wüster’s view, writing
overview
9
thesauri is a terminological activity because it focuses on the characteristics and structuring of content. Thesaurus descriptors are terms and characteristics at the same time, and the relationships established by terms in documents are considered to be logical relationships. Finally, terminology is closely linked to the special subject fields. Terminology is not an end in itself, nor can terminological work be concerned with simply providing compilations of a series of concepts with their corresponding names. Terminology is at the service of science, technology and communication; as a result, it must work within the limits of providing a service to other disciplines. Subject specialists and general and applied terminologists work together to carry out the ordering and standardization of concepts and terms for each special field. The current development of terminology is the result of advances in technology and the ever increasing need for specialized communication among communities with different languages. 2.2–Terminology, a new practice As with any relatively new concept, terminology is subject to change, which depends on the theorists and specialists that practise it. The first reason for change derives from the relationship that terminology has with its preceding disciplines, especially with semantics, lexicology and lexicography. Some people believe that terminology is nothing more than a new perspective adopted by these older disciplines. Terminology is currently seen as an art or practice rather than as a science. Even though it has a well-defined aim, namely to satisfy the expressive needs of its users, its working methods are mainly empirical. Theoretical research and the refinement of the processes of recognition, analysis and creation of terms must improve before terminology can be placed among the sciences deriving from linguistics.—Dubuc (1985)
If we agree with Dubuc we must acknowledge that there are a variety of views as to the nature of terminology, its purpose, and its position within the complex panorama of the sciences. In Dubuc’s opinion, the theory of terms and the practice of terminology lead to different positions. We can identify the two extremes: one position claims that terminology is a separate discipline with its own theory, while the other claims that terminology owes its theoretical assumptions to other, more consolidated disciplines. The latter position is taken by Sager (1990), who, although not granting terminology a full independent status, acknowledges that the field has theoretical foundations upon which terminological practice is based: There is no substantial body of literature which could support the proclamation of terminology as a separate discipline and there is not likely to be. Everything of import that can be said about terminology is more appropriately said in the context of linguistics or
10
terminology
information science or computational linguistics. We see terminology as a number of practices that have evolved around the creation of terms, their collection and explication and finally their presentation in various printed and electronic media. Practices however well-established, do not constitute a discipline, but there is no denying a long history of methodologies which themselves require theoretical underpinnings to justify their distinctive nature. Disciplines establish knowledge about things and as such are justified in their own right; methodologies are only means to an end, in the case of terminology, how to do things.—Sager (1990)
Defined as the process of compiling, describing, processing and presenting the terms of special subject fields in one or more languages, terminology is not an end in itself, but addresses social needs and attempts to optimize communication among specialists and professionals by providing assistance either directly or to translators or to committees concerned with the standardization of a language. This more pragmatic, applied view of terminology is more attuned to the requirements of contemporary society where epistemologic reflections have given way to more pragmatic attitudes. Dealing with real communicative needs in the fastest and most effective way has replaced both thought about the principles behind terminological issues and about the ways to address these issues. The reasons for this change in direction can be attributed to a series of changes that have occurred in scientific disciplines and in society itself. We have moved from specialists’ concerns about the ‘‘right form’’ (i.e., the standard form) of the first third of this century, when Greek and Latin roots were preferred, to a more pragmatic and functional approach. This change has been brought about by the new approaches in theoretical linguistics, which have abandoned prescriptive grammar, by the growth of language policies aimed at addressing situations of language conflict, and by the leading role terminology has come to play in non-European countries like Canada, and within Canada, in Quebec. In addition, the enormous advances in computer science have altered not only the work carried out by large institutional bodies, but also that of individuals. Even Sager, who does not believe that terminology is a separate discipline because it does not have its own epistemology, recognizes the importance of separate principles and methods suited to the purposes terminology processing wants to achieve: facilitating communication among specialists, or the self-assertion of technologically non-dominant languages by means of suitable intervention in and planning of language usage. 2.3–The functions of terminology As an intersectional and multidisciplinary science, terminology is located at the crossroads of a large number of subdisciplines of linguistics (semantics or differential lexicology,
overview
11
among others), but it is not their preserve. In L. Guilbert's words, ‘‘the essential aim of the terminological lexicon is not the language itself’’. In fact, terminology is closely linked to an activity carried out within the field of knowledge and thus it is inseparable from its social context and its obvious applications.—Goffin (1985)
Terminology can only be understood in relation to special languages and communication and addresses a variety of purposes, all of which are related to communication and information. There is, consequently, a wide range of approaches to the theory and practice of terminology. We can, however, establish a series of basic assumptions shared by all approaches. We first need to identify four different points of view which in turn lead to different focuses for terminological work and applications: a. For linguists, terminology is a part of the lexicon defined by subject matter and pragmatic usage. b. For subject field specialists, terminology is the formal reflection of the conceptual organization of a special subject and a necessary medium of expression and professional communication. c. For end-users (either direct or intermediary) terminology is a set of useful, practical communication units which are assessed according to criteria of economy, precision, and suitability. d. For language planners, terminology is an area of a language requiring intervention in order to reaffirm its usefulness and survival and to ensure its continuity as a means of expression through modernization. Mindful of these four points of view, we can now identify two major user groups of terminology: users of terminology for direct communication or communication through intermediaries, and terminologists, who write glossaries, facilitate communication, or mediate in some other way. According to the needs of these two groups, terminology can be said to have two dimensions which are closely related: a communicative dimension and a linguistic dimension. For the first group, terminology is a tool for communication. For the second, it is the target of their work. Users Two main groups of people use terminology as a communicative tool: direct users, and intermediaries who use terminology to facilitate communication for other users. The direct users of terminology are the specialists in each subject field. For them, terminology is a necessary tool for communication and an important element for conceptualizing their own subject matter. This two-fold function that terminology has for them accounts for their interest in standardization as a process for determining the definition of concepts and fixing the corresponding names. Specialists use
12
terminology
terminology regardless of whether a term is appropriate within a particular linguistic system or not. Their communicative needs start from the knowledge of the concept and from the need to communicate it; their interest in terminology focuses on concepts and how they can be named clearly and unambiguously. Terminology is primarily the business and the responsibility of several groups of specialists. Terminologists, with their working methods and knowledge, are merely technical aids in a multidisciplinary field.—Corbeil (1982)
Terminology intermediaries are language professionals like translators, technical writers, and interpreters who need terminology to carry out their profession of facilitating communication. They need glossaries and specialized dictionaries because they assist in technical writing or in translating a text from one language to another. Terminologists Terminologists, terminographers and neologists, language planners and information scientists must be both specialists in language, information and documentation and in an appropriate subject field. Their work consists of compilation, description, processing and creation of terms. 2.4–Schools and working methods in terminology As stated above, systematic interest in terminology arose simultaneously in several European countries (Austria, the former Soviet Union, and the former Czechoslovakia). It is from these three centres that terminological practice first expanded to the West (France, Canada, Quebec) and North (Belgium and Scandinavia) then in a second, more recent period to the South (Northern Africa, sub-Saharan Africa, Central and South America, Portugal, Spain) and, even more recently, to the East (China and Japan). Developments in each one of these regions are characterized by the context in which terminology is studied and by the purpose it aims to accomplish. According to Auger (1988), we can establish three major orientations in terminology processing defined by their main objectives: terminology adapted to the linguistic system (the linguistic approach), terminology for translation (the translation approach), and terminology for planning (the aménagiste approach, as it is known in Canadian French). a. Terminology adapted to the linguistic system This orientation is represented by the three schools of Vienna, Prague and Moscow. The Vienna school of terminology, which is the best known, is based on the work of E. Wüster and adopts his principles formulated in his ‘general theory of terminology’. This school’s importance stems from the fact that it has developed a
overview
13
systematic corpus of principles and methods that constitute the basis of much theoretical work and modern practice. Its most salient feature is that it focuses on concepts, and steers terminological work towards the standardization of terms and concepts. The Vienna school arose from the needs of technicians and scientists to standardize the terminology of their fields in order to ensure efficient communication and transfer of knowledge among specialists. The principles of this school are reflected in standardized documents on the vocabulary of terminological work, and terminology as a discipline, on the field’s methodology and data transfer, and on the presentation of finished terminological products. Most of the countries in central and northern Europe (Austria, Germany, Norway, Sweden, and Denmark) work within this framework, in which the subject specialists themselves are responsible for specialized terminologies. The Czech school of terminology, of which L. Drodz was one of the major proponents, arose as a result of the functionalist linguistic approach of the Prague school of linguistics. It is almost exclusively concerned with the structural and functional description of special languages, in which terminology plays an important role. Special languages are considered a ‘‘professional style,’’ which exists alongside other styles such as the literary, journalistic, or conversational styles. This school conceives of terms as units that make up the functional professional style. It arose as a result of the multilingual nature of its geographical area. It was very interested in the standardization of languages and terminologies, and its terminological work was linked to the Czech Language Institute (a part of the Academy of Sciences). The Russian school of terminology, was based on the work of Caplygin, Lotte, and their co-workers; it was also aware of Wüster’s work from the time it first appeared. Consequently, it was mainly interested in the standardization of concepts and terms in the light of the problems connected with the multilingualism in the former Soviet Union (now the Commonwealth of Independent States). These three schools of terminology, which all share a linguistically based perspective (they all consider terminology a medium of expression and communication) have given shape to the theoretical basis of terminology and the methodological principles governing its application. In addition, they are one of the main forces behind language planning and terminology as subsequently developed in Quebec and by the federal government of Canada. b. Translation-oriented terminology This second orientation in terminology, which supports translation, is highly developed in institutionally bilingual or multilingual provinces or countries, e.g. Quebec, the Walloon part of Belgium, and forms the basis for the terminological activities carried out by multilingual international bodies (e.g. un, Unesco, eu, fao). It also represents the most important motivation for the creation of termino-
14
terminology
logical databanks (termium of the Canadian government, eurodicautom of the eu, btq of the Quebec provincial government, and others). This orientation establishes terminological equivalents in the various languages which are used as points of reference by translators and which contribute to the quality of a translated text. c. Terminology oriented towards language planning Language planning as an institutional activity was started in the 1960s and was initially intended to introduce policies supportive of the use of minority languages inside larger sociolinguistic areas. For example, in Quebec policies were implemented to secure a ‘‘normal’’ status for French and its full development in all spheres of usage. Similar plans are currently being implemented in many countries with language situations similar to those of Quebec. The underlying belief of this type of language planning is that the use of an unstable language can change with systematic, strategic intervention carried out by official bodies, with the right legislation and appropriate measures aimed at implementing the change. To attain the desired change, the language in question must have up-to-date, coherent terminology to ensure professional communication in all fields. The objective is to replace terminology imported from languages spoken in technologically dominant countries, thus fostering word-formation in the native language. 3.–Organizational aspects Terminology was not the result of the desire to create a different field. It is the enormous development of technology and the growing needs of communication among different communities that lies behind its development as an autonomous discipline.—Dubuc (1985)
As Dubuc states, the circumstances of reality and the needs resulting from technological development in our society are the factors behind the emergence of terminology as a separate field of study and activity. The few works on the subject of terminology written in the 18th and 19th centuries as well as the systematic work of the first third of the 20th century can be seen as a direct response to this need. Today terminology continues on the course laid out by society’s needs. In full awareness that terminology is essential for specialized communication (because good terminology ensures its users that the communication will be precise and effective) countries in the developed world have set up centres and technical committees to provide standardized terminology for science and technology. Developing countries have likewise understood that terminology is one of the tools that can give them access to the industrialized world because it is through terminology that the exchange of knowledge and technology transfer occurs. As a
overview
15
result, they have invested time and effort in creating, unifying and internationalizing terminology that responds to the needs of modern society and that allows them to maintain efficient international relations. I do believe that terminology is a practice with three distinct, although interrelated, conceptual characteristics: it is a practice of a cognitive, linguistic and social nature. It is cognitive since that is an essential aspect of terminology, if understood as a scientific project. This cognitive aspect, subject to field variation, is basic because it relates the linguistic aspect to the social practice.—Rey (1995)
Interest in terminology, however, does not arise uniformly nor does it take the same shape in all places. Since interest develops because of a variety of causes and with a variety of goals, depending on circumstances and context, terminological activities are organized differently in each country. The characteristics, goals and possibilities of each country vary, as do the language situations and structures and the technical and human resources involved. This situation can be exemplified in the case of Quebec by the following quotation: The teaching of terminology must take into account the social context within which the future terminologist will work. What is this context? The best way to describe it is to recall the circumstances in which terminology appeared in Quebec. It was first conceived by translators at the time when Canadian and Quebecois language policies were being defined. In my view, terminology is a discipline at the service of the language policy carried out in Quebec and is essential for applying the policy on bilingualism that has been undertaken by the Federal Government. The abovementioned situation results in the realization that terminology was born within a bilingual context. The type of terminology practised here is mainly contrastive. (. . .) In Quebec, terminology also plays a role in the process of updating the lexicon, and it takes on two different aspects. First, there is lexical readjustment, that is, the Quebecois are now using French terminologies that already existed but which were unknown because the industrialization process had been carried out mainly in English. Second, there is lexical updating proper, which is an activity defined in terms of its relationship to translation.—Maurais (1987)
3.1–Different orientations, according to spheres of influence Not all terminological activities are considered equally important. Initiatives for collecting and disseminating terminology have been taken by regional or national centres or agencies or by international organisations which, in turn, have created specific areas of influence. Each one of these spheres of influence may represent a separate orientation of terminology. The three orientations identified by Auger (1988) are conditioned by the characteristics and factors of each specific area of application. This diversity of objectives is the reason for the different types of terminological activities around the world and
16
terminology
for the coexistence of different models and methods that are suitable for the specific purpose. Although, in these general reflections, we are simplifying a reality that is in itself complex and varied, we will establish three points of reference for terminology as currently organized. a. The central and northern European approach. The best known representative of this school is Infoterm, the International Information Centre for Terminology in Vienna. It was created by Unesco in 1971 as a part of the unisist program and with the cooperation of the Austrian Institute for Standardization. Infoterm has disseminated Wüster’s work and theories on terminology, and its sphere of influence includes Northern Europe, North Africa, Latin America, as well as China, Japan, Portugal and Spain. The recent creation of the International Institute for Terminological Research (iitf) in Vienna is noteworthy. b. The approach of multilingual translation departments in federal and international institutions, e.g. the United Nations and the European Union. The Federal Canadian government is also active in terminology because, since Canada is a bilingual country, the government is responsible for safeguarding the language rights of all its citizens. c. The approach of government agencies of countries undertaking language standardization. This approach is inspired primarily by the language planning policy of Quebec, introduced in the 1970s to foster the strengthening of French in Canada. The interventionist policies of successive governments of Quebec, have reversed the former trend towards Anglicization and today French is not only the language of the provincial government and the official language in all institutions and professional documents but it is widely used in all forms of communication. Strong legislation protecting the rights of French-speaking citizens is the basis for government intervention. This third approach has been the most influential for the terminological activities in Catalonia, in the Basque region of Spain, and in other areas where there is a policy to change the status of a regional language. Many newly independent countries in northern and central Africa have adopted institutional bilingualism in which the use of each language is complementary to the other in different areas of communication. The need for these developing countries to exchange knowledge and technology with more developed countries, combined with the goal of promoting the use of the native language, results in an approach to terminology that has a double objective: one directed towards standardization, in the sense of establishing a single form (the central European approach), and the other directed towards standardizing the use of the language (the Quebec approach).
overview
17
These different approaches can be seen in variations in the priorities of terminological work, in the terminographic methodology adopted, in the agencies involved in terminological intervention, in the products of terminology and, finally, in the interaction with other disciplines. The following quotation exemplifies these differences: In an African environment, and as far as terminology is concerned, the lexicographer cannot only be a linguist or a specialist in search of the terms that constitute the vocabulary of any special subject field, but also needs to be an ethnographer, and as such must address not only the language but especially the society that speaks it.—Halaoui (1990)
a. The priorities of terminological work Since the goal of the central European school of terminology is effective and precise communication between professionals, its main priorities are scientific and technical language. A standardizing stance only needs to be taken with reference to special subject fields, and specialists are the targets of standardization activity. It is only recently that work has begun on the scientific disciplines in the humanities and social sciences. In contrast, neither the translation-oriented nor the language planning approach can afford to restrict itself to scientific and technical subjects. With their very practical goal of facilitating translation, or preparing standardized terminologies for a language with terminological gaps, these approaches become involved in all areas of knowledge and human activity. In planning their terminological work they must therefore bear in mind the situation and importance of the various communication contexts to provide their users with an efficient special language. b. The terminographic methodology In any activity the methodology used has to be adapted to the possibilities of the environment. Terminology is obviously no exception, and consequently must focus the work chosen on the nature of the subject, the linguistic and terminological situation of the environment, the purpose of the work, and the language of the specialized communication. In the case of terminology aimed at standardizing concepts and terms, the process for choosing a standard name or the process of fixing a concept is usually carried out by committees of specialists who, in the case of alternatives, choose one of the various available terms in order to establish a single national or international designation that is valid for each concept. In contrast, terminological work linked to language planning pursues a rather different goal. Since this approach is aimed at reinforcing and extending the use of a language by making it viable in all contexts, choices among alternative variants must be avoided. Although the capability of a language to describe a specific situa-
18
terminology
tion is beyond doubt, the social and economic imbalances in countries, combined with their historical circumstances, have resulted in some languages developing naturally in consonance with their technological and commercial development, whereas others have lagged behind. A language that cannot be used in all types of communication is doomed to disappear, and a language cannot be used in all situations unless it has the necessary terminology. Countries with non-dominant languages and which are technologically dependent on others must create terms based on neology.6 On the other hand, countries that produce science and technology in their own language only compile and standardize their own terminology. c. the agencies involved in terminological intervention The different ways of conceiving terminology and its functions also condition the composition and organization of research teams and the scope of their activities. Dubuc was very insightful on this point: Fireman, safeguard, grammarian on duty, walking dictionary, language agent, Gallicization force, high-quality communications promoter, a terminologist must be a bit of each... Until now, we have mainly trained office terminologists. It is probably the right time to train terminology commandos, without leaving aside the technical expertise, in order to build a society with French language and culture.—Dubuc (1987)
What, then, should terminologists do and what role should they play in a specific socio-political situation? The macro- and micro-sociolinguistic environment in which the work is carried out provides us with the answer. Terminological activities and the role of the terminologist are conditioned by the nature of the environment, which also determines the approach to be adopted. One issue, however, seems uncontroversial: subject field experts must always participate to some degree in terminological work. The role of specialists is the third element that differentiates the three spheres of influence in terminology. In the central and northern European approach, subject field specialists are the real protagonists; they are involved in the development of methodology, in training, in systematic research and in standardization. If necessary their work is backed up by that of linguists trained in general terminology and terminography, who help them to present the material and investigate possible means of naming concepts or choose the appropriate canonical forms. But only subject specialists can devise conceptual systems and recognize the feasibility of designations. Without them, it is impossible to conceive of a systematic approach to terminology. Systematic terminologies can only be prepared by the subject specialists of the subject field in question. Wüster emphasized that systematic work should be the domain of terminologists, i.e. of subject specialists with an additional training in terminology . . . . Therefore,
overview
19
modern terminology work requires the work of terminology commissions which should be composed of the following types of specialists: 1) The specialists of the subject field concerned . . . 2) A specialist for the application of terminological principles and of methods in terminography . . . He should be a subject specialist or a linguist with a specific training in terminological principles.—Felber (1984b)
In the language planning approach to terminology, on the other hand, linguists play a much more salient role due to the fact that the focus of work is on more general areas which do not entail a highly specialized formal structure. Nevertheless, specialists must take part both in answering questions on concept structure or naming that linguists might pose and in choosing the most suitable designation in each case. This intervention by end-users in terminological decisions clearly favours the subsequent acceptance and spread of the new forms. Users are both the source and the target of the research process and of its results. The participation of users in the production of terminology is considered essential, not only in order to ensure the scientific and technical reliability of the work, but also to facilitate the implementation of these very terminologies.—Cayer (1990)
d. The products of terminology The three approaches also differ with respect to their products. Terminologies only addressed to specialists are usually presented following ISO recommendations, with very precise definitions, little redundancy and usually with equivalents in other languages. The priority ordering of terms is systematic, since the specialists are the end-users and they know the subject field. However, when the users of terminologies are not so well defined or when the purpose of the terminology is to extend the use of the proposed forms—with the subsequent abandoning of the forms deemed unsuitable—the way the term collections are presented is almost as important as the content of the material itself. Terminological products aimed at language standardization must be adapted to each target group. So Auger rightly observed: It is well-known that people do not read dictionaries and, consequently, distributing lexicons is not sufficient to initiate a change in the language attitudes of users. We should accept the fact that these terminological products are not always designed for or addressed to well-defined target users. Conversely, they are too frequently conceived without taking into account users' needs.—Auger (1986a)
e. The interaction with other disciplines Finally, the central and northern European view of terminology, as stated previously, focuses on the relationship of terminology as a discipline to logic, computer science, the theory of communication, and information theory. This orientation is consistent because terminology is considered above all the study of concepts and the study of conceptual systems. Work in terminology consists simply of representing
20
terminology
conceptual fields and establishing precise designations to ensure efficient professional communication. In contrast, within the language standardization approach, terminology is conceived as a part of applied linguistics, because its object of study is a language and its goals are practical in nature. Terminology is a special part of the lexicon of a language. In a full description of a language, terminology cannot be ignored. 3.2–Organization of terminology A comprehensive treatment of the organization of terminology implies analysing the aspects it covers and then establishing how each of these aspects is organized and who is entrusted with the final responsibility in each case. Countries with a substantial amount of terminological activity organize work according to their own social and political structure, the objectives that they want to attain, and the resources they are willing and able to spend. Thus, in a specific socio-political environment a comprehensive terminology agency must be equipped to deal with the following aspects: a. planning, coordination and management of terminological resources, including the planning of work for a specific language and country, the coordination and supervision of the application of this plan, the assessment of the results and the management of the allocated resources. b. terminological research, including the preparation of systematically structured terminology. c. standardization of terms, involving the evaluation of alternative terms used to designate a single concept. d. dissemination of standardized terms and issuing of decisions on consultations about terminology and miscellaneous aspects related to terms. e. implementation of terminology among professional groups and specialised areas of activity. f. training in terminology, involving the education of specialists able to prepare terminologies or participate in a part of the preparation process. A complete training programme includes both the training of terminologists who will take charge of systematic work and the training of other specialists involved (e.g. scientists and technicians, translators, interpreters, technical writers, teachers, specialists in documentation, computational linguistics, lexicographers) as well as the training of teachers of terminology. Even though it may seem obvious that in certain contexts, e.g. in the case of domi-
overview
21
nant languages, some of these functions are less important than others, well-organized terminological activity should comprise a rational distribution of responsibilities and tasks, and the acceptance of responsibilities on the part of the agencies best suited to deal with each task. Consequently, even though a country may have a terminology centre responsible for the management, planning and coordination of research, it is a good idea for the initiatives in terminological research to be carried out by the centres of economic, administrative and research activity, because they can also play an important role in standardizing new terms and in defining priorities. It also seems logical for centres of higher education to take charge of the training of terminologists and other terminology professionals, with practical internships in the workplace, so that students can experience real work firsthand and become acquainted with the conditions in which they will later exercise their profession. Many countries have organized terminological research and have created either administrative agencies or complete centres for work on special terms. The institutionalization of practical work on terminology has the advantage of allowing a better coordination of work and an increased rationalization of economic and human resources. Terminological activity, particularly if it aims at covering all specialized areas, requires a substantial investment of resources. If government agencies take charge of terminological activities, they can give new terms a legitimacy that they could never obtain from a non-governmental body, no matter how highly organized or well researched. In most countries with an active language planning policy the administrative agencies responsible for implementing it also take charge of terminology, whether directly (as in the case of Quebec or the Swiss canton of Berne) or indirectly via the creation or support of a specialized centre (as in the case of Catalonia or the Spanish Basque provinces). Despite the advantages of a government presence in terminological activities, official bodies cannot ignore academic institutions, semi-official centres or the private sector, because they have expertise in special subjects. Allowing non-governmental bodies—the direct users of terminology—to become separated from the terminology process does not benefit the standardization of a language, which requires the direct involvement of all speakers and a militant attitude towards using the language. Users of terminology must thus feel they are a part of the process in order to ensure the quality of work and the effective use of the terminology produced. End-users must be members of standardization committees; they should head research teams, and participate in the planning stages of future work. At the same time official terminology centres should not ignore initiatives from universities and from the business world. They should likewise organize the various facets of terminology (planning stages, research, standardization, dissemination, implementation, and training) by counting on the support of specialized centres for one or more of these tasks.
22
terminology
As an illustration of the above, we will briefly discuss the organizational model for terminology that has been adopted in Canada and Quebec (OLF 1989, Cayer 1990). The federal government of Canada and the provincial government of Quebec are the leading centres for terminology dealing with the problems and social needs arising from bilingualism in Canada and the increased official presence of French in Quebec. To approach these issues, both governments have created specific agencies within their respective administrations. The Secretary of State of the Canadian government includes an administrative agency that is in charge of all aspects related to the translation of all official texts from French to English and from English to French, and of specialized terminology (terminology development and the termium data bank). Research in terminology is exclusively aimed at translation, and consequently bilingual glossaries are regularly published and the termium data bank is constantly being updated and added to (it can be accessed by modem and is available on cd-rom). The government of Quebec organizes terminology through the Office de la Langue Française (olf), which is concerned with all issues relative to the defense and promotion of French. The language and terminology services are in charge of the planning, implementation, standardization, dissemination and implementation of new terminology, especially that related to the most advanced special subject fields in which French has gaps that need to be filled. As a result of the goal to promote French, the OLF’s terminological research compares English with French but does not express opinions on English. Rather, its purpose is to update the French vocabulary and to take prescriptive action, in the sense that it aims to direct the use of French by counting on the participation of users, who are organized into standardization committees. The terminology commitees of the olf deal with the troublesome linguistic problems terms can create. Universities in Quebec, particularly the University of Laval and the University of Montreal, offer associate-degree and bachelor-degree programmes (and a Ph.D. at Laval) in terminology oriented towards linguistics or translation; they have standing agreements with the olf through which students have free access to the btq, the Quebec government’s data bank. The language services of private businesses address all language-related issues (text writing, translation, terminology) in the company and deal with the specialized terminology necessary for the activities of the company in question. 3.3–International cooperation An analysis of linguistic behaviour in contemporary society reveals a paradox between two apparently contradictory tendencies. On the one hand, there is a trend
overview
23
towards unification, towards using a single language of communication for certain formal purposes. In this sense, English has become the koiné, the international language, for scientific and technical communication. On the other hand, a growing number of countries insist on using their own language in all communicative contexts. They thus defend international multilingualism based on the mutual understanding of neighbouring languages and the use of advanced technological resources to aid mutual comprehension, without having to limit the use of some languages in favour of others. This paradox also occurs in terminology. On the one hand, we are faced with specialized vocabularies that allow us to maximally reduce subjectivity and which require a high degree of internationalization. This aspect, which is the unifying side of special languages, aims at attaining precise and efficient communication. But terms, albeit to a lesser degree than words, reflect a certain world view, and express the culture of a people. This latter aspect is the diversifying side of terminology, which respects the cultural identity of communities that import technology developed by others. The predominant behaviour patterns and the most widespread types of organization of our society which tends towards internationalization in all spheres, have contributed to a certain neutralization of the cultural peculiarities of specialized terminology used for international communication. In contrast, they have encouraged the adoption of measures to protect the linguistic and cultural identity in communication within a community. The technological resources used for information and communication, the amount of money required to carry out these research programs, the increasing value attached to information and the need to be cost-effective have made international cooperation essential for the further development of information and communication. We must not forget that all cooperation between countries with varying levels of development almost automatically involves some degree of subordination, although the collaborators themselves may not be conscious of this. The cost involved in the research of communication technology and the resulting need to make this investment profitable have pushed societies towards various types of international communication: • the exchange of terminological, documentary and background information; • the cooperation in applied research (preparation of multilingual terminology, development of computer programs for terminological work, development of databases and other tools to facilitate access to information, etc.); • the collaboration in the defense and balanced development of non-dominant languages; • the training of professionals in terminology.
24
terminology This need for international cooperation has resulted in real projects and activities:
a. The creation of international centres and associations that favour information exchange and interrelationships. Among the important events to note are: the creation of Infoterm (International Information Centre for Terminology) in 1971 by Unesco; the creation of termia, a terminology association that has since disbanded; and the creation of the Terminology Commission of the International Association of Applied Linguistics. On a different level we should also mention the founding of the Unión Latina, with headquarters in Santo Domingo, the Dominican Republic, which includes all countries speaking Romance languages, the creation of the International Institute for Terminology Research, and the creation of Lexiterm, an association made up of French-speaking countries, founded in Geneva in 1988. b. The creation of multilingual databases by official bodies which, even though they are aimed at meeting the needs of a specific organization, have become a point of reference for various countries and favour international exchange. In this respect eurodicautom, the data bank of the European Union, stands out because of its important role in Europe. c. The creation of networks for cooperation among countries with the same language or programs for collaboration among countries with similar characteristics and whose terminological work is directed towards attaining similar goals. In this regard we note the Réseau International de Néologie et Terminologie (rint), for French-speaking countries from the Americas, Europe, and Africa; the Red Iberoamericana de Terminología (rit), for Spanish- and Portuguese-speaking countries; and, arabterm, for Arabic-speaking countries. TermNet, the international terminology network proposed by Infoterm and chartered as an international association in 1989, deserves special mention. In 1991 it had 39 members, mostly from Europe. d. The setting up of joint programs in terminology training (Boulanger & Reguigui 1987). The numerous programmes sponsored by the Centre International de Recherche en Linguistique Française Appliquée (cirelfa) of Quebec to train terminologists and language planners of French-speaking African countries as part of a North-South collaborative effort in terminology stand out in this respect (Rondeau & Sager 1986). e. The production of specialized journals in Terminology, e.g. Terminology, published by John Benjamins of Amsterdam.
Chapter 2 Terminology, an interdisciplinary field
Only after having thought about it in depth have I chosen such a pretentious title: ‘‘The General Scientific Study of Terminology, at the crossroads between linguistics, logic, ontology, computer science and the sciences of things’’ . . . In comparison with other sciences, the term crossroads implies considering terminology as a scientific field in itself.—Wüster (1981)
Like all interdisciplinary fields in science, terminology is a discipline that is defined in relation to the other fields from which it takes a specific set of concepts. Nevertheless, we must consider that an interdisciplinary subject does not define its field of study as the sum of the concepts from the disciplines that comprise it, but rather that, firstly, it chooses from these fields only a specific set of concepts and elements and, secondly, that it elaborates from these concepts its own object and field; only by doing so does it acquire the status of discipline. Wüster considers terminology as being located at the intersection of linguistics, logic, ontology, information science, computer science and individual disciplines. This interdisciplinarity of terminology is determined by the characteristics of terminological units, which are simultaneously language units (linguistics), cognitive elements (logic and ontology, i.e. part of cognitive science) and vehicles of communication (communication theory). Terms appear in specialized communications (information science) and computers are usually employed in terminographic activity (computer science).
1.–Terminology and linguistics The general scientific study of terminology is largely influenced by its relationship to applied linguistics, of which it is a branch. In Gunther Kandler's words, applied linguistics can be described as follows: ‘‘That which goes beyond linguistics in order to collect linguistic knowledge from every domain of life and to make it useful for each one of these domains.’’—Wüster (1981)
Linguistics is the study of language; there is, however, some divergence as to what
26
terminology
is understood by ‘‘language,’’ and there are many different paths for arriving at a definition of language. We can see these different paths by comparing the two major linguistic trends of the 20th century. Structural linguistics describes specific languages and indirectly allows us to draw conclusions about some aspects of linguistic behaviour and facilitates a typology of languages. Generative linguistics focuses on describing an individual’s linguistic capability (competence) and not on the description of specific languages. Structural linguistics observes and describes linguistic facts, whereas for generative linguistics a description alone is not enough; linguistics must also explain how and why these facts occur. Current linguistic theory is dominated by generative linguistics. A linguistic description of a language involves accounting for its substance and structure, its use and its acquisition. A complete study of a language, therefore, requires three different theories: first, a theory of competence; second, a theory of usage; and third, a theory of child language acquisition. A theory of competence must include the description of the various elements that account for the internalized linguistic knowledge that speakers of any language have and which allows them to construct and understand utterances in their language. This theory, which is also known as the theory of language structure, accounts for the units, rules and restrictions contained in the grammar of all languages. It is in this theory that the various modules—lexicon, syntax, phonology and semantics—are described. The grammar of a language is the set of modules whose interaction accounts for the user’s production and comprehension of utterances. Each module is an organized system of rules and restrictions limiting the application of rules. The grammar also includes explicit mechanisms to connect the various modules to one another. Abstraction from specific grammars of those aspects which are shared by all or a group of languages (or, more appropriately, the proof in specific languages of the hypotheses concerning the principles that govern language in general) takes us to language universals, which are a part of the universal grammar that is shared by all languages. Languages differ from one another because of the different applications of the parameters of universal grammar. All linguistically competent speakers necessarily have a grammar simply because they are speakers. This grammar comprises all the necessary elements to account for their competence. Up to now we have used competence in its most restricted sense. As a result, to explain what speakers know when they construct an utterance like (1) Why don’t you just ignore him, he’s often quite indiscreet. they must be able to handle the following linguistic information:
an interdisciplinary field
27
• the words that make up the utterance, • the collection of restrictions of words like ignore, impose on the field they dominate, • the phrases that the words can be grouped into in order to produce utterances, • the projections of the words onto syntactic structures, • the phonological rules that allow speakers to pronounce the words and their combinations, with the necessary stress, • the semantic rules that allow listeners to interpret each word with the meaning and each thematic or logical role within the complete structure of the sentence, • the rules that imply that there must be a referent outside the utterance, who has the characteristics of someone indiscreet and which is represented in the utterance by the pronouns he/him. But this is not all. The utterance would be absurd unless the referent is known to the speaker. This information, which the speaker obviously has, is independent of the linguistic structure and does not belong to the speaker’s linguistic knowledge, but rather to what the speaker knows about the real world. In this sense, we must distinguish between the knowledge speakers have about language and the knowledge they have about the world. What speakers know about language constitutes their linguistic competence; what they know about the world is their extralinguistic competence. Let us now look at (2) below, in contrast to (1): (2) Forget him! That guy is a loudmouth. Even though speakers of American English know that (1) and (2) basically mean the same thing, they are able to differentiate between the situations in which one or the other can be used; to be more precise, they know in which situations (2), the marked utterance, can never be produced. If this is the case, it must be because in addition to the knowledge about the language and about the world, speakers also have information about the rules of language usage, which constitute their pragmatic competence. Theoretical linguistics focuses on the description of language competence, and leaves the description of this other type of knowledge to pragmatics and discourse analysis. 1.1–Theoretical linguistics and applied linguistics Until relatively recently it was common to use the word linguistics, without modifiers, to refer to theoretical linguistics. There were no linguists outside linguistic the-
28
terminology
ory; at most there were language teachers, speech therapists, style editors, translators, lexicographers, etc., but no linguists. There is an explanation for this situation. First, only a few language specialists not concerned with linguistic description had a solid knowledge of the principles of linguistics and of the general rules of language structure. Consequently, their interest was almost exclusively practical in nature. Due to this lack of theoretical knowledge, they often derided it and deemed it useless for solving linguistic problems. Secondly, due to a strict interpretation of scientific methodology, the connection linguistic theory makes between language and competence, made it difficult to study language usage systematically. But language use, with its great variety of manifestations, is much more complex than linguistic competence. If we want a complete account of language, this cannot be given in terms of formal and semantic structure alone. It must also be analyzed from its social aspect. People express themselves, communicate and socialize with one another primarily by means of language; we use language to transmit most of our traditional beliefs and to organize thought. As such, describing language not only involves accounting for the competence of speakers in general, but also for speakers’ usage in specific communicative situations within a complex society. Applied linguistics studies language in its social function as a structure and a tool for communication, as a system and a way to solve the communicative and informative needs of society. The increased importance given to applied sciences has played a role in this change of attitude, as has the pragmatism that presently dominates our behaviour. This new social attitude has favoured the development of various branches of applied linguistics, such as language teaching, language therapy, computational linguistics, lexicography and terminology. Undoubtedly, terminology as a discipline is one of the privileged branches of applied linguistics. From an epistemological point of view, it could be classified among the ’applied sciences’ because it has to develop both a theory and a praxis. This praxis is evaluated through scientific principles, mainly through the so-called glossaries or lexicons, which are its end results.—Goffin (1985)
1.2–Variety of the language system For technical reasons, the notion of competence presupposes an abstract and idealized notion of speaker-hearer, an abstract person outside space and time who has completely internalized the data of his native language. In reality, on the other hand, speakers are far removed from this ideal; they are influenced by their environment and the epoch they live in and only partially know the data of their language. They
an interdisciplinary field
29
belong to a linguistic community that may also suffer interference from contact with other languages. The system of a real language, then, is not as homogeneous as that proposed by the linguistic theory of competence. From the standpoint of performance, it can only be conceived of as a system composed of various interrelated subsystems, such as dialects, sociolects, and historical variants. All speakers are influenced by their geographic origin, the social group they form a part of and the generation they belong to. In addition to these natural varieties, all speakers also mould their utterances to specific communicative situations by means of various functional variants or registers required by the communicative situation. These registers are marked by several different criteria: a. The medium used to transmit the information (oral/written) b. The subject matter involved (general/special) c. The communicative purpose or functional value (inform/assess/influence/ persuade) d. The degree of formality among the participants and the level of abstraction desired for the information (formal/informal; more abstract/less abstract) The terms of these dichotomies do not enter into absolute opposition, but rather allow varying degrees of nuances and intersections. Applied linguistics views language as a heterogeneous system of dialects and functional varieties, and allows us to place terminology as one of its branches since it is a part of one of the functional subsystems determined by subject specialization.1 1.3–Lexicology One of the basic components of any language is the lexicon, which consists of the words of the language and the rules accounting for a speaker’s creativity. Words are also units of reference to reality and connect us to the real world. For a number of years generative theory paid little attention to the description of how the lexicon works, as it was considered too irregular and asystematic. This situation has changed in the past few years and the lexicon is now being studied in depth from the viewpoint of its regularities. The lexicon of a speaker has been defined as the set of lexical units containing phonological, morphological, syntactic and semantic information, the appropriate set of word formation and readjustment rules, the set of possible projections on syntactic structures and a set of restrictions on rule application. In a linguistic theory centred on the description of speakers’ competence, a coher-
terminology
30
ent lexical theory must be able to account for everything speakers of a language know about the words they use and could use. Understood in this way, the goal of lexicology is to construct a model of the lexical component of a language which includes speakers’ implicit knowledge of words and their use as well as systematic and appropriate mechanisms to connect the lexical component with the other grammatical components. It must also account for the capability speakers have to create new units by following systematic structural models. The information about words must account for speakers’ lexical knowledge, independent of the language they speak. Speakers do not, however, limit their knowledge of words to linguistic aspects, because, in addition to linguistic information, they have paralinguistic, communicative and extralinguistic information which determine the real usage of each word. For a word like inquisition, for example, in addition to linguistic characteristics like: • • • •
pronunciation [˘ın’kw˘ı-z˘ısh’ən] related to the verb ‘inquire’ usually occurs in count noun contexts meaning ‘action of inquiring’
speakers also have extralinguistic information such as: • Historically this word refers to the ecclesiastic court that defended the ‘purity’ of the catholic faith by condemning those who deviated from the rules set by the Catholic Church. • It is a semantically marked word in the sense that it is associated with an organization reproached by all who defend human rights. • The word is semantically related to inquire because it refers to the name of a court based on the interrogation of the accused. Extralinguistic knowledge ensures that speakers do not use this word when they simply want to refer to the action of inquiring or investigating. Finally, it is important to point out that a word is not an isolated unit within the set of lexical units of a system, but is closely tied to the other units of the same level that constitute the lexical system of a language. Each word is the centre of a network of relationships with other words, and the lexical system becomes a complex network made up of groups of interrelated words. 1.4–Lexicography Lexicology describes the words of a language and explains how speakers operate
an interdisciplinary field
31
lexically. In contrast, lexicography deals with the principles and methods of writing dictionaries. A dictionary is a linguistic product that brings together a chosen set of words (or other language units) and provides information about them. The way the entries are selected and ordered in a dictionary constitutes its macrostructure and the information about the entries, its microstructure. Dictionaries vary both according to the linguistic information they contain and the purpose they serve. Typologies of dictionaries are usually based on the deviation of the various types of dictionaries with respect to the basic lexicographic pattern established by general language dictionaries. General dictionaries are usually classified according to the following criteria and contain the following information: a. Sources for the information: a selection of various source materials, most of which are written b. Choice of entries: the most usual forms c. Form for entries: e.g. the lexeme d. Order of entries: e.g. alphabetical e. Information accompanying each entry: • grammatical category • definition • semantic uses determined by the field of usage or by change of meaning processes • examples illustrating usage f. Primary purpose: e.g. descriptive g. Type of reader: e.g. reasonably educated speaker h. Purpose of the dictionary: e.g. increasing the competence of the user and/or resolve linguistic vacillations or gaps Any dictionary that deviates from this pattern is a ‘special’ dictionary because of any of the following variations of structure and content: a. The sources: e.g. a dictionary of the languages used by one author. b. The choice of entries: e.g. a dictionary of physics; a basic dictionary, if the criterion for selection is the extension of a form; a dialect dictionary. c. The nature of entries: e.g. a dictionary of collocations or fixed expressions. d. The order of the entries and the arrangement of the information: e.g. a thesaurus. e. The type of specific supplementary information that systematically accompanies
32
terminology
the entry: e.g. the origin and evolution of the word (etymological dictionaries), geographic varieties (dialect dictionaries), etc. f. The social function it aims at: e.g. a dictionary of standard usage. g. The target group: for example, a school-age dictionary. h. The specific uses aimed at: e.g. a dictionary for editors and technical writers; a bilingual dictionary for translation (Cabré & Lorente 1991). 1.5–Terminology The word terminology refers to at least three different concepts: a. The principles and conceptual bases that govern the study of terms b. The guidelines used in terminographic work c. The set of terms of a particular special subject The first concept refers to the whole field, the second, to its methodology, and the third to the sets of terms on a specific topic. In its first sense, terminology is generally seen as a interdisciplinary field that deals with the naming of concepts of special subjects, and their realization in linguistic or other forms. Even though, as we have seen in Chapter 1, the autonomy of terminology as an independent discipline has been disputed, we accept that terminology is a field with its own theoretical principles (terminological theory) and its own applied purposes (the writing of vocabularies, glossaries and dictionaries, and the standardization of designations). The concepts constituting the theory are not original, but, as in other interdisciplinary subjects, borrowed from the neighbouring disciplines, in this case linguistics, logic, ontology, and information science. If we accept that terms constitute a subcomponent of the lexicon of a language, since a speaker’s competence cannot exclude a specialized vocabulary (and even less so in the case of ideal speaker-hearers who know everything about their language), terminology clearly forms a part of linguistics. From the point of view of its base material, the terms, terminology is also a part of applied linguistics. However, terminology does not make use of all linguistic concepts; it chooses according to its objectives, i.e. it takes elements from morphology, lexicology and semantics and only operates with a limited number of concepts from these branches of linguistics. We conclude that terminology is an interdisciplinary field of enquiry whose prime object of study are the specialized words occurring in natural language which belong to specific domains of usage. Finally, considering that terminology is the result of the intersection between
an interdisciplinary field
33
linguistics and the other disciplines from which it takes its terms we have yet a third angle from which the subject can be studied. Consequently, there is no single scientific approach to terminology, but several. The three most important of these are: a. For linguistics terminology is a part of the special lexicon that is characterized by subject and pragmatic criteria. b. For scientific-technical disciplines terminology is the formal reflection of their conceptual organization and thus an essential means of expression and communication. c. For the user (either direct or intermediate), terminology is a set of useful communicative units which must be evaluated from the point of view of economy, precision and suitability of expression. 1.6–The specificity of terminology If we start from the principle that terminology requires a different approach to words than is usual in linguistics, it must have elements that clearly show this difference. Wüster points out that linguistics and terminology approach their object of study in different ways. These differences lie in the way language is conceived which can be shown in two specific elements: the approach to term formation and the characteristics of dictionaries written from a linguistic or a terminological viewpoint. Regarding the perception of the nature of language, lexicology is based on words and does not conceive of meaning unless it is related to the word; terminology, in contrast, considers that the concept, which is its main focus, is prior to the name and can be conceived of independently from the name or term that represents it. In addition, lexicology is always linked to grammar. Words in dictionaries are described with respect to their use in context; they are considered as elements of discourse. For terminology, on the other hand, terms are of interest on their own account, and neither inflection (provided by the morphological form appropriate for its use in context) nor syntax (which inserts them in the proper grammatical context) are of consequence. Finally, whereas linguistics distinguishes between and includes both synchronic and diachronic features of words, terminology is only concerned with synchronic aspects. Regarding the formation of terms or words (what Wüster calls the ‘‘conscious formation of language’’), general linguistics and thus lexicology defend the free evolution of languages and reject any sort of intervention. They are not concerned with prescriptive aspects of language and exclude consideration of standardization
34
terminology
(in the sense of establishing one form by deprecating others) because this reduction involves impoverishing the language. Terminology, on the other hand, permits intervention and has as one of its objectives the establishment of standardized forms. In this respect it differs from the purely descriptive views of linguistics. Terminology also conceives of terms in an international sense, and as a result gives priority to those methods of term formation that bring historical languages closer to one another. This leads to the adoption of international criteria for term formation and guidelines that are broader than those customary for a single language, e.g. the preference for terms formed with Greek and Latin roots and affixes. It is the task of terminologists to find a happy medium between authenticity and internationalization.2 Given this international outlook, terminology is concerned with the written rather than the spoken form of words (both the full form and variants such as initialisms, short forms, abbreviations). The second specific characteristic of terminology identified by Wüster refers to the fact that lexicology and terminology present their inventories of words or terms (dictionaries, lexicons, vocabularies) in different ways because they start from different viewpoints: terminology starts with the concept and lexicology, with the word. These differences can be observed, for instance, in the definitions that both disciplines use for their units, in the sequencing of articles and in the fact that only terminology uses symbols, especially from logic, to indicate the relationships among concepts. Since it starts from the concept and then proceeds to the designation, terminology must be absolutely sure that it is naming a specific concept and not a similar one. As a result, terminological dictionaries favour exhaustive descriptive definitions of concepts which often also indicates the relationships among related concepts. Lexicography of the general language is less explicit, and is chiefly concerned with avoiding identical definitions, unless words are completely synonymous. Since terminology starts from concepts, and concepts in the same field form a structured set, and because technical dictionaries of a special field are addressed to those people who are already experts, terminological reference works often present entries in a systematic order as opposed to the alphabetic order of general language dictionaries. Systematic ordering is in itself an attempt to reproduce the conceptual system of a special subject. Ordering by concept also allows a better layout for multilingual dictionaries, since in theory the concept is the same for all languages, and the difference lies in the designation.3 1.6.1–Terminology and lexicology The large number of characteristics shared by lexicology and terminology allows us to treat them as closely related fields, because:
an interdisciplinary field
35
• both deal with words • both have a theoretical and an applied side • both are concerned with dictionaries. The overlap between the two is not total, however. Some people consider one (terminology) to be a part of the other (lexicology), while others claim the two can be differentiated. Some of the characteristics that can be listed as being distinctive are (a) the domain; (b) the basic unit; (c) the purpose; and (d) the methodology. a. The domain As already stated, lexicology deals with the analysis and description of the lexical competence of speakers. To account for this competence it is assumed that all speakers have a repertory of words (allowing them to exchange information with other speakers of the same language) a set of word-formation rules (allowing them to form new words) and linguistic and encyclopedic data about each word (thus explaining how they can use them correctly, precisely and properly in each communicative situation). If lexicology deals with all the words of a language, terminology only focuses on the words belonging to either a specific field (such as physics, chemistry, anthropology or drawing) or to a professional activity (such as business, industry, sports, etc.). The domains of lexicology and terminology, then, do not coincide: the domain of lexicology is wider and includes that of terminology. By this criterion terminology would be a part of lexicology. b. The basic unit Lexicology deals with the study of words, whereas terminology deals with terms. Terms and words are similar and different at the same time. A word is a unit described by a set of systematic linguistic characteristics and has the property of referring to an element in reality. A term is a unit with similar linguistic characteristics used in a special domain. From this standpoint, a word of a special subject field would be a term: thermometer (metrology)—an instrument used to measure temperature thermometry (technology)—the technology of temperature measurement Linguistic analysis of the above examples does not provide any specificity of the terms to differentiate them from words in the common lexicon. From a linguistic point of view, a word is a unit characterized by having a phonetic (and graphic) form, a simple or complex morphological structure, grammatical features, and a meaning that describes the class to which a specific object belongs. A term is also a unit presenting the same characteristics.4
36
terminology
Once the terms of a terminological inventory are analyzed and contrasted with the words in a dictionary, some specific differences emerge. For example, the methods used in term formation do not have the same frequency as those of words. In terminology, units made up of learned formatives and set phrasal constructions are usually much more productive than in general language word formation. This does not mean that terms cannot use the same morphological elements as words, nor that the lexical formation rules cannot be the same, but the presence of certain Greek and Latin combining forms and the frequency of compounds in terminology introduce a differentiating factor. There are other differences as well. Whereas a terminological inventory usually contains only nouns, in a general language dictionary all grammatical categories are represented: nouns, verbs, adjectives, adverbs, determiners, pronouns, prepositions, conjunctions, and even interjections. Regarding coverage of word classes lexicology and terminology are clearly different. Other aspects also reinforce the difference between terms and words. Words are not just linguistic units that can be described solely from the standpoint of the system of the language; they are also communicative (pragmatic) units that identify speakers by the way they use words in specific expressive or communicative situations. Pragmatics is the factor that most significantly differentiates terms from words. Pragmatically, terms and words differ with respect to their users, the situations in which they are used, the topics they communicate, and the type of discourse in which they usually occur. Users of words are all speakers of the language. Users of the terms are the professionals that deal with the special subject field in question. Words are used in widely varying situations, whereas the use of the terms of a special subject field is usually limited to professional circles. Terms are normally used to talk about specialized topics. In contrast, words are used to refer to any subject in daily life, to express feelings and give orders, and to refer to language itself. Finally, the types of discourse in which words are employed are less restricted than those in which terms are used. Terms usually appear in technical and scientific discourse, i.e. in texts that are predominantly objective in nature (Cabré 1991c). c. The objectives Lexicology deals with words in order to account for the lexical competence of speakers. Terminology, on the other hand, deals with terms in order to establish a reference to concepts of the real world. The purpose of general language dictionaries is to identify the lexical units of a language and to present them in a conventional form for practical use. Terminology does not explain terms within the framework of theoretical linguistics, nor does it pretend to describe the terminological behaviour of experts. On the contrary, it only
an interdisciplinary field
37
attempts to provide theoretical constructs and principles governing the choice and order of terms of special subject fields in order to standardize their form and contents. As a result, the objectives of terminology are clearly different from those of descriptive lexicology, because terminology does not attempt to provide an explanation of the knowledge that experts have of terms (which they obviously have, otherwise we could not explain how they could use terms properly). Instead, terminology tries to identify segments of a specialized professional reality. Terminology, then, is aimed at identifying and naming the concepts belonging to a specific subject. d. The methodology Finally, the methodology used also differentiates lexicology from terminology. Lexicology works from theoretical hypotheses, which it refutes or confirms by analysing samples of speakers’ discourse. Terminology, on the other hand, does not explain human behaviour, but rather looks for terms to fill in a previously established conceptual grid. 1.6.2–Lexicography and terminology Lexicography, as stated above, is the applied branch of lexicology and is concerned with the writing of dictionaries. Comparison of the pragmatic aspects and goals pursued by terminology (naming and standardization) with those of lexicology (description) leads us to separate the two subjects; comparative analysis of lexicography with practical terminology (terminography) leads to a more confusing situation. The output of lexicography and terminography takes the form of compilations of lexical or terminological units—i.e. dictionaries.5 From this point of view, then, their objectives overlap or even coincide. Despite this similarity, lexicography and terminography are different due to the other aspects that make a dictionary of commonly used words and expressions different from a special subject field terminology. a. The linguistic aspects of dictionaries Writing dictionaries is a multi-stage process, and at each point the author makes a series of choices that eventually lead him or her to different types of dictionaries. Terminological lists differ from lexical lists because at some stages of the process certain possibilities are chosen and others which are more appropriate to general language dictionaries are discarded. As a result, in terminology, specialized documents, whether oral or written, are the only source material. Entries, which are always based on lexemes (even though they may be compounds of one or more words), are chosen as a function of the subject. The only information that is retained is what is considered pertinent, and this information, particularly when it comes
terminology
38
from scientific and technical subjects, is preferably presented following international guidelines. b. The purpose of terminographic work In terminology the writing of dictionaries implies the standardization of the terms of a specific, specialized domain. Terminography is not limited to collecting the terms of a particular domain for informative or descriptive purposes, but rather the aim is to establish certain terminological units as standardized forms, as reference forms, thereby discarding other variants for the same concept. The final goal is achieving precise and unambiguous professional communication. In general lexicography there are also prescriptive dictionaries which have the same function. The words included in them are the correct forms accepted by the academic institution that issues the dictionary; in terminology, on the other hand, the standardized forms are preferred because their users (subject field experts), formally organized into standardization committees, have agreed on them. c. The working procedure in terminography Writing a general language dictionary starts with establishing a list of words constituting the inventory of entries for the dictionary. The lexicographer then describes them semantically by means of the definition. This process is semasiological—it moves from the form to the meaning. The procedure used in terminography is precisely the opposite. The terminologist first establishes the list of concepts that constitute a field. This list will be more or less restricted, depending on the subject field. The concepts in this structure are related logically and ontologically to one another, and the entire set constitutes the conceptual system of a discipline or special activity. The terminologist assigns a certain designation to each concept, and this designation is the term that specialists commonly use when referring to the concept. When several different designations exist for a single concept, either one is chosen and all the others are discarded, or more than one is accepted, but one form is given priority over the others. Working from the concept to the name is known as an onomasiological process. semasiological process lexicography word concept term onomasiologic process terminology
an interdisciplinary field
39
2.–Terminology and cognitive science Every cognition is the result of a psychic process, which leads to knowledge. This process is no state but an activity of a subject. Cognition is in the same way as knowledge, something psychic, attached to individual. There is no objective, detached cognition possible.—Felber (1985)
Terminology is based on semantics, and terms are units that relate language to the real world and represent objects in the real world. Specialists use terms to express themselves and exchange thoughts and organize the structure of their disciplines. Terms can be described in three ways: linguistically (already discussed in the previous section), cognitively, and with respect to communication. 2.1–The analysis of terms Any sign possessing a meaning can be represented on three axes: the form, the meaning, and the referent it represents. By placing it on one of these three axes, any term can be analyzed first in relation to the other terms of the same type with which it forms a specific subsystem; secondly, its relationships can be examined on each axis. Through the form we gain access to the formal system that allows us to form new words and expressions. This formal system consists of a series of units (word elements and words) and a set of term formation rules which restrict their combination. Similarly, through the meaning of the sign we gain access to the semantic system of a language. Meanings of individual signs are not isolated in the speaker’s mind, but form ordered semantic sets together with other meanings. Because of the ordering of the concepts into sets, speakers are able to store a large amount of data and find them immediately when they require them to express themselves. The referents of concepts, which in the real world are found in the form of concrete and abstract objects, do not constitute a chaotic, unstructured set, but occur in a specific order—obviously variable and multi-functional—which guides specialists when they establish the conceptual structure of a field. Ontology is the discipline that deals with analysing objects of the real world, their location and the relationships they have with each other. This systematic description of the real world, however, is still viewed differently by psychologists, philosophers, and linguists. The referent-concept relationship, which is one of the most difficult points to approach scientifically because we can only work with hypotheses and indirect empirical proofs, is assumed to describe how individuals know the real world. This topic is still quite controversial today. The controversy over whether reality exists in its own right objectively, or whether it only exists through an individual’s perception, has yet to be resolved. It is certain that individuals employ a variety of cognitive
40
terminology
strategies to understand the real world and as a result they behave according to this understanding. For instance, they avoid whatever they believe presents danger, they recognize the food they like, they orient themselves in a specific space, etc. Similarly, they use expressive systems, of which language is the most important, to refer to this real world. The signs they use to refer to it, however, are not reality itself but rather only its mental representation organized into concepts. Theoretically, speakers could form an individual concept of each individual object. However, if the human mind, which has limited capacity, had to store each unit of the real world as a different conceptual unit, memory, which is also limited, could not hold the amount of data that would be necessary. The most plausible starting hypothesis is that speakers store concepts representing a class of objects and not individual concepts. By a process of abstraction, speakers put aside the accidental and inessential features of objects and arrive at an abstract idea of a class of objects that is identified only by its relevant characteristics. The reason why speakers form one class and not another (thus basing themselves on a certain group of characteristics) is difficult to explain scientifically. The relationship between form and referent of signs raises a second question. Although this relationship has proved to be of interest to specialists in stylistics and poetics, in terminology it has not generated any interest because it does not appear to be essential to the discipline. The relationship between concept and form, on the other hand, is the most seriously studied topic of the three relationships, because it concerns the question of the possible ambiguity of signs. In terminology this question is of major importance since the absence of ambiguity and the single reference of a term to a concept are essential elements for effective communication. If we compare the vocabulary of general language and terminology in this respect we see that the two systems differ significantly. Most words in the lexicon of general language have multiple meanings. Each linguistic form is associated with numerous meanings (some of which are clearly related to one another: e.g. neck: referring to an animal, a part of a bottle, a part of clothing; in other cases the relationship is more remote: e.g. bank: an entity where money is kept, a set of machines in a row, a seat). Theoretically, terms should be unambiguous and have one meaning and only one designation corresponding to one form. The polysemy of the common lexicon is treated as homonymy in terminology. A term in the system of a subject field should identify only one concept (e.g. in anatomy, the term neck has only one meaning). In practice, however, even within a single subject, we note that a single form can be related to more than one meaning. Wüster’s model of the sign, originally inspired by Ogden and Richards’ (1923) triangle, illustrates the three dimensions of the term as a sign and shows the relationships that terminology deals with:
an interdisciplinary field
B
A meaning
symbol
B1
B2 b1
a1 and a2 A1 and A2 A B B1 and B2 b1 and b2
41
b2
A1
A2 a1
a2
represent individual objects in reality are the individual concepts representing a1 and a2 is the abstract concept representing A1 and A2 is the concept of the symbolic representation of A are the individual abstract representations of a phonetic or graphic form are the individual phonetic representations of B1 and B2
This diagram of the term, showing its referential, conceptual and symbolic dimension, allows us to identify three basic aspects of terminology: the cognitive aspect, the linguistic aspect, and the communicative aspect. A term is a unit referring to a reality that is expressed by means of a form and used for intercommunication. Of the three, the cognitive aspect is the most complex to describe. Cognition is the result of a mental process that leads to knowledge. The problem of how human thought understands objects, and by abstraction, constructs concepts, is at the root of the theory of terminology. Cognition is a mental process that consists of understanding a reality. A cognitive theory of terminology should provide an explanation of three key issues related to knowledge: a. How individuals conceive of reality and structure knowledge. b. What concepts exist, how they are formed, how they are related to one another, and how they are ordered within the structure of knowledge. c. How concepts are related to terms. These fundamental theoretical questions have not received much attention from terminologists because most of them are almost exclusively interested in applications.
42
terminology
2.2–Concept formation A concept is an element of thought, a mental construct that represents a class of objects. Concepts consist of a series of characteristics that are shared by a class of individual objects. These characteristics, which are also concepts, allow us to structure thought and to communicate. In order to communicate concepts and their supporting propositions, speakers use written or oral linguistic signs made up of a term or groups of terms, or some other type of symbols. What they express, however, is not the real world as it is but rather how the individual and the community have internalized it. Language does not reflect the real world exactly, but rather interprets it. This explains why a single segment of the real world (a special subject field) can generate different structures simultaneously (different scientific theories) or successively (scientific changes). Concepts are mentally independent of terms and exist before they are named, as opposed to meaning which, as Saussure stated, is inseparable from its sound image (‘signifié’).6 Within logic, conceptology deals with the nature of concepts, their representation and the relationships between concepts. Ontology is the field dealing with objects, how they are ordered in the real world and the relationships between objects. If we grasp a segment of reality by a process of cognition, and turn it into a concept by a process of abstraction, the relationships established between objects in the real world and the similarities among these objects reflect the relationships and similarities established between concepts. Current linguistic theory, as discussed in Chapter 1, 2, considers language as one of the systems of knowledge of the world alongside other cognitive systems such as touch, hearing, and sight. The knowledge speakers have about language is systematized in an internalized grammar which explains how utterances are produced and understood. This grammar includes all the knowledge speakers have about the language they use—the words, syntactic rules, pronunciation rules, and rules of semantic interpretation. An utterance is a logical opinion related to the surrounding world, a predication about the world in which we live. Utterances have a semantic structure made up of the semantic content of words7 and that of sentence functions. The lexicon consists of words that all speakers know and the set of rules that permits them to create new words in order to increase the size of their lexicon. All speakers have words of a general nature they can use in any situation; they have other words and terms which they use in more restricted situations and can adapt to various functional situations such as professional contexts. From the point of view of terminology, therefore, the lexicon of a language consists of the many separate subsystems representing the knowledge structure of each subject field or discipline. Each knowledge structure consists of variously interlinked concepts.—Sager (1990)
an interdisciplinary field
43
If we accept the above reasoning, we have to support the hypothesis that all speakers have a lexicon which is made up of a large number of words shared with most other speakers of the same language, but simultaneously consists of more restricted sets of words they only share with individuals with whom they have some other characteristics in common, for example, those working in the same profession. Professionals in the same field share a subset of specialized vocabulary which they acquire naturally as their knowledge of a certain field advances. Terms, then, are not semantically isolated units, nor is the knowledge of the specialized world produced by means of isolated terms. As speakers become more familiar with a special segment of the real world, they turn their knowledge into conceptual structures in which each concept occupies a specific place and acquires a functional value. Terminology thus is the basis for the structure of thematically specialized knowledge. These variable conceptual structures reflect the knowledge that one or more specialists have of a subject; it is, therefore, often difficult to freeze conceptual structures in time and there is room for a wide range of viewpoints among specialists about certain relationships among concepts. Accordingly, the cognitive approach to terminology presupposes that the work of the terminologist begins with establishing the structure of concepts which summarizes the knowledge of a discipline and then searches for the names of each concept in this structure. Traditionally a concept is, as stated above, an element of knowledge that represents a class of objects of the real world,8 consisting of a set of the characteristics shared by all the individual objects.9 Concepts are thus grouped together into sets (conceptual fields) and share some characteristics. There are, however, some theories of concepts that are more specific. Sager (1990), for example, presents a model of knowledge in the form of a multi-dimensional space made up of a series of intersecting axes, each of which represents a class of conceptual characteristics or dimensions in an intersecting relationship. The choice of one dimension for a concept can limit the choice of another different dimension, even though it may not do so necessarily.10 In this space, a concept is a unit of knowledge that is represented and identified only by the reference to the coordinates on these axes. Placing a concept in a position in this space of knowledge means giving it a value, and this value is expressed in its definition. In Sager’s (1990:15) model ‘‘the value of a concept with respect to an axis is generally defined as a range and only exceptionally as a point.’’ Most concepts do not have rigidly set borders but rather approximate, diffuse limits. With standardization the limits of concepts are artificially defined. The structure of knowledge can change for two reasons: a new axis may appear, which means that all the previous concepts have to be redefined in function of this new dimension, or a new way of seeing things may arise, which would not change the axis for preexisting concepts, but would mean that their position on the axis changes.
44
terminology
Social norms determine the knowledge structure specialists are expected to have of special subject fields. Becoming a specialist means acquiring the terminology which represents this knowledge structure. 2.3–The relationships between concepts Of the three key topics in terminology related to thought and knowledge—the acquisition of knowledge, the formation of concepts and the organization of concepts, the last one has attracted the most interest and work by terminologists. The first topic has been studied primarily by psychologists and is currently the object of study by specialists in artificial intelligence and expert systems. The second topic has been studied mostly by logicians and philosophers, who have been more interested in thought than in cognition. Finally, the third topic has been analyzed by specialists in scientific and technical fields, who have been interested in establishing the conceptual structure of their field. Terminologists have also had to play a role in one facet of this third topic, namely the organization of conceptual systems, in order to establish the structure of concepts in a field, because this structure, prior to the naming of concepts, constitutes the starting point for the onomasiological process of applied terminology. The criteria for organizing concepts that make up a single field, as well as their membership of a specific set, depends on the way objects in the real world are understood. Within each field concepts are related to one another on the basis of two major types of relationship: logical relationships, based on similarities, and ontological relationships, based on the proximity or contact in space or time. 2.4–Subject classification Terminological tools, based primarily on the concepts contained in a document and expressed by means of names or terms, are used in documentation and information, especially in order to retrieve stored information. Professionals in the field of information, who classify information and store it in computerized systems which allow quick and selective retrieval, need to have classification systems for the content of documents. These systems have been primarily based on two features: on the identification of the main concepts contained in a document and on the subject classifications of the domains of knowledge. Thesauri, which are collections of semantically related terms, are essential tools for organizing and retrieving information. All documents transmit information based on a specific arrangement of the concepts they contain. Each concept conveys information, and various types of relationship exist among concepts. The frequency of appearance in a single document of
an interdisciplinary field
45
certain sets of logically or ontologically linked concepts allows us to identify the contents of a document because the concepts are expressed via linguistic forms. In this way, the analysis of the conceptual content of a document allows us to identify and describe the topic under discussion. Thesauri are used to control the names of the concepts and the relationships among them, and thus enable more efficient retrieval of information. The main advantage of thesauri, even though they cannot reflect the complicated relationships of objects in the real world, is their three-dimensional nature. The classification of knowledge by subject areas or special domains also permits the classification of technical documents into fields and the retrieval of the documents relative to each field. A more or less specific classification corresponding to each subject or domain reflects the organization of the contents of the field in question. The most common type of classification is hierarchical. Although the usefulness of this sort of classification based on a single dimension is useful, it cannot reflect the structures of the real world because they are much more complex and multidimensional in nature. To facilitate consultations, hierarchical classifications must be accompanied by alphabetical indices with all the names that appear in a classification. This allows for access to a hierarchical structure via both the form and the content.11 3.–Terminology and communication Every society—which may or not may constitute a language community—comprises various social strata. All existing subcategories such as regiolects and sociolects go to make up the LGP (language for general purposes) of the language community. This stratification is reflected in phonetic, lexical, syntactic and semantic variations. Other sociolinguistic distinctions exist besides those directly connected with social class. . . Communication with these groups calls for the relevant special languages. This is the social aspect of LSP (language for special purposes).—Picht & Draskau (1985)
The ordering of thought and the conceptualization represent the cognitive side of terminology, the transfer of knowledge constitutes its communicative side. Terminology is the most important characteristic of specialist communication because it differentiates special languages from the general language and also the various special languages from one another. Experts use terminology not only to order thought, but also to transfer specialized knowledge in one or more languages and to structure the information contained in specialized texts. 3.1–Specialized communication In principle, subject-specific communication among specialists is not very different from general communication. The restrictions imposed on the elements involved in
terminology
46
special communication that give it a specificity not found in general communication are of a different sort. As defined by Jakobson, all communication involves the following five elements: reality encoder
medium
decoder
language The act of communication results in a sixth element which simultaneously links the five elements, detailed above, to one another—which, considering only communication via natural language, is either a written or oral message. The encoder and the decoder start from knowledge acquired prior to the act; i.e., they already possess: a. knowledge of the real world they want to communicate b. knowledge of the social norms of usage of the language, i.e. the effect of certain linguistic forms (pragmatics of variation), and the conventions of expression (text type, standards of graphic representation, standardized forms, etc.). In addition, both participate in the communicative act with certain intentions and expectations. The encoders’ intentions (to inform, describe, ask, assess, order, etc.) condition the form of the message, and they attempt to mould the message to the expectations they assume the decoders to have. Decoders assess the message in accordancewiththeirexpectations.Ifthetextmeetstheseexpectations,communication is successful; if not, a decoder can indicate, either by means of interrogative language or nonlinguistic expressions (gestures, facial expressions, movements, etc.), that the communication was unsuccessful and the encoder can them make another attempt. If there is a lack of interest on the part of the decoder, the encoder stops trying. In special communication, as opposed to what we have just outlined, there are a series of restrictions of the scope of each factor in communication. First, the two participants in the communication are to a greater or lesser degree experts in a subject field12 and communicate with each other presupposing that they share a certain amount of information about the area of knowledge at issue (Sager, Dungworth & McDonald 1980, Varantola 1986).13 Second, the reference world of their communication is limited to that of the special field, which is more formally conceptualized than the world expressed by general language. The communication system also includes general language, which supplies the syntax, morphology and a part of the lexicon. Finally, the text types generated in scientific and technical communication
an interdisciplinary field
47
are fundamentally informative and descriptive in nature, and their main function is referential.14 Nevertheless, there is nothing in principle that prevents a text of this sort from containing elements that indirectly lead to other communicative purposes such as causative, expressive, and metalinguistic. 3.2–Terminology and specialized communication Specialized communication differs from general communication in two ways: in the type of oral or written texts it produces, and in the use of a specific terminology. The use of standardized terminology helps to make communication between specialists more efficient. The criteria they use to evaluate specialized texts are not the same as those used to evaluate general texts. In general texts, expression, variety and originality prevail over other features; in specialized texts, concision, precision and suitability are the relevant criteria. A scientific text must be concise because concision reduces the possibility of distortions in the information. It must also be precise because of the nature of scientific and technical topics and the functional relations among specialists. Finally, it must be appropriate or suitable to the communicative situation in which it is produced so that, depending on the circumstances of each situation, every text is adapted to the characteristics of the interlocutors and their level of knowledge about the topic, introducing more or less redundancy according to need. Terminology plays a major role in achieving these three objectives. On the one hand, the original terms used to name a specialized concept are usually concise;15 the use of a term instead of a paraphrase contributes to precision; and finally, the use of standardized terminology is the best resource that specialists have to refer to an area of specialization which they share. 3.3–Terminology and translation Beside specialists, translators and interpreters are also major users of terminology when they mediate communication between specialists. Translation is a process aimed at facilitating communication between speakers of different languages. Multilingual terminological activity supports technical translation. Translation implies understanding the source text and this requires knowledge of the specific terms of the source and target languages. This means, in turn, that technical translators must have some familiarity with the subject matter they are translating. A good technical translation not only has to express the same content as the source text, but it also has to do so in the forms that a native reader of the target language would use. In the case of specialized translation, the reader will be a spe-
48
terminology
cialist in the field. As a result, good technical translators must choose the topics they want to work in and attain minimal competence in some specific field in order to be certain to respect the content and form of the two languages they work in. To do their job translators depend on bilingual or multilingual vocabularies of the terms occurring in the text. This does not mean, however, that translators do not prepare terminology themselves. On occasion they have to act as terminologists to find equivalents for those terms that are not listed in the available vocabularies nor in specialized data banks. Besides, the time constraints within which translators often have to work may not allow them to hand the task over to a terminologist. In addition to the equivalents in other languages, terminology prepared for translators must contain contexts that provide information on how to use the term, and, ideally, provide information about the concept in order to ensure translators use the precise form to refer to a specific content. Terminological equivalence is the key to multilingual terminology. The authors of bilingual or multilingual technical dictionaries often start from the principle that terminologies reflect objective structures of the real world. In fact, the way the real world is reflected in the structure of a special language may not be the same across languages, especially in fields that are not highly structured such as arts and social sciences. 3.4–Terminology and language planning The communicative nature of language is also the driving force behind the language and terminology planning that is currently being carried out, more or less systematically, in a large number of countries around the world. The health and survival of a language depends on its being appropriate for all contexts of communication identified by a society. A language reduced to informal usage only begins to lose its prestige and in the end disappears. In this sense special languages are key parts of the real capabilities of survival of a language. Developed or developing societies need to have at their disposal an up-to-date language they can use for international contacts. This language must have modern terminology—either created or borrowed—that allows it to refer to all the new concepts needed for the development of technology. When the language of international communication does not coincide with a country’s own language, there must be a language policy to determine in which cases the native language should be used and where the international one is better suited. Linguistic intervention in support of a language is necessary wherever the language of a region is not the same as the official language of the country it belongs to and within which it is in a minority situation. Policies to protect native languages are also needed in countries which have recently become independent and in which
an interdisciplinary field
49
the language of the former colonial power has the role of official language and lingua franca for the whole country. The standardization of a language must be a planned process that begins from an explicitly defined situation, has goals to be met within a period of time, suitable channels for dissemination and resources for implementation, and, finally, a legal framework that favours this process of change. According to Auger (1986), terminological standardization only makes sense within a framework of general language planning, with six basic functions: research, standardization, dissemination, implementation, evaluation and control, and updating. In a dynamic process these six functions can be successive stages going from the initial conception of a terminological product to putting it into operation in a specific environment. We can identify the following eight successive stages in the process: a. Analysis of the terminological needs of a situation in accordance with the overall situation, and selection of the most suitable strategies for intervention b. Preparation of a terminological research plan adapted to the needs of the environment in question c. Preparation of the terminology with the participation of relevant users d. Standardization of the prepared terminology e. Choice of the most suitable format and presentation for the prepared terminology f. Implementation of the terminology in practice by suitable policies g. Monitoring the use of the terminology h. Constant updating of the terminology A language standardization plan can be deemed to have been successful when the situation of a language has been changed for the better because the success of a terminology plan does not end with its preparation, but rather depends on its acceptance by real users. Well prepared products alone are not enough to attain this goal. Products must be adapted as much as possible to each situation to favour their use. The dissemination of terminology must be accompanied by effective actions to encourage usage and by indirect measures to change the customs, attitudes and prestige of the language being standardized. From the point of view of language planning (and also of the terminological development of language), we must dispose of the traditional idea that conceives of terminological products (such as dictionaries, lexicons and vocabularies) as being useful for the translation, encoding and writing of specialised texts. Especially conceived for translators and technical writers, these products play an essentially static role which is reflected by the
50
terminology
position they occupy on desks and, only in the best of cases, on shelves. It is well-known that people do not read dictionaries and, consequently, distributing lexicons is not sufficient to initiate a change in the language attitudes of users.—Auger (1986a)
Teaching standardized terminology to future specialists during their university studies is one of the surest ways of modifying usage, but it is slow. Training these specialists in the creation of neologisms and methods of systematic designation of terms, and teaching them to use the resources of their own language when faced with a gap, provides a language with a strategy that is much more appropriate than substituting a term that has already taken root.
4.–Terminology and documentation We cannot separate terminology from documentation because every terminological work requires, directly or indirectly, a great amount of specialized documentation. This principle is widely accepted by all public and private bodies with a certain interest in terminology.—Rondeau (1983)
In addition to being the basis for the structuring and the transfer of knowledge, terminology is also the basis for the writing of technical texts (technical writing), for the translation of specialized texts (technical translation and interpretation) and for the description, storage and retrieval of specialized information (technical documentation). A document is a unit of information that can be described both by its form (types of document, format, etc.) and by its content (what it transmits). In an efficient retrieval system a document can be accessed both via its form and its content. The form of a document is described by bibliographical data. The content is represented by indexing terms which refer to the content by means of descriptors (key words, notations, symbols, etc.) Both types of information make it easier to access and retrieve information stored in manual or electronic files. Documentation is a relatively new field that deals with the collection, analysis, assignment, classification and storage of documents in order to make the information retrievable for various uses, users, and purposes. Like terminology, documentation is cross-disciplinary in the sense that it can be applied to any branch of science or human activity. It is practice-oriented because its aim is to provide information users with suitable documents. Documentation centres also disseminate information in the form of secondary (bibliographies and directories) or tertiary (bibliographies of bibliographies) publications. In order to provide access to information, information scientists analyze documents formally and semantically in order to describe them. Their working procedures can be summarized, albeit simplifying their complexity, as follows: formal
an interdisciplinary field
51
description of the document, description of its contents, and storage of both descriptions in a file or database. The automated information on documents is assembled in databases which are normally a part of a data bank that also stores the documents themselves. 4.1–The relationship between terminology and documentation Information scientists, i.e. indexers, librarians, abstractors, etc. describe the contents of documents by using terminology. Indexing consists of summarizing the substance of a document by means of key words or descriptors, which are terms; they are usually controlled by a thesaurus. Terms are used to express and represent all relationships among concepts in thesauri. The use of a standardized terminology to index documents makes documentation systematic and unambiguous, which would not be the case if words from the general language were used randomly. Standardized terminology is also an element of control for information scientists. In this sense terminology is at the service of documentation. The introduction of a thesaurus into a document database involves three operations related to terminology: a. Design of a classification of concepts b. Selection of concepts and declaration of the relationships among them (this involves controlling the hierarchical relations established) c. Selection of terms by eliminating synonyms, and, if possible, homonyms. While terminology is at the service of documentation, documentation is also at the service of terminology. Technical documents are the basis for terminological work. Terminologists do not invent designations for concepts in a previously established system, but rather identify and collect terms that specialists use in documents. Terminologists use documents in order to acquire knowledge of the subject and its conceptual structure, to search for the terms used in a subject field, to confirm the quality of the primary data they have collected, to study the data from the various viewpoints represented by different documents. To this end, terminologists study each term with reference to its semantics, function, form and status as it occurs in documents. It is only when various designations for a single concept are found, i.e. when there is a lack of agreement among specialists on the current form or when the term in question is linguistically not entirely appropriate, that terminologists can attempt to intervene and suggest unification or standardization. Documentation is present not only at the beginning of a terminological task, but throughout the entire process. The final product, the prepared terminology, in turn, constitutes a document.
52
terminology
Work in terminology results in a wide range of documents: glossaries, dictionaries, lexicons etc. The terms in these documents constitute a terminological database, which is also a document. 5.–Terminology, computer science and knowledge engineering Terminology and its applications are not a goal in itself. Modern terminology work is closely linked with documentation and informationscience. High quality terminologywork results in reliable, multifunctional terminographical data, which are primary elements of information and knowledge management systems. Advanced terminology documentation assisted by computer, therefore, by itself is knowledge engineering at the level of conceptual logic. It can be called terminological knowledge engineering.—Galinski (1990)
The use of computing resources in applied terminology will be discussed in Chapter 5. In his key article on the definition of terminology as a point of convergence of other disciplines, Wüster mentions computing as one of the fields intersecting with the various special subject fields to constitute the basis of terminology. For this reason we examine here the ties between the two subjects. At first sight, it seems obvious that we must make a distinction between the relationship that terminology has with fields like linguistics, logic or ontology (from which it takes theoretical elements for building its own domain of study) on the one hand, and the one which it has with information science, and computing on the other, as these are fields with which it has an bilateral relationship. Terminology provides computation and information science with conceptual structures and terms, and computer science produces useful tools for the processing of terminology. Terminology relies on computer science to provide tools and, as a result, is increasingly able to eliminate manual working procedures. In its turn, terminology provides computer science with concepts useful for the development of artificial intelligence and expert systems. The link between terminology and cognitive science is created by concepts. Concepts are the units constituting the basis of knowledge, and concept systems describe the way each field organizes knowledge. In this sense, the theory of terminology and the theory of knowledge are closely related. Concepts are the basis for both the theory of terms and that of artificial intelligence, the branch of cognitive science dealing with the design and creation of expert systems. The main goal of the theory of terminology is to establish the bases to account for the formation and use of concepts in specialized fields. The main goal of the theory of knowledge is to describe the formation of knowledge, the various types of knowledge and the interrelationship between speakers’ knowledge and its use in real situations. Both theories constitute the basis for a new field: knowledge engineering. Knowledge engineering, which is primarily an applied field, uses knowledge to
an interdisciplinary field
53
construct expert systems among which those related to language and communication are essential. The central topic of study and work in this new applied field is the identification, representation, treatment, transformation and transfer of knowledge. Knowledge engineering relies on the basic elements from the various fields that make up terminology (logic, ontology and documentation), and uses the development of cognitive science to implement knowledge in expert systems. Artificial intelligence starts from the assumption that an expert system that can perform ‘‘human’’ operations must possess the knowledge that humans possess. Speakers have acquired this knowledge through experience by using the various cognitive systems they naturally have. Following this argument, if we want a technological construct ‘‘to work like a human being,’’ it has to be equipped with the knowledge required to recognize and analyze a situation so that it can then act accordingly. However, since situations are not always the same and the actions taken by humans adapt to each type of circumstance, an expert system, if truly intelligent, must also be able to distinguish situations from one another and act accordingly. In other words, it should be able to reason. 5.1–Computer science at the service of terminology In 1974 Wüster wrote the following: Of all the sciences mentioned until now, computer science is the most recent one. It is the science concerned with building and using computers, which are not only used to solve mathematical problems. Computers are also a great technical aid for another science which only slightly pre-dated computer science: the science of documentation and information, thus the German term ‘‘Informatik’’.—Wüster (1981)
In Wüster’s days, computers were only tools for storing and later retrieving information. Since then the situation has changed dramatically, and the services computers can provide for terminology go well beyond being simply a way of storing information. Although the collection of documents, the study of concepts, placing the information in files and dissemination are essentially the same now as twenty-five years ago, working methods to achieve these goals have diversified, and tools have become easier to use and more powerful, since Wüster’s time. We can identify several different phases in the evolution of computer science as applied to terminology: 1. At first, computers were used as tools to store information. They were first applied to bibliography, and later to terminology. During this phase the first data
54
terminology banks of documents and bibliographies were created, followed in the 1960s by the first terminology data banks, which were similar in nature to a dictionary used for translation.16 The introduction of microcomputers reversed the trend of this first phase of large-sized data banks, and led to the creation of minibanks that are more specialized in terms of information and target users. During this first phase the computer systems were incompatible with the classification systems.
2. In a second phase, as people continued to work on making computerized data compatible, tools were designed to facilitate person-machine and machinemachine communication. Natural language interfaces were developed which changed some of the conditions of using large databases, making them more userfriendly. The possibilities for accessing information were also diversified: information can now be obtained remotely via a personal computer and a modem, via optical disks (CD-ROM).17 Finally, in this second stage the structure of data banks and concept classification systems was refined, and users were able to obtain selective information from the bank according to their interests. In addition, electronic versions of dictionaries and terminologies from information contained in the data bank were developed. 3. The third stage includes work on expert systems in linguistics and terminology developed within the framework of artificial intelligence. Some of the operations previously carried out by terminologists are now performed by computer, or a terminologist is aided by a computer. Computers directly participate in some activities such as the choice of documents for a task, the search for these documents, the preparation of automated records, the writing of definitions, the creation of terms, the preparation of standardization files, or the detection of new terms.18 Optical and acoustic systems for entering data, which are alternatives means of storing texts noticeably improve the services computers provide to terminology and documentation. The integrated work stations for terminologists consist of a personal computer containing a word processing program, several electronic dictionaries, a terminology database management system, several checkers, a program for the creation of databases from texts, a semi-automatic or automatic search program for terms that also allows semi-automatic writing of terminological files, a program for publishing data, access to several terminology, bibliography and other information databases, etc. Computer-assisted teaching has also changed the training possibilities in terminology.
an interdisciplinary field
55
4. The fourth phase of development of terminology-oriented computer science began with the incorporation of knowledge into expert systems; this is the stage of knowledge engineering. The function of a ‘‘cognitive engineer’’ (Lapalme 1988) consists of working with a human expert to establish the knowledge and basic mechanisms that all humans use in decision making. To do this, this knowledge must be codified in a formal system that can be understood by a computer in the form of decision rules that consist of a series of conditions for rule application followed by a decision or action. The rules use an inference module to combine with one another. Finally, an expert system of this type usually has a justification system made up of a series of rules allowing it to explain why one decision and not another has been made. The effectiveness of an expert system depends on the quantity and quality of the knowledge it possesses and the capabilities of the system of rules. At present expert systems can only act if they have previously been given information on the situation. They are consequently far from behaving like people, who know how to deal with unforeseen situations. 5.2–The usefulness of terminology for computer science Computer science also benefits from terminology because all computer programs on any aspect of language inevitably need terminology. Computers process lexical items or terminology to develop systems based on aspects of natural language processing. Processes like machine or machine-assisted translation, checkers, information retrieval, assisted writing and database programs developed in natural language, interfaces in natural language and expert systems based on knowledge all use terminology. All these systems require a dictionary or thesaurus containing the units the computer will understand. Systems that carry out activities on language (writing, correction, translation, etc.) need dictionaries as reference point for each operation. The spelling checker for a text, for example, is based on a reference dictionary containing a wide range of words in fully inflected form. Any lexical item not included in this dictionary is considered an error. The human operator must decide if the software has correctly identified a mistake or marked it because it did not recognize the form. Knowledge-based systems also require a dictionary to control the designation of concepts. In this type of system, thesauri are preferred because they contain information on the relationships among concepts, and these relationships are also structures of knowledge. Finally, systems that facilitate communication with computers in natural language (systems that access databases, systems that retrieve information, etc.) also require reference dictionaries. In this case the dictionary is not aimed at correction but rather at comprehension.
Chapter 3 The foundations of terminology
1.–Special languages The nature of language is such that general language and special languages can be accommodated within one natural language: the fundamental characteristics of language are manifested both in English and in the language of chemical engineering, both in French and in the language of physics. The difference between general and special languages is a difference of degree rather than kind: the degree to which the fundamental characteristics of language are maximized or minimized in special languages. Special languages are used more self-consciously than general language and the situations in which they are used intensify the user’s concern with the language. It is therefore on the level of use that we look for more specific differentiating criteria.—Sager, Dungworth & McDonald (1980)
A language is not a homogeneous system of expressive possibilities, nor is the grammar describing it a monolithic set of rules and units accounting for all the sentences its speakers produce. Language is a complex, heterogeneous system made up of interrelated subsystems, each of which can be described at the phonological, morphological, lexical, syntactic, and discourse levels. These five levels are the most common ones in linguistic description and the units representing them are the most important elements in the characterization of a language: phonological level morphological level lexical level syntactic level discourse level
phoneme morpheme lexeme sentence text
The fact that we have established these fundamental levels does not, however, mean that we cannot also distinguish other, complementary levels, nor does it mean that we cannot identify other units. Morphophonemes and stereotypical phrases or sayings are examples of units that participate in more than one level of description.
the foundations of terminology
57
And phrases, clauses, and discourse fragments are units we have in addition to the five basic ones mentioned above. The heterogeneous nature of a linguistic system is not only limited to the various levels of description, but is also evident from the series of language modalities known as dialects and functional varieties, as we stated in Chapter 2. A linguistic system is thus not only vertically heterogeneous and varied in terms of levels of grammatical units in relation to one another, but also presents a series of complementary varieties within this vertical order that are determined by the nature of communicative situations.1 1.1–Types of discourse A specific communicative situation requires appropriate forms of discourse. These forms are determined by the specificity of the elements of the communicative process (see Figure 1). Individuals A and B, the participants in a communicative situation, act alternatively as encoders and decoders of messages, constructing a text when they are encoders and deconstructing it when they are decoders. They use feedback to act on their own message. Both individuals have idiosyncratic characteristics (linguistic as well as nonlinguistic) that condition not only their interpretation of reality, but also the possibilities and probabilities of constructing and interpreting a message. Individuals construct and interpret messages in a particular way depending on where they are and what they are thinking at the time. A text is not just a linguistic unit but a mode of social and cultural expression allowing individuals to relate to one another. Therefore, many aspects besides those that are purely linguistic must be analyzed in order to characterize a text correctly. First, texts are complex linguistic units and conform to the rules of combination of each language system. Secondly, texts are complex pragmatic units, because they are produced by people who are neither psychologically transparent nor ideologically neutral. Thirdly, texts are complex sociolinguistic units because a language is a system for social communication that occupies a certain place in the society that uses it and has a relationship to other languages and their societies with which it is in contact. Finally, texts are complex cultural and anthropological units which reflect and communicate a system of cultural and ideological values by means of discourse. This network of elements and complex relationships comes into play in all communication processes. In addition, beside the characteristics of the society and language they use, individuals inevitably have their own social and cultural and linguistic idiosyncrasies.
terminology
58
ENVIRONMENT (the real world) interpretation intentionality CODE selection combination
MESSAGE
MEDIUM exchange
individual A encoder/decoder speaker/addressee
individual B encoder/decoder speaker/addressee
Figure 1. Elements in the communication process according to Jakobson We can therefore consider texts as a point of interaction between language and the real world because texts refer to a many-faceted, multidimensional, and changing reality. Texts are based on this reality and reflect a possible view of it. At the same time, they are units that reveal the complicated network of multilateral relationships among the factors participating in any act of communication. 1.2–General language and special languages A language consists of subcodes that speakers use according to their expressive needs and the nature of the communicative situation. Despite all this diversity,
the foundations of terminology
59
however, all languages have a set of units and rules that all speakers know. The set of rules, units and restrictions that form part of the knowledge of most speakers of a language constitutes the common or general language. The units of the general language are used in situations we call ‘unmarked’. In contrast, we speak of special or specialized languages to refer to a set of subcodes (that partially overlap with the subcodes of the general language), each of which can be ‘specifically’ characterized by certain particulars such as subject field, type of interlocutors, situation, speakers’ intentions, the context in which a communicative exchange occurs, the type of exchange, etc. Situations in which special languages are used can be considered as ‘marked’. The general language—the langue tout entière, in Kocourek’s words—containing both marked and unmarked varieties can be imagined as a set of intertwined, interrelated sets. What all the sets share is the general language. Each one of the subsets can be a special language. But, the description of a language cannot be limited to establishing a system of rules and units; it must also consider how speakers use the language. Therefore, special languages must take account of the elements that play a role in an act of communication: the participants, the communicative circumstances, and the purposes or intentions associated with the communication. 1.2.1–The concept and scope of special languages In order to define special languages, we first need to analyze the idea of specialization in reference to language, and will then clarify what we understand by special languages. Finally, we will define its characteristics and contrast them with those of the general language. AccordingtoKocourek(1982),speciallanguages,conceivedofassystemsofcommunication, can be approached in two ways: via semiotics, and via natural languages. The first approach is based on the fact that a special language is a system for transmitting and exchanging information that employs various codes at the same time, of which human language is undoubtedly the most important, but not the only one. Other systems that are three-dimensional (models, reliefs, etc.), two-dimensional (maps, drawings), iconic (ideographic signs, photographs, diagrams, or illustrations), or symbolic (nomenclatures, tables of symbols, numbers, etc.) share with human language the function of means of communication in technical and scientific contexts. The other approach, which relates special languages to natural languages, raises questions about whether special languages function autonomously from general language, about the difference between special languages and the common language, and about the relationship between special languages and the so-called artificial languages, especially symbolic languages.
60
terminology
1.2.1.1–Artificial languages and special languages To delimit special languages we must first consider whether we are dealing with systems that form a part of natural language or of artificial systems. The answer to this question will guide our discussion in the following sections. The following are the salient features of artificial languages: • • • • • • • • •
They are ‘‘invented’’ languages. They are constructed using natural language as the point of reference. They have a previous, controlled conceptualization, They are unable to admit new units that have not been previously set and conceptualized. They are unambiguous, and thus do not have synonyms nor polysemous terms. Their syntax is reduced. They have a reduced inventory of signs that is first established in written form. They are valid wherever they are used. They have no scope for meaning variation.
When measured against these characteristics special languages cannot be considered to be artificial: • They cannot strictly be considered ‘‘invented’’ but rather to a large extent they participate in the general language. • They have not been constructed with language as a point of reference, as they form a part of language. • They can admit new units, even if, in fact, it is desirable to control their growth. • They permit synonyms. • They have all the syntactic possibilities offered by the general language at their disposal, even though they use only a part. • They have an inventory of signs that grows steadily. Alongside these characteristics, which distance special languages from artificial languages, there are other shared ones, but they seem insufficient and not representative enough to allow us to consider special languages as part of the class of artificial languages. Special languages, similar to artificial languages: • have a prior conceptualization that is relatively controlled; • do not usually admit (at least in theory) new terms unless the concept has been defined and named;
the foundations of terminology
61
• do not have, in theory, polysemous terms; polysemy in general language is homonymy in special languages; • the written form of terms takes priority over the spoken form; • many terms are valid internationally; • have the same limited functions. This comparison shows that special languages exhibit a mix of characteristics of artificial and natural languages. 1.2.1.2–Definitions of special language In our view, a special language is a sublanguage of what is known as natural language; a sublanguage enriched with brachygraphical items, that is, acronyms and ideograms integrated into the language according to its grammatical constraints.—Kocourek (1982)
There is still no single and clear definition of the concept ‘‘special language’’.2 De Beaugrande (1987) provides a synthesis of some of the most representative positions: 1. Special languages are linguistic codes that differ from the general language and consist of specific rules and units. Definitions like the following are representative of this approach: A complete set of linguistic phenomena occurring within a definite sphere of communication and limited by specific subjects, intentions and conditions.—Hoffmann (1979)
Hoffman’s position does not delimit the concept with respect to the general language. On the other hand, he mentions extralinguistic and communicative factors as determining the specificity of special languages. 2. Special languages are variants of the general language. In this view, the languages of physics or computer science are just as different from each other as they are from a sociolect or a dialect. Theoretical or descriptive linguistics generally support this position. For example, Rondeau, Rey and Quemada define special languages as lexical variants of the general language. It must be noted that the terms ‘special language’ (specialised language) and ‘common language’ only refer to a subset of a language as a whole, that which consists of lexemes.—Rondeau (1983) Strictly speaking, the language of law does not exist by itself but rather only as a part of the French language, and it consists of the vocabulary of law and, undoubtedly, of some particular syntactic constructions.—Rey (1976)
62
terminology Descriptive linguistics rejects the terms technical and scientific ‘‘language,’’ both of which are equally inappropriate. Rather, we should talk about vocabularies, understood as the collection of particular uses of the French language and its varieties, pronunciation, morphology and syntax, which are all part of the common language. It is here where our analysis begins to establish functional, albeit non-hierarchical, relations.—Quemada (1978)
3. Special languages are pragmatic subsets of language as a whole. This is the position taken by Sager: Special languages are semi-autonomous, complex semiotic systems based on and derived from general language; their use presupposes special education and is restricted to communication among specialists in the same or closely related fields.—Sager et al. (1980)
Picht and Draskau also support this view: LSP is a formalized and codified variety of language, used for special purposes and in a legitimate context—that is to say, with the function of communicating information of a specialized nature at any level—at the highest level of complexity, between initiated experts, and, at lower levels of complexity, with the aim of informing or initiating other interested parties in the most economic, precise and unambiguous terms possible.—Picht & Draskau (1985)
This position is based on the view that it is difficult to define special languages entirely in linguistic terms, as most specialists in the field have noted: Special languages are readily recognized as pragmatic or extra-linguistic subdivisions of a language. Certain difficulties arise when we attempt to explain special languages satisfactorily in linguistic terms.—Sager, Dungworth & McDonald (1980) One can of course select a domain and run through its linguistic resources in terms of some familiar scheme like phonology, morphology, lexicon, and the like. But surely the major aspect of LSP is its communicative potential, and the latter can be described only in a roundabout and incomplete fashion within such schemes.—de Beaugrande (1987)
Despite all these differences, special languages have many features in common with the general language, as Kocourek (1982) aptly states, so that a consensus definition of special language must include the following characteristics: a. The distinctive elements of special languages are not isolated phenomena, but rather interrelated sets of characteristics. b. The purpose of communication is more important than other, complementary functions. c. The special nature consists of differences in subject field, user knowledge, area of usage.
the foundations of terminology
63
1.2.1.3–Broadening the concept of special subject Although this is a simplification of the issue, for the moment we will distinguish two views of specialization: a. Specialization by subject field. b. Specialization by pragmatic circumstances (users, type and occasion of communication. Scientific fields such as experimental sciences, mathematics, social sciences, economics and law, technical fields like engineering, construction and communications, specialized activities like sports, commerce and finance, all generate texts that diverge to some extent from the texts considered typical of general language. A specialized topic can thus determine the specialization of a language or a text. Nonetheless, it is difficult to determine which texts are considered special and which are general by only considering the subject field, for at least two opposing reasons: a. Our daily existence is full of contexts that are specialized to one degree or another, even though this everyday quality makes it more likely for the specialization to go unnoticed. b. In daily life we often mention specialized topics. This phenomenon has been called ‘banalization’ by some (e.g. Galisson 1978).3 The second view defines special languages with reference to the circumstances of their use. Following this view, we can establish an unmarked model of language usage that would be a reference point to determine the degree and type of deviation a specific situation presents in relation to the model. Each deviation (in the statistical sense) would constitute a criterion of specialization. Texts representative of standard or unmarked situations would belong to the general language because they do not present special lexical or pragmatic characteristics. Taking general language texts as our point of reference, any type of text that varies from this norm can be considered a special language text. General, unmarked texts are spontaneous, oral, semi-formal, covering subjects that all speakers of the language know and are primarily intended to provide or exchange information. In contrast to the above, specialized texts are all types of discourse that diverge from the general characteristics in subject matter, peculiarities of the interlocutors and the communicative situation, the communicative function and the channel for transmitting the data. A text considered specialized by any of these criteria requires language rules that are somewhat different from those required for general communication. Subjects
64
terminology
that are the object of usual conversation and language exchange in the general situations of people’s everyday life, that do not require speakers to learn a specific vocabulary, are not special. Beside dialectal and sociolectal variations the discourse of most speakers is not special. The situations shared by most members of a social community are not special and the principal purpose of general communication, i.e. to inform, is not special either. Finally, the channel of general communication is also general. 1.2.1.4–Narrowing the concept of special subject The broad view of special languages outlined in the preceding section makes it difficult to delimit the concept and must therefore be rejected. If, however, the subject field is the only factor taken into consideration when defining special language, it is hard to decide which subjects are special and which are not, since it seems clear that all human activity requires some degree of specialization. Sager et al. express this cogently: The fact that humbler occupations like nursing, book-keeping and cooking and even hobbies also involve special areas of human interest and therefore also require and indeed have their own special language is much less often acknowledged. Since practically every human activity can be attributed to one subject or another, all language could be split into so many subject languages and the word ‘‘special’’ would be superfluous.—Sager, Dungworth & McDonald (1980)
The best way to define special languages is to combine the subject criterion with other, pragmatic criteria and conditions such as types of user. In the view of Sager et al., special languages are limited to communication between specialists: Special languages, or more precisely special subject languages are usually thought of as the means of expression of highly qualified subject specialists like engineers, physicians, lawyers, etc. and are often derogatively referred to as ‘‘jargon.’’—Sager, Dungworth & McDonald (1980)
Picht and Draskau, on the other hand, state that Sager’s restriction is irrelevant: It may not be true to say, with Sager et al., that ‘‘the use of LSP presupposes special education and is restricted to communication among specialists in the same or closely related fields;’’ communication between experts is, as we have seen, only one of the higher levels at which LSP may be used. LSP is also used for the purposes of initiation and instruction, training and development at a lower level of abstraction and specialization.—Picht & Draskau (1985)
These authors hold that a special language can be divided into different levels of specialization, the highest of which corresponds to communication between experts,
the foundations of terminology
65
and the lowest to general purpose information meant for the layman. Picht and Draskau believe that what defines a special subject text is the topic, and that a text should still be considered as belonging to a special language when it is aimed at a general audience, even though its degree of specialization, and thus of abstraction, is lower. 1.2.1.5–Criteria for defining special languages In this book we use the term ‘‘special languages’’ to refer to the subsets of language that are pragmatically characterized by three variables: subject field, type of user, and type of situation in which communication takes place. We therefore assume the following: 1. Special subject fields are those that are not a part of speakers’ general knowledge; they are the object of a specific learning process. 2. Speakers who have this type of knowledge are users of special languages, in other words, subject field experts, although here we make a distinction between originators and recipients of specialized communication. Originators who produce specialized communication must have knowledge of a specific subject field which they have acquired through training. In contrast, recipients can be other experts or the general public, which passively receives special communication while acquiring knowledge. 3. Communication in special languages is usually formal and occurs in situations governed by professional or scientific criteria. 4. Special languages are characterized by a number of language-based features (units and rules) and text-based features (text and document types). 5. A special language is not a structurally monolithic subset, but rather permits the following variations depending on usage and the communicative situation: a. The degree of abstraction, which depends on the subject field, the recipients of the information, and the sender’s communicative purpose. b. The communicative purpose, which determines variations in text type. c. Geographic, historic, and social dialects. d. Personal style. 6. Special languages share a number of pragmatic and language-based characteristics, thus allowing us to refer to them as a subset of the general language that presents a degree of unity. 7. Special languages are a subset of the language as a whole. They intersect with the general purpose language, with which it not only shares features but also
terminology
66
LANGUAGE General language SL1
SL3
SL2
Figure 2 maintains constant exchange of units and conventions. We can visualize this as Figure 2. 1.2.1.6–Types of special languages In order to deal with language in all its diversity, it has been useful to divide it into more homogeneous subsystems. One of the resulting subsystems may be the ‘‘external’’ yet fruitful division of language as a whole into the so-called functional sublanguages. Kocourek (1982)
The most widely held classification of special languages is based on subject fields and the style and degree of abstraction with which the contents are presented. By subject field we establish groups of subjects, subjects, sub-subjects, etc. according to the hierarchical classifications used in documentation. Let us look at two examples: Production and economics Corporate organization of production Factors in production Prices; Setting prices; costs Price types Regulation and control of prices Highest price Lowest price Margins or differences in price Chemistry Theoretical chemistry Experimental chemistry
the foundations of terminology
67
Analytic chemistry Organic chemistry Inorganic chemistry Nonmetals Noble gasses Metal alloys with 6 subgroups Water Heavy Water
Special subjects can also be seen as forming a continuum in which the various fields can be identified. As Varantola notes, each subject field constitutes a specific special language, and the set of special languages forms the special language or specialized discourse as a whole: On the continuum formed by different SL’s, we would discern the prototype areas of various fields such as science, technology, law, religion, etc., and also determine interface areas where the different fields fuse.—Varantola (1986)
Secondly, in the terminology of any one special language we can distinguish various levels of abstraction which allows us to identify several different discourse types which are determined by the degree of abstraction with which the topic is represented or by the style used in a particular communicative situation. The possible classification of special languages will be discussed later. Here we first want to deal with another question, namely whether it is legitimate to discuss the idea of special language in the singular. The answer to this question is affirmative if we conclude that, after studying several special subject languages, all are found to share enough characteristics for us to speak of a single type of discourse. Experts have generally considered technical and scientific discourse to be a unified type and typical of special language texts. Others have talked about special language in the singular by placing a number of special subject field languages within its boundaries. This hypothesis assumes that a special subject field is a set of expressive varieties that can be considered as variants of a common special language code, which is only subcategorized by the degree of abstraction and the communicative functions that are most important in each situation. Picht and Draskau (1985), who support this unified view, list a series of characteristics shared by all special languages, thus allowing them to refer to them as a single type: a. Special languages have a single purpose, in the sense that they are used in a specific social setting and for communication.
68
terminology
b. They have a limited number of users. c. They are acquired voluntarily. d. They are autonomous with respect to the general language, in the sense that variation among special languages does not bring about variation in the general language. De Beaugrande also supports this single-variety view: I should prefer to speak of variety of LSP, because while we speak of terminologies in the plural, LSP varieties from various areas of specialism have many shared characteristics. To speak of ‘‘a variety’’ of LSP allows both this character of genus or superordinate term, and at the same time, of an extensive term covering many categories and lending itself to classification on another level e.g. the LSP of chemistry. Any taxonomy would soon show that no clear borders exist between different LSP in any sense comparable to those between, say, English and Danish.—de Beaugrande (1987)
In this view there are three types of factors that tie special languages into a common class: pragmatic, functional, and linguistic factors; subject fields creating only subordinate varieties of ‘special language’. From the standpoint of pragmatics, there seem to be enough reasons to claim a common denominator for all special languages, because all exhibit the same characteristics in terms of: a. Users, both quantitatively (a limited number) and qualitatively (defined by their profession or expertise). b. The communicative situations in which these languages occur (formal situations of a professional nature). c. The main functions (which are primarily informative). Despite these shared features, detailed analysis of these features in each of the various special languages reveals that not all of them conform to the overall scheme to the same degree. Rather, they offer a very wide range of possibilities going from highly marked special languages to ‘‘languages’’ that are perhaps better described as special aspects of the general or common language. Figure 3, taken from Rondeau (1983), illustrates this complex classification which, like all classifications, has fuzzy edges between the various types. In addition, units from one type of LSP are constantly crossing over to some other type. Rondeau also defends the hypothesis of a single special language in the fields of science and technology, which he considers the most representative of all special languages. He concludes that we can only talk about special language as a subset of the general language that can be differentiated from the general language if we
the foundations of terminology
69
L
Area of highly specialized SLs G
Middle ground for SLs Transition area between general language and SLs
SL3
SL2
SL1
(L) Overall lexicon of a language (G) General language (SL) Set of SL areas
Figure 3. Position of special languages, according to Rondeau (1983)
consider specificity on the levels of the text, the lexicon and the pragmatic and functional characteristics of the texts. On the textual level, special language texts are characterized by distinctive features that clearly differentiate them from journalistic, religious, literary, advertising, etc., texts. We also recognize a distinctive technical and scientific style, with some variations due to subject fields and the various levels on which subjects can be discussed. Subjects can be contrasted with each other, for example pure science vs. engineering, theoretical science and technology vs. their applications; experimental sciences vs. mathematics; vs. economics; vs. the humanities; etc. Kocourek recognizes five main classes or types of special subject fields according to their degree of abstraction. a. pure sciences b. experimental sciences c. applied sciences & engineering
70
terminology
d. subjects viewed from the production standpoint e. subjects viewed from the consumer standpoint4 With regard to the lexical level identified by Rondeau, special languages are specific in the terminologies they use. These terminologies can reflect various degrees of specialization, depending on the type of subject and the level of abstraction being dealt with. From the standpoint of pragmatics, the argument for speaking of a single special language is supported by the ease with which expert interlocutors understand texts written in a different language, even when they are incapable of maintaining an oral conversation about a trivial topic or understand a question about everyday life. The reason behind this situation is that special languages have relatively controlled units and rules that are international and voluntarily set by users themselves. These factors, beside the topic restriction, facilitate mutual comprehension. The more specialized a language, the more restricted its number of users and the more international its units and rules will be. This situation, which does not occur in general language, makes special languages different and is a factor unifying them. De Beaugrande (1987) expressed this succinctly: LSPs tend to share much of their resources not merely with LGPs, but also often have common cognate resources. LSP thus tends to be more ‘‘international,’’ or indeed ‘‘universal,’’ than does LGP.—de Beaugrande (1987)
Sager et al. made a similar observation: While in no way wishing to diminish the importance of language as an instrument of thought, it is in communication that differentiation between types of languages is required and made. So we have to explain the fact that a French physicist can read an English research paper on physics or even understand a lecture on physics delivered in English while at the same time being incapable of reading an English newspaper or asking his way around London. The communication process itself may differ between one special language and another or between special and general language since different groups of specialists make different demands of communication and the circumstances surrounding such uses of language also vary considerably.—Sager, Dungworth & McDonald (1980)
Finally, the various special languages have the common function of transferring information. To summarize, we can state that scientific and technical communication generated from special languages is characterized by three particular features: • Texts are concise (they tend not to be redundant), precise (they tend to avoid ambiguities), and impersonal (they are not emotive). • In the vocabulary nouns and nominal groups (as opposed to verbs and adjectives) predominate, both qualitatively and quantitatively.
the foundations of terminology
71
• The discourse of special languages prefers written language over oral language, and also uses symbols from other semiotic systems. 1.2.2–General language and special languages The difference between general language (in the sense of language common to all users) and special language is difficult to establish. Definitions of LSP or SL versus GL abound; none is universally applicable, for obvious reasons. Basically we are dealing with two intuitively correct assumptions that are good as working concepts but which resist clear-cut definition and delimitation. Varantola (1986)
While some authors have declared special subject languages to be wholly different from general language, the two texts presented below demonstrate that there is considerable overlap between general language and special languages: A Texts (general language) Computerized axial tomography—popularly known as the CAT scan—has truly revolutionized medicine and what we know about many normal structures in the human body, particularly the brain and spine. By combining X-rays taken from many directions, it has become possible to make images representing cross-sectional slices of the living brain similar to those observed in actual cut sections from the autopsy table. The CAT scanner, which was developed in the early seventies in England, has eliminated the need for many dangerous invasive procedures. Today’s scanners use threedimensional imaging techniques, which are especially useful in providing information on the alignment of bones. Metamorphic rocks are those rocks that have altered their form as a result of changes in their physical environment. In general, these alterations, or recrystallization, are brought about either because hot magma comes into contact with cooler rocks, or by large-scale movements of the Earth’s plates that change the pressure and temperature of the rocks. Minerals within the original rock respond to these changing conditions by reacting with one another to produce a new combination of minerals that is stable under the new conditions. Since metamorphism represents a response to changing physical conditions, those regions of the Earth’s surface that are the most active in terms of earthquakes, tremors, and volcanoes, like the Pacific rim, will be the places where these processes are easiest to observe. And because the composition of minerals and the pressure and temperature they are subjected to can vary a great deal, there are many types of metamorphic rock.
72
terminology
B Texts (special language) MRI studies of the living brain are equivalent to actual anatomic slices and thus can provide new anatomic and pathological information. An added advantage derived from the use of more powerful magnets is the ability to perform spectroscopy within the living body. MR spectroscopy requires a homogeneous magnetic field around the region of interest and special coils and radio-frequency currents. Although hydrogen nuclei have been exploited most in MR applications to medicine, potentially useful nuclear signals can also be elicited from isotopes of biologically important elements such as phosphorus. The importance of phosphorus spectroscopy is that it reflects changes in energy and breakdown of tissues. For example, identification of the phosphorus spectrum within the living brain may provide information on the depletion of adenosine triphosphate and the accompanying accumulation of lactic acid in stroke victims. Phosphorus spectroscopy might also prove able to distinguish benign from malignant tumors without resort to invasive methods. However, it must be remembered that the equipment associated with MRI is very expensive, and the additional cost of protecting the environment from the MRI scanner is a major obstacle. Because protolith compositions and the pressure-temperature conditions under which they may be placed vary widely, the diversity of metamorphic rock types is large. Despite the wide range of processes involved in the recrystallization of sedimentary and igneous protoliths into metamorphic rocks, there are only three important variables that effect metamorphic change: temperature, pressure, and the original chemical composition of the protolith. Temperatures at which metamorphism occurs range from the conditions of diagenesis (approximately 150°–200° C) up to the onset of melting. Rocks of different compositions begin to melt at different temperatures, with initial melting occurring at roughly 650°–750° C in rocks of granitic or shaley composition and approximately 900°–1200° C in rocks of basaltic composition. Above these temperatures, metamorphic processes gradually give way to igneous processes.
The two types of text share linguistic characteristics that are important enough to show that they are both based on a common system. They also present significant differences that may lead to the conclusion that in some aspects they come from different codes. The code unifying them is the general language; what separates them are the elements of the special language. We can therefore say that the two codes intersect. To differentiate the elements of general language from those of special languages in these texts, we discuss three types of features: linguistic (lexical, morphological, and syntactic) pragmatic (situation, originators and recipients of the message) and functional (the communicative intention).
the foundations of terminology
73
1.2.2.1–Linguistic features The features that can be analyzed and which reveal differences are the choice and usage of linguistic units at each level of the text grammar and the overall meaning of the text, the choice of words, the differences in frequency of certain units and structures, the presence and absence of some units and the use of alternative codes, etc. Common features of both types of text features are: a. The same graphic system of expression, the same alphabet. b. The same phonological system; if we read the texts aloud, the phonetic units are the same in both cases. c. The same morphological system. d. The same syntactic rules for the construction of phrases and sentences. e. The same sentence types. The greatest divergences are found in the vocabulary. The words in the general language texts are much easier to understand for most speakers of the language than those in the special texts. This observation allows us to identify three groups of lexemes in the A and B texts: a. General language lexical items, e.g. brain, medicine, slice, pressure, rock, temperature b. Specific lexical items that can be attributed to a borderline area between general language and special language: imaging, invasive, scanner, chemical composition, metamorphic, recrystallization c. Lexical items specific to special texts: adenosine triphosphate, lactic acid, spectroscopy, basaltic, diagenesis, protolith Certain structures and categories appear more frequently in special texts than in general language texts: a. Morphological structures based on Greek or Latin formatives: diagenesis, igneous, pathological b. Abbreviations and symbols: MRI, C c. Nominalizations based on verbs: accumulation, identification, recrystallization d. Straightforward sentence structure with little complex subordination. Certain units and structures characteristic of the general language are not found in special texts: certain affixes (those which are deemed colloquial), some verb forms (the second person forms, imperatives, etc.), some pronouns (second person pronouns), and some sentence types (exclamations, etc.).
terminology
74 1.2.2.2–Pragmatic features
The content of special language texts concerns scientific, technical, or professional topics, even though the concept of profession permits the inclusion of many subject fields, e.g. sports and business; if special is restricted to science and technology, these topics would not be considered specialized. 1.2.2.3–Functional features People generally use language not just for communicating with one another, but also for many other purposes. Following Jakobson (1963), we assume that language has six main functions, each of which is primarily linked to one of the six factors that take part in a communicative act: • • • • • •
the referential function the expressive, or emotional, function the conative function the phatic function the metalinguistic function the poetic function
code (metalinguistic)
message or text (poetic) encoder
medium
decoder
(expressive)
(phatic)
(conative)
reality (referential)
Figure 4. Jakobson’s functions
the foundations of terminology
75
Some of Jakobson’s functions are rare in specialized communication. While the basic purpose of technical and scientific languages is to inform and exchange objective information on a specialized topic, this is much less the case for areas such as administration or services. In the first three types of subject identified by Kocourek, in Section 1.2.1.6, most communication is referential. Text types such as description, definition, classification, enumeration, calculations, reasoning, argumentation, citation, reference, etc. utilize the syntactic and textual resources that provide information as objectively and impersonally as possible. On the other hand, they do not use general language text types such as narration, quoted dialogue, non-rhetorical interrogation, etc. The characteristics of scientific and technical texts (doctoral dissertations, technical reports, formal lectures, specialized articles in learned journals, etc.) reflect this tendency towards impersonalization and objectivity by using elements like: • • • • • • • •
first person plural as a means of expressing modesty the present tense absence of exclamations short sentences avoidance of unnecessary redundancy frequent use of impersonal formulae noun phrases other systems of representation, e.g. drawings, tables, in the body of the text
Special language texts can also present some other features which distinguish them from general language: • They often represent an implicit dialogue between the writer and the recipient of the message. • They do not implicitly present personal positions; when they do occur they are indicated by such phrases as e.g. according to the author, in our opinion, we believe that, etc.). • They attempt to persuade the reader indirectly, although it might not be done explicitly, by providing arguments, citing data, providing examples, explaining, etc. • They introduce metalinguistic elements such as explanations, definitions, parenthetical material, synonyms, etc. The number of these functional resources used in each text depends on the degree of specialization of the communication, and on the prior knowledge of the readers of the specialized communication. The less expert the reader, the more redundant the text will be and the more metalinguistic elements it will contain.
76
terminology
• Specialized messages do not ignore the elegance of language, the appropriateness of the forms used, or the advantages of the right format and layout. 1.2.3–Variation in special languages Up to this point we have shown the unity of special languages, differentiating them from the general language. We conclude that, from a theoretical and methodological standpoint, it is important to establish the concept of special language in the singular. We then showed how this abstract concept could be subdivided into types of special languages by subject matter. We also showed, in a second mode of diversification, that different levels of abstraction in the vocabulary chosen can lead to quite different texts. We shall now examine whether, in addition to subject field and degree of abstraction, there are other criteria that allow us to postulate several varieties of special languages and if so, where they occur. We start from the premise that any special language, by virtue of its being a subset of the general language uses a single code which permits variations. Depending on the pragmatic function and the context of situation, including the epistemological factor, the same topic within a special field lends itself to discussion at different levels of abstraction. The degree of specialization reflects the pre-knowledge the receptor is assumed to possess. —The continuum of text types within a LSP is perhaps best seen as a circle with a central common core. The core would then expand in all directions with various realizations of text types at differing levels of specialization and abstraction.—Varantola (1986)
The variations occurring in special languages are of the same nature as in general language, i.e. dialect, register and style. Because their main purpose is communication between experts, the range of dialectal variation is narrower than in general language. Two scientific texts on the same topic produced at two different points in time can present interesting differences with respect to how the discipline is seen, which undoubtedly influences the terminology used and how the text is expressed. To every action there is always opposed an equal reaction: or, the mutual actions of two bodies upon each other are always equal, and directed to contrary parts. Whatever draws or presses another is as much drawn or pressed by that other. If you press a stone with your finger, the finger is also pressed by the stone. If a horse draws a stone tied to a rope, the horse (if I may so say) will be equally drawn back towards the stone; for the distended rope, by the same endeavour to relax or unbend itself, will draw the horse as much towards the stone as it does the stone towards the horse, and will obstruct the progress of the one as much as it advances
the foundations of terminology
77
that of the other. If a body impinges upon another, and by its force changes the motion of the other, that body also (because of the equality of the mutual pressure) will undergo an equal change, in its own motion, towards the contrary part.—(Sir Isaac Newton, Axioms or Laws of Motion, in Encyclopedia Britannica, 1989 Yearbook of Science and the Future, Chicago 1990.) Newton’s third law states that the actions of two bodies upon each other are always equal and directly opposite; i.e., reaction is always equal and opposite to action. The proposition seems obvious for two bodies in direct contact; the downward force of a book on a table is equal to the upward force of the table on the book. It is also true for gravitational forces; a flying airplane pulls up on the Earth with the same force that the Earth pulls down on the airplane.—Vol. 8. Micropaedia, of the 15th edition of the Encyclopedia Britannica, Chicago 1988 Those who favour relying on a haphazard self-expression in the use of English will find little support from practised writers. The fact that we ourselves know perfectly well what we mean is simply not enough. Indeed, it is not enough even if our addressee is only ourselves, as for instance in making notes for future reference. To the ‘I’ who writes them now, they are clear enough; but what about the ‘I’ who reads them in a month or a year hence? If self-communion across even a short interval of time can be incomprehensible, we see the need to be all the more careful when what we say or write is addressed to other people.—Randolph Quirk & Gabriele Stein, English in Use, London: Longman, 1990. The book appears, like its predecessor, after more than ten years of active fulltime preparation. It is hardly necessary to observe that no one editor could harmonize all the diverse and disparate matter by reading and criticizing every line or even determine and keep firm control over editorial policy, nor could an editorial board of fixed membership. Instead the editor in chief has used his editors one by one and has delegated multiple responsibilities to them individually as occasion required.—Philip Gove, Preface to Webster’s 3rd New International Dictionary, Springfield, Mass.: Merriam Webster 1961 The main purpose of special languages, i.e. allowing objective, precise, and unambiguous exchange of information particularly between subject field experts and professionals, makes dialectal variation very minor. The issue is not one of affirming one’s own geographical origin, but rather one of communicating unambiguously.5 In this sense, presenting a highly specialized scientific text in a written article or a conference paper does not require the same discourse as a spontaneous oral communication on the same subject. Finally, the intentions or purposes of the communication, both in general and
78
terminology
special language, also condition the syntactic, morphological, and textual devices used, e.g.: Text 1: descriptive Work in terminology at the School of Economics was first undertaken in the late sixties. It was not until autumn 1974, however, that common interests and research objectives within the various different language departments achieved any organized consolidation. Since then, considerable developments, both qualitative and quantitative, have led naturally to a division of labour and to specialization within the Centre itself, while at the same time the Centre has made a significant contribution at national level towards bringing together qualified persons with an interest in the field of terminology.—(Picht & Draskau, 1985)
Text 2: argumentative Usage doctrines must change with the times, of course. The fundamental linguistic virtues—order, clarity, and conciseness—are unassailable, yet they must be constantly reinterpreted against an evolving social background. It is no longer permissible to pretend that the English language is a club for gentlemen who have ‘‘their’’ Latin authors at their fingertips. The language is too important to be left to nostalgic reveries. Nevertheless, it is also too important to be excused from responsible critical review.—(Geoffrey Nunberg ‘Usage in the American Heritage Dictionary’, The American Heritage Dictionary, 3rd edition, 1992.)
Text 3: directive You may deduct only the part of your medical and dental expenses that exceeds 7.5% of the amount on Form 1040, line 32. To the extent you were not reimbursed, you may deduct what you paid for: • Prescription medicines and drugs, or insulin. • Medical doctors, osteopathic doctors, dentists, eye doctors, chiropractors, podiatrists, psychiatrists, psychologists, physical therapists, acupuncturists, and psychoanalysts (medical care only). ‘Forms and Instructions’ Internal Revenue Service, US Department of the Treasury, 1995.
1.3–Special language documents The types of specialized texts will be discussed in section 3 of this chapter. For methodological reasons we shall here differentiate between two basic types of specialized texts: specialized texts, i.e. any communication occurring in special
the foundations of terminology
79
language; and specialized document, i.e. specialized texts requiring a very elaborated formal codification from the standpoints of design and linguistic expression. Specialized documents are characterized by factors which affect the layout of the document and the way information is distributed, like the following: • they generally constitute closed lists; • they are usually representative of several speech acts (in the sense of Austin 1962); • they are normally not spontaneous; • they only occur in written form. Such formal written documents require formal rules of composition in addition to the use of fixed expressions according to the purpose of the document. Each type of document, besides having a specific format, can also have a specific terminology, syntax, and phraseology that distances it from what is usual in general language.
Example 1. Call for a meeting in a university DATE: January 6, 1997 FROM: John Smith, Chair TO: All Department members RE: Departmental meeting The first meeting of the new semester will take place on Friday, January 13 at 3 p.m. in Lindley Hall 224. Items on the agenda are the five-year plan and hiring plans for the coming year.
Example 2. Receipt May 14, 1997 I hereby acknowledge receipt of the amount of $300 (three hundred dollars) for medical services performed in this office on the date printed above.
(Signature) Robert Lawrence, M.D.
terminology
80 Example 3. Dictionary entry
elephant (el’ə-fənt) n. 1. Either of two very large herbivorous mammals, Elephas maximus of south-central Asia or Loxodonta africana of Africa, having thick, almost hairless skin, a long, flexible, prehensile trunk, upper incisors forming long, curved tusks of ivory, and, in the African species, large, fan-shaped ears. 2. Any of various extinct or living animals related to either of these two animals. (American Heritage Dictionary, 3rd edition) The rules for writing specialized documents are not learned in the same, spontaneous manner as general language is learned. Creating certain documents requires knowledge of the rules governing the phraseology and the textual, syntactic, and lexical form. 1.4–The role of terminology in special language texts and documents It is difficult to establish discreet divisions between the terminologies of allied fields, just as it is difficult to draw a line between terminologies and the common language lexicon. What at one point is specialized may become part of the common language a short time later,6 and there is a constant, uninterrupted flow of terms between disciplines. This is why the distinction between terms belonging to a special subject field and terms used in a special subject field is generally accepted. A term belongs to only one special subject field; it can belong to several subfields of this main field, and at the same time it can be used in many other special subject fields.7 Reality is a continuum that is only artificially divided into subject fields, and all scientific phenomena can be analyzed from many different standpoints and different scientific perspectives.8
2.–Terminological unit As the product of terminology as a science, every terminology is defined as a group of terms either sharing the same traits of extension or delimitation, or belonging to the same domain. This definition entails two complementary definitions: a definition for term and a definition for extension or delimitation common to all the terms constituting a terminology. Gouadec (1990)
Terms, like words in the general language lexicon, are distinctive and meaningful signs which occur in special language discourse. Like words, they have a systematic side (formal, semantic, and functional) since they are units of an established code;
the foundations of terminology
81
they also have a pragmatic side, because they are units used in specialized communication to refer to the objects of the real world. Terms do not seem to be very different from words when we consider them from the formal or semantic point of view; they differ from words when we consider them as pragmatic and communicative units. The most salient distinguishing feature of terminology in comparison with the general language lexicon lies in the fact that it is used to designate concepts pertaining to special disciplines and activities (see Figure 5). concept
designation
reality
discourse
Figure 5
2.1–Terms as systematic units A terminological unit, or a term, is a conventional symbol that represents a concept defined within a particular field of knowledge. According to this idea, drawn directly from Wüster's teachings, each word must be given importance . . . because it is a term since: 1. 2. 3. 4. 5.
its expression is a symbol it results from a convention its content is its most important trait the concept it represents is explicitly defined the description of a concept leads to the field of knowledge in which it occurs. Lérat (1989)
As Lérat states, terms are units with form and meaning that belong to the linguistic system of a particular language, within which there exist several alternative specific subsystems. These subsystems are not complete subcodes from the standpoint of levels of specificity, but represent partial aspects of variations with respect to the common code, as in Figure 6.
terminology
82
GRAMMAR LEXICON SL2 SL1 SL3
Figure 6. Relationship between grammar, the lexicon, and special language lexicons Like all other meaningful units of a linguistic system, then, terms are part of a structured system. They occupy at a specific level (that of lexical units) within that system, and are related to the other units on the same level, as well as to the units on the other levels with which they build the discourse. The relationships terms establish both with other lexical items and with the other components of the grammar do not show any overall specific trait that would make them systematically different from the relationships words establish. Nevertheless, it is important to note that within a language’s lexical component, terms are much more closely related to the other terms in the same discipline, with which they form specific structures, than to other terms from different subject fields. Terminology makes use of the same rules for building sentences and discourse as the other lexical units. Consequently terms are signs that can be analyzed from three points of view: the formal (the designation), the semantic (the concept), and the functional (grammatical category and distribution). Formally terms are phonological units that can be articulated phonetically (and represented graphically) with an internal structure made up of morphemes. Semantically terms are units of reference to objects of the real world and, as such, have a meaning that can be described as a set of distinctive features. Functionally terms are distributional units requiring a certain linguistic environment, and in discourse they are often found combined with other terms (phraseology). 2.1.1–The designation Dictionaries of Linguistics and reference books often ignored one of the three parts of morphology, as defined in the Trésor de la Langue Française: ‘‘The study of the different categories of words and of the forms they may take in a language (inflection and deriva-
the foundations of terminology
83
tion).’’ The first part is important because it relates an aspect of discourse to the linguistic analysis of words by means of correlations found in the grammar of the language itself. The second part will be left aside here since it is of a more general nature. Since the beginning of structuralism, much has been written on the third part: lexical formatives have been the cause of terminological growth, even though we still do not have a taxonomy that is any more coherent than that of Darmstetter.—Lérat (1990b)
The formal side of the terminological unit is called a designation or term. From the point of view of linguistic theory, terms are base-level phonological representations which have a phonetic form. All speakers of a language possess an underlying phonological form for each term, which they can phonetically interpret applying morphophonological and phonological rules. Morphologically, terms are structures of constituent morphemes which form the basis of their meaning. However, even though linguistics does not account for the graphemic representation of words—this is not a form of knowledge belonging to the natural competence of speakers—in terminology this aspect plays an important role for description. For terminology the written form of terms is of prime importance, because it is this written form (and not its pronunciation) that standardization acts on. 2.1.1.1–Pronunciation and written form In linguistics, any unit above the level of a distinctive feature has a phonological representation that takes shape as a phonetic form. In terminology, the form of a term is considered a sequence of sounds or letters representing a phonological base form, in accordance with the rules governing the phonetics of the language, or a written form, in accordance with the orthographic conventions of each language. thermometer phonetic form [thər-m˘om'˘ı-tər] written form thermometer keystroke phonetic form [k¯e'-str¯ok'] written form keystroke bronchitis phonetic form [br˘on-k¯ı'tis] written form bronchitis Phonetically, terms are not obviously different from words: both types of unit respond to the same rules of the linguistic system, although in terms we can observe particular traits that we do not often see in the general language lexicon. However,
84
terminology
these traits do not derive so much from the fact that they are terms, but rather from the fact that in terminology the presence of units formed with Latin- and Greek-based derivatives (affixes or combining forms) is greater than in the general vocabulary. The written form of terms does not differ from that of words. It is conditioned by the general rules of orthography. The morphological structures that predominate in terminology yield more words with some special written signs (such as hyphens or diacritics) or an increased presence of certain combinations of letters, which is the result of a larger number of neoclassical words in the special languages of certain subject fields.9 Little else is different. 2.1.1.2–Morphology Formally a term is a unit that can be broken up into separate elements and may be made up of smaller, identifiable and meaningful units known as morphemes. The morpheme is the smallest unit of language that meets both formal and semantic criteria. The descriptive unit of the level immediately below the morpheme, the phoneme, has no meaning. From the morphological standpoint, lexical units are simple if they contain only one morpheme, or complex if they have more than one morpheme. The meaning of a compound unit is in principle the combination of the meanings of its constituent morphemes but this is not always the case. If we compare the units of the general lexicon with those of terminology, we see that the basic morphological structure of terms coincides with that of the words of general language: (affix)1 – lexical root n (lexical root)1–n (affix)1–n The lexical root is the only morphological unit that is essential for a term.10 Affixes combined with the root or several roots combined with each other yield complex terms. All lexical roots have a stem. A stem differs from an affix in that it can function as a term without the presence of an affix: body duodenum commute stain On the other hand, affixes—both those situated before the stem (prefixes) and those occurring after the stem (suffixes)—require a stem in order to form a term: antiantibody -al duodenal -tion commutation counter- counterstain
the foundations of terminology
85
Stems and affixes usually have one form; they may occur with variants (allomorphs or written variants, depending on whether we are dealing with oral or written data). The occurrence of a variant depends on the morphophonological context: derm -oid flute mega -cephaly ferr -ic
dermat flaut megalo ferro
-oid -ist -cephaly -electric
2.1.1.3–Types of terms Terms are generally classified by the criteria of form, function, meaning, and origin. 1. From the point of view of form, terms are classified according to the following criteria, which are not mutually exclusive: a. The number of constituent morphemes. Terms, like words, can be either simple or complex: acid, molecule, niter, nimbus, enzyme, cell; acidification, intramolecular, nitrify, nimbostratus, enzymologist, cellular. b. The types of morphemes that play a role in forming a complex term. Terms can be derived words (formed by adjoining an affix to a lexical base) or compounds (combining lexical bases or a combining form and a lexical base, to which affixes can then be adjoined): ulcerous, illness, receiver, emotional, telephotography, transmission, antibody, Afro-Caribbean, presuppose, countercyclical, infrahuman, prenatal, afterburner Simple as well as complex (both derived and compound) terms can be used as the base for forming new terminological units: microfilm microfilming overfeed overfeeding appendix appendectomy megalomania megalomaniacal synthesis photosynthesis pharmacology pharmacological c. Complex terms can be made up of a combination of words that follow a syntactic structure. These structures—terminological phrases—are more frequent in terminology than in the general language lexicon: batch processing optical character recognition cardiopulmonary resuscitation dilation and curettage boiler hatch
terminology
86
These phrases are governed by the same rules that combine free phrases and are not formally distinguished by any specific feature. This is why it is both difficult to differentiate between terminological phrases and freely constructed phrases, and also hard to delimit exactly what segment corresponds to the terminological phrase, since phrasal terms can sometimes have the form of their own description: internal revenue service taxable income flat-glass water gauge high-vacuum electron beam fusion welding d. Finally, from the standpoint of form, there are terms that appear to be simple, but upon further examination turn out to be complex: initialisms, acronyms, abbreviations, and short forms, all of which are examples of truncation processes. 1. Initialisms are units made up of the combination of the initials of a longer expression. They often correspond to the name of an organization, document or process, and in many cases they become lexicalized in the common language: MIT Massachusetts Institute of Technology CIA Central Intelligence Agency WHO World Health Organization Initialisms can be phonetically realized in different ways, although usually one form will take precedence over the others: • it can be pronounced as a word: GUI [g¯oo¯ '¯e] RAM [rˇam] PET [p˘et]
(graphical user interface) (random-access memory) (positron emission tomography)
• it can be pronounced by reading the letters: UFO (unidentified flying object) AC (alternating current) UPI (United Press International) • or the above two possibilities may be combined: CD-ROM—[s¯e’d¯e’ r˘om]—(compact disc read-only memory) 2. Acronyms are words formed by combining segments from a developed phrase which are pronounced syllabically:
the foundations of terminology
87
• they can combine the initial segments of the first and second elements of the phrase: Amex American stock exchange hazmat hazardous material FORTRAN formula translator COMSAT communications satellite • or, they can combine the initial segment of the first element and the final segment of the second: bit binary digit 3. Abbreviations are forms that are usually established by consensus. They reproduce a part of a word and practically act as a symbol for the word: vol volume l liter 3rd third 4. Finally, short forms or clippings are often used to make discourse more economic. They are based on using the first part of a longer word, or the first word of a phrase: super superintendent hypo hypodermic syringe hyper hyperactive macro macroinstruction 2. From the standpoint of the function they have in discourse, terms can be classified into various functional groups: nouns, adjectives, verbs, and adverbs. In terminology, as in the general lexicon, the proportion of units belonging to one grammatical category or another is clearly different. This imbalance is perhaps even more true of special languages, because the number of nouns is highly disproportionate in relation to the number of adjectives or verbs. 3. From the point of view of meaning, terms can be classified by the class of concepts they designate. Concepts can be grouped into classes and subclasses in accordance with the features they share (which ultimately are the defining traits of a conceptual class) and the relationships they have with one another. Several types of relationships arise from the interaction among concepts, processes, and their resulting states, which are also concepts. With this in mind, we can initially identify four major conceptual classes: • objects or entities • processes, operations, actions
terminology
88 • properties, states, qualities • relationships
Each one of these classes of concepts is predominantly expressed by a particular functional class.11 • • • •
objects or entities processes, operations, actions properties, states, qualities relationships
nouns verbs, nominalizations of verbs adjectives adjectives, verbs, prepositions
Within a special subject, concepts are organized in structured sets that are called conceptual systems, and these systems reflect the view of the real world held by the experts in the discipline or activity. Each conceptual system is a structure that can cover several classes of concepts: objects (car, airplane, helicopter, ferry, etc.); properties of these concepts (visual, auditory, perceivable, material, shared, etc.); relationships (equivalent, subordinate, derived, etc.); operations (weld, solder, dissolve, etc.), and so on. In each specific special subject field every conceptual class is internally organized into several subclasses as a function of the types of concepts included in the class, their properties, relationships, and/or functions. 4. Finally, from the standpoint of their linguistic origin,12 terms can be created by means of applying the rules of the language itself (derivation and conversion), or they can be borrowed from another code or subcode. In this respect we are speaking of borrowings in the broadest sense. However, the idea of borrowing (or loan word) is not used by all specialists in this broad sense, so it is convenient to discuss its limits. Following the discussion in Chapter 2, 1 and Coseriu (1962) we recognize a distinction between two different ideas of the term language: on the one hand, the idea of a historical language and, on the other, the idea of a functional language. Coseriu argued that a historical language is made up of several functional languages. From a historical viewpoint we can identify different types of borrowing, depending on the code from which the borrowed units comes: • loan words from Greek or Latin, often called neoclassical borrowings • loan words from another, contemporary language, i.e. true borrowings • loan words from other geographic or social dialects, or other subject fields from the same language, which are usually not considered borrowings.
the foundations of terminology
89
The first two types of borrowing share a trait that unites them: they come from a foreign code. Despite this similarity, they also exhibit aspects that clearly differentiate them: society identifies them differently, and their dissemination is not the same. Borrowings from Greek and Latin are numerous in many languages, particularly in the Romance languages, but they are also commonplace in languages like English, and they are considered natural in these codes. As a result, they often go unnoticed as borrowings, which is exactly the opposite of what occurs with borrowings from present-day languages, even if the borrowing is from a cognate language.13 In terminology, neoclassical derivations and compounds are more frequent than in general language, especially in the natural sciences. International terminology standards explicitly recommend the use of neoclassical stems and affixes to form new terms in order to encourage the international nature of designations. International standards also recommend that neoclassical terms be formed from elements of a single historical language, although this recommendation is not always followed. We often find neoclassical words consisting of formatives from both Latin and Greek: bigamy, pluviometer, tripod, mammography, as well as words combining a neoclassical formative with an English formative: infrared, lumpectomy, microwave. In addition, in English blending of the first syllable(s) of one term and the final syllable of another is used to create terms that sound neoclassical: ballute (balloon + parachute) electrocute (electro + execute) tangelo (tangerine + pomelo) As opposed to neoclassical forms, borrowings or loan words are lexical units that come from a foreign language system which have been incorporated into a language either consciously or not. Societies which depend on importing technological and scientific knowledge need to designate the new concepts and therefore tend to use a large number of terms from other languages which, once a part of usage, are very difficult to displace. Scientific and technological transfer is the most frequent cause behind the high number of borrowed terms from a language in which the product or idea was created. English terms widely used in other languages: camping, cashflow, compact disc, mouse. Leaving aside the linguistic considerations of the arbitrariness of the sign, speakers generally consider that there is a natural tie between designations and the concepts they represent and that this should be maintained. As a result, the first form used for designating a concept is usually deemed to be the most natural one, and it
terminology
90
is hard to replace it. Only appropriate intervention can reverse this very human feeling towards words. Some words that have been incorporated into English with little or no change from their original language are:14 coup d’état, kudos, pianissimo, protégé. Once a borrowing has been fully accepted by the importing language, it can generate derived words that follow the morphological rules of this language: Blitzkrieg (from German) karate (from Japanese) machine (from French)
blitz (verb) karateist machinist, machine (verb)
In many cases borrowings are unnecessary because most languages have enough resources to create their own terms. For example English has borrowed many foreign expressions for which there are widely used native equivalents, e.g. hoi polloi (common people, terra incognita (uncharted territory), opus (work). Replacing borrowings with genuine forms of a language is indicative of the vitality of a language, of the attitudes its speakers have with respect to language survival, and the hierarchical relationship that exists among languages, which all express political, economic, and social relations. 2.1.1.4–Phrasal aspects of terminology The morphemes that make up terms, like those that make up words in general language, and the words forming terminological phrases do not combine with each other in a linear fashion, but rather form morphological and lexical constituents that follow the basic patterns of a language’s lexical system and respect the binary principle of combination (see Figure 7). Complex forms in general—derived words, compounds or phrases, set or free forms—follow a hierarchical structure (which in linguistic theory is known as X-bar theory) which postulates that all combinations of lexical elements adhere to the following structure: X′′ X′
→ →
X′′ X′
— the sewing machine sewing machine
X′′ X′
— batch processing batch processing
X′′ X′
the days
Specifier X X′ X (YP) (ZP)
days of the week of the week
the foundations of terminology gastrointestinal ulcer ulcer
essential amino acid
gastrointestinal gastro
91
amino acid
essential
intestinal intestin(e)
essenc(e)
-al
amino
acid
-al
multilingualism multi
lingualism lingual
-ism
Figure 7 When we consider the phrasal aspects of terms, we must first distinguish between phrasal terms, which are lexical structures with a terminological value (even though they are made up of words), and free structures, which are not lexemes but rather simply phrases that occur in discourse. These two types of structure do not formally differ from each other, but an analysis of their lexicological or terminological nature reveals that they do not behave in the same way in discourse. Between these two poles there are a series of constructions that are neither phrasal terms proper nor totally free syntactic formations. These are combinations that occur frequently in the discourse of a special subject field (which makes them similar to terminological phrases), but they do not correspond to established concepts. We refer to these collocations as phraseology.15 All special languages have this type of phrasal unit. For example: • computer science: press the key, edit a text, copy a block, insufficient memory, digitize a text, reboot the computer • administrative law: propose an amendment, provide documented proof, fill out a form, adjourn a session
92
terminology
• linguistics: parse a sentence, well-formedness condition, explanatory adequacy. 2.1.1.5–The systematic nature of designations The elements that make up terms follow the rules of the subsystem of the language to which the terminology of a special subject belongs. Terminological designations are systematic in two respects: a. with regard to the general lexical system; b. with regard to the terminological system of each special subject field. In general, to create a new designation terminology has available the same resources and mechanisms that the general lexicon has to form new words. Terms conform to the kinds of structure the system allows and are subject to the same rules of combination and restriction. Terms can be created by specialists, or simply borrowed, which involves incorporating a unit that has already been formed in another language directly into the speaker’s own language. 2.1.1.5.1–Term formation New terms that are the result of a formation process (compounding, conversion, truncation) are usually classified according to the strategy speakers have used to create them. These strategies can be formal, functional, or semantic in nature. 1. Formal methods Among the formal methods available for creating new terms we distinguish two methods: the combination of morphemes and words, including derivation, compounding, and creation of phrases; and the formal modification of existing units by means of truncation processes, including initialisms, acronymy, and clipping. a. Derivation consists of the addition of affixes to lexical bases to form new words. In English three possibilities are known: prefixing, suffixing, and mixed cases in which both a prefix and a suffix are added to the same base: a. prefixation: reopen, untie b. suffixation: processor, transmitter c. prefixation + suffixation: unconstitutional, disarmament b. Compounding consists of combining two or more lexical bases to create a new lexeme. Three types of combination are used: a. combination of native contemporary forms: greengrocer, roadsweeper b. combination of neoclassical forms: electrolyte, lysosome c. combination of contemporary and neoclassical forms: bio-science, megacity
the foundations of terminology
93
c. The use of phrasal units involves forming a new unit from the organized syntactic combination of words, e.g. random access memory, magnetic resonance imagery d. Truncation is a formal device consisting of reducing a unit to one of its parts. It includes the formation of initialisms, acronyms, and clippings. a. initialisms are abbreviated complex terms made up of the first letters of the term element: FBI (Federal Bureau of Investigation), EU (European Union), IOC (International Olympic Committee) b. acronomy: ALGOL (algorithmic language); bit (binary digit) c. clipping: jumbo (jumbo jetliner), narc (narcotics agent) 2. Functional methods Functional methods comprise two processes: a. Conversion or zero derivation, i.e. changing the category of an existing word without changing its form (except that in this grammatical conversion any characteristics of inflection belonging to the former category are eliminated and replaced by those corresponding to the new category). For example: recordnoun recordverb pigeon holeverb pigeon holenoun b. lexicalization, by converting one of the inflected forms of a lexeme into a new word with a different grammatical category. For example: VERB NOUN harden hardening weld welding 3. Semantic methods Semantic methods modify the meaning of a term.16 The form on which the new form is based can belong to several subcodes. Two prime criteria are used to classify the terms that result from this kind of formation process: provenance of the base form, and the type of semantic modification produced in the process. a. By provenance of the base form, the two most frequent sources are the general language lexicon and the terminology of other special subjects. b. Modification takes one of three forms: 1. extending the meaning of the base form • pin (straight piece of wire with a sharp point) • pin (pegs on a printer platen engaging holes at the edges of paper)
terminology
94
2. narrowing the meaning of the base form • printer (one or something that prints) • printer (the part of a computer system that produces printed matter) 3. changing the meaning of the base form • washer (one who washes, extended to appliances) • washer (flat disk used to relieve friction, prevent leakage, or distribute weight) Beside the formal, functional, and semantic methods, all languages also have other ways of creating new terms. Borrowing and loan translation or calquing incorporate units that belong to one language code into another. For example: a. borrowings, i.e. taking over a word or expression with its spelling: hors d’oeuvre, coup d’état b. loan translations, i.e. translating the components of a foreign word: English wintergreen, from Dutch wintergroen; Spanish rascacielos, from English sky scraper 2.1.1.5.2–Term formation rules The ability to create words and terms is a natural manifestation of a speaker’s linguistic competence. All speakers of a language are capable of proposing new designations that refer to new perception of the real world or of proposing alternative names for a previously named concept. For this purpose speakers use processes governed by linguistic rules that are part of the lexical component of a language’s grammar. When speakers form a new word or term, they often recombine known morphemes, a process which requires less effort than the creation of original units. Consequently, the mechanisms leading to the formation of new words or terms, in contrast to those that lead to borrowing, are governed by linguistic rules that are a part of a language’s grammar. We can identify the following types of rules for word or term formation:17 1. derivational rules using a prefix [prefix + [base]x ] (x)(y) 2. derivational rules using a suffix [[base]x + suffix] (x)(y) 3. compounding rules [[base]x # [base]y](x)(y) 4. conversion rules [base]x → [base]y Terminologists must know the word formation processes of the language. The des-
the foundations of terminology
95
ignations proposed for new concepts made by terminologists and experts must be supported by a good knowledge of all the devices a language offers. This applies both to the first designation proposed for an original concept and to proposals to change designations because the existing form is not deemed suitable. In addition, new designations must also be mindful of the formal tendencies of the subject field they belong to. For example, in chemistry all enzymes have the suffix -ases. 2.1.2–Concepts It is very important for terminographical practice to consider concepts as elements of conceptual fields and, more particularly, of taxonomies understood as groups of related concepts. It is only within the scientific and technical domains that the structuralist conception of semantic or lexical fields corresponds to the nature of linguistic objects.—Lérat (1989)
A term has a linguistic form and a content representing the concept. ISO standard 704 (1987) Principles and methods of terminology defines concepts as ‘‘mental constructs that are used to classify the individual objects in the external or internal world by means of a more or less arbitrary process of abstraction.’’ This definition of concept allows us to differentiate clearly between concepts and the objects in the real world the concepts represent. Concepts are the result of a selection process of the salient characteristics defining a class of objects and not the individual objects themselves. By means of the process of designation we use terms to refer to concrete and abstract reality, a reality that is external and internal as well as individual and collective. 2.1.2.1–The analysis of concepts A concept is a unit of content consisting of a set of characteristics. As ISO/R standard 1087 states, the characteristics of concepts are the properties that describe it. The expression of the set of characteristics of a concept is its paraphrase or definition: • a microsome is: – a small particle – found in the cytoplasm of a cell – made up of fragmented endoplasmic reticulum to which ribosomes are attached • formaldehyde is – an aldehyde – colourless – gaseous
96
terminology – used for manufacturing melamine, phenolic resins, fertilizers, dyes, and embalming fluids – used as a preservative and disinfectant in liquid form
Concepts differ from each other in that they possess different characteristics. A good conceptual description must therefore include this opposition of distinctive characteristics that distinguishes concepts: An ordinary light beam is: • one of the two beams into which light is divided • one that hits a solid material • refracted according to the normal laws of refraction An extraordinary light beam is: • one of the two beams into which light is divided • one that is incident to a birefringent solid material • one that manifests itself in a direction other than that of the optical axis of the material An incident light beam is: • one of the two beams into which light is divided • one that is incident to a birefringent solid material • one that runs in a direction other than that of the optical axis of the material An emergent light beam is: • one of the two beams into which light is divided • one that is incident to a birefringent solid material • one that crosses the surface The presence or absence of a characteristic instead of another within a set of interrelated characteristics differentiates between the concepts in a conceptual system. We can thus draw the conclusions from the above examples. The four concepts all present some of the same characteristics that make them all pertain to the same field: a. light beam b. one of two c. follows certain patterns of refraction However, they have some characteristics that set them in opposition to the others
the foundations of terminology
97
d. one that manifests itself in the same direction or in a different direction from the optical axis of the material 1. in the same direction = ordinary light beam 2. in a different direction = extraordinary light beam e. that has a trajectory 1. reaching the surface = incident light beam 2. crossing the surface = emergent light beam 2.1.2.2–Types of characteristics In accordance with the definitions of concept cited, the characteristics constituting concepts do not belong to a single group but belong to various classes established by several criteria: a. Depending on the relevance they have in shaping a concept, the characteristics may be essential or complementary. The essential characteristics of a concept describe its essence and are thus necessary to define it: polyimide
• set • of polymeric compounds obtained by reaction of dianhydrides and diamines
polyisoprene
• polymer • made from the union of isoprene molecules in reverse order
polymastigina • order of flagellates • subclass Zoomastigina • possessing four or more flagella and axial organelles The non-essential, or complementary, characteristics of concepts add elements that are not relevant to the description: polyimide
• the most important of which commercially is obtained from pyromellitic anhydride and 4,4 diaminodiphenyl ether
polyisoprene
• is produced synthetically in several forms • the most important forms are stereospecific
polymastigina • have highly complex cytoplasmatic structure • many are parasites The attribution of essential or complementary characteristics to a concept depends on the special subject field in question and the objectives of the definition. The same object of the real world can give rise to different concepts belonging to different special fields:
terminology
98 polyhalite
a. in chemisty: Hydrous sulfate of calcium, magnesium, and potassium b. in mineralogy: Mineral crystallizing in the triclinic system, with a hardness of 3–3.5 and a density of 2.77 and which is generally found in fibrous masses
sugar
a. in chemisty: generic name given to glucides, which can be subdivided into monosaccharides and oligosaccharides b. in food preparation: Sweet, crystalline water-soluble substance that is present in the sap of many plants.
b. Depending on the relationship they have with the object they represent, the characteristics of concepts can be intrinsic or extrinsic. The intrinsic characteristics of concepts describe the concept as a representative of a class: shape, colour, or size in the case of the names of objects; types of actions, feelings, or processes, in the case of verbs, etc. The extrinsic characteristics of concepts are external to the definition as a class: purpose, origin, target, place, inventor, etc. polyhalite a. intrinsic characteristics: • hydrous sulfate of magnesium, calcium, and potassium • mineral • usually found in fibrous masses • hardness of 3–3.5 • density of 2.77 • red, white, yellow or gray b. extrinsic characteristics: • found in Ischgl (Austria) and Texas odometer a. intrinsic characteristics: • device • coupled to the engine or axis of the traction wheel by a flexible hose • provides a series of figures, usually six in number, in the same colour • may include a seventh number in a different colour
the foundations of terminology
99
b. extrinsic characteristics: • used to count distance travelled by a vehicle • the first six figures indicate miles or kilometers • the seventh figure indicates tenths of the unit of measurement used The essential characteristics of concepts do not necessarily correspond to their intrinsic characteristics, nor do complementary characteristics always coincide with extrinsic characteristics. Rather, these features interact with one another because they belong to different criteria of classification. 2.1.2.3–The description of concepts Concepts are sets of interrelated characteristics that describe a class of objects of the real world. We can obtain this description by at least two procedures: intension and extension. The description of a concept by means of intension, which is the more common of the two procedures both in terminology and in general lexicography, consists of the ordered listing, from more general to more specific, of all the characteristics describing the concept. This logical process moves from genus to species. For example, the intension of the concept wolf is the set formed by the following characteristics: wolf: mammal → carnivore → fissiped → canine Describing a concept by means of the process of extension consists of listing all of its possible realizations, all its species. The extension of the concept halogen or that of regular polyhedron corresponds to the set of all the possible species of halogens and regular polyhedrons, respectively: halogens: fluorine, chlorine, bromine, iodine, and astatine regular polyhedrons: tetrahedron, hexahedron, octahedron, decahedron, etc. 2.1.2.4–The relationships among concepts Considered as parts of sets, terms are not isolated units occurring outside a specific context, but rather elements that form part of a specialized linguistic system. They occur in a concrete environment corresponding to a specific field of specialization. In this sense, terms are related to all the other terms that form part of the same special subject, with which they constitute a conceptual field. For example, the conceptual field of chemistry is the set of concepts constituting this domain. This set of concepts is structured into other, more specific subsets, and so on. The general set of concepts of chemistry is thus made up of the concepts referring to the appara-
100
terminology
tus used in laboratories, the substances with which chemists work, the processes and reactions produced, the components of substances, etc. The concepts of a conceptual field are connected to each other by logical and ontological relationships. A. Logical relationships The logical relationships existing between concepts are based on the similarity of concepts, i.e. on the fact that they share one or more characteristics. Logically, a concept can be more general than another; i.e. a more general concept lends all its characteristics to the other more specific concept, but the specific concept has, in addition to the characteristics of the generic concept, at least one other characteristic that differentiates it from the generic concept. In this case we speak of logical subordination (for example, mineral-anthracite). Equally, there may be two specific concepts subordinated to the same generic concept; i.e., both concepts have the characteristics of the generic, but each one has at least one particular characteristic that differentiates it from the other. In this case we speak of logical coordination between two specific concepts (for example, anthracite-pyrite). The combination of coordination and subordination establishes the hierarchical structure of a subject field. A hierarchical structure of concepts consists of a vertical sequence of concepts that usually enter into a relationship of inclusion. At each level of the sequence there is a horizontal series of concepts at the same level that enter into a relationship of opposition. For example: ANIMALS mammal (reptile, amphibian) carnivore (rodent, insectivore) fissiped (pinniped) canid (felid, mustelid) The most important hierarchical relations are those of inclusion, which cover generic relationships (species-genus): building house, apartment building, firehouse house detached house, split-level house wall load-bearing wall, curtain wall, cement-rendered wall, etc. In genus-species relations, the most specific concepts belong to the extension of the generic concept. The inclusion relationship concepts have with one another is reflected in the description of each concept by the explicit statement of its characteristics. A specific concept has all the characteristics of the generic concept and one or more other particular characteristics.
the foundations of terminology
101
a. emergent light beam: light beam that crosses a reference surface incident light beam: light beam that reaches a reference surface b. reflected light beam: light beam emerging from a surface where the beam has undergone reflection refracted light beam: light beam emerging from a surface where the beam has undergone refraction Not all concepts, however, are the result of the combination of characteristics; in many cases they are complex products of the combination of simple concepts. From the standpoint of their description, complex concepts are formed by the combination of structured sets of characteristics, each of which describes a simple concept. The combination of simple concepts making up complex concepts that are not in a relationship of subordination nor logical coordination can lead to three cases, all of which are based on the transfer of meaning: a. The combination of two concepts may result in a third, which is specific to one of the two starting concepts (machine → sew: sewing machine). In this case we speak of one concept determining another. b. The combination of two concepts may result in a third, which is specific to both of the starting concepts (clamp → spring: spring clamp). In this case we speak of conjunction of concepts. c. The combination of two concepts may result in a third, which is only characterized by the differentiating features of the two starting concepts, and thus not specific to either of the two (Indo → European: Indo-European). In this third case we speak of disjunction of concepts. The conceptual relationships and combinations can be expressed by symbols. Symbols for relationships: the same sense as = sense similar to ≈ generic to > specific to < Symbols for combinations: conjunction of concepts ∧ disjunction of concepts ∨ determined by -+ In the formation of new terms by means of combining morphemes or words, the conceptual relationships established between the units constituting a new term are
terminology
102
a
b
Conjunction de notions
Disjunction de notions
1 - Vis1
Forme de base; Type de forme; Configuration1 Aspect extérieur
1.1 Vis11 (Filet en relief)
1.2 Écrou (Filet creusé)
1.11 Vis111 Vis, d’assemblage
1.12 Vis de mouvement
Forme1; Configuration11
Taille
Forme et taille; forme111 Configuration111
c
d
Intégration
Intégration et disjonction de notions
Vis11
Corps lisse; Partie lisse
Tige
Partie filetée
Corps de boulon; Écrou Vis111 Cors de boulon (=vis) avec écrou; Ensemble vis-écrou (=boulon)
Conceptual relations according to Wüster; taken from the Dictionnaire multilingue de la machine-outil. London, 1968.
the foundations of terminology
103
the same as those that obtain between the concepts. In term formation by transfer of meaning, we speak of restriction to refer to the move from the generic to the specific; and, we use extension to refer to the same conceptual relationship in the opposite direction, from the specific to the generic. In linguistics these relations— restriction and extension—are known as metaphor. B. Ontological relations As we discussed in Chapter 2, terminology shares with ontology the interest in the nature of beings of the real world and the relationships with other beings. Ontological relations are not based on the similarity between concepts, but rather on the proximity of beings to each other in the real world. From this standpoint, there are two types of ontological relationships based on the contact among concepts: a. Coordination relationships (part-whole relationships), which describe two types of relationships: those established between a whole and its parts (car—wheels, seats, doors, steering wheel), and those among the various parts of a single whole (wheels, seats, doors, steering wheel). These relationships are based on the contiguity of objects in space, and are thus produced simultaneously. b. Chain relations, which are based on the succession of objects in time (cause-effect relationships), which are thus sequential. Ontological relationships can also exist between the formatives or words that make up complex terms. 2.1.2.5–Concepts as systematic units A concept belongs to a structured set representing a conceptual field and acquires its value with reference to this set. A conceptual field can be structured in a variety of ways, each of which reflects a particular view of the real world. A conceptual field can have various structures at the same time within a single special subject field, as can be seen in the examples below: Pharmacology Option A 1. products 2. substances they contain 3. process to obtain them
Option B 1′. products 2′. formulae for commercial presentation 3′. costs and profitability
104 Animal zoology Option A 1. habitat 2. migratory stages 3. types of reproduction
terminology
Option B 1′. relationship to human world 2′. reactions and behaviour 3′. learning capacity
As will be seen in detail in Chapter 4, terminological research benefits from establishing the conceptual structure of a special subject field. The conceptual structure of a field reflects a specific cultural and scientific view of the real world. In this sense, it allows for more than one scientific approach to the same object, and more than one cultural approach to the same reality. Structuring a conceptual field allows us to produce more controlled, more coherent terminologies because it provides a systematic approach to a special subject field. We can control the relationships between the various concepts in a single field, check what is equivalent among terms in different languages, produce definitions in a systematic fashion, and retrieve information more efficiently. Finally, the existence of a conceptual structure allows the naming of new concepts to be more consonant with the other designations in the same field. 2.1.2.6–The representation of concepts A concept can be represented in dictionaries either by a definition or by an illustration. Illustrations are iconic units that reproduce the idea that individuals have of a certain class of objects in the real world. Definitions are linguistic formulae that, like illustrations, are aimed at describing the concept. The International Organization for Standardization (ISO) has fixed the nature and types of definitions in two different standards, and in each one a different aspect of the same concept prevails. ISO standard 704 (1987) states: definition: complete description, usually in language, of a concept using other, already known concepts. ISO standard 1087 (1990) adds: definition: statement which describes a concept and which permits its differentiation from other concepts within a conceptual system. The second definition is more useful for terminography because, theoretically speaking, it is important to differentiate three types of definition: a. linguistic definitions b. ontological definitions c. terminological definitions
the foundations of terminology
105
These three types of definitions can be differentiated with respect to the object they describe and the contents they express. Each type of definition refers to the same reality seen from a different perspective. The object of linguistic definitions is the linguistic sign; that of ontological definitions, the object in the real world; and, that of terminological definitions, the concept of a special subject. Likewise, the contents expressed by the three types of definition are different from one another. Linguistic definitions do not usually include all the characteristics of a concept but rather only those that are most important for differentiating it from another concept in the language. The main goal of linguistic definitions is to clearly distinguish different concepts. Ontological definitions include all the particular intrinsic, extrinsic, essential, and complementary aspects of a concept, regardless of whether they are relevant to defining it as a class or not. We often find encyclopedia-like features in definitions. Terminological definitions are more descriptive than contrastive; they describe concepts in exclusive reference to a special subject field and not to a linguistic system. Nevertheless, the difference between these types of definition is clearer in theory than in practice. We only have to consult a non-specialized dictionary, for example the American Heritage Dictionary of the English Language, Third Edition, to see that lexicographers use a mixed process somewhere between a linguistic definition and a terminological one to define terms: hammer n. 1. A hand tool used to exert an impulsive force by striking, consisting of a handle with a perpendicularly attached head made of a relatively heavy, rigid, material. 2. A tool or device similar in function or action to this striking tool, as: a. The part of a gunlock that hits the primer or firing pin or explodes the percussion cap and causes the gun to fire. b. Music. One of the padded wooden pieces of a piano that strikes the strings. c. A part of an apparatus that strikes a gong or bell, as in a clock. 3. Anatomy. See malleus. 4. Sports. A metal ball weighing 16 pounds (7.2 kilograms) and having a long wire or wooden handle by which it is thrown for distance in track-and-field competition. 5. A small mallet used by auctioneers . . . Definitions can be intensional or extensional. Intensional definitions, as stated above, are based on bringing together the characteristics describing the concepts. Extensional definitions enumerate the specific objects that a concept represents. In terminology, definitions (and illustrations) are often produced in accordance with widely accepted conventions and principles. General principles: definitions must
106
terminology
• describe the concept (i.e., they must be true); • allow differentiation of the defined concept from similar concepts in the same or in different special fields; • bring together the dimensions pertinent to each special field; • be located in the perspective of the conceptual field a concept belongs to; • be appropriate for the aims of the project in which they are presented. Specific practices: Definitions also have to be written according to a number of recommendations: • they must be compatible with the type of definitions used in a specific field and as such must start from the pre-existing structure of this field; • they must collect all the essential characteristics of each concept, in accordance with the established structure of the field; • they must reflect the systematic relationships a concept establishes with other concepts in the same field; • they must include all the characteristics that are important for a complete description of the concept, even if they are not essential. From the standpoint of expression, definitions must meet certain conditions: • they must be correctly expressed: e.g. butter knife: knife used to cut butter • they must adhere to the formal standards for writing definitions: e.g. nitriding: a process of case-hardening in which nitrogen is introduced into the metal by keeping it at a suitable temperature in the presence of a nitrogen source. • they must use language that is suitable for the intended readership; compare: migraine: severe continuous pain in the head, often with vomiting and difficulty in seeing migraine: a paroxysmal, extracranial headache usually confined to one side of the head, in which vascular headache is combined with other bodily disturbances such as changes in autonomic nervous system function. • they must be in the form of a single sentence; • they must adhere to the lexicographic principles dealing with formal presentation. These principles of presentation can be summarized as follows:
the foundations of terminology
107
• Definitions must be written so that the initial descriptors are of the same grammatical category as the head term and are in an inclusive relationship semantically with the head term: oxygenate: treat, combine, or infuse with oxygen circulation: movement or passage through a system of vessels, as of water through pipes muddy: full of or covered with mud • Definitions must use known words; if more specific words are used, they must be terms defined in the same glossary or dictionary: kymogram: graph or record made by a kymograph kymograph: instrument for recording variations in pressure, as of the blood, or in tension, as of a muscle, by means of a pen or stylus that marks a rotating drum. • Definitions should not be circular: dense: having relatively high density density: the quality or condition of being dense • Definitions should not be defined by negating something else: unequal: not equal true: not false insatiable: impossible to satiate • Definitions should not include unnecessary paraphrases that only provide information that could be derived from the term itself: Huntington’s disease: disease identified by George Huntington, American physician tricolour: having three colours • Definitions should avoid metalinguistic formulae: circulate: verb designating the action of moving or passing through . . . ISO standard 704 includes additional criteria for the formal preparation of definitions in terminology. 2.1.3–The term-concept relationship In theory, terms are unambiguous (the relationship between form and concept is
108
terminology
unique) as well as mono-referential (one term names only one object). In general language most words are polysemous which means that they have more than one definition in a dictionary. For example: gorge, n. 1. a narrow cleft with steep, rocky walls, esp. one through which a stream runs. 2. a small canyon. 3. a gluttonous meal. 4. that which is swallowed; contents of the stomach. 5. a choking mass. 6. Fort. the rear entrance or part of a bastion or similar outwork. [. . .] In addition to polysemy, there is often synonymy, which occurs when a concept can be expressed by more than one designation. Synonymy, as opposed to polysemy, cannot be detected from examining a single dictionary entry, but rather becomes evident upon examining several entries and by looking up the cross-references. 2.1.3.1–Polysemy In terminology, polysemy is treated quite differently from the way it is treated in lexicography. Terminology is based on the principle that one designation corresponds to one concept, but this univocal relationship does not always occur in practice. The semantic value of a term is established solely on the basis of its relationship to a specific conceptual system. Identifying a term as belonging to a special subject field involves placing it in a specific conceptual system, and as a result what in lexicography is considered polysemy, in terminology becomes homonymy. A. Lexicography key. n. 1. a metal instrument for operating a lock by moving its bolt. 2. a means of attaining, understanding, solving, etc. 3. a book, pamphlet, or the like, containing the solutions or translations or material given elsewhere, as testing exercises. [. . .] 12. Music. a. (in a keyboard instrument) one of the levers that when depressed by the performer sets in motion the playing mechanism. b. (on a woodwind instrument) a metal lever that opens and closes a vent. c. the keynote or tonic of a scale. d. the relationship perceived between all tones in a given unit of music to a single tone or a keynote. e. the principal tonality of a composition. [. . .] B. Terminology key. n. cartography. list of map’s symbols and their meanings; legend. key. n. telecommunications. individual dialing button on touch-tone phone; switch that opens or closes communications circuit, moving from one line to another. key. n. technology. device that operator uses to send telegrams.
the foundations of terminology
109
In spite of this principle, in a single special subject field there can be identical terms with different meanings. Their independent conceptual status may be justified by the fact that they belong to different branches of the same field. Polysemy is one of the most productive ways of extending a language’s lexicon. The origin of most polysemous terms is analogy of one concept to another, which allows the designation of one concept to be used for designating another. A new term is thus created from partial semantic overlap. virus: a. Medicine: an infectious agent, especially any of a group of ultramicroscopic, infectious agents that reproduce only in living cells. b. Computer science: self-replicating program intentionally designed to disrupt operations of one computer or computer network thaw: a. Hydrology: melting of snow and ice that occurs when the temperature rises above the freezing point b. Political science: improvement of political relations after a period of hostility or tension reaction: a. Chemistry: a change or transformation in which a substance decomposes, combines with other substances, or interchanges constituents with other substances b. Physics: an equal and opposite force exerted by a body against a force acting upon it c. Medicine: the response of cells or tissues to an antigen 2.1.3.2–Synonymy Even though theoretically a concept is expressed by a single designation, in reality there are alternative designations for a single concept and the designation of two different concepts can coincide, even within the same special field. Broadly speaking, two units designating the same concept are synonyms. In this very general sense, synonymy exists between units at various levels: a. Between a designation and its definition alternating current = electric current that reverses direction in a circuit at regular intervals b. Between a designation and an illustration of the same concept c. Between equivalent terms in different languages Eng. design, Fr. dessin, Sp. diseño
110
terminology
d. Between designations of different functional languages correctional centre, penitentiary, prison, jail e. Between alternative designations in the same historical language tocology/obstetrics; windshield/windscreen; computed tomography/computed axial tomography/computerized axial tomography/computerized tomography Strictly speaking, terminology only considers synonyms to be semantically equivalent units that belong to the same historical language and to the same formal register. Synonyms for a single concept, however, do not always correspond to absolute equivalents, but rather manifest a range of possible cases. Sometimes there is synonymy between two semantically equivalent units in which one form is derived from the other. This sort of relationship occurs between: a. Initialisms and their full form laser = light amplification by stimulated emission of radiation b. Abbreviations or clippings and their full form metro = metropolitan railway In other cases, two designations are synonymous only in a very narrow linguistic sense and are not synonymous in a pragmatic sense. There are many cases of two synonymous units that belong to two different registers of the same language, but this does not usually appear in a single specialized dictionary. This dissymmetry occurs in cases like the following: a. Between a scientific name and its popular name Diapheromera femorata = walking stick Antirrhinum majus = snapdragon b. Between a standard form and dialectal forms sweet pepper = mango frying pan, skillet = fry pan, spider Two synonymous forms can also coexist in a single special language and represent different subcodes. This is the case for the nomenclature related to terminology and for the symbols related to terms. Finally, two synonymous units can simply be variants of the same word or term: materiel = matériel data bank = databank guerrilla = guerilla archaeology = archeology In those cases in which two synonymous units might affect the unambiguous sense of terms, terminology tends to reduce concurring forms designating the same concept to just one designation. This tendency is justified in order to achieve unam-
the foundations of terminology
111
biguous communication among subject field experts. Here again the goals of terminology and lexicography diverge. Terminologists search for semantic clarity, the elimination of factors that disturb communication, and the updating of reference in specialized discourse; this leads them to avoid multiple forms, and as a result, to a reduction in the expressive richness of a language. 2.1.3.3–Homonymy Homonyms are terms that have the same form but represent entirely different concepts. This definition of homonymy, in which terminology and lexicography coincide, is clearer on paper than in reality. In the first place, we must note that homonymy is much more frequent in terminology than in the general lexicon. This difference in frequency is more apparent than real because it is based on the fact that in terminology each special subject field is considered a closed domain. Terminology conceives of each subject code as a system that is to be differentiated from the others that with it make up the global linguistic system (historical language). Consequently, any term from a special field that is extracted by analogy and applied to another field will be a homonym.18 In contrast to this approach, general language lexicography is based on the historical language as a whole and not on the subsets that make up the whole. As a result, the number of homonyms in the general language will be fewer than in terminology. Conversely, there will always be much more polysemy in lexicography than in terminology. Traditional linguistics draws a distinction within homonymy between phonological homonymy (homophony) and orthographic homonymy (homography). Homophones are units that are pronounced the same but are spelled differently: e.g. night/knight, here/hear, bare/bear, whereas homographs are words that have the same spelling but differ in origin, meaning and sometimes pronunciation. Perfect homonyms are those that present both homophony and homography: e.g. match (counterpart)/match (narrow piece used for ignition); bore (to drill)/bore (past tense of bear)/bore (provoke boredom). 2.1.4–Function Terms are combined with words from the general language and other terms of the same special field to constitute specialized texts. Terminological units follow the patterns set by their basic grammatical category and subcategories. All terminological units are contextually restricted by their category and speakers know what the categorial descriptions are in each case. From the standpoint of their linguistic function in discourse terminological units can be nouns, adjectives, verbs, or adverbs.
112
terminology
Seen from the perspective of the language system as a whole, the proportion of the various grammatical categories is clearly uneven. Nouns represent two-thirds of all terms. One of the peculiarities of specialized terminology is the tendency to prefer noun-based structures, as opposed to verb-based or adjective-based structures, to designate the concepts in a subject field. The cells in the conceptual structures of a domain take on the linguistic form of nouns. 2.2–Terms as pragmatic units Terminology is not independent from social practice: it occurs within the discourse of particular interlocutors and it partially reflects the ideology conveyed by the sciences and technology. If terminology is to have a theory, that must be reflected in the issues it deals with and the functions it performs—(Rey 1985). Only its foundations, its à priori, its assumptions and its premises will be questioned here. Is there any contradiction between the premises of terminology and its working methods and practices? Can they be overcome? Are terminologists only responsible for developing tools to collect and retrieve terms, and for their own training? Should not they also deal with the problems posed by language policy and somewhat conflictive language interaction, as well as with the communication networks through which concepts and terms emerge? Gambier (1991a)
Besides being formal, conceptual and functional grammatical units, terms are also pragmatic units of communication and reference and as such have certain discourse characteristics and occur in well-defined communicative situations. Terminology has paid relatively little attention to usage in communication. Much linguistic theory generally ignores the diversity with which any one language manifests itself in acts of communication. Alongside this linguistic view of language, alternative proposals have been made for analyzing language as a tool of expression and communication (Cabré & Payrató: 1990). Pragmatics, one of the fields of applied linguistics, deals with the description of language usage and is related to—if not situated somewhere between—sociolinguistics, psycholinguistics, anthropology, and discourse analysis. Pragmatics is very important in characterizing the different nature of terms. The five most important pragmatic factors that allow us to differentiate between general language and terminology are: a. the basic purpose b. the subject dealt with c. the users d. the communicative situations in which both codes are found e. the types of discourse in which terms or general language words appear
the foundations of terminology GENERAL LANGUAGE LEXICON
TERMINOLOGY
BASIC PURPOSE
BASIC PURPOSE
performative, expressive, communicative, and others
referential
SUBJECTS
SUBJECTS
generic
specific
USERS
USERS
general
specialists
COMMUNICATIVE SITUATION
COMMUNICATIVE SITUATION
– structured
+ structured
DISCOURSE
DISCOURSE
general
professional and scientific
113
Figure 8. Pragmatic factors allowing us to differentiate general language lexicon from terminology
The chart (Figure 8) summarizes the differences between the two viewpoints, although it represents a simplification of the issue. Terms are used to name a specialized reality and are thus different from words in the general language because they have a primarily referential purpose. Other purposes of language, such as performative, expressive, communicative, poetic or metalinguistic uses, which are so pervasive in communication as a whole, are usually quite rare in terminological discourse.
114
terminology
Terminology is used to designate concepts of a restricted field or activity. In this sense a unit can only become a term if it is located within a special subject field. The general language lexicon, on the other hand, is used to refer to all types of activities shared by speakers. Terminology is used by experts in a specific subject field. If we extend terminology to include professional or sports vocabularies, or those vocabularies related to some human activity, the number of terminology users increases and diversifies substantially and at the same time the degree of specialization of these uses decreases considerably. The seemingly clear dividing line between the general language lexicon and terminology based on user-types (speakers of a language vs. experts in a specific area of knowledge, respectively) becomes blurred. In theory terminology is used above all in structured communicative situations and not in colloquial, everyday situations in which general language vocabulary would be used. Nor is terminology used in texts that are not referential. In reality, experts in a given domain can use the specialized terminology in a variety of communicative acts and at several different levels of abstraction, thus blurring the classic, well-defined view of what terminology is. Terminology, as both a discipline and an activity, but especially as a discipline, needs a new orientation which stresses its social and pragmatic aspects. Terms are only units in a system if they are used in communication, and therefore we must reconsider the importance of relating them to their natural speakers and social groups.
Chapter 4 Terminology in practice: Terminography
1.–The foundations of terminological practice One of the basic principles of terminological work is that it must abide by language rules. If the language must benefit from improvements, if the aim is to learn about it and to enrich it, terminological work can never occur without meeting the condition of complete respect for its structures. The practice of terminology is thus required to present language the way it is; to exploit its original content to the extent possible; and to shed light on the system and the lexicon that is at its base as precisely as possible.—Halaoui (1990)
Terminography involves gathering, systematizing, and presenting terms from a specific branch of knowledge or human activity. The relationship of lexicology to lexicography is parallel to that which exists between terminology and terminography, the application of terminology that deals with special language dictionaries. Even though it is practical in nature, terminography is not an independent activity that individual specialists can deal with on their own but is governed by a series of technical, formal, and procedural recommendations that have been internationally agreed. 1.1–Theoretical principles Terminography must not be confused with translation. Translators need specific terminology for specialized texts, but this does not imply that they themselves must develop the terminology, nor that they have to deal with all the terms in the special subject field in question. Working in terminology does not mean translating a term from one language into another based on supposedly equivalent designations, but gathering the designations that users of a language use to refer to a concept andultimately, if necessary, proposing alternatives in those cases where speakers’ designations are unsatisfactory. While translators are not terminographers, in daily practice the distinction between these two groups of professionals is often blurred.1 Terms for a special language glossary must be ‘‘collected’’ from real texts, and not ‘‘invented’’ or ‘‘created’’ by terminologists. When specialists have to discuss
116
terminology
a concept, they do not stop because of a lack of terminology. They always use specialized terms to express themselves, so the terms are there. Terminographers can always gather the terminology specialists use to communicate with each other, regardless of whether it is the most rigorous or appropriate or whether some designations are the best way to refer to a concept or not. The fact that a specific terminology has a real presence in specialized texts, however, does not mean that it is necessarily the best one possible for a particular discipline. The absence of a certain topic in communication and specialized texts obviously implies a gap in terminology, and if detected by a terminologist, it is up to him or her to fill the gap with a neologism. In addition terminography is guided by the facts that terms are indivisible units with form and content; that the form and content of terms tends towards an unambiguous relationship; that terms have a fixed place within a specific conceptual field, without which reference the term has no specific meaning; and that concepts are related to other concepts in the specific field they together constitute. 1.2–International standards In an attempt to harmonize work in terminography and to make it easier to transfer knowledge and data, terminography is also guided by recommendations made by international committees like TC 37 of the ISO.2 There are guidelines, published as standards, for unifying designations and concepts in terminological theory and practice, and for the methods to be applied for the presentation of terms and their respective data. (See Chapter 6.) 2.–Materials used in terminography A terminological record is designed according to the type of information it should contain. It cannot be a definitive document included in an infallible orthodoxy. It is rather a sort of snapshot of a particular usage found in a certain time and place. It is thus important to give the record the simplest possible form, without sacrifying the rigour required for the validity of the given information.—Dubuc (1985)
Each terminology processing centre uses the source material suitable for their purpose. Three types of material are generally used: • Reference works, which provide information about the various aspects of a project; • Specific documents, which constitute the material basis for a project; and, • Support materials, which facilitate and complement the work.
terminology in practice: terminography
117
2.1–Reference materials Reference works are the documents terminologists use to obtain background information on theoretical, methodological, practical or bibliographical aspects on the subject. This information can refer to the conceptual system of a subject domain, its system of designation, or complementary aspects of the professional or scientific activity related to the field in question. They may be terminological works on the same or on a related topic, dictionaries that cover a part of the terminology on the topic, handbooks and other background materials, etc. Once this information has been examined, terminologists begin the terminographic process itself.3 They first evaluate the information and choose those source materials they believe to be the most important. Then, using this documentation and consultations with specialists in the field, terminologists acquire information on the contents of the subject field and structure a concept system. Third, they isolate and extract terms in the documents chosen for this purpose. Fourth and finally, all the documentation is examined to recheck the data and complete the data obtained from the extraction process. There are four main types of documents, depending on the topic covered: • • • •
documents on documentation documents on the special subject field documents on the topic documents on methods
2.1.1–Documents on documentation The three most common ways of obtaining information about searches and of finding what documentation is going to be available during the whole terminographic process are: • consulting secondary sources (bibliographies) and tertiary sources (bibliographies of bibliography) • consulting document databases • consulting terminology management centres and subject specialists Terminology centres often have specialized libraries. Their documents can be used to help guide a search and make the initial choice of materials. 2.1.2–Documents on the special subject field Terminological searches require a combination of knowledge, regardless of how it has been acquired. Terminology teams need three types of knowledge:
118
terminology
• Expertise in terminology: theory, methodology and practical experience. • Expertise in the subject field (typically possessed by specialists in the field). • Expertise in the language or languages in question (typically possessed by linguists). Terminography is usually undertaken jointly by teams of experts in a subject field and linguists specialized in language structure and terminological methodology (Cabré 1989b). Practical work in terminology requires that a terminologist knows enough about the field in question.4 Subject specialists progressively acquire knowledge about their field throughout their training, and consequently do not begin from zero when they start a terminographic task. They probably have to acquire, however, knowledge about language structure and specific training in the principles of and methods used in terminology. Terminography can only be carried out by someone with a background in linguistics if he or she has a solid familiarity with the subject, in addition to experience in the principles and method of terminology. One of the most useful types of document prepared at terminology centres is the special topic file. A topic file brings together different materials on a single topic; it usually includes articles from newspapers and specialized journals, references to a more complete bibliography, expert opinions, photographs, and other illustrative material. Topic files are not limited to information on contents, but may also include background information about centres specializing in the topic, the names and references of experts etc. 2.1.3–Documents on terms The most common sources for those working in terminography are dictionaries, other types of lexical lists and terminological databases. Until recently, printed dictionaries were practically the only instruments for consultation on terms, and still remain the most consulted tools.5 The reasons for consulting dictionaries and lexical and terminological databases are many, and they correspond to a variety of needs: • • • • • •
to resolve doubts about the existence of a term in a language to know how to use a term grammatically, to know its meaning and its spelling to know its equivalents in other functional or historical languages to find out the name of an object to find alternatives to a designation to find out how many terms have been collected in a specific domain, etc. Both general and specialized lexicographic works are particularly useful if their
terminology in practice: terminography
119
perspective coincides with that of the user. From the standpoint of usefulness for terminology, we can identify seven types of works: a. General language dictionaries, which include the most basic terms in specialized terminology known by the general public. b. Encyclopedias, besides containing a lot of terminology, often present conceptual data in classifications, charts, illustrations, etc. c. General science and technology dictionaries, which in principle only include specialized terminology. d. Specialized dictionaries, which give equivalents, and definitions, explanations or classifications. They are often more precise than encyclopedias or general science and technology dictionaries. e. Specialized visual dictionaries, which allow users to find the designations in a subject field starting with a concept represented by an illustration. f. Lexicons and vocabularies arranged by subject matter, but without definitions. These works can provide information about the conceptual structure of a special subject domain.6 This group also includes thesauri and classifications. g. Terminological databases, the most complete and usually the best, most up-todate sources because the machine-readable format not only allows for constant updating of information but also for multi-tier retrieval of the information depending on the purpose of each search. Other works of a more restricted nature, such as dictionaries of scientific and technical neologisms or bulletins of standardized terms, are also basic sources to be consulted by terminologists. 2.1.4–Documents on the research method and presentation of work Basic knowledge of terminography is usually acquired at training courses with specific aims and addressees in mind (Cabré 1992b). In addition to being familiar with terminology manuals, methodologies and articles dealing with some aspects of the field, terminographers must also be aware of the agreements that have been reached internationally and which are published as standards. The standards on terminology of the ISO Technical Committee 37 refer to the following five aspects: a. the assumptions to be taken into account in terminology: ISO 704 (1987) Principles and methods of terminology b. the vocabulary used in terminology: ISO 1087 (1990) Terminology—Vocabulary.
120
terminology
ISO/DIS 1087–2 Terminology work—Vocabulary—Part 2: Computer applications [Partial revision of ISO 1087:1990] c. the work process: ISO/R 860 (1968) International unification of concepts and terms ISO/DIS 860 Terminology work—Harmonization of concepts and terms [Revision of ISO/R 860:1968] ISO/TR 12618:1994 Computational aids in terminology—Creation and use of terminological databases and text corpora d. the presentation of vocabularies in general and of the information about terms: ISO 10241 (1992) International terminology standards—Preparation and layout ISO/DIS 1951 Lexicographic symbols and typographical conventions for use in terminography (Revision of ISO 1951:1973) ISO 639 (1988) Code for the representation of names of languages ISO/DIS 639–2 Codes for the representation of names of languages—Part 2: Alpha–3 code ISO/DIS 12620 Terminology—Computer applications—Data categories e. the format for exchanging terminological data: ISO 6156:1987 Magnetic tape exchange format for terminological/lexicographical records (mater) ISO/DIS 12200 Terminology—Computer applications—Machine-readable terminology interchange format (martif) Countries with specific organizations for coordinating and directing terminological work usually publish instructions that aid specialists and allow the standardization of information. These recommendations aim at achieving a degree of homogeneity in the methods used and at standardizing the symbols used in classifications and the presentation of data. Some language or terminological organizations whose authority is recognized in a specific field periodically prepare documents to formally settle problematic cases which the standard codification of the general language has not foreseen, for example, the use of graphic symbols. Finally, terminologists must be aware of the terminological policies set by standardizing and administrative organizations. This is particularly important in settling questions about neologisms in special subject fields. These guidelines are usually made available in specific documents that must be widely disseminated among specialists and terminologists in order to effectively control new units in special languages.
terminology in practice: terminography
121
2.2–Specific materials for terminographic work Oral and written technical and scientific communication is the basic source material for extracting terms. Terms are usually created by the specialists responsible for the concept, who introduce a certain designation into a special subject field. Experts then communicate with each other and discuss their subject matter using designations they consider suitable to the system of their language. When terminologists collect the terms of a special subject field, they have to examine the existing terminological units and, if necessary, propose a change. The source material chosen for a specific terminographic task constitutes the document corpus from which the designations that will make up the initial list for a terminological search are extracted. These sources must be assessed for their quality, namely: • they must be representative of the subject matter, in accordance with the objectives of the task and the delimitation of the topic, so that they permit the preparation of a sufficiently large initial list of units; • they must be up-to-date both regarding the designations that experts really use and the topic; • they must be explicit enough to allow retrieval of the identification and information at any point in time during the process or during dissemination of the terminological data. The sources of documents must be duly indicated and referenced in a bibliography. 2.3–Support materials Records are the starting point for organizing the information of a terminological search. They are structured so that they usually include a protocol for use. Three types of records are used in systematic searches, depending on the objectives of each stage of work: extraction records, terminological records, and correspondence records. In ad-hoc searches, i.e. searches directly responding to user queries, a specific record, known as a query record, is normally used. 2.3.1–Extraction records Extraction records must contain the terms that have been identified in the document sources. The structure of an extraction record allows the representation of complementary information directly or indirectly present in the text in which the terms occur.
terminology
122 1 2
3
4
5
6
7
Fields identified on the record 1–Entry 5–Definition/Context 2–Grammatical category 6–Reference 3–Subject field 7–Author/Date 4–Description of contents Figure 9. Sample extraction record from the Catalan Language Service, University of Barcelona. An extraction record normally has the following fields: • the entry, corresponding to the terminological segment found in the text, as it appears or in a regularized lexicographic form, its ‘canonical’ form • the grammatical category, which is deduced from the form in which it is used in the text • the context in which it appears • the complete reference of the source document. In addition, an extraction record may also include other information, such as: • indication of the language the document is written in, if simultaneously working with different languages • the motivation for the record (context record, source record, definition record, etc.)7
terminology in practice: terminography
123
1. EN ENTRY a. Overall indications All expressions recorded with their own entry shall be considered documented neologisms in the dictionary; neologisms clearly identified as subentries in a more generic entry shall also be considered documented (this is the case of phrases and some complex terms). The entry shall be recorded following the natural order of the sequence, and must be recorded in full, with no alterations or abbreviations of any sort. The entry shall always be written in lowercase letters; if a term appearing with an initial capital letter is to be recorded (e.g., a word deriving from a proper name), it should also be recorded in lowercase, maintaining the original form in which it was found in context. When the author of the record finds two or more formal variants of a single term, complete records must be filled out for each variant, as if they were different terms. When a single lemma is found under two different grammatical categories (especially noun and adjective, but also masculine noun and feminine noun), it shall be considered as if it were two different terms and therefore two records, one for each term, shall be made.
Figure 10. Part of the protocol for extraction records, Catalan Language Service, University of Barcelona • information management data (record author, date of production of the record, etc.). The protocol for using the extraction record provides information on the type of data that should appear in each box, the possibilities of the information on the record, and the recommendations or standards for presentation for each item of information (see Figure 10). 2.3.2–Terminological records Terminological records contain all the relevant information about a term (the information extracted from the extraction records or the reference documentation), and are presented according to certain set criteria. There are several different models for terminological records, each of which is designed to meet the needs and objectives of a particular task or organization.
124
terminology
Within this diversity we can first distinguish among monolingual records, monolingual records with equivalents, and bilingual or multilingual records. Monolingual records contain information about a special language term from a single language. If the record includes equivalents for this term in other languages, we speak of records with equivalents. These records have a double nature: they are monolingual because they have been the object of research in a single language, but they are also multilingual because subsequent consultation of multilingual lexicographic documentation allows us to assign equivalents in other languages. Bilingual or multilingual records must contain complete information about two or more languages, respectively. Standard terminological records usually contain the following information: • • • • • • • • • • • • • • • • •
identification of the term entry term source of term grammatical category subject area(s) definition source of the definition context(s) source of the context cross-reference to synonymous terms concept of the cross-reference other types of cross-reference concept of each type of cross-reference author of record and date written miscellaneous notes for unanticipated information equivalents in other languages, indicating the language source of each equivalent
In addition to these data, terminological records may also include other items of information depending on the reason why the information is being obtained. For instance, if data is being collected on the standardization or status of a term, the record would include: • the status code of the entry term • the status code of each one of the designations on the record
terminology in practice: terminography 1
1′
3
4
5
3′
125 2
4′
5′
6 7 8 9 7′ 8′ 9′ 10
11
12 13 14 15
Identification of the fields on the terminological record 1. 1′. 2. 3. 4. 5. 3′. 4′. 5’. 6. 7.
Catalan entry Spanish entry Number of record Reference (Catalan entry) Grammatical category (Catalan entry) Status code (Catalan entry) Reference (Spanish entry) Grammatical category (Spanish entry) Status code (Spanish entry) Subject fields Definition (Catalan entry)
8. 9. 7′. 8′. 9′. 10. 11. 12. 13. 14. 15.
Reference (Catalan definition) Cross-references (Catalan entry) Definition (Spanish entry) Reference (Spanish definition) Cross-references (Spanish entry) Equivalents (French, English, third/fourth language) References (Equivalents) Illustration Author of record Author’s working group Date
Figure 11. Sample Terminological Record, Catalan Language Service, University of Barcelona
terminology
126
• the status code of the other information • the standardizing body, if appropriate If the geographic scope of the term is to be indicated, the record will also include information about the country or region in which the term is used. If data management is to be controlled on a more detailed basis, then the record must contain other information about the record’s preparation and its authors, or data about the people who have participated at some stage of the writing, etc. If the relationship of the base term to other terms and its place within the conceptual network of the field needs to be more precise, there will be information on the conceptual structure of the field. Finally, if the record is multilingual, then it will include equivalent designations in several languages. Terminological records are usually accompanied by guidelines for usage in which the fields of the record and the exact characteristics of each field are carefully presented: • • • •
the reference code for the record (manual/automatic) the identification code the full name of the field the definition of the field
5 STATUS LABEL
SL C
Status labels for Catalan are made up of a number indicating the degree of reliability of the term, and may be followed by a letter or letters in cases in which the term has been proposed or backed by a recognized body. The letters (up to three characters) refer to the institution. 1. Standard term 2. Term considered standard by the Supervising Committee of TERMCAT 4. Neologism (non-standardized term) (letter = ‘‘proposed by’’) 4 TC TERMCAT 4 SLC Catalan Language Service, U. of Barcelona 4 UAB/UPC Language services of other universities 4 . . . Other language services 5. Unresolved term
Figure 12. Section of the protocol for terminological records, Catalan Language Service, University of Barcelona
terminology in practice: terminography
127
• the representation (including the various possibilities for each field and the corresponding representation code) • indications on the type of field (numerical, alphanumerical, textual, etc.) and its conditions (obligatory or optional; repetitive or non-repetitive) • notes and comments on the field • examples Terminological records, like extraction records, are filled out following a guideline (for example Figure 12). 2.3.3–Correspondence records In some bilingual or multilingual databases in which the information is stored on separate records by language, a correspondence record can be used to correlate all the designations for a single concept. In these cases, the authors must ensure the correlated records refer to the same concept; the best way to do this is to start from the definition contained in the records themselves. 2.3.4–Query records In ad-hoc terminology searches resulting from a query made by an end-user, the 5. Reply to an ad-hoc query The data in the reply must be filled in as concisely as possible. Formulation The answer provided to the end-user must be summarized, as in the section on the query. The key word is obviously the same. References The code for the sources used must be included. Follow document DD2. Comments Any pertinent observation in summarized form. Provisionality code Indicate if the response is definitive or provisional: Definitive Provisional
D P
Figure 13. Section of the protocol for query records, Catalan Language Service, University of Barcelona
128
terminology
1. PERSON OR BODY PLACING THE QUERY Address
/
Name Telephone/Ext
Contact
Date
Time
2. QUERY DATA Type of query
Response time
Org.
3. QUERY TYPE Subject No.
Subject field
/
4. QUERY Key word
Grammatical category
Equivalents Formulation Definition/Context
5. RESPONSE Formulation
P/D
References Comments 6. RECORDS CONTROL Operation
Record
7. EXIT DATA Author’s initials
Date
Time
Figure 14. Sample query record, based on those of the Catalan Language Service, University of Barcelona
terminology in practice: terminography
129
records contain the query data, allow for indication of the data from the search and contain the solution proposed to the end-user. Terminology query centres usually also have guidelines for using query records that ensure homogeneous treatment and harmonious presentation of the data from queries. The machine-readable query records of a centre constitute a separate database that can be used to respond to subsequent queries about the same topic.
3.–Working methods Lexicographical and practical work that is more or less terminological, has given rise to the development of certain working methods which are often indiscriminately designated ‘‘terminological,’’ although many of them do not merit the description in that not even the most elementary conditions of terminology are fulfilled. Many of these methods originated in LGP lexicography—in which sphere they often rightfully belong—or were developed by translators seeking to improve the tools of their trade. The major defect of these methods lies either in the deficient theoretical foundation, or in working conditions, which make it impossible to apply the basic principles of terminology.—Picht & Draskau (1985)
Terminography is an ongoing process which can be separated into several stages. Before presenting these stages, and the problems that accompany each stage, it is first necessary to classify the various types of terminological search possible. There are two general criteria for characterizing terminological searches, or searches in general: the number of languages involved, and whether the search is systematic or not. By the first criterion, searches can be either monolingual or multilingual. By the second, they are either systematic or ad-hoc.8 Systematic searches cover the terms of an entire special subject field or a subpart thereof. Ad-hoc searches are restricted to a single term or a small set of terms belonging to a subsection of a subject field, or to a group of terms belonging to different fields. There are four types of terminological searches: a. systematic monolingual searches b. systematic multilingual searches c. ad-hoc monolingual searches d. ad-hoc multilingual searches 3.1–Systematic searches Our working methods are based on the well-established principle in terminology that since the vocabularies of special subjects are collections of interrelated, terminological,
130
terminology
structured units, it is necessary to use them by considering them as such. Our approach takes this consideration into account and is equally systematic since we always go from terms to concepts, and not vice versa. It is also oriented towards the terminological work carried out with two languages that we consider equally valid for naming technical and scientific realities, even though each one of these languages reflects a particular division of reality. Acknowledging that there is terminological allomorphy between two different languages (such as French and English) clearly shows the need for separate lexicographic descriptions of their terminological universes. It is not until the final stage of the terminological work is reached that the two descriptions will be brought together, when equivalents between terms and concepts will be established. Whenever the target language has a gap in designations, it will be filled in by means of neology. These gaps reveal differences between English and French in terms of how they divide up reality. The most important advantage of this systematic option is that it avoids qualitative gaps in the terminologies of the languages in question.—Auger & Rousseau (1987)
3.1.1–Systematic monolingual searches Terminological work done on a large set of terms belonging to the same subject field in a single language has at least the following six stages, as shown in Figure 15. a. Defining and delimiting what is to be done b. Preparing for the task c. Writing the terminology d. Presenting the project e. Supervising the project f. Resolving problematic cases 3.1.1.1–Defining and delimiting a search Before starting a terminological search, the research topic, the user of the work, the purpose it is supposed to fulfil and the amount of data to be gathered have to be determined. Authors must first define the topic they want to address; they often do not pay sufficient attention to the conceptual and terminological scope it might have, nor to the diversity of scopes a single subject field can have. As a result of this lack of definition, they often have to re-orient their work halfway through by either limiting its scope or redefining the scope of analysis, both of which involve additional costs. Secondly, all terminological work defines the end-user. While the set of terms belonging to a special subject field are the same, no matter what terminological work is to be done, the exact choice of what has to be done, the principal corpus chosen as the source of extraction, the way the data about each term are presented,
terminology in practice: terminography Stage 1
Definition and delimitation of task topic addressees purposes size
Stage 2
Preparation of search acquisition of information choice of consultants choice of information selecting an extraction corpus structuring the field proposing the work schedule
Stage 3
Preparation of the terminology extraction extraction record terminological record
Stage 4
Presentation of work
Stage 5
Revision of work
Stage 6
Treatment and resolution of problematic cases
Figure 15. Stages followed in a systematic monolingual search.
131
132
terminology
and even the type of data gathered all vary with the characteristics of each situation and each group of end-users. A non-expert public requires an user-friendly end product. A highly specialized public, on the other hand, does not need basic information. Thirdly, not all terminological work has the same purpose. We must distinguish between descriptive work, which is limited to gathering the terms in a special subject field, and work with a prescriptive purpose, whose aim is to influence use by specially indicating terms whose use is recommended (labeled as recommended, standard or standardized). Within prescriptive work, we can further differentiate between work that is strictly prescriptive about the form of terms, and work that directly prescribes the concept (which then, albeit indirectly, affects designations). Assigning labels to a term in order to guide usage is based on decisions made by authoritative bodies. Decisions about the form of a term, which in general come from language organizations, are primarily addressed to society as a whole and place standardization of a language above other objectives. Decisions about concepts (and indirectly about terms), which tend to be of scientific, business or professional interest and which are addressed to a specific audience, usually come from technical organizations. Since their main priorities are to facilitate and ensure the exchange of ideas, services and products among experts, their most important goal is the concept, although they may indirectly prescribe designations. Finally, all terminological work must consider the scope of the task being undertaken, not only because of the above factors, but also because of more pragmatic factors such as the timetable for finishing the task, the number of people involved, the material and financial resources, the terminological needs of the subject, etc. 3.1.1.2–Preparing for the task The second stage of terminological searches consists of bringing together the information about a topic and the available documentation. This preparatory stage includes the following activities: a. Acquiring the necessary information about all aspects of the task and about the material, professional, conceptual, and scientific framework in which the topic is located b. Choosing the consultants for the project c. Establishing the corpus d. Drawing up the conceptual structure of the field e. Proposing an outline for the project
terminology in practice: terminography
133
a. Acquiring information about the topic and its context Subject field experts or linguists who want to work on systematic terminological searches must know the content matter of the topic, the available documentation, the professional context in which the topic exists and the sociolinguistic situation of the domain in question. The reference bibliography is the best way for a terminologist who is not an expert in a technical or scientific topic or in a very specific activity to obtain a general idea of the subject field. If the search is done directly by a subject field expert, this first step is, of course, simpler, since experts can be assumed to have an adequate understanding of their own special subject. Their training allows them to structure the subject field, choose the best material for extraction, and detect terminological segments of the chosen texts better, more easily and more reliably than a terminologist with a linguistics background. If linguists have to carry out this first step, then they must have acquired a minimum of information about the field and must be able to count on the constant advice of subject field experts who will be able to verify if the linguist’s idea is appropriate and who can resolve the occasional problems that inevitably arise during a search. Subject field experts can give advice as to the best documentation available, not just for acquiring information but also for extraction. Thorough knowledge of the content matter of the field and its conceptual structure leads to the structuring of the conceptual field of a specific domain or activity. This conceptual system can be represented by a diagram (usually, although not necessarily, a network representation) which allows easy visualization of the complex system of conceptual relations. Terminologists must also have information on the documentation available about the topic, not only from the standpoint of terminological research that might already have been done in this or in a related field, in the language they are working in or in other languages, but also from that of more representative works that might be sources of information about the field or that can be used as the most appropriate documentation for extracting terms. To find out about this documentation, terminologists with a background in linguistics will again need to rely on experts in the field and on experts in documentation. In addition to the information about the topic and the documentation, authors of systematic terminological work must have pragmatic data about the organizational and linguistic reality of the professional medium in which the terminology in question will be used. Finally, the prestige of a language in a professional context or the simultaneous presence of several languages also determines the strategy used for approaching a special subject field. The most prestigious language in a special
134
terminology
subject field is usually the one supplying the majority of terms in the field. Terms are created in a language because the ideas, processes or objects that these terms represent have been created or invented in the society using that language. New ideas and new products are naturally associated with their original designations and when they become a part of a different language system the designations tend to maintain their original form. When terms are regularly borrowed from another language, it is usually only by an interventionist language policy that the linguistic identity of the receiving language can be preserved. Nevertheless, each professional context is a particular case with respect to the countries and languages from which it borrows, the linguistic attitudes of its leaders, the degree to which its users will resist change, etc. Consequently, good knowledge of all these factors must guide the proposals terminologists make for new designations. At the end of the second stage of information acquisition, the terminologists can reconsider the initial objectives to see if they are appropriate, and then reconfirm, change, or better define them in accordance with the knowledge they have acquired about the topic and the context. They now prepare a first draft of the work schedule, which is the final point of this second stage in the process. b. Choosing consultants for the project What the authors have learned about the professional context in the previous substage allows them to choose the people that will act as consultants. If the initiative is led by a subject field expert, one or more terminologists should constantly advise; if, on the other hand, it is led by a terminologist, then there should be one or more subject field experts acting as consultants. c. Selection of information and determining the corpus Two types of information are needed: extraction documentation and reference documentation. The extraction documentation is the corpus from which terms will be selected. It must meet a series of conditions, without which it would be difficult to ensure the reliability of the results: • it must be pertinent, i.e., representative, of the field being analyzed, and, if possible, written by a highly regarded author • it must be complete, and as such include all aspects of the terminological task to be performed • it must be up-to-date, so that the list of terms obtained will be useful • it must be original, i.e., written in the language in which the terminological work is being carried out.9
terminology in practice: terminography
135
The reference documentation provides the members of a team with the information about various aspects of the subject field and the research project. d. Conceptual structuring of the field Once all the information has been obtained, terminologists, if necessary with the aid of a subject field expert, can draw up the concept structure of the field and represent it graphically. A concept system is made up of a structured set of concepts organized into concept classes. The major concept classes and sub-classes, as well as concepts of the same class, are related on the basis of the characteristics they share or by their use in reality. Specialized thesauri (or the part on a specific topic in a general thesaurus) and subject field classifications are very useful models for structuring a field. It is not easy to give general rules for structuring a special subject field, since there are many factors that can affect this operation (the subject matter, the degree of consolidation of the field, tradition, etc.); but nevertheless all appropriate structuring must have the following characteristics: • It must cover the entire field being researched (and, if possible, must link it to the immediately superior field, so that the precise field of research is seen as a part of the more general field). • It must contain all the structured branches of the sector which is to be covered by the terminology. • It must present the relations between branches and those inside each branch. • It must not include irrelevant concepts. • It must be structured so that a single concept is not found on different branches. Structures are usually represented in tree diagrams. The more detailed and rigorous the conceptual structure is, the easier will be the subsequent work on terms and other aspects related to the conceptual side of terminological work, including: • • • • • • • •
delimiting the topic and its scope controlling the extraction of terms selecting terms attributing terms to a subsection of a domain writing definitions controlling key descriptors incorporating terms into a database establishing new concept frames for related subjects.
136
terminology
e. Proposing a work schedule If necessary, researchers can now revise the original schedule and draft a definitive schedule of work, establishing: • • • • • • • • • •
the title of the work the delimitation of the research field the objectives set the users the information about the terminological and bibliographical situation of the subject field the information about how the professional context is organized the information about the language situation of the context the structuring of the subject field being researched establishing the corpus proposals for the organization of the work itself: – who will do it: authors and consultants: names and work assignments – working methodology: general stages, specific operations in each stage – schedule – resources necessary for carrying out the job: budget estimate, materials, infrastructure, etc.
3.1.1.3–Preparing the terminology The third stage consists of preparing the list of terms and providing the information about them that is appropriate for the type of work being undertaken. The following operations are involved: extraction, writing the extraction record, and writing the terminological record. A.–Extraction Terminological extraction (sometimes called excerption) consists of selecting those elements out of the corpus that are considered terms of the special subject field which is the object of the terminology. First, the linguistic segments in the corpus that correspond to a concept in the special subject field have to be recognized and delimited. Experts in a particular subject recognize these terminological segments representing concepts in their discipline much more easily than non-experts. They can be guided by the tree structure prepared in the second stage. The second step is to select those terms that should be included in the terminology, given the topic and objectives. Not all terms occurring in the corpus need to
terminology in practice: terminography
137
appear in the terminology. Some terms do not belong to the topic at issue but rather to another field; in other cases, they belong to the general language. The decision whether a term is pertinent is not straightforward because the project’s objectives and users are factors that can alter the choice of the units appearing in the terminology. The recognition of phrasal structures offer particular difficulties, e.g. ‘head of household’, intangible asset’. In order to form new terms, subject field experts turn to the lexical resources available for the formation of transparent units, namely phrases.10 Though the transparency of most phrases makes them readily acceptable, they cause other problems; they are difficult to recognize formally, and to delimit since there are no apparent differences between a free combination and a bound structure.11 A number of tests have been proposed to determine the limits of terms and phrases which, applied together, prove quite helpful. Indicators of termhood are: • a phrase is lexically organized around a single base (random access memory, central processing unit, communication adaptor unit) • other linguistic elements cannot be inserted into the terminological phrase (*head of household but not head of the household) • none of the parts of the phrase can be modified individually (*power of attorney but not power of many attorneys) • the term can be replaced by a synonym (sonogram [sonograph, echogram, ultrasonogram]) • an antonym exists in the same special subject field (even parity vs. odd parity, serial port vs. parallel port) • the frequency with which the same terminological phrase occurs in texts of a particular special field • the phrase is a single lexemic unit in other languages (Spanish ‘tipo de letras’ = font, Spanish ‘portaaviones’ = aircraft carrier) • the meaning of the expression as a whole cannot be deduced from the separate meanings of its parts, e.g. a ‘slide rule’ is not a rule; foxglove, • the presence of certain linguistic units inside a phrase indicates that the phrase is most likely a freely combined string (The performance of this jazz or soul singer). B.–Writing the extraction record Once found and delimited, the selected terms are included in an extraction record with the following details: • the canonical form of the term or phrase
138
terminology
• the context, i.e. the part of the text in which the terminological unit functions grammatically • the reference of the document in which the term is found: its formal and typological characteristics and its contents • the grammatical category and subcategories (gender, conjugation class, etc.) that are evident from the form the term has in the context in which it appears • other, more varied and irregular information, such as an equivalent form, an illustration, a complementary definition, etc. • the information on an extraction record is usually complemented with information about the management of the extraction process: author, data, position • miscellaneous data Extraction records usually contain the term as it occurred in the corpus but in canonical form. This form is kept even when it is not considered to be the most appropriate term. The written form should only be changed in cases in which there has been a spelling mistake or a grammatical error in the text (of which, obviously, one must make a note), or when the written form is determined by the placement of the term in the text (such as a capital letter after a period, use within quotation marks, italicized forms, etc.). Terms are written in lowercase letters, except in the case of those containing proper nouns, which begin with a capital letter. Extraction records can be the source of some significant doubts for terminologists. Two of the most frequent problems are the criteria by which the contexts have been chosen and the number of records that must be filled out for each term. We distinguish between several types of context for terms: a. testimonial contexts, which simply illustrate that a term occurs in a text; no other information is provided . . . since the terminals have to be equipped with a printer, and if we do not want to have to type the information ourselves, they also need to be equipped with a scanner. b. defining contexts, which provide information about the meaning of the term This technique (telephotocopy or facsimile) allows transmission of a copy of a document (facsimile) from an issuing centre to a receiving centre by means of a telecommunications network. c. metalinguistic contexts, which provide information about the term as a unit within a formal system . . . laundering used to refer only to clothes, but in some parts of the coast it refers to illegal financial activities.
terminology in practice: terminography
139
The best types of context in terminology are defining contexts.12 There is no standard practice as far as the number of records that must be prepared for each term. However, it would seem that two records are sufficient to demonstrate the existence of a terminological unit. More than two are only justified if the new contexts provide additional, pertinent information, either formal (special usages) or conceptual (clarifications of meaning). C.–Writing the terminological record Terminological records are based on the information in the various extraction records made for a single term. A terminological record is a structured guide that allows us to assign information about a term in an ordered fashion. Machine-readable records can be consulted in many more ways than manual records, and they make it much easier to process and reorganize data during subsequent stages of work. There are several models for terminological records, depending on the characteristics of the specific task. Some basic details are normally found in all: • • • • • •
the entry the reference of the term the grammatical category of the term the subject area to which the term belongs the definition or context or both equivalents in other languages, if equivalents are going to be included in this terminological job • cross-references • management data for the record: the record’s author, the date • miscellaneous comments • status code or label of the term or record, if one has been proposed. All of this information should be presented according to a set of standardized conventions, without which it would be quite difficult to later retrieve the information and exchange it. Below we analyze some of the data that are usually found in a terminological record and indicate the most frequent types of representation for each class of information. The document containing the instructions the terminologist must follow to fill out a record is the protocol for using the record. If systematic terminological work is to be harmonious and consistent, these instructions must be followed.
140
terminology
i. The entry The entry of a terminological record is the term. It is presented according to the general conventions for lexicography: • nouns in the singular. In languages with grammatical gender, nouns with more than one gender are usually represented as if they were adjectives in the same entry (Spanish profesor -a, ‘teacher’) • adjectives in the singular masculine form (in languages with overt gender marking on adjectives), followed by the feminine ending if one exists, as in most general language dictionaries, e.g. Catalan sintètic -a (‘synthetic’) French ondulant -e (‘ondulating’) • verbs are represented by the infinitive, e.g. solder, harden, format When the entry involves more than one word, the order is always that which occurs in natural language, e.g. Pythagorean theorem and not theorem, Pythagorean positron emission tomography tomography, positron emission ii. The reference of the term The document from which a term has been extracted is its reference source. Sources are usually coded following a regular system of referencing that allows users to retrieve all the information from a document at any time and locate a term at any point in the retrieval process. The representation of sources must be simple, easy to memorize and short. Coded documents constitute a source record in which each document is completely specified and catalogued. iii. The grammatical category The grammatical category of terms is shown on the record by previously set codes, which are those usually used in lexicography (n, pl, v, vt, vi, adj, adv) iv. Subject field Each term is assigned to a subject field, i.e. • the general field to which it belongs • the specific subfield to which it belongs, within the larger general field • the specific sub-subfield within the subfield, if necessary. The subject field on the terminological record must be assigned according to a previously established classification. If possible, it should follow an authoritative classification appropriate for the structure of the field.
terminology in practice: terminography
141
v. Definition A definition is usually a complex sentence that is the semantic equivalent of the term being defined. It must follow the conventions governing definitions and the specific features of the domain. (See Chapter 3, 2.) Definitions must respect the concept structure previously developed and particular attention should be given to: • the parameters describing each branch of the conceptual structure • the order in which the parameters are presented in the definitions • the level at which the definitions are presented, depending on the users and the purpose of the project. Definitions have to follow certain conventions of presentation: • they must be written in lowercase letters, except for the first letter of the first word, which is the most general descriptor; • they must form a complete sentence; • they must be written in a formal impersonal style. For examples, see Chapter 3, Section 1 and Chapter 2. A definition is considered adequate if it is semantically equivalent to the term it defines. It is useful to indicate the source from which the definition has been taken. If the definition comes from a field expert either orally or in writing, the name of the specialist will be the source. If a terminologist is the author of the definition, the abbreviated name (or some other identification code must be entered as the source of the definition.13 vi. Context The contexts of a term is taken from the extraction corpus. The contexts that best describe the concept and are most representative of its use in the text are chosen. In terminological records including definitions, usually one or at most two contexts are enough to illustrate how a term is used, except in rare cases of multiple usage. The context must be recorded without alteration so as not to affect the meaning of the term illustrated. Ellipses should be indicated in the conventional manner. The source of the context must also be indicated (in coded form). vii. Equivalents in other languages Dictionaries, encyclopedias and other terminological collections are used to find equivalent designations in other languages. These sources, like all others, must be coded and fully referenced. If the terminologist is not a subject field expert nor fully competent in the other languages, it is preferable to rely on dictionaries containing definitions or illustra-
terminology
142
tions rather than on vocabularies that only present a list of equivalent words. It is also preferable to use multilingual texts that have been standardized by an international body. If the record layout does not foresee separate boxes for each language, the equivalent designations are preceded by a symbol indicating the language.14 viii. Cross-references Cross-references indicate reciprocal relationships; one term refers to another to which it is related for some reason, and vice versa. Cross-references can be: • informative: a term refers to another in order to broaden the information about its designation or concept, or to specify its relationships with other forms and concepts in the same field • prescriptive: a term refers to another in order to favour its use, to deprecate a designation, or to indicate the existence of alternatives on the same sociolinguistic level a. Informative cross-references Informative cross-references can be of various types, depending on the reason for the cross-reference: • Synonyms, i.e. a term refers to another that designates the same concept. Synonymy can exist between: – a standard form and any formal, orthographic, or phonological variant; – an abbreviation and its full form; – a term and its shortened form; – a term and its scientific name; – a term and the symbol representing it. The type of synonym being cross-referenced must be explicitly indicated, e.g.: mammaplasty ISO tel. macro Juglans regia ®
VAR INI ABBREV SHORT SCI SYMBOL
mammoplasty International Standardization Organization telephone macroinstruction English walnut tree Registered trademark
Geographic dialect forms are a special type of synonym. As opposed to what happens with synonyms from two historical dialects, in which the current designation replaces the archaic name, synonymous terms belonging to two geographic dialects are used at the same time in their respective territories. In some subject fields, dialectal synonymy must be respected in certain functional situations.
terminology in practice: terminography goosefish T
DF
Any of several anglerfishes of the genus Lophius.
SYN monkfish
DIAM p. 122
MAC 19/12/90
monkfish X
R
SYN goosefish
DIAM p. 215
Figure 16
RMV 23/02/90
143
terminology
144
Synonyms can be classified as main or complementary in order to indicate that one form is preferred to another. In practice, the two terms are either synonyms on the same level or one of the two terms is the main or favoured term whose usage is to be encouraged, and the other form is secondary and its usage is to be avoided. Specific codes like the following are used in these cases: SYN synonym on the same level MN SYN main synonym SC SYN or COMP SYN secondary or complementary form sonography SYN ultrasound erythrocyte SC SYN red blood cell Terms linked in this way should indicate this fact on both records. • Antonyms, e.g. myopia • Superordinate
ANT
hypermetria
In the same conceptual field, two terms can be cross-referenced to indicate what the superordinate and subordinate concepts are, as in: standard cost
SUB
cost accounting
b. Prescriptive cross-references Cross-references to alternative terms can be qualified so that the nature of the reference becomes clear, e.g. incide sulphur
OBSOLETE DEPRECATED FORM
for make an incision for sulfur
In the rare cases of homonymy, cross-references, in addition to referring to the corresponding term, must clearly specify to which of the two records they are referring. To solve this problem, all records should be numbered so that in cases of homonymy the number of the record can also be referred to. Examples: 152 telephotocopy see fax 71 [fax 71: message] 153 telephotocopier see fax [fax 72: machine] ix. Author and date of record The coded identification of the person or team that has written a terminological
terminology in practice: terminography
145
record must be indicated. Usually there is a complementary file which explains the codes used. The date when the record was prepared permits controlling the progress of the search. x. Notes Additional information which cannot be attributed to any other datafield can be recorded here. Notes can also contain encyclopedic information about terms. xi. Status label Some descriptive projects include all the data collected by terminologists from the corpus, regardless of reliability. Projects designed to provide information about the quality of terms use a code that is known as a status label or reliability code.15 For standardized terms, ISO recommend using the names of standardizing agency as status label. The status label can apply to the entry term, the whole record, or only some of the information on the record. If it does not affect the entire record, the scope must be expressly indicated. There are a number of widely used labels, e.g. • • • •
standard or standardized term deprecated term neologism pending approval term pending further research, etc.
For computer processing each record must be divided into separately labelled fields. Each block of information constitutes a field characterized by: • • • • • • •
identification coded label full form of label definition representation nature of the field (open/closed; recursive/non- recursive) length, where appropriate In closed fields, the number of markers is listed with their code, e.g. Field Code Name Definition
020 GC Grammatical category Functional class listed on the extraction record
146
terminology
Use Obligatory, not repetitive Representation Code of lowercase letters representing the shortening of the corresponding label noun n plural noun n pl adjective adj adverb adv transitive verb vt intransitive verb vi reflexive verb vr phrase phr Computer-based processing is vastly superior to manual processing of terminology because it facilitates the control of terms and cross-references, changes, and reorganization of information, ordering of entries and preparation of indices, etc. It also permits much better retrieval facilities. For these reasons, it is preferable to work directly on computer and create machine-readable terminological records. 3.1.1.4–Delivering the work The fourth stage of a systematic search consists of preparing the text of the terminology for the user. The definitive version of a text is normally the result of several prior documents (each of which is itself the product of stages of work). The final product should follow a series of principles and internationally established conventions of presentation. It is important to distinguish the four types of documentation produced in the course of the research; each type resulting from a stage in the process has certain information needed for the following stage, and this must be preserved so that it can be reviewed at a later point. The four types are: a. The extraction documentation, i.e. the set of extraction records b. The working documentation, i.e. the set of terminological records c. The draft final document, consisting of the terminology to which the authors and reviewers make adjustments d. The final document, which is the definitive text given to the user The extraction documentation contains all the information extracted from the corpus which will be reduced in successive stages of the work during the preparation of the terminological record. The working document consists of the set of terminological records that have been prepared on the basis of the information corpus and data from other sources,
terminology in practice: terminography
147
e.g. bibliographic references and consultations of lexicographic works and subject experts. The terminological record brings together for each pertinent term all the data considered necessary in accordance with the type of data collection desired. The minimum information it must contain is: the term—the grammatical category—the subject field—the source of the term—the definition with its source, or a defining context—cross-references to other terms—the author and date of the record. If the project is multilingual, the equivalents of the term in other languages are also obligatory, and they must be referenced with their corresponding sources. If the project is prescriptive in nature, the status label of the term or its information items must be included. The draft final document contains an initial listing of the terms prepared based on the terminological records. This draft is analyzed by field experts and supervising terminologists in order to detect any possible gaps in terms, errors in content, inappropriate forms or other methodological and organizational aspects that have yet to be resolved. This text, which then serves as the basis for writing the definitive version of the terminology, includes an additional section on the problematic cases that have not been solved, so that they can be considered by an appropriate authority. Troublesome cases can be the object of a standardization file in which the various alternatives for designating the concept in question are presented. The structure of the draft final document should be the same as that of the definitive text of the final document. The final document should be presented according to the recommendations made by international bodies and general lexicographic practice. Some terminology agencies produce their own criteria for presentation that are quite useful when preparing the text for publication (Tackels 1990, termcat 1990a). The final document should contain sections with the following information: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.
Title and authors Table of Contents Preface (optional) Introduction and justification of the subject studied Instructions for use Conceptual organization of the subject field Glossary of terms presented in dictionary entry form Indexes of terms by language Other indexes (optional) References
These ten types of information can be organized in different ways and with slightly different structures, but they must all be present.
148
terminology
1. Title and authors The name of the work and its authors are usually on the cover and repeated on the title page. The author’s (or authors’) name(s) (an individual, group, institution) is not only needed for the sake of intellectual property rights, but also serves to lend the project a certain value. The names of the experts who have participated in the work also allows users to have an idea about the degree of acceptation the new proposals might have. Even though there is no standardized international classification for the names of terminological works, we can distinguish several types of works according to their characteristics. The name lexicon is usually reserved for collections of terms without definitions and with equivalents in one or several languages (usually one language). Works including definitions (and usually equivalents also) are usually called vocabulary or dictionary depending on the diversity of information included in the terminological articles and the scope of the subject field. The more exhaustive the area covered, the more specialized the target users, the more specialized the subject field, and the more specialized information items that are regularly included, the more likely the title is to be dictionary as opposed to vocabulary (Cabré 1989b). 2. Table of contents The table of contents should contain the names of all sections and indicate the page numbers. 3. Preface Some documents have a brief statement written by an independent expert on the merit of the work and the author(s). 4. Introduction and justification of the subject studied The introduction explains the motivation for the work, comments on its scope, defines the objectives and the end-users and generally includes the justification for choosing the topic. In addition, if the author is an institution or team, then the introduction includes a reference to the people who have taken part in the research. 5. Instructions for use A terminological work, like all lexicographic works, usually includes a section on the methodology that has been followed and on how to use the collection. There is usually a list of symbols used in the work. 6. Conceptual organization of the subject field Systematic terminologies usually include a presentation of the conceptual organization of the subject field analyzed. This information is usually presented in a chart or diagram, and provides information on how the concepts are structured and on the
terminology in practice: terminography
149
conceptual scope of the project. If the terminology is organized systematically, the order of the data collected should follow the semantic network representation. 7. Glossary of terms The terms are presented as dictionary entries, each of which includes two parts: the entry and the information that accompanies each entry. Entry terms are presented in their canonical form, as on the corresponding terminological record, and may or may not be preceded by the number that identifies them in the indexes at the back of the work. The entries can be listed either alphabetically, or systematically according to the structure of the concepts. The alphabetical order has two possibilities: continuous alphabetical ordering (alphabetical order that does not count the spaces or other nonalphabetical signs), or discontinuous alphabetical ordering (which does take the spaces and other signs into account, counting them as separators). If the terms are ordered by subject, the conceptual organization determines the order. The terms are ordered in sub-blocks, if they exist, and within the same block and sub-block, alphabetically. Terminological information consists of selected data about a term, depending on the type of work (lexicon, vocabulary, or dictionary), the number of languages included, the purpose, and the target users. The information should be presented systematically, following a prescribed order that varies according to the data considered most important to the particular project. The languages in multilingual vocabularies are usually ordered following the recommendations of ISO standard 639, according to which the languages belonging to the same family are listed first, followed by the other languages organized by family and in alphabetical order. Each type of information can be preceded by an identification code. 8. Indexes of terms by language Terminological works containing designations in more than one language should include, after the body of articles, an index of terms for each language with the corresponding equivalent in all the other languages. The indexes should be in discontinuous alphabetical order. The index is cross-referenced to the corresponding article. To avoid possible problems with homonymy, the identification number of the corresponding entry is also listed. 9. Other indexes Work that systematically contains illustrations, abbreviations or symbols for terms, or preferred and deprecated terms, must have selective indexes for this types of information. All entries in these indexes use identification numbers to refer back to the entry.
150
terminology
10. References The list of references used usually differentiates between types of document, e.g. lexicographical works, terminological works and special subject field documents; or: reference and consultation documentation and extraction documentation. It is important to design the layout of the bibliography following generally accepted practices. 11. Other information Finally, if the authors of a prescriptive work want to stress the fact that they were unable to decide on some terms, these terms will be marked as unresolved or as pending further study. 3.1.1.5–Supervising the job The supervision of a terminological project is the responsibility of the terminologists working for centres specializing in terminology (or other, authorized bodies) and recognized experts. Special subject experts review all aspects relating to the subject field, which include: the extraction corpus used, the conceptual structuring of the field of work, the list of terms included, the validity of the definitions and equivalents, cross-references for synonyms, and the viability of proposals for neologisms. They can also detect gaps of terms in the subject field, and the lack of relevance of terms included in the work. The supervisors trained in terminological methodology, who usually have a background in linguistics, review all the data referring to the overall structuring of the project, the application of methodological principles, and the presentation of data, which should follow official and international recommendations. At the end of the analysis, a supervisory report is issued that is addressed to the work’s authors so that they can make the appropriate corrections. This process also allows supervisors the opportunity to point out the doubtful points that need further consultations. 3.1.1.6–Dealing with problems Even after a terminology has been written, not all queries are necessarily resolved. There may remain questions about the delimitation or representation of a concept, even after consulting the references and subject field experts. The designation of a concept may be uncertain or no suitable equivalents in other languages, may have been found. There may be designations for which there is no clear concept, or there may be a concept for which there is no designation in the language being worked on. It may also be the case that only a non-standard designation has been found, or that for a single concept several alternative, non-standard designations have been
terminology in practice: terminography
151
found. Finally, there may exist a standard designation that is not actually in use while, at the same time, several, non-standard designations may be in use. Each case has to be resolved differently: To resolve concept-related problems, other reference works or subject field experts must be consulted; ultimately it is the experts who know their own field best. To resolve problems of designating concepts, multilingual, specialized lexicographic works, specialized databases, and subject experts who are qualified to deal with multilingual texts must be consulted. To deal with cases of designation, official standardization bodies can be consulted, as can field experts and representative endusers. The working text that forms the basis of the draft final document should include those aspects of the project for which the authors request a decision. This section may also include the terms for which the authors present one or more suggested designations. Each of these cases can be presented as a standardization file. A standardization file is a document consisting of a set of proposals for designating a single concept, with assessment of each proposal’s chances for acceptance (Boutin-Quesnel et al. 1990). Standardization files generally consist of a succinct presentation of the problem, the available information such as definition, usage notes, equivalents, alternative proposals with their supporting documentation and a proposal for a particular course of action. It is advisable to present a standardization file for cases in which there is a conflict between the term used and the one that seems most appropriate. Standardization files make the study of difficult cases by special commissions and standardization bodies easier, and make the entire decision-making process about new designations much smoother. 3.1.2–Systematic multilingual searches Terminological research is multilingual if all the steps explained above for a monolingual search are followed for each language involved. In this way there is an extraction record and a terminological record for each language. The starting point for multilingual searches is multiple, whereas the end point is unitary. The various monolingual records are combined in order to turn them into a single, multilingual record. Correspondence records can be used to bring all this information together. Once the respective terminological record has been established for each language, the terms corresponding to a single concept in the various languages can be related to one another. The definition or illustration from the concept structure established are the basic elements for verifying the correspondences. An initial multilingual list of equivalent terms results from this culling of terms for a single concept. It is useful for this list to include the definition of the concept in one language,
152
terminology
preferably the one that is considered the most important for this particular project. Gaps in lists must be filled by consulting other sources, broadening the scope of the corpus, if necessary, and by looking up terms in specialized dictionaries. In addition, experts can be consulted on specific issues, if those preparing the terminology are not fully versed in the field themselves. 3.2–Ad-hoc searches Ad-hoc terminological searches, carried out in the empirical sense, have existed for centuries. In fact, everybody—any author, writer, translator or any other language user—has at some point had to make the effort to verify what the meaning of a word was, find the term to express a particular concept or try and find the equivalent for a foreign term in his/ her own language. In doing this, that user has, in a way, carried out an ad-hoc search. Célestin, Godbout & Vachon-L’Hereux (1984)
By ad-hoc search, as opposed to systematic search, we mean work on an isolated term or a limited set of terms in a single special subject. This approach to work is usually the result of a query that a user addresses to a terminological service. Classifying terminological work as ad-hoc or systematic fundamentally depends on two main criteria, one of which relates to the number of terms covered, and the other to the reason why the search is being undertaken. Ad-hoc terminological searches concentrate on any issue relating to a single specialized term. Work on small groups of terms (usually no more than 60) that are homogeneous in the sense that all the terms belong to a single subject domain is also considered ad-hoc, although in this case it is also called a subject mini-search or ad-hoc subject search. A subject mini-search does not follow the tracks of a single term or answer a query about a single concept works on a set of related concepts that belong to the same subject field. This type of work falls somewhere between systematic and adhoc research: subject mini-searches use the same methodology as systematic searches, but are similar to ad-hoc searches regarding the range and scope of the object of work as well as the way the process is usually organized. While in systematic terminological searches terminologists collect terms designating specific concepts in a subject domain or subdomain in a structured way, adhoc searches may address a specific problem or a terminological doubt of a user, or may complete the terminology of a certain special subject subarea. Users who make queries are usually translators, technical writers, professors, etc. who have to give an exact name to a concept in a specialized text, and the word required is not to be found in lexicographic or terminological reference works. Ad-hoc searches usually involve three stages: the query, the search, and the response:
terminology in practice: terminography
USER
153
QUERY SEARCH
TERMINOLOGIST RESPONSE It may also include some other steps, depending on the particular case. Users of a consulting service are faced with a terminological problem they do not know how to solve, and the lexicographic and terminological resources they have consulted provide no help. In this situation users formulate the question as a terminological query and take it to a centre or specialized terminologist, who interprets the exact nature of the query and, if necessary, works with the user to reformulate it. The terminologist then checks to see that the term does not exist in a dictionary or database, and if not, whether it is a query that has been made and answered previously, in which case it would be present in restricted-access files. Finally, if it is definitely a new case that cannot be dealt with immediately because of its complexity, the terminologist does a search and communicates the result to the user. Queries made by end-users are usually assigned to query records that then become a part of the database. They can then be the source for later queries on the same problem. Query records usually have three separate parts containing the data collected from the query, the data collected during the search, and the data given in the response. 3.2.1–The query A user might formulate any one of the following questions, all of which would then be the object of searches: • What is the name in English for the coin-operated machine that lets people take their own blood pressure? • What is the name in Spanish for companies that partially or completely control interest in another company or other companies? • What do you call the equipment used by jai alai players to propel the ball in English? • How do you say hostile takeover in French? What is standard Spanish for ballpoint mouse? How should you say cash management in standard Catalan? • Is there a Spanish term for dumping? All of these cases have one thing in common: they deal with one concept and ask for the corresponding term. They can start from different points; the user can either
154
terminology
start from a conceptual characterization or from a term that exists in another language. Based on prior assumptions, users often ask questions that hide one or more questions, thus requiring a multiple response, which must be reformulated and usually narrowed down in scope in consultation with terminologists. Each of the following, as well as a combination of various elements, can constitute both the starting and end points for a query: • • • • • • • • • •
the name or entry the source of the term the grammatical category the subject area it is used in the meaning the context equivalents in other languages its relationship to other designations its status code; is the term standard or not its geographic scope, etc.
The starting point of a query (i.e. the information users know) can be: a formal designation, a variant, a synonym, an equivalent term, etc. It can also be a concept, expressed by a precise definition, by a semantic approximation, a context or an illustration. It can be a field of usage more or less determined by subject or semantics. A query can deal with one or more languages; in this latter case one of the most important issues in multilingual terminology arises: terminological equivalents. Finally, users can make queries in order to verify data they already know or to request data they do not know. The reason for engaging in a search is illustrated by the following questions: • • • • • • • • • • • •
What is the Catalan term for definition x? What is the Spanish term for the French term x? What is the standard term for the dialectal variant x? What does the Catalan term x mean? What does the Spanish/French/English/German term x mean? What does term x used in special subject field y mean? How is term x used in a text? What special subject field does term x come from? What special subject field is term x also used in? Has term x become standard/accepted? What organizations have accepted/use term x? Which one of the following designations is considered standard for naming
terminology in practice: terminography
155
concept x, or standard for naming what in language y is designated by z? • Is the term used to refer to a number a databases spelled data bank or databank? • Is the term data only used in the plural? This list of queries, which obviously is not exhaustive, can be organized according to the following four variables: a. The type of query (what is unknown and therefore being asked) b. The starting point of the query (what is known) c. The number of languages involved d. The purpose of the query (why the query is being made). 3.2.2–The search Since each type of query situation may require a specific work process, there is no single search strategy for all queries users might bring up but ad-hoc searches usually fall into three stages. The first two stages are concerned with preparation for the search and the search itself and the third involves synthesizing the results. A.–Preparation for the search The first stage of punctual searches consists of precisely delimiting the terminological problem behind the query, whether it is a question about a designation or about a concept, and placing it in the right subject domain. The data collected by the terminologist in a query are usually kept in a query record. In this process terminologists delimit the special subject field in which the term or concept is located, delimit the exact conceptual segment that, within this specialized field, deals with the problem, correlate the specific concept in the question with other related concepts, thoroughly characterize the concept that is the object of the query and formulate a precise definition for it. Narrowing down the object of the search and preparing the pertinent documentation involves analysing the data in the query and assessing the material that will be used during the search. i. Analysis of the query data After analysing the data contributed by the query itself (data which has been recorded on the query record), terminologists collect as much information as possible about the term. With these data in hand they proceed to identify and precisely characterize the consultation. Careful attention must be paid to the following aspects: • identification and delimitation of the term (clarify the form of the term that is the object of the query and determine what its limits are)
156
terminology
• the exact source of the term (oral or written) • the precise situation of the term in the subject classification (the subject field it belongs to and the subfield within this field) • clarification of the concept represented by a designation (by means of an illustration, a definition, or an explanation) • the exact linguistic context in which the term occurs • identification of the language the term comes from, if this is pertinent • identification of the grammatical category of the term • verification if it is a term as opposed to a lexicalized sequence in discourse • identification of the sociolinguistic characteristics of the term (linguistic level or register it belongs to) • identification of possible synonyms (other terms, initialisms, abbreviations, variants, etc.) • analysis of whether the term is considered standard in the source language, etc. ii. Assessment of the material to be used for the research There are three types of support used by terminologists engaged in punctual research: written documents, data banks, and consultations with experts. We distinguish the following types of written documents: lexicographic works and general encyclopedias; specialized dictionaries; thesauri and classifications; standards; formal specialized documentation; and non-conventional documentation. General and terminological data banks are one of the most important sources of information. Finally, experts and members of terminology and standardization committees are clearly most useful for resolving terminological problems, especially those dealing with new, specialized concepts. B.–The search In the second stage of ad-hoc searches, the data included by the end-users in the query are checked, and new information about the issue is collected in order to define the form of the term, its meaning, the subject field and subfield in which it is used, the grammatical category, synonyms, its status with respect to standardization, sociolinguistic characteristics, equivalents in other languages, the specific contexts in which the term in used. If it cannot be classified as definitive, the term will be tagged as a provisional proposal in the corresponding space. C.–Synthesis of results The third and final stage of ad-hoc research involves synthesizing the results obtained in the second stage and preparing the response to the query. Queries about
terminology in practice: terminography
157
terms that have already been treated in lexicographic works or resolved by standardization commissions receive an immediate response from a terminologist at a query service, but most queries must be researched, and at times the amount of work involved can be significant. Not all questions posed by users have a solution. Even when solutions exist, they are not always accurate or unique. Not finding a standard designation for a concept, forces terminologists to formulate proposals for designations that fill a lacuna in a special subject field, or replace a designation considered unsatisfactory, or favour one form over other competing alternatives. Consequently, if the response is provisional, this must be explained to end-users, so that they in turn can take the measures necessary if they need to use the term. The results of ad-hoc terminology are formulated as responses to end-users. The search is finished when the term in question becomes part of the set of terms belonging to a particular class. Terms considered definitive are then included in the free-access record of terms, and those for which a solution has been proposed become part of a provisional record which may be submitted to a standardization committee. 3.2.3–General process of ad-hoc searches All ad-hoc searches have common features (see Figure 17). If the information requested does not pose any problems, i.e. the term exists, the terminologist can analyze the case, consult the material and subject experts, if necessary, and respond immediately. However, if the status of the term or the information requested is not so clear and involves problems with standardization, nonexistence of a term, or hesitation among several terms, the terminologist may need more time. The steps of his research may then be as follows: • • • • • •
analyze the case consult the material consult subject experts, if necessary make a proposal provide a provisional response communicate the proposal to the pertinent bodies.
The results of ad-hoc terminology searches are usually presented in a report that includes the term, the subject field, an explanation of the problem that is the object of the query and a linguistic analysis of the term. The conclusion states the term that has been chosen, its proposed definition and the references supporting the solution.
terminology
158
Overview of an ad-hoc query User: A. Starting point: – locating a terminological problem: • finding a term • verification of a term • verification of aspects of the term:
– correctness – meaning
– usage – suitability, etc.
▼ B. Actions and strategies: – individual research: • consult general language dictionaries • review subject field files • consult specialized dictionaries • review material on the subject • consult terminology data banks • review materials on criteria for • query subject field experts creating and borrowing terms ▼ C. End point: – make a decision – places an ad-hoc query Figure 17. Stages followed in an ad-hoc search.
terminology in practice: terminography
159
CONSULTATION SERVICE Terminologist: A. Starting point: – receives a query – fills in data on a record – assesses the difficulty of the response – answers/gives a time period for responding ▼ B. Actions and strategies: – preparations: • analyze the data in the query • assess the material available to do the search
– the research • verify that data provided by the end-user • look for new information – in documentation – in data banks – from subject field experts ▼
C. End point: – synthesize results of the search – respond to user within specified time period Figure 17. Stages followed in an ad-hoc search.
Chapter 5 Computerized terminology
In a futuristic scenario, terminologists will have access to huge data (or knowledge) banks; from these banks, they will download items belonging to their corpora; they will go through these items without having to previously tackle the text manually; they will automatically establish their working terminologies, contrasting the incoming terms with the semantic descriptors that will later be used in the automatic writing of definitions; they will classify, choose, merge and edit data bases reducing their intervention to a minimum. Their work station (. . .), equipped with advanced and intelligent office tools, will allow them to control, by themselves and throughout the whole process, the elaboration of their product and carry it out under the best conditions.—Auger (1989a)
Although the development of computer applications for language processing in general and for terminology in particular has been truly spectacular, the situation has yet to reach that described by Auger above. The progress of computer science, particularly with the latest generations of personal computers and their penetration into the world of language, has brought about a change in the way many fields are viewed and in the tasks based on language processing. The ties between linguistics and computer science have evolved gradually, and have given rise to various applications that can be classified according to the degree of complexity of the computational treatment applied. We can identify several stages in the development of this relationship.ə In the first stage there were applications that were limited to using linguistic data as mere forms, without submitting them to any manipulation or analysis. Word processors (the most widespread of all microcomputer applications), spelling checkers, hyphenation programmes and communications programmes belong to this first stage. The second stage includes machine-based linguistic tools designed for users working in language and communication: database managers, electronic dictionaries, systems to aid writers, translators or terminologists (computer-assisted systems for translation, writing, correction or learning). In the third stage we find systems that manipulate data, either by analyzing it or by converting it into data with other characteristics. Included here are analyzers, lemmatizers, classifiers, programs for processing statistics, etc.
computerized terminology
161
Finally, in the last stage we find what are known as expert systems, which act with a certain amount of ‘‘intelligence’’ and to some degree attempt to replace human intervention. This stage includes automatic term identification and extraction, machine translation, systems for automated learning, automated indexing, text generation, etc. It is only in these last two stages that we enter the actual field of computational linguistics. The development of computer applications has yet to produce the multipurpose, flexible products that can meet all the needs of language users and researchers. Nor has it been able to create systems that are intelligent enough to satisfactorily replace human intervention and thus process unrestricted dialogue in natural language. Nonetheless, research in artificial intelligence on cognitive computers that actually possess knowledge about language and the world is making progress.
1.–The concept and scope of computerized terminology Computational linguistics has been developing strong interests in sublanguage studies in recent years, and one now finds lexical, syntactic and semantic descriptions of several sublanguages, for various text types and for various natural languages. Also, there are tools becoming increasingly available that help in the building of e.g. morphological analyzers for specific sublanguage and text type processing. Although there is no complete package of tools or systems available that will allow fully automatic detection of terms in running text, there are nevertheless many useful tools and systems that can be used to provide potential terminological units to the terminologist for verification, or which will identify potential defining phrases and explanatory contexts.—McNaught (1988)
The relationships between computer science and terminology is moving in two directions—computer science assists and changes terminological activities and its methodology, and terminology helps research in computational linguistics. It has become a key factor in knowledge modelling and is used to transfer knowledge and technology, since it is by means of terminology that structured thought is transmitted. Terminology is at the basis of knowledge, and as such it is necessary for its representation. Cognitive science and knowledge engineering need terminology in order to advance. Computer-aided terminology, which was characterized by Auger (1989a) as the ‘‘weighty component of modern terminology,’’ can be located between computational linguistics (computer science applied to the processing of language data) and the industrialization of derived products (the language industries). The creation and development of computer-aided terminology can be accounted for by a series of factors related to computer science as well as to the needs of society itself. On the one hand, computer-aided terminology has benefited from the progress and widespread use of text-based computer products using microcomputers. Today it is
162
terminology
impossible to conceive of a language or communication-related task being performed without computers, especially since the prices of these machines have become so low that they can be afforded by the smallest office. On the other hand, contemporary society, being dominated by the power of information and the need to communicate, places a great deal of importance on anything which facilitates communication. Terminology, an essential element for specialized communication, is thus increasingly important as a means of transferring thought and technology. With these new values in mind, developed societies have created information management organizations not only to facilitate communication, but also to deal with information in the most unobtrusive, standardized way possible in order to eliminate ambiguity and ensure flexibility. Terminology has benefited from advances in text-based computer science because it concentrates on the processing of written texts. As we have seen, terminology doubly depends on these texts: texts are used extensively because only a variety of specialized texts can yield the conceptual structures for special subject fields and the designations of the concepts therein; and texts are used intensively in order to extract the units that are the object of onomasiologically oriented research.
2.–Contribution of computer science to terminology There is no such thing as terminology without computer science. Computers have become essential for terminological activity. To some people, this statement sounds hackneyed; to others, it is a priority, a goal to be attained in the short term.—Otman (1989)
We identify two different types of possible influence of computer science on terminology, the introduction of computers in terminology has changed both the methodology of terminological work and the actual processing; and research in artificial intelligence which has permitted terminologists to design expert systems that can perform some terminological tasks. 2.1–Computer science and terminological methodology The effect of computer science on the methodology of terminology can be seen in the use of previously recorded electronic corpora and the exploitation of terminological, and knowledge databases. Although the onomasiological approach is necessary for researching designations, in practice, especially in bilingual and multilingual contexts, terminologists usually begin with lists of terms in a specific language which can now be produced by a computer and then look for their equivalents in one or more other languages. Thus, the principles on which terminology compilation is currently based have changed markedly with respect to those used
computerized terminology
163
in pre-automation days. Computer-aided text analysis and the possibility of processing large amounts of information have changed the bases of terminology compilation, as well as how the appropriateness of terms is conceived, and the degree of human intervention in the whole work process. The result is that the two disciplines that were seen by many as being different from one another—terminology and lexicography—have come closer together. Terminology compilation now also has a greater communality of principles and methods with lexicography than ever before; but both activities still lack a sound theoretical foundation.—Sager (1990)
As Sager so aptly states, terminology and lexicography are in theory two quite different disciplines, yet in practice they are headed towards a single methodology because of their common use of computer applications. Traditionally, a search for a single term that was not listed in a dictionary involved selecting a bibliography and consulting specialists in the field. Nowadays the fact that terminologists can access terminological databases and specialized text databases has markedly changed their working methods. Today systematic terminology is based on researching a representative corpus of texts. This corpus can be processed systematically and compared with other, previously recorded texts from the same subject field that have been quantified in terms of frequency of units and controlled in relation to periodic appearance of forms. And all these operations are carried out by computers or are computer-assisted. In addition, the selection of recorded texts does not depend on user requests, thus resulting in a more coherent, broader-based corpus that, in turn, benefits the search for terms. Analysis of large corpora of data independent from terminology research can contribute new ideas for research and even detect structures that we might be unable to discover in corpora limited to terminology alone. These contributions from computer science, then, imply a change in the way terminological methodology approaches texts. They also give an increasingly important role to the linguistic context of terms. The ability to access large databases has therefore brought about a change in the focus of terminological work. Terminologists consult data banks as a first source of information to find out if a term is listed or to extract supplementary data or information, before proceeding with any further terminological research. The change in methodology the Office de la Langue Française of Quebec proposes in this sense is representative of these new approaches: Two situations may occur when compiling a terminology: there may either be a lack of terminology, or there may be a preestablished terminology. At an initial stage, there is no terminology in either of the two languages used in the research. The terminologist must come up with two different terminologies, one for each language, by going through the
164
terminology
corpora. In the Office, to put it plainly, this practice has been abandoned due to its timeconsuming requirements.—Office de la Langue Française (1989)
The use of machine-readable corpora and large data banks offers terminological research significant advantages over traditional systems that were in use until quite recently. The huge amount of data terminologists have available allows them to obtain well-founded information about terms and to have much other information at hand. On the whole, this provides a more solid foundation to the decisions terminologists have to make throughout the process, and gives terminological activity increased flexibility and the possibility of responding better to user groups. Thanks to computer science subject specialists can now also intervene in the research process by contributing other viewpoints, giving their opinions on the terminological decisions and detecting gaps in terms. Automation reduces the methodological differences between terminological and lexicographical activities. In both fields information is now gathered from corpora of texts in machine-readable form, both accept the principle of a threefold separation of source material, database and separate user products, both can build up the content of their fields in any order they want, because the computer can take care of the ordering and in both areas is there a trend to reduce the human intervention and interpretation in favour of machine inferences on the basis of existing factual information.—Sager (1990)
Computer science also enables terminologists to access databases of images and to use images in word-processing programs. This has signified a major innovation in terminological methodology. For some types of concepts, particularly those referring to objects, images are the best mode of representation because they are easy to understand and allow users to access information starting from a general concept when other characteristics of the term in question are not known. Research on terms used for designating concepts in a special subject field, as we stated above, begins with identifying these concepts in texts. A concept, however, has no material presence except through the way it is symbolized, i.e. it can be represented graphically or by a linguistic expression, an abstract symbol, a definition, a paraphrase or an image. With computer graphics images can be integrated into the description of terms and the best mode of using images can become a new topic for research in terminology. 2.2–Computer science and terminological practice Computer applications have had an even greater impact on how work is carried out. In practically all steps of terminological work, both in specific queries as well as in regular, systematic work, computer science has provided resources and tools that facilitate the most repetitive tasks, thus making information searches more reliable
computerized terminology
165
and faster. Given the large amount of information computers have made available to them, terminologists can now be much more confident about the decisions they make about terms. Interactive computer programs have made it possible for specialists and end-users to help terminologists, thus increasing the quality and reliability of the results. The ability to access huge amounts of information stored in distant data banks also makes terminological work much easier, although it makes it more complex. As Sager states, terminology has gone from being an art to being a technique. There are five basic points in terminography at which computers can play a highly significant role for terminologists: a. Selecting documentation, prior to beginning work b. creating the corpus and isolating and extracting data c. writing the entry d. checking the information in the entry e. ordering the terminological entries. For selecting relevant documentation, terminologists now have access to various types of data banks with general, background information such as directories of research centres, existing databases, research that has been done or is being conducted about the subject, directories of researchers specializing in the subject and of the specialists available, etc.; document banks consisting of publications and dictionaries about the subject; text banks of corpora of technical texts; and, terminological banks with lists of terms about the subject and linguistic and interlinguistic information about each term. For the creation of the corpus, terminologists can select the machine-readable texts they believe to be most representative and incorporate them into a text data bank; if texts are not in electronic form, they can be put into machine-readable form with the aid of an optical scanner or can be entered manually. Once the text is machine-readable it can be analyzed by a semi- or fully automated extraction program to isolate and extract those units which presumably are terms.1 An automated or semi-automated indexing program for texts allows the terminologist to detect the presence of conceptual descriptors and to initially analyze the contents of each text. For drafting of term records, terminologists use computer files for writing terminological records by transferring some of the information from the reference text files, e.g. entry, source, and context or definition depending on the type of source, etc. The software used in the previous stage to index texts will be of great help in writing definitions. Once the record is complete, it can be edited in its entirety or just partially, or it can be combined with records from other sources. Terminologists can also control information about cross-references and equivalents by computer.
166
terminology
For checking and finalizing the entries, terminologists go back to databases and transfer any missing information to the record. For publishing, computers allow terminologists to present the information in various formats, e.g. on paper, disk, tape, file, optical disk, to include the data that are best suited to each case, e.g. context, equivalents in one or more languages, and to take advantage of a variety of layouts and different fonts, different order of information, etc. The way the final result is issued can also vary: there are various electronic forms and even for paper production there are advantages; for example, desktop publishing software makes it possible to obtain immediate paper output at reasonable cost. There are, nevertheless, still problems, of both a specific and a more general nature. The main weaknesses of the use of computers in terminological work are related to one or more of the following factors: a. lack of integrating computer resources in work methods b. lack of compatibility among the resources themselves c. the limited degree of computer processing available by each resource at present, as human intervention is constantly required d. the lack of a smooth interface between people and computers because of the difficulties associated with communicating via natural language e. the inadequacy of such hardware items as text scanners f. the limited number of existing corpora, especially in languages other than English Despite these shortcomings, research in computerized terminology continues and increasingly attracts the attention of those investing in and working in computational linguistics and artificial intelligence. 2.3–Artificial intelligence and terminology At present most computer applications for terminology (and in linguistics as well) are at best machine-assisted and still require a significant amount of human intervention. Despite these limitations, much work is being done on tools that are capable of acting and reasoning, so that they can increasingly replace terminologists in repetitive, time-consuming tasks, such as: • selecting the corpus for identifying and extracting information about a topic • selecting and extracting terms from a corpus • attributing terms to one or more subject fields, in accordance with their frequency in texts
computerized terminology
167
• preparing definitions from analysing the content of the text from which the terms come and their morphological structure, if they are complex forms • identifying the conceptual structure of an area • morphological analysis of terms in order to offer proposals for both formal and semantic neologisms Obviously, an expert system able to do any of these functions requires a great deal of ‘‘intelligence.’’ One of the fastest developing fields today in artificial intelligence is knowledge engineering—the intersection of the theory of knowledge and computer science. Here terminology and documentation play a very important role. The construction of ‘‘intelligent’’ computer systems is based on computers having some ability to reason based on knowledge and experience. Specialists in the theory of knowledge, psychology, and theoretical linguistics are attempting to describe how human beings acquire, structure and process knowledge in theory and how they learn through experience in order to build a model that can account for human thought (Parent 1989). Experts in artificial intelligence and computer programming propose to implement a computer model that acts intelligently, i.e. that is able to reproduce what is observed in reality, interpret it, communicate what it thinks, reason about situations with people or machines and make decisions based on the multiple variables suggested by each situation. To reach this point, specialists must first have a detailed description of how reality is organized and a model that can satisfactorily account for the acquisition and representation of knowledge and of how the mind processes knowledge. Terminology and the theory of knowledge are closely linked to one another by the notion of concept, which is the foundation of the theory of terms and the starting point for terminological research. The set of terms in a special subject field is at the same time a structure of concepts reflecting the organization of knowledge about the field in question. Terms thus become the key to the representation of specialized knowledge. An intelligent expert system, which must know these concept structures in order to process information and act accordingly, internalizes structured knowledge by means of terms. Each term is an item of specialized thought, and the relationships among terms in a single special field reflect the field’s concept organization. Artificial intelligence borrows from the theory of terminology a series of key theories for representing knowledge: the theory of concepts, understood as units of thought and as units of communication, the theory of concept representation, the theory of concept designation, also including non-linguistic symbols, which are increasingly important in the representation of concepts, the theory of objects, understood as ontological units and the theory of the structure and order of concepts, which relies on tools like thesauri.
168
terminology
Within the language industries, knowledge industries deal with designing and creating, usually multilingual, products that address the problems arising from the relationship between the representation of knowledge and terminology in order to facilitate the storage of specialized information and its subsequent retrieval. In other words, they are designed to make communication across languages and the transfer of knowledge and technology between countries possible. To date, however, expert systems with all these types of knowledge have not been attained, and only the future knows how far we can go: Let's dream a little . . . The translator sits in front of the work station: the keyboard, the screen divided into windows, the printer, the screen to view graphical and iconographical data, the scanner. The translator starts working by scanning a text (or by capturing it...) and he/she can then relax and find out about the news of the day or watch his/her favourite programme on one of the screens. Meanwhile, the computer chooses terminological units, phraseological units, units of textual articulation and even contextual units conveyed by the text. A sound indicates that the choice has been made and that the results are ready for the translator to look at, while a variable amount of terms is being paired and a variable amount of phraseological units is pre-paired with terminological units. Once the chosen items and those still not chosen are known, the translator prompts a documentary search on the terms to be retrieved by connecting into a complementary data base. The translator will then display the data on one of the screen windows, together with an illustration, animated image, film, etc. that may give more exhaustive information. Next, the relevant or potentially useful data are exported into the programme the translator works with. A search is sent via telematics to a centre of specialised documentation to solve the cases that remain unknown... Is it an utopian idea? Surely, but it is also an idea where humans go as far as the orders they can give go. Is it extremely costly? Maybe, but is it not the case that a lot of money is spent on the writing of documents on paper which may never be consulted? Is it a dream? Undoubtedly, if we let the illusion that the computer can think delude us and if we substitute human intelligence with the computer's mechanical algorithm.—Gouadec (1987)
3.–Terminology and data banks Although the analysis procedure will differ for human and machine translation, one result, in either case, will be a list of updates to the terminology files. Human-oriented and machine-oriented terminology files contain quite different information and a machinetranslation dictionary will contain common words not needed in a human-oriented terminology file, but ideally, they will be synchronized to contain records for the same technical terms. This will allow the human-oriented files to contain the documentation of how the translation was chosen and how the term is used by human experts, while the machineoriented files contain the syntactic and semantic codes needed for machine processing.—Melby (1989b)
computerized terminology
169
Data banks are by far the most widely used computer aids and have become primary reference tools both for research on the description of general and special languages, as well as on work applied to any branch of linguistics and the design of expert systems. The need to be able to access information in all fields of specialization and, in addition, have it continuously updated, has brought about the development of large information banks to meet the needs of end-users. Over the years, the design of data banks has changed due to the evolution of computer technology and software but also in response to our widespread use of computers in all fields. 3.1–The creation of a data bank A data bank consists of information organized into records, each of which is subdivided into data fields. Creating a data bank involves a systematic process that goes from designing the form of the bank desired to implementing it on a computer. This process cannot be improvised, but must follow a number of steps. The planning starts with a definition of the major features expected of the data bank. This stage includes the following operations: a. Identification of needs, usually by means of a needs survey consisting of: • identification of target users • delimitation of the needs of each user group • comparison and coordination of the needs identified • setting priorities; this operation is easy if the bank’s target users are homogeneous, and more complex if they are heterogeneous. b. Assessing the strength and weaknesses of the situation from which you are starting. This is particularly important when the project’s purpose is to change the model of bank. c. Identifying the obstacles that might arise throughout the process: i.e. time, human resources, budget, psychological attitudes towards the new project or project change. d. Defining the parameters that will be used in a feasibility study, carried out by a group of experts, and market research. e. Choosing the working group, which must be made up of specialists and representatives of future users, and drafting the work schedule—identification of tasks, assigning responsibilities, calendar, and resources required. The next step is a feasibility study which will permit the bank’s creator to determine what solution is the best for a particular situation. It must cover aims and functions of the bank, its hardware and software requirements, its size, organization,
170
terminology
location, major channels of dissemination as well as its profitability. Once the feasibility study has been approved the basic design of the data bank can begin. This operation covers the description of the overall structure of the bank and the processing of the data. There are two steps in this stage: training the team for the project, and analysing the general features that must be planned ahead: files, data conversion, system compatibility, size criteria, presentation of information to users, etc. Next, suitable computer hardware and software must be chosen. With respect to this point, we must stress the importance of adapting the software to the hardware. It seems more reasonable in this respect to first choose the software and then the hardware, and not the reverse, because it is the software that will determine if a database effectively serves the users it was designed for. The choice of software, however, is not an easy operation, but rather immediately brings to the forefront a fundamental choice that has to be made: whether to acquire a standard system or specially create a system. There are pros and cons to each option. Conceiving and developing one’s own software has the advantage that it will always be better adapted to the specific needs of the bank and the target users; however, this requires a significant amount of time for analysis and programming before it can be put into operation and involves a major economic investment. On the other hand, choosing commercially available software, which is more likely to be compatible with other systems, never really satisfies the specific needs of a particular project. Once the equipment is in place, the detailed design of the bank can be started. The characteristics of this bank will be known from the preliminary studies, i.e. how it will work, how information will be organized and processed, and how that data will be retrieved. It is at this stage that the following features must be decided upon: • Entry: type of information, sources for the data, entry system, organization of the entry, structure of the information, entry protocol, etc. • Storage: type of records, relationships among the records, structure of the records, protocols of representation and use of records, etc. • Retrieval: types of queries that are to be answered, ways of retrieval, formats for retrieved information, protocols for consultation, typology of users, etc. Before beginning a full-scale operation it is advisable to design a pilot project or prototype for the bank. At this point the bank is created as a pilot project, including: analysis and programming, the writing of manuals and protocols, entering of data, verification of data, preparation of training materials, etc. If the experience of the pilot project is positive, the data bank is finally created for future use.
computerized terminology
171
Updating a data bank is a continuous process, and is essential if the information bank is to be truly useful. It must be reviewed periodically and users should be surveyed for their opinion to determine their degree of satisfaction. This survey can be automated, as stated previously. In conclusion, a reasonable project for creating and making use of a data bank must be the result of planning. Its utility and effectiveness must be reviewed periodically. It must be based on rigorous, appropriate documentation, and it must be designed with the participation of representatives of real users, as in the end it is they who can best define what is needed. 3.2–Organization of the data Creating a data bank has to begin with the idea that all banks consist of several related databases—the main database, containing the essential information, and secondary databases, each of which has different but interrelated information. Each of these other databases has its own logic, and can be accessed independently. In a terminology data bank the main database contains the terms, and the secondary one the sources, authors, classification of thematic areas, etc. Once the overall structure and the number, type and organization of the various databases comprising it have been determined, the structure of the information in each database must be defined and organized into fields. The set of related fields in a single information unit constitutes a record, and each record is composed of a number of data fields. Each field is identified by a field name or tag and its contents must be described, i.e. the types of data it can contain, its length, if it is of fixed length, and the set of characteristics related to its nature as a closed or open field, the recursiveness within a single record, and its potential for indexing for consultations must also be described.2 Information can be accessed in the traditional way, via a query language, or in conversational mode, using menus the user chooses from. If a query language is used, the end-users need the help of an intermediary or must have prior training in how to query the bank. The menu-driven approach results in more fine-tuned queries relating to the quantity of data, but it must provide the user with a reference manual that can be accessed via the help programme. The constant appearance of menus on the screen can make this type of consultation rather tiresome. Unrestricted queries using natural language are still far from being satisfactory. Many data banks also have a data management system to report on how the bank has been used and whether it meets user needs. These systems can produce almost entirely automated reports on the status of data. Despite the progress in this field, however, the current status of data banks is still not well-defined. We might say that they are at the point between an initial stage,
172
terminology
in which they were designed and used for storing large amounts of information, and a latter stage that has just begun, in which the goal is for them to contain some degree of intelligence so that they can take a more active role in processing information needs and in adapting to user requests. At the same time we can see a move away from the single, isolated data bank towards a multi-bank environment. The idea is to simultaneously access several data banks that are all integrated into an overall working structure that not only includes the databases, but also the other computerized tools and resources. 3.3–Data banks of interest to terminology Work in terminology, as part of specialized communication, depends on bibliographic and document databases, text databases, and knowledge databases. Terminological databases are described in section 5 of this chapter. A.–Bibliographic and document databases A bibliographic or document database is of interest to terminologists because it contains information on special subject publications and documents. Access to the information in a document database from any of the fields in a record that is a part of the document’s description is easy because the characteristics are delimited and obtainable. In contrast, access to the information content is much more complex. Whereas the database’s designer has no problems in automating discrete items of information (author’s name, year of publication, type of book, language, etc.), the description of the document’s contents is quite another issue. Documents are usually described by subject. They are assigned subject codes from a pre-established, and usually hierarchical or nested, classification that is expressed by means of descriptors. This classification is accessed via a thesaurus that indicates the relationships between the various descriptors, thus making access to the documents somewhat easier. In addition to this classification, the description of a document’s contents also usually has a brief summary or, alternatively, a series of key words. Both the descriptors of a classification and the elements in a thesaurus are terms with form and meaning, in addition to being meta-scientific operational units. Information about documents is retrieved through terminology from the standpoint of the subject field. B.–Specialized text data banks A text database is a machine-readable collection of texts organized by a specific criterion. Each text constitutes a processing unit which can be described formally, semantically, and pragmatically. Data banks of texts from a single special field constitute, in the words of Corbeil (1989), a supertext.
computerized terminology
173
Text data banks today are one of the tools most used in terminology and constitute one of the most promising elements in terminology research. Terminologists use text banks to find the terms of a particular special subject field. Analysis of texts is the first step in determining the contents. Extraction of terms from texts can be computer-assisted. The machine can identify terminological units on the screen from which terminologists select the ones needed for setting up the conceptual structure of a subject domain. Finally, text databases allow terminologists to establish the definitive list of terms in a collection. This operation, which would be quite simple if all the terms were single units with only one written form, becomes more complicated because phrasal units are very common in terminology. Moreover, these phrases can be very varied in structure and length. The statistical analysis of the frequency of certain word combinations helps to identify the groups of forms that might constitute a terminological unit in a particular subject field (Auger 1988, Boulanger 1988a, Goffin 1979). C.–Knowledge bases Artificial intelligence applied to the creation of expert systems must be founded on a corpus of knowledge. Otherwise, these systems could not intelligently deal with situations nor provide the information asked of them. An expert system consists of two components: a database of specialized knowledge and a computerized system for processing information. A thematically specialized database for experts is a structured set of information that can be focused on a specific area or on several interrelated ones. In either case the thematic field of reference must be explicitly identified and well-defined. A database of specialized knowledge is a systematic set of information that represents the sum of knowledge a specialist has about a specific field. This information comes from subject field experts themselves, who explicitly define their knowledge about the field in question, and from the extraction of documentation on the topic that has a database of specialized texts. An abstract model is constructed with the information from these two sources. In theory, this model has the same knowledge that specialists in the topic have, and can act in their place. Terminology is at the root of knowledge representation because terms, as we have stated, represent concepts, which in turn represent classes of objects in reality. Terminology and knowledge bases are thus closely related to each other. 3.4–The evolution and limitations of data banks We can identify two major stages of development in computerized applications for processing specialized data, bearing in mind the main objective of research and professional activity in terminology and special languages. The first stage
174
terminology
covers approximately 1970–1985, and the second from 1985 to present. In the first stage, 1970–1985 according to Feigenbaum and McCorduck (1983), efforts and funding were concentrated on building up large masses of information. At this time the large text, lexical and terminological, bibliographical and document data banks were created. The main goal was to make up-to-date, specialized information available to professionals in order to make knowledge transfer feasible. This period, even with the constant increase in the amount of information, did not—as might have been advisable—pay special attention to the quality of the information in the banks, and ease of access to the data by end-users was even less important. As a result, access to the computer was based on formal languages which are quite different from natural language, and intervention of an intermediary or learning the systems for making queries became necessary. Moreover, each bank had its own system for storing and consulting information which was incompatible with other systems. The information from one could therefore not be transferred to another. There is a third problem with these data banks. Even though users have large amounts of information available, the fact that most banks are not highly specialized means that the data does not meet user needs. Because they were more concerned with the quantity of information, they paid less attention to its quality or depth. Consulting a data bank gives users a large amount of data that requires very careful selection. Nevertheless, there is no doubt that in this initial stage data banks were responsible for increased spread of information, and this directly contributed to the spread and availability of knowledge. Data banks, however, are still too centralized, too costly in terms of time, personnel, and budget, very slow, difficult to access and quite simply not easy to use. The large number of consultations made to banks during this stage reveals aspects that are still not satisfactorily handled, e.g. incompatibility of data, lack of coherence in the various banks’ purposes. At the same time, needs that had previously gone unnoticed became evident, e.g. the range of on-line simultaneous consultation, user involvement in updating the bank, the ability to obtain data that can be used directly in writing tasks, translation or interpretation, etc. The most salient characteristic of the second stage of data bank development is that banks come closer to addressing the variety of end-user needs. They do this by: • providing selective, high-quality information that meets users’ purposes • allowing simultaneous access to interrelated and complementary banks • making access to information as natural and uniform as possible • integrating access to information and use of automated systems in the same work station
computerized terminology
175
In this second stage of computer applications to special domains we can identify three ways in which research is aimed at improving the earlier shortcomings. Attempts are made to make the systems compatible with each other, to harmonize methods of gathering and describing data, to make the query systems compatible, to provide simultaneous access to several banks, to develop interchange systems, to increase the quality of information, to enable users to obtain more selective information, etc. There is research on improving user access and on making compiling and storing information easier; this involves natural language interfaces, voice-based entry systems, systems for optical character recognition, compilation of terminology from text bases, etc. In spite of being an excellent mental exercise, the use of available resources cannot ignore either the terminological data or the limits of present expert systems. This is so because, useful as they may be, these ‘‘computerized thinkers’’ do not have real intelligence: they are unable to learn or draw from their own experience, unable to act with common sense, and unable to deal with new situations for which they have not been programmed. In some ten or twenty years, research on artificial intelligence will produce expert systems with cognitive abilities that are more similar to those of human beings, such as the ability to learn and develop based on experience, to compare present and past experiences in order to establish significant relationships, to discover new information on their own and to come to their own conclusions, to explain the reasoning behind their ideas and identify the underlying thought processes.—Pavel (1987)
4.–Terminological data banks Lexical or terminological databases and the associated technology open up possibilities for research and applications. As increasing amounts of machine-readable text collections are available it is now practicable to base dictionaries and glossaries on text corpora rather than on the individual choice of the lexicographer or the terminologist. Compilation techniques can be assisted by question-and-answer systems, thus ensuring greater consistency. The interactive mode of dictionary consultation can also benefit from question-and-answer systems to guide a searcher through the full range of information available. The most exciting new developments are, however, likely to occur in the modes of representing terminological relationships. Systematically structured glossaries are not new but their realization on a larger scale has always been hampered by cost and the sheer complexity of the task. The representation of terminological fields and networks has advanced considerably in recent years. Specialist translators in particular are likely to welcome the possibility of diagonal searches of databases which is, of course, also the most reliable test of the quality of a terminological database.—Sager (1983)
Terminological data banks, along with document and text data banks, have played a very important role in resolving issues that fall between language and special
176
terminology
subject fields. Bilingual and multilingual data banks, which were first developed as translation tools, have provided and continue to provide information about the terms used in science and technology. They have also been basic elements in the general standardization processes of languages because they are used as references for ‘‘correct’’ forms. The ability to store large amounts of terms with all their respective information, to easily update information at a lower cost than with printed matter, to disseminate information on a large scale using the latest systems, have all made terminological data banks essential tools for language and subject specialists. The first systematic terminological data banks appeared at the beginning of the 1970s when changes in computer science allowed the design of a structured system for storing information (Rondeau 1983, Sager 1990, TermNet News). What was initially the most troublesome issue, the storage capacity of computers, yielded in a second stage to how to retrieve information, both in the sense of how much time a query required and how the consultations could be made. At this point another important change occurred: we went from document data management systems, which had been adapted for processing terminology, to designing systems expressly built for terms, which are better suited to the needs and objectives of the field. Finally, we have moved from the creation of large, centralized term banks that can only be consulted on mainframe computers to small data banks for very specific topics, often held on personal computers. The panorama of term banks, then, has changed substantially in twenty years. Two different trends have characterized the growth of terminological banks and databases. First, a struggle between different spheres of influence or a race to see who can build the largest bank. Second, we must admit that it was also the result of a legitimate interest to obtain profits from relatively complex, and thus costly, computer systems. Plus, the personal computer had not yet appeared . . .—Gouadec (1987)
4.1–Definition A terminological data bank is a structured collection of information about the units of meaning and designation of a special subject field addressed to the needs of a specific group of users. It usually consists of a main database, which has the terms, and a variable number of databases related to the main database and sometimes to each other. Each of these databases has information on some aspect of the terms. Traditionally terminology data banks were created to meet very specific needs. Most of the data banks that currently exist were designed to aid translation and usually contain terminological information from the lexicographic and terminological literature (lexicons, dictionaries, encyclopedias, vocabularies, glossaries, etc.). Their primary purpose is to facilitate translation by giving translators a one-stop, user-friendly tool for queries that includes several dictionaries and is capable of
computerized terminology
177
providing reliable suggestions. With data banks translators can have at their fingertips larger quantities of information in a single reference work. Translators, who still today constitute the largest group of data bank users, consult a term bank for the same reasons they consult a dictionary, and the specific queries they make are often of the same nature, e.g. the spelling of a term, its meaning, its grammatical category, the field in which it is used, its synonyms, pragmatic or standard usage, equivalents in other languages, etc. Sager (1990) defines modern terminological data bank design as a set of special language vocabularies with the following characteristics: • The information is stored on a computer. • They include nomenclatures, special terms and phrases, with the information necessary for their identification. • They can be used as monolingual, bilingual or multilingual dictionaries. • They offer on-line access. • They are the basis for dictionary production. • They are used to monitor the vitality of a language and the creation of terms. • They are ancillary tools for information and documentation. A data bank with these characteristics is not just a reference source for answering queries like those made in the initial stages, but rather a basic tool for editing dictionaries and transferring knowledge. From being single-purpose, electronic versions of dictionaries term banks have become a flexible, multipurpose tool for the various groups of end-users who turn to banks for the information best suited to their needs. Sager (1990) divides data banks into two major classes, first generation banks and second generation banks, based on the type of software they use. First generation banks, which started at the beginning of the 1970s, have changed little over the years. They have a relatively simple structure, and most were built with customdesigned data management software. Their data is organized lexically rather than conceptually. Access is restricted because of the difficulties involved in requesting information, and thus the bank is normally accessed via an intermediary. Since 1980, however, specialists in computational linguistics and computer science have worked on improving these banks with respect to the quality of the data and the ease of access to the information. Clearly, the ability to make on-line queries represents a major step forward. The inclusion of a data bank in an information distribution network makes it easier to disseminate data to the general public. Putting entire data banks on cd-rom places them at the disposal of centres that otherwise would not be able to access the information in the bank because of the costs involved in connection or a lack of the right resources.
178
terminology
The second generation of term banks has improved on the earlier designs from the standpoint of the software used to create them. Many small, highly specialized data banks have been developed separately from the large, official centres. The designers of these newer banks have not had to deal with modifying a previously established system, as was the case with the first large banks. As a result they have been able to design banks that are more flexible and better adjusted to meet user demands. 4.2–Classification A typology of terminology data banks that reasonably accounts for the multiple features of these banks cannot start from a single criterion of classification. The result is not a single but rather a multiple classification, depending on the parameters chosen in each case. Rondeau (1983) classifies terminological banks using the following six criteria (see Figure 18): 1. the objectives the bank is designed to meet 2. the nature of the data contained 3. the target users 4. the organization of data 5. the approach to language used 6. the ways the data contained are disseminated Some of the factors proposed by Rondeau, such as 3. and 6., do not really classify data banks but rather simply present some of the possibilities they offer. Other relevant criteria have been suggested by Sager (1990) and Felber (1985), and in accordance with their proposals, we distinguish the following types of term banks: 1. Banks defined by objectives: • informative banks, designed to disseminate terminology • prescriptive banks, designed to intervene in term usage 2. Banks defined by their entries (Felber 1985): • banks based on terms • banks based on concepts 3. Banks defined by subject matter: • banks containing information about several subject fields • banks on a single special subject 4. Banks defined by their size: • large banks, usually of administrative bodies • terminology minibanks, developed by a professional or a centre specializing in a subject field
computerized terminology
179
translation text production
circulation of CSTs 1. Objectives dissemination of terminologies
teaching of languages for special purposes sociological implementation of terminologies
2. Users
translators terminologists and terminological lexicographers technical writers specialists in scienti c and technical elds teachers of languages for special purposes the general public
3. Linguistic approach
prescriptive (standardized data) informative (non-standardized data with quality ratings) purely descriptive (all types of data, without quality ratings)
4. Nature of data
terminological textual translation-based grammatical encyclopedic lexical from general language visual (charts, illustrations)
morphological syntactic orthographic
document-based
bilingual records
5. Organization of data
term-based
immediate 6. Dissemination methods batch
languagespecic
multilingual records computer terminal remote printer CRT telephone (through an intermediary) immediate printout automatic typographic composition
Figure 18. Classification of terminological data banks according to Rondeau
180
terminology
5. Banks defined according to the main interest of the data they contain: • term banks (usually with definitions) • phrase banks (terms in context) • banks of terms in documents (specialized texts) • encyclopedic banks (terms with encyclopedic information) • visual banks (images with captions) 6. Banks defined according to the choice of contents in relation to their objectives (Rondeau 1985): • standard banks, containing only correct information • descriptive banks, containing all types of information • informative banks, containing all types of information but indicating their relationship to a standard 7. Banks defined by the number of languages in the terminological information:3 • monolingual banks • multilingual banks • monolingual banks with information about equivalents in other languages 8. Banks defined according to how the data is organized (Rondeau 1983): • banks organized by document • banks organized by terms without context 9. Banks defined according to the hardware used: • banks developed for mainframes • banks developed for microcomputers 10. Banks defined by system organization (Sager 1990): • first-generation banks • second-generation banks The various criteria in this classification are not mutually exclusive. They combine with one another to result in banks with a mixture of characteristics, the sort of banks that function everywhere today. The target users of terminological data banks are also varied in nature. Sager (1990) and Rondeau (1983) present the following typology, which clearly reflects this diversity: • • • • •
subject field experts, scientists, and technicians communication professionals experts in lexicography and terminology information and documentation professionals language planners
computerized terminology
181
• publishers, language teachers, applied linguistics researchers, etc. • the general public that consults dictionaries Each one of these groups has terminological needs it expects to be met by a data bank. Custom-building term banks to the specifications of restricted groups of target users, however, is only possible if we are talking about small banks. Otherwise, to get the most out of the substantial resources that have to be invested, banks are designed to be very flexible so that they are multi-purpose. 4.3–The design of a terminological data bank All terminological data banks, regardless of their purpose or size, have a main database containing terms. This term collection, like all ancillary databases, is made up of entries, each of which contains information about the term. The creation of a terminological bank must follow the same general steps for building data banks outlined in Section 4.1 of this chapter. For a term bank we can break the process down into three stages: compilation, storage and retrieval (see Rondeau 1983). 4.3.1–Compilation The first stage involves decisions about three fundamental aspects, in accordance with the bank’s feasibility and characteristics set forth in its overall design: 1. The source material. Compilation starts from texts and not terms. Decisions have to be made on: • the nature of the material: pertinent subject matter, type of texts that should be chosen, evaluation of texts, data that must be collected about each text • its location: where the material is and in what format • the method of extraction: how to go from texts to terms and to their initial information • the materials for selecting texts: records of text descriptions, protocols for using records, manuals on the criteria for assessing materials, documents describing the operations, etc. • the organization of the compilation process: staff, priorities, identifying and assigning tasks, etc. 2. The information that must be extracted from texts to create the extraction record (see Figures 19 and 20): • the description of the term, context, reference information for each unit and the representations of all the information on the corresponding extraction records and protocols for use
182
terminology
know-how m AV 11-10-90 SECON L’expansió sobre les bases preexistents significa que les activitats del negoci giren sempre entorn d’una activitat bàsica de la qual es denomina el *know-how i es posseeix un profund coneixement de la competència. . . CTS 14-11-91 kremlinòleg m DB 05-02-89 SILL Una distinció capaç de deixar perplexos fins i tot els més experts *kremlinòlegs. ESS 06-02-89 laboralista adj AV 09-1190 MÓn Així com les eleccions de Bootle, a Liverpool, van passar sense penar ni glòria, donada la gran majoria de votants *laboralistes en aquell districte, les eleccions de Bradford van centrar l’atenció política en el transcurs de la campanya. . . MAC 10-12-90 laisser faire v intr DB 12-03-89 ECON . . . Bush no hi va voler intervenir, de manera que va adoptar la mateixa posició de *laisser faire que Reagan amb la vaga de controladors aeris de 1981. PVL 20-06-89 Fitxes de buidatge automatitzades.
Figure 19. Sample automated extraction records (in Catalan, with loan words from English and French).
computerized terminology 951 CM Fr
De
En
952 CM Fr De
En
953 CM Fr
De
En
183
BTLWFT790083220 MG5 SIB fraise de forme; fraise de profil outil pour le fraisage de l’entre-dents dans le fraisage dent par dent. Le profil de l’outil correspond à l’entre-dents CIRP Voc Tech Production Mécanique 8: Taillage Formfraeser; Profilfraeser Werkzeug zum fraesen von Zahnluecken im Teilverfahren mit dem de Zahnluecke entsprechenden Profil CIRP Wb Fertigungstechnik 8: Verzahnen form cutter; profile cutter tool for cutting tooth gaps by indexing method which are identical to the shape of the cutter CIRP Dic Production Enginer ing 8:Gear manufacturing BTLWFT790083221 MG5 SIB fraise de forme d’ébauche; fraise d’ébauche CIRP Voc Tech Production Mécanique 8: Taillage Zahnformforfraeser; Zahnschruppfraeser Fraese zum Vorbearbeiten von Verzahnungen CIRP Wb Fertigungstechnik 8: Verzahnen form roughing cutter; gear roughing cutter; gashing cutter cutter for roughing gears CIRP Dic Production Enginer ing 8:Gear manufacturing BTLWFT790083222 MG5 SIB fraise de forme fraise de forme dont le profil correspond à l’entre-dents à tailler; fraise à disque ou fraise à queue CIRP Voc Tech Production Mécanique 8: Taillage ill.p.188 Zahnformfraeser Zahnformschlichtfraeser scheibenfoermiger oder fingerfoermiger Formfraeser mit dem Huellprofil der zu schneidenden Zahnluecke CIRP Wb Fertigungstechnik 8: Verzahnen Ill. S. 188 finishing gear milling cutter; form cutter; gear cutter form cutter having the shape of the generated tooth gap; disk-shaped or finger-shaped CIRP Dic Production Engineering 8:Gear manufacturing ill.p.188
Figure 20. Automated terminological records (EURODICAUTOM).
184
terminology
3. The information that must be included in the identification and extraction process to make up the term record: • the description of the information, sources for completing the extraction information, how it is represented, what criteria will be used to evaluate the quality of the information, work procedures, etc. The end result of this first stage is a term record that will be the basis for the data bank. When information on a main term record is codified, there must be a related record in which the code is explained and which gives information about the other codes in the same field. There is international agreement that certain information must appear on a terminological record: • • • • •
the entry term with its source data on the concept it names, usually a definition a reference number for the term in the database the special subject field it belongs to usage notes
In addition to this information, there can be, and usually are additional items of information such as: • • • • • •
the source for the term the context in the language equivalents in other languages status label relationships to other terms other grammatical information
Even though not all of this information is obligatory for most term banks, only the specificity of the bank being built can determine the necessity of certain data. 4.3.2–Storage Information extracted from texts is organized into database records containing all the information about a term. A record is a set of several fields, each of which contains a certain type of information. As in any database, the fields of a terminological database usually have a field name or tag that identifies them. Most terminological records consist of the following fields or blocks of fields: • the term, which is the headword • the information source, which is usually coded • conceptual data: definition, special subject field, other related concepts and the
computerized terminology
•
•
• •
185
type of relationship (see Chapter 3, 2), notes on the conceptual domain of the term, other domains in which the term is used, etc. linguistic data: grammatical category and subcategories, phonetic transcription, variants, abbreviated forms, synonyms and other related terms, collocational information, language and regional code, etc. pragmatic, normative and sociolinguistic data: contexts and collocational information, usage notes, indication on the correctness of the term, notes on field of usage, connotations, deprecated forms, etc. equivalents in other languages administrative data: author of the record, date recorded, internal notes, etc.
In order to provide information about the reliability of data, it is advisable to indicate the source in each field of the record if it is not the same as the source for the term. Sometimes information about the source of the term gives indirectly information about its reliability. A terminological data bank generally includes the following ancillary files (see Figure 21): • term file • source file • subject field file • grammatical category file • status label file • author file
SUBJECT FIELDS FILE
ADMINISTRATIVE DATA FILE
Figure 21
MAIN FILE
SOURCES FILE
AUTHOR FILE
186
terminology
These files contain all pertinent details, and are used to clarify questions about data from the main record and to further help users. A terminological data bank may also be linked to other, ancillary databases which can be accessed independently for other purposes: • Bibliographic and document databases, in which each document is described completely. • Subject databases, containing a set of records on the subject field, each of which contains the characteristics and indicators of the content of each subject and the relations existing among them. Thesauri could also fulfil this role of concept specification. • Text databases, containing the texts themselves with a physical, formal, semantic and pragmatic characterization of each text; these texts, each of which is a record in the text database, are the basis for research in computational linguistics. • Knowledge databases, in which the basic concept structures for each special subject area are specified. They might also include illustrations. The internal structure of a terminological data bank is determined by the logic behind the programming. The organization of a bank into files follows a logical arrangement that is set before data is entered. It responds both to the organization of the work process and to the filtering desired for the information. This internal structure can have several levels of perception depending on who is using the data. When contacting the data bank to make a query, users only see the data that the system has selected according to the elements the user has provided. External users never see all the information in the bank or its structure. 4.3.3–Retrieval Terminological databases must be designed to address whatever requirements users have. But retrieval of information from a bank is constrained by the designer who has developed the database and structured the information in it. Information retrieval is further constrained by the nature of the query systems. A user can either consult the database directly (direct query) or through an intermediary who acts as a bridge between the query and the answer expected. Although traditionally banks had intermediaries, direct queries are becoming more and more commonplace. To make a direct query the user must be able to communicate with the system without much prior training. Consequently, the information system is usually supplied with query menus, help manuals or with natural language interfaces allowing a less cumbersome dialogue between the person and the machine. A query can be made in a formal language—the most frequent procedure—or in
computerized terminology
187
natural language. If it is made in natural language, there is an interface connecting the user to the system. Users can make on-line queries to databases or they can use more traditional methods such as telephone, correspondence, fax, etc. The number of different types of query allowed by a database depends on the information it contains and the power of the system used to develop it. In general, all terminological databases accept a variety of queries: queries about a single item of data or about a set of data, queries based on a single criterion or a combination of criteria, queries starting from a specific data item or from approximate data. 1. A query can be made about a single item of data, usually starting from a designation in the language the database was written in or in one of the languages for which equivalents are given, for example: • grammatical category of term x • definition of term x • term corresponding to French (or some other language) x • status label for term x • source for term x • author of the record for term x • context in which term x is presented • date on which the record for term x was made • synonymous term for term x Other information about the same term that could be requested is: • definition, equivalent and context for term x • equivalents in other languages In this type of query the program normally goes straight to the record of the relevant term and chooses the information requested. This is usually very quick. Queries can also refer to a set of terms, e.g.: • list of terms in quantum physics (all of which possess code x) • list of terms with status label x • list of terms with no equivalent in a specific language • list of terms from source x In these cases the system gives information from several different records. It analyses the databases and chooses those records that comply with the requested
188
terminology
condition. The result is a list of terms that may or may not include the information asked for. 2. Queries can also be made with a combination of criteria. In this case the search conditions for information, consisting of the characteristics a record must possess in order to be chosen, must be specified. This is done by indicating the relationship that unites them, expressed in Boolean or other types of relations. Relationships such as and, and not, or, greater than, between, less than, etc. can be used, as can be seen in the following sample queries: • list of terms in subject area x that are not verbs 1st condition: belongs to subject area x 2nd condition: is a verb relationship:and no • list of terms that are nouns and come from source x 1st condition: is a noun 2nd condition: comes from source x relationship:and • list of terms written by author x or by author y 1st condition: is written by author x 2nd condition: is written by author y relationship:or With the right programming a database can also be used to provide information on linguistic structures. In this case a sample query might be: • list of terms in subject area x that end with character y This would provide information about suffixes of terms in a specific subject area. Likewise, a query like: • list of terms in subject area x containing the string ‘‘by’’ between spaces would provide a set of terminological phrases in a certain subject area. 3.
Finally, queries can also be made when the exact term or group of terms are not known. This is particularly frequent when a designation is not known but, on the contrary, information about the meaning of a term is known. In these cases thematic or visual dictionaries allow us to arrive at the concept more easily. But, given the ease with which they obtain information, databases can provide the person making the query with successive lists of information that bring him or her progressively closer to the term in question. Let us look at an example.
computerized terminology
189
Someone is looking for a term in a specific subject area that names the handle of a tool used for a specific purpose. To arrive at this term the following steps can be followed—even though here we use simple queries, the successive stages can be joined together in complex consultations: • terms in subject area x (list 1) • of the units from the first selection, those which are nouns (list 2) • of the units from the previous selection, those which contain the terms ‘‘instrument’’ or ‘‘tool’’ or ‘‘device’’ (list 3) • from the previous selection, the units in which the purpose is indicated in the definition (list 4) • from the previous selection, the units in which the segment ‘‘handle’’ is indicated in the definition (list 5) Lists of information that progressively come closer to what is being searched for can be examined in this way in order to find the answer. According to the diversity of possible queries there are an equally wide range of answers possible. When queried, the system makes the pertinent search, chooses the answer that seems right and offers it to the user either directly or through a previous stage which permits that users themselves determine, whether the answer provided corresponds to the query. This second option is what the system does when it is not sure whether the user needs the information requested because of the scope of the answer. It therefore gives users the opportunity to choose again, but this time with a more restricted range. This usually occurs when the query affects more than one record. Let us look at an example: 1st query:
List of term records from source x
1st answer: There are 535 records in the database that meet this specification. With this information the system offers the user various options: – display the 535 records in summary form – display the complete records screen by screen – print them or transfer them to another format (indicate which one) – make another selection (if so, indicate what criterion will be used). Retrieval systems offer the results of a query in a variety of formats: • • • • •
on the computer screen on microfiche on a file for electronic transfer in a printout on a photo-composition disk
terminology
190
• on a microcomputer disk • on magnetic tape • on cd-rom (when a file of considerable size must be consulted but not manipulated). The layout for information does not have to coincide with the way it is represented in the database. A flexible system allows users to select data and represent them according to: • • • • •
the order of terms the selection of information about a single term the order of information about a single term the layout and typeface used the presentation of the material
With these possibilities, and desktop publishing systems, texts can be prepared from the database itself. It is even possible to generate texts so that they can go directly into publication by means of photolithography. Finally, a good database, in addition to information about data management, should have programs which identify the services provided to users and provide information about the elements that need improvement. It is therefore advisable to have a control over which queries obtained answers and which did not, and to provide its managers with periodical reports on this issue. Systems can also include an electronic mailbox allowing users to express their opinions on their degree of satisfaction and on the additional information that might improve the quality of the term bank. 4.4–Limitations and problems The initial idea of terminological data banks as large warehouses of centralized information that could be constantly updated and consulted over and over again has proved to be beset by several persistent problems. On the one hand, centralizing information has not turned out to be as beneficial as it first seemed. At the time when the large term banks were developed, computers were not powerful enough to process the very large amounts of data they had to contain in a short period of time, although they were able to store the data in a classified structure. On the other hand, the continuous updating of such large term banks, the data of which areconstantly changing, is beyond thescope of researchprograms and requires resources that are very often unavailable. The lack of specialists makes this updating even more troublesome. Often gaps are noticed and filled when the issues have already been dealt with directly by users who are not professional terminologists.
computerized terminology
191
The size of these data banks also requires that priority attention must be allocated to work and resources, with the subsequent relegation of other features such as attention to users. Queries to data banks have often been answered in too uniform a fashion, paying little attention to the specialization of users and their prior level of knowledge. Although the retrieval systems of term banks have evolved very rapidly in the past few years, the services they offer are still not entirely satisfactory. Query languages increasingly resemble natural language and query systems are becoming more user-friendly, as menus have been introduced to guide the query and improve the choice of information. Finally, the widespread dissemination of the data promised by computer systems is substantially reduced in practice because the professionals who are the potential users often lack the right equipment. Nonetheless, the development of systems allowing direct remote access by modem has increased the number of data bank users greatly, and the availability of complete terminology banks on CD-ROM (as in the case of termium) has also helped to foster use of existing banks. Even though the advantages of having large, automated banks of information in terminology are numerous, there is one aspect that to date has been unsatisfactory. In theory, term banks should assist the transfer of knowledge and lead to cooperation among countries with varying degrees of development. In practice, however, large databases have been developed without consideration for the compatibility of the system with those developed by others. Large information centres therefore cannot exchange data with other centres and much work has therefore been unnecessarily duplicated. The incompatibility of the large data banks is not only due to computers. The way data is classified substantially adds to the problem. Currently each large term bank uses a different thematic and document classification, which, according to them, is adapted to the specific needs of the context and its end-users. Theoretically an interface should be able to solve the communication problems between systems, and even permit communication between mainframes and microcomputers in order to transfer data from one system to another despite the different structures. But in practice little money is spent on this issue because it is not considered a priority, and sometimes this sort of information transfer is not considered to be cost-effective. Currently several relatively large data banks are incompatible with one another and with the ever-growing number of small databases that usually use standard systems that permit information transfer to take place. 4.5–The future of terminological data banks The experience of the past twenty years shows us that the idea of a single term bank is not feasible because of language and cultural diversity (Fortin 1988). The end-
192
terminology
users who must be satisfied, have quite different needs, depending on their own work environment. The overall computerization of modern society and the widespread use of microcomputers have both contributed to opening up access to data banks and removing some of the barriers between people and machines. But data banks are still not widely used and use of terminological data banks in particular is quite limited. The largest data banks have only fulfilled their passive role of being large information stores and reference sources for queries. But the fact is they could, and should, be used for many other purposes in current research, such as artificial intelligence. There are, however, three areas in which research is currently being carried out. They can be summarized as follows: 1. The relationship between users and computers: more rational systems for entering data, e.g. optical scanning and voice recognition, and construction of natural language interfaces. 2. Complete text processing: computer-assisted and automatic writing, correction, translation and publishing. 3. Creation of expert systems to automatically or semi-automatically effect certain terminological tasks such as identification and extraction of terms, preparation of definitions, construction of conceptual systems, etc. The choice is between leaving behind the use of ‘‘stored data’’ with all its limitations, or addressing the development of systems of terminological and linguistic information integrated into the various stages of terminological work. These stages go from the initial research to the transfer of information, which is adapted to different groups of users. This choice, however, requires revision of the long-term strategy to develop terminological databases and also of the consequences that those resources have on theoretical research as far as the activities related to artificial intelligence, natural language processing and basic research on terminology are concerned.—Fortin (1988)
There remain many fundamental issues that require research and development. The possibilities of data exchange and transfer from one system to another must be enhanced. Both the hardware of current banks and the software the banks have developed are so different from one another in their overall conception that information exchange is not only difficult but also costly. Users should be given better access to the data. Current research in computer applications in terminology should concentrate on resolving the practical issues of improving information retrieval, such as building simpler, faster systems that are user-friendly, providing more concise information that is easily transportable and available in various formats for use with different
computerized terminology
193
systems. It should be possible to offer responses from several databases simultaneously in a window-based system. The users of data banks will change. The growing integration of these systems in telecommunications networks and the development of telematics (mainly of videotext) will allow for the introduction of a larger, less specialised, number of users. Undoubtedly, translators will continue to be the main users of data banks, but the appearance of new users will influence the nature and the evolution of these banks.—Andersen et al. (1987)
Since information science is continually making new possibilities available, users should be increasingly demanding. Dialogue between people and computers, the possibility of gaining access to large data banks without problems and being able to have small databases for personal, restricted use will undoubtedly change work in terminology.
Chapter 6 Terminology and standardization
Progress in science, technology and economy is heavily dependent on communication of information. This communication of information, however, is strongly impeded by difficulties which arise because of ambiguous terminology. Unambiguous communication is only possible if the concepts—the elements of thinking—have the same meaning for all who participate in the communication process at the national or international level.—Felber (1984b)
There is general agreement that special communication demands a higher level of precision than that required in general communication. Communication without ambiguity would require each designation to correspond to a single concept and each concept could only be designated by a single term. This is clearly not the case for general language, in which words are usually polysemous and meanings can be expressed by several alternatives that are synonymous to one another. The constant and rapid progress of science and technology has brought about a spectacular increase in the number of concepts that have had to be named, either because of new creations or adaptations of something already in existence. As a result the terminology in several subject fields has often experienced an uncontrolled increase in the number of designations. Organized terminological standardization is a way to combat the diversity of names and thus ensure communicative precision among specialists. Ambiguous terminology based on polysemy, synonymy and homonymy obviously presents obstacles to communication among specialists and inevitably frustrates efforts to order thought. For this reason as early as the 19th century, scientists and, at the beginning of the 20th century, technicians, felt it was necessary to regularize terminology in their respective areas and thus became directly involved in the standardization process.
1.–General standardization If we take etymology as starting point, the standard language eliminates to the extent
terminology and standardization
195
possible the ‘‘higher’’ and ‘‘lower’’ varieties of language or the language practises that the descriptive or prescriptive linguists works with.—Guespin & Laroussi (1989)
There is some disagreement around the world as to the use of normalization (e.g., German Normung, French normalisation, Spanish normalización, Catalan normalització) or standardization to refer to the concept of ‘‘setting a form up as a model or type.’’ Some argue that normalization is the best way to express the action of reducing several concurrent possibilities to a single norm. Proponents of this view claim that words based on normal do not have the authoritarian connotation associated with words based on standard. For others standardization is the best term available to refer to decisions issued by authoritative bodies and does not lead to the unnecessary polysemyof normalization. Normalizationseems toimply two ratherdifferent meanings: one referring to extending the use of a language, e.g. a language in a minority position returning to a ‘‘normal’’ situation, and the other referring to determining which form is the most suitable for a particular concept. In English standardization has both meanings and is the accepted term for referring to both the correction of a sociolinguistic situation and the choosing of a specific term as a reference form. 1.1–Basic features The International Organization for Standardization (ISO), which was created in 1947 to develop universal standards to improve communication and cooperation among nations and to reduce the barriers to commercial exchange on the international level, defines the term standardization as follows: The process of formulating and applying rules for an orderly approach to a specific activity for the benefit and with the co-operation of all concerned, and in particular for the promotion of optimum overall economy taking due account of functional conditions and safety requirements.
The end result of standardization is a consensus decision made by a representative commission. It is presented in a document called a standard in which the aspects that have been agreed upon are presented and the contexts in which it is to be applied are specified. Standardization is used to reduce the several different varieties of a product to a single one, in order to allow the exchange of this product with the knowledge that it complies with the same conditions of quality, measurement, operation, compatibility, etc. as others. Standardizing terms that designate products means that specialists using terms do so in accordance with the agreements that they themselves have made, i.e. they will use one and only one designation for a well delimited concept. The earliest activities related to standardization began in the 17th century, when some companies established rules for production due to the trade brought on by
196
terminology
incipient industrialization (Manu 1984). The standards were first internal to the company and a way of optimizing production; later they were extended beyond the firm to facilitate business relations. These rules were the clearest precedents of what later would become standards. And very early on people realized that isolated rules about products were inoperative unless there were also rules governing the language used. Industrial standardization, then, was closely tied to terminological standardization. The industrial revolution of the 19th century, which is the definitive force behind international cooperation, explicitly considered the need to unify some systems of production, measures, weights and other characteristics, including terminology, which could be exchanged. Standardization, which above all is directed at eliminating unnecessary differences in order to facilitate exchange, is mainly applied to products, production methods and treatment processes, or product manufacturing in industry and public services. The following can thus be the object of standardization: • • • •
products and processes, which are primarily but not exclusively industrial units of measurement safety of people and goods terminology and the representation of symbols
Within these parameters standardization is based on a series of principles that must be followed (Rondeau 1983): a. It is an act of simplification that reduces variety and increases uniformity by choosing one alternative over others.1 b. It is a group activity that must be achieved via consensus, not by imposition. c. It is an activity whose only value lies in its application; the publication of a standard is meaningless unless the standard is applied. d. It is an act that can be revised, but it should be stable enough so that those who have to comply with it take it into account; decisions must be the result of detailed, reasoned arguments. e. It is a social activity with economic repercussions. Today there are two very marked attitudes present in the world in relation to standardization: that of Western European countries and North America, and that of the countries that until recently formed the Eastern Bloc, hereinafter Eastern Europe. In the opinion of the former, standardization is more like the expression of a voluntary decision rather than a prescriptive imposition. Therefore, the decisions always have to be based on the consensus of the representative bodies, and before a standard
terminology and standardization
197
is considered definitive it must be used in specialized contexts for a period of time in order to verify its suitability. In this view standards are usually directed at bodies, rather than at individuals, and they usually take the form of recommendations. In the case of noncompliance, sanctions are not imposed because the bodies themselves have decided to standardize a product or process for operational purposes. In Eastern European countries, in contrast, standards were until recently considered enforceable. Their requirements had to be complied with, and noncompliance was penalized. This philosophy can be explained by the language situation characterizing these countries: their multilingualism makes teaching scientific and technical subjects and communication in special subject fields very difficult. 1.2–Standardization bodies At present standardization takes place on three different levels: international, national, and regional. In international standardization, there are two wide-ranging international bodies that for historical reasons wield the greatest authority: the IEC and the ISO. The International Electrotechnical Commission (IEC), headquartered in Geneva, was founded in St. Louis, Missouri in 1904. In 1906 it began work on unifying methods and developing guidelines for regulating electrical engineering production internationally. Today the IEC is the sole international body of reference for electrical engineering and electronics. It has 73 national committees and 111 subcommittees that cover the entire field of electrical and electronic engineering and other related fields such as nuclear powerplants, electronic data processing, electroacoustics, etc. It has 43 member countries and to date has published more than 1,500 international standards. The IEC has had a committee for terminology (Committee No. 1) since 1910. In 1938 it published the first edition of a dictionary of electrical engineering in 50 parts, each of which covered a specific subfield. This first edition included 2000 terms defined in English and French, with equivalents in German, Spanish, Italian and Esperanto. In 1970 the IEC published a second edition of this dictionary in 24 parts in English and French containing 8,100 concepts and equivalents in six languages. Russian became the third official language of the IEC in 1960. The dictionary currently has 34 volumes separated into 321 parts and has 10,000 entries and 11,000 terms in 9 languages.2 In 1926 a group of national standardization bodies created after World War I founded an International Federation of National Standardization Associations (ISA). This was the direct predecessor of the International Organization for Standardization (ISO). The ISO was created in 1946 to replace the ISA, which had ceased operations due to World War II. The purpose of the new body, headquartered in Geneva,
198
terminology
is to facilitate international coordination and unification of industrial standards, as stated in the original by-laws, which were signed by the first 25 member nations. Today the ISO has some 170 national committees—one committee per country— which act as full members or corresponding members. The ISO’s range of activities is clearly differentiated from that of the IEC. The ISO is responsible for all technical specializations except electrical engineering and electronics, which are the exclusive domain of the IEC. An agreement between the two bodies establishes the respective domains of each body and the distribution that has been set in all fields of standardization on the international level. The first international committee on terminological standardization was created at the ISA in 1936 by request of the Soviet Union, which had adopted the proposals made in Wüster’s work. This committee (ISA 37) operated in four fields from 1936 to 1939, namely rules for naming concepts, the international unification of terms, monolingual and multilingual vocabularies. World War II also ended the ISA’s terminological activities, which were not taken up again until the founding of the ISO in 1946. Wüster took charge of the secretariat of the new standing committee 37 ‘‘Terminology’’ in 1952, as between 1947 and 1949 the committee had been inactive. The committee began to operate within the Austrian Standardization Institute (Österreichisches Normungsinstitut). What was formerly committee 37 ‘‘Terminology’’ became Technical Committee 37 ‘‘Terminology (Principles and Coordination).’’ It was thus clearly stated that its purpose was not to prepare terminology but rather to unify principles and coordinate terminological work. Ninety countries currently participate in the ISO, representing 95% of worldwide industrial production. There are 190 technical committees on different subjects, and they publish the results of their work in the form of various types of international standards or technical reports, e.g. standards for products, processes, control, services, measurements, etc. Besides the two major worldwide standardization bodies, there are also other organizations that deal with general standardization. They may go into further detail about aspects that the ISO and IEC committees only treat generally, or act in a particular geographical area, in addition to, or as a complement to, the international bodies. We can identify four types of standardization bodies that complement one another and that deal with terminology in addition to their specific subject matter: 1. National and regional bodies that represent the ISO and IEC3 2. Multinational bodies that act as standardization bodies 3. Major international organizations that have standardization sections 4. Specialized bodies in a particular subject matter In Europe there are two rather important regional standardization bodies that
terminology and standardization
199
bring together areas composed of several different countries: cenelec, a body with 15 member countries that develops standards on electrical engineering in agreement with IEC principles;4 and the European Committee on Standardization, which was created in 1961 for fields other than electrical engineering and electronics. In addition to official bodies, companies play a significant role in all stages of developing standards, from their preparation to their revision, and most particularly in their application.
2.–Terminological standardization In the light of glottopolitics, the answer to the question ‘‘should we standardise terminology?’’ will be yes. But this ‘‘yes’’ must be precisely defined: terminological standardisation should entail standardisation at a certain, still undefined level. It is a ‘‘yes’’ for a language policy which reduces the terminological mess to a minimum, although it must take a certain amount of variation into account, as without that the language would become sterile. If that were the case, the language would be unable to assume the tasks that any living language must assume in relation to the emergence, transfer and evolution of information.—Guespin & Laroussi (1989)
The term ‘‘standardization’’ as applied to language presents yet another ambiguity, even if we discard its possible interpretation as expansion of usage. As Auger (1984) noted, the term standardization can refer both to establishing some forms of language by means of self-monitoring and to the intervention of an appropriate organization in order to establish preferences for some forms over others. As a result, ‘‘terminological standardization’’ has at least three meanings: • Institutional standardization is a process by which a body sets the preferred use for a designation. • International standardization is a process by which an international body sets the characteristics or conditions that certain products should comply with together with the terms appropriate for them. • Non-interventionist standardization is a process by which a certain terminological system monitors itself by mutual accord of its end-users. Terminological standardization in the first two senses is a part of general standardization, which includes both standardization of special terms and standardization of the principles and methods of terminology as a theory and practice. 2.1–Standardization of terms Standardization of terminology plays a primary role in the standardization of
200
terminology
industrial and commercial products and is an essential part of the process because specialists express ideas and conceptualize reality through terms and also need terms to establish and express product standards. Standardization of terms is a complex process that entails a number of operations: the unification of concepts and concept systems, the definition of terms, the reduction of homonymy, the elimination of synonymy, the fixing of designations, including abbreviations and symbols, and the creation of new terms. This process of standardization is based on a series of implicit postulates which derive mainly from the practical experience of recent years. They clarify the limits within which standardization is valid and set the criteria to be adhered to. The purpose of terminological standardization is to aid communication in special languages, and is not applied to the vocabulary of the general language. Terminological standardization is a concern of all special fields, i.e. humanities and social sciences as well as scientific-technical subjects, both in technical or scientific contexts and in activities such as commerce where it has been practised the most. Proposals for standardization of terms must be mindful of sociolinguistic factors (usage,medium,languagepolicy,userneeds,etc.),psycholinguisticfactors(idiosyncrasies, customs, morals, aesthetics, inhibitions, etc.) and formal linguistic criteria (well-formedness, morphological motivation, possibilities for derivation, etc.) Terminological standardization cannot be carried out without the intervention of subject specialists, who, after all, are the real end-users of its products. Though it is regularly revised, standardized terminology must give the impression of stability. Proposals for terms, especially those for special fields with international scope, must reconcile national and international viewpoints. Terminological standardization requires prior preparation so that decisions can be firmly based and rigorous. Approval of a terminology standard must be given by an authoritative body and must include measures that either support or require usage, depending on the context) to ensure application. Standardized terminology must be the object of a standard that must be disseminated among those who need to use it.5 Even when all these factors are taken into account, standardization does not always attain what it sets out to do. Only when the complex reality of languages is fully appreciated and when there is a clear attitude of respect toward speech communities does the intervention acquire a certain amount of legitimacy, as Guespin and Laroussi elegantly state: Terminological standardisation is a legitimate aim, but it requires taking a great number of factors into account. When the standardisation process does not take real language uses into account, it develops separately from real language. In order to succeed, standardisation must be given a previous glottopolitical diagnosis. It would then be able to describe:
terminology and standardization
201
• the conditions applicable to the production of scientific discourse and the evolution of the relationship between sciences and technical fields; • the communicative conditions required in the transfer of knowledge and technology; • the forms deriving from language conflict, which is unavoidable in the countries concerned.—Guespin & Laroussi (1989)
2.2–Standardization of principles and methods of terminology The second field of application for terminological standardization are the principles of terminology as a discipline and the methods of terminography as a practical endeavour. This is the task of ISO Technical Committee 37. Some countries systematically organize standardization of terminology on the national level by making the creation of terms comply with their guidelines by means of a standardization policy for the language as a whole. This process, even though it is used for the whole standardization process for terms, namely defining concepts, designation and relationships between concept and designation, is mainly focused on establishing designations. In order to undertake this process, specialized commissions of experts and terminologists are usually created according to subject fields. The commission members coordinate their work by following guidelines that usually come from a central standardization body. Terminology commissions acting within a language planning policy require recommendations that are particularly aimed at making existing designations consistent and at creating new words. They therefore usually base themselves on written criteria for the creation of terms and the treatment of borrowings and loan translations from other languages. These criteria approach the issue from the linguistic and the extralinguistic standpoint. A policy on terminology must possess ‘‘doctrinaire’’ documentation if it wants to be coherent. Having a doctrine does not mean taking rigid actions or applying mechanical, artificial treatment to language issues. Rather, it is the only way for terminological standardization to become a unifying factor for a language and to contribute to improving communication. Official decisions on standardized terminology are usually disseminated as a standard or recommendation, depending on the needs and characteristics of each sociopolitical context. 2.3–ISO Technical Committee 37 ISO Technical Committee 37 Terminology (Principles and Coordination) is charged with establishing the principles and methods for terminology, and its
202
terminology
specific objective is to standardize the methods for creating, compiling and coordinating terminologies. It is made up of 15 participating members (P-members) and 38 observers (O-members). It liaises with other ISO technical committees and with 32 other organizations (Galinski & Nedobity 1989). ISO/TC 37 is structured into working groups (WG) and technical subcommittees (SC) that deal with the following topics: • • • • • •
terminology vocabulary coordination of international terminological standardization operating procedures for ISO/TC 37 principles of terminology representation of names of languages computerized resources in terminology To date, ISO/TC 37 has published the following standards:
ISO 639: 1988 ISO 704: 1987 ISO 1087: 1990 ISO 1951: 1973 ISO 6156 ISO/DIS 639–2: ISO/DIS 860 ISO/DIS 1087–2 ISO/DIS 1951 ISO 10241:1992 ISO/DIS 12200 ISO/DIS 12616 ISO/TR 12618: 1994
Code for the representation of names of languages Principles and methods of terminology Terminology—Vocabulary Lexicographical symbols especially used to classify vocabularies with definitions (reconfirmed in 1984) Magnetic tape exchange format for terminological/lexicographical records (mater) Codes for the representation of names of languages—Part 2: Alpha-3 code Terminology work—Harmonization of concepts and terms (Revision of ISO/R 860:1968) Terminology work—Vocabulary—Part 2: Computer applications (Partial revision of ISO 1087:1990) Lexicographic symbols and typographical conventions for use in terminography (Revision of ISO 1951:1973) International terminology standards; Preparation and layout Terminology—Computer applications—Machine-readable terminology interchange format (martif) Translation-oriented terminography Computational aids in terminology—Creation and use of terminological databases and text corpora
terminology and standardization ISO/DIS 12620 ISO/R 919:1969 ISO/R 1149:1969 TEI/LISA/ISO-TIF:1992 ISO/CD 12–199:1996
203
Terminology—Computer applications—Data categories Guide pour l’élaboration des vocabulaires systématiques Présentation des vocabulaires systématiques multilingues Terminology Interchange Format (tif) Alphabetical ordering of multilingual terminological and lexicographical data represented in the Latin alphabet
The structure of ISO/TC 37 can be visualized as follows: ISO/TC 37 ÖNORM
WG 5 ÖNORM
SC 1 GOST
SC 2 SCC
SC 2/WG 1 DIN
SC 3 DIN
WG 6 ÖNORM
WG 7 ÖNORM
Abbreviations: A WG Working group SC Subcommittee ÖNORM Österreichisches Normunginstitut GOST CIS State Committee for Standards SCC Standards Council of Canada DIN Deutsches Institut für Normung
3.–Terminology and neology The concept of neology cannot easily be defined, especially because it cannot be reduced to a linguistic curiosity. Neology is part of a current interactive subgrouping about which the general purpose language and the technolects have reached an agreement that establishes a close relationship between society, the dictionary makers, the political situation and, nowadays, the huge sector of language industries.—Boulanger (1989)
The appearance of a new concept normally coincides with the appearance of a new designation. This new name arises in the language of the society that created the
204
terminology
new concept. Neology, seen as a way of creating new designations, is obviously necessary in special subject fields in which the emergence of new concepts entails constant neological activity. There are two types of situation that require specific intervention. First, when two or more designations converge in a concept and this results in a negative impact on communication; and, secondly, when a special language does not have the designation required for expressing a concept. In the former case, the various designations must be reduced so that only one remains for the benefit of accurate communication. In the latter case a new designation must be created to express the new concept. 3.1–An overview of neology Neology is the field of study that deals with new phenomena that appear in languages. In the opinion of classical lexicologists like Rondeau (1983) or Guilbert (1975), neology deals with the study of the linguistic phenomena that appear at a given time in the development of a language in use. These phenomena can be seen at all descriptive levels of language, in the phonetics, phonology, morphology, syntax or the lexicon. When we speak of lexical neology we are referring to the appearance of new words, or lexical neologisms. In this sense lexical neology covers all those lexical units that can constitute a dictionary entry, regardless of their form. This traditional definition of neology, which only covers linguistic processes for forming new words, is today not comprehensive enough to describe all the activities centred around linguistic innovations. Rather, the linguistic side of neology is just one of the aspects of a whole series of approaches that place neology on three levels —linguistic, cultural, and political. With the emergence of computerized terminology at the beginning of the 1970s and the development of language planning for minority and disadvantaged languages, the notion of neology has broadened. As Boulanger (1989) states, neology currently refers to at least five different activities: • the practical process of creating new lexical units using resources that may or may not be sensitive to the usual mechanisms of linguistic creativity in a language • the theoretical and applied study of lexical innovations: the processes of creation, the criteria for recognition, acceptability or spread of neologisms, the social and cultural aspects of neology, etc. • institutional activity that is systematically organized to gather, assign, disseminate or implement neologisms within a specific language policy
terminology and standardization
205
• the task of identifying entirely new or recently developed special subject fields, or fields that have gaps requiring intervention • the relationship between the new item and dictionaries, especially in two regards: the use of dictionaries as a filter for recognizing neologisms and the analysis of how neology is treated in dictionaries 3.2–Neologisms Neology as a process is a clear and unquestionable concept, even though it is complex and thus difficult to describe; conversely, a neologism, as the result of a process, is extremely vague. It is more often than not a controversial label given to a lexical item as opposed to a clearcut notion. I have called it a ‘‘pseudo-concept’’ because it is the result of a relative, even subjective, opinion that is related to its definition, and this definition is not based on objective newness but rather on a feeling of novelty.—Rey (1988a)
As Rey’s comments clearly indicate, in theory, neology is a clearly delimited concept. Nevertheless, the characterization of a specific segment as neological is not easy unless certain points of reference, which are to some extent arbitrary in nature, are taken into account. There are several possible parameters to determine if a unit constitutes a neologism or not: a. diachrony: a unit is a neologism if it has arisen recently b. lexicography: a unit is a neologism if it is not in dictionaries c. systematic instability: a unit is a neologism if it exhibits signs of formal instability (e.g. morphological, graphic, phonetic) or semantic instability d. psychology: a unit is a neologism if speakers perceive it as a new unit These criteria, which are not mutually exclusive, do not yield the same results nor can they be applied in the same contexts. Neologists tend to favour the lexicographic parameter as the most systematic one to determine whether a lexical unit is a neologism or not. Under this criterion a term is a neologism if it does not occur in the lexicographic corpus chosen as the reference corpus, also known as the exclusion corpus, in a search. This shifts the problem from one of identifying neologisms to one of choosing the reference lexicographic corpus, which varies not only with the subject of the terminology but also with the type of neologism, since several different types exist. A classification of neologisms cannot be based on a single criterion but rather must be multidimensional. Different types of neologisms are identified below according to what they are based on. From the standpoint of their belonging to the general language, there is a distinc-
206
terminology
tion between two large groupings of lexical neologisms that behave differently: those present in the general language (true neologisms) and those present in special languages (also known as neonyms). They differ from one another in several features: • • • • • • •
in their creation in their primary function in their relationship with co-occurring synonyms in the resources favoured for creating the word in their continuance in the language in the way they coexist in the system in the way they relate to other systems
Lexical neologisms in the general language, as opposed to neonyms or terminological neologisms, can be characterized by the following series of factors: 1. They are usually more spontaneous, i.e. they arise for no apparent reason, they appear to be frivolous and are generally short-lived; neonyms, on the other hand, arise because of a need for a designation and are usually more stable. 2. They are not affected by synonymy but usually coexist with synonyms and acquire a certain stylistic value as a contrasting feature. Neonyms, in contrast, reject synonymy because it can distort communicative efficiency. 3. They tend toward formal conciseness, whereas many neonyms are phrases. 4. They often appeal to old and dialectal forms of the language and to borrowings, rather than to compounds based on neoclassical languages. 5. They do not usually spread beyond the language in which they have been created, as opposed to neonyms, which are designed to be international. In contrast to lexical neologisms, neonyms cannot be separated from the features expected of terms: lack of ambiguity, single reference, belonging to a special field, stability, conformity to existing term formation patterns. Despite these differences, and even though Rondeau states that the difference between the two types of neologism is clear, it is difficult to clearly separate the two types. From the standpoint of their function, neologisms can be classified as either referential or expressive. Referential neologisms develop because they are required, i.e. there is a gap in a specific special field that must be filled. Expressive neologisms develop simply to introduce new forms of expression into the discourse. From the standpoint of their formation four basic types of neologisms can be identified:
terminology and standardization
207
1. neologisms in form, including the following structures: • derivations (with prefixes and suffixes) • compounds • phrases • shortenings (using initialisms, acronyms, clippings) 2. functional neologisms, including cases of lexicalization of an inflected form and those formed by syntactic conversion 3. semantic neologisms, including three types of processes: broadening or narrowing or change of the meaning of the base form 4. borrowed neologisms, which are true borrowings and loan translations (see Chapter 3, 2) 3.3–Linguistic characteristics of neologisms Terminological neologisms do not differ from other terms with respect to their formal linguistic characteristics. In contrast to words of the general language, like all terms, neonyms have an unambiguous relationship between the designation and the concept named, i.e. they are monoreferential. They are created because of the need to name a concept, and they are usually more descriptive, and consequently longer, than single words. They are more stable in terms of duration, and are based on set series of derivations within each special field. From the standpoint of form, neologisms use the same resources as other lexical items, but they generally favour one resource or another depending on the type of creation in question. Spontaneous creations are usually short, planned neologisms are more often expressed as phrases. Forms can also depend on the special field the term belongs to. For example, in technical fields phrases are more common than in scientific fields, where terms morphologically based on Greek or Latin are very frequent. 3.4–Pragmatic and sociolinguistic aspects of neologisms The study of neology is not limited to analysing the linguistic characteristics of new words but besides the linguistic aspect, i.e. whether the language accepts the use of specific resources to name new concepts, it considers the cultural aspect, i.e. to what extent neologisms reflect the evolution and status of the technical and cultural development of a society; and the political aspect, i.e. to what extend neologisms strengthen or weaken the survival of a language. In general, applied neology is primarily oriented towards ensuring that a language of culture is capable of being used in all types of communication, both from the
208
terminology
standpoint of subject matter and from that of new types of information exchange. Both lexicography and terminology reflect the fact that a language possesses the necessary codified linguistic tools (which reassure its users) and enjoys sufficient vitality to allow the creation of the forms necessary to adapt to new realities. Within the field of applied linguistics, the study of neologisms is particularly linked to language planning, language standardization, and language codification. Research into neology affects lexicography, because general language dictionaries must be updated, and terminology, because designations in special languages must be created or adapted, and finally language planning, because criteria guiding lexical formation must be set. In the specific context of language standardization, neologisms become the object of intervention by official bodies. Insofar as loanwords are concerned, the ten-year CILF policy establishes the need to economize as much as possible, but not at any price. The French language is rich enough to be able to name most new realities.—Diki Kidiri, Joly & Murcia (1981)
For the effective implementation of any neologism, intervention, which is legitimate in special language terminology, must ensure that the new terms comply with both the linguistic and sociolinguistic conditions that govern the creation of neologisms in the particular language and society. The linguistic conditions that a neologism must exhibit in order to have a reasonable chance of being accepted, especially if it is to replace a borrowing that is already in use, are the following: • It has to designate an explicitly delimited, stable concept that already exists so that the new term is unambiguous. • It has to be as brief and concise as possible, even if phrasal forms are more prevalent in some fields. • It has to conform to rules of grammar. • It has to be as transparent as possible. • It has to be able to be the basis for possible derived forms. • It has to conform to the phonological and graphic systems of the language. In addition to these linguistic criteria, new or preferred terms also have to have certain sociolinguistic characteristics. They have to result from necessity, i.e. they must name a new concept, avoid co-occurring forms, and replace unsatisfactory forms. They should not have negative connotations or unfortunate associations and fit in with any existing language policy so that they are easy to remember. Finally, the methodology employed in neological activity must be rigorous and the established criteria for research, decision-making and organization should
terminology and standardization
209
be adhered to. For this to succeed, specialists who can guide the proposals for neologisms should be involved so that the new creations do not contradict the rules followed by the other units in the special subject field in which it will be used. It cannot be presupposed that forms already in use cannot be uprooted, if considered unacceptable. Above all, a term cannot be standardized without bearing in mind the system of concepts and designations it forms a part of. Even if all these principles and recommendations are adhered to and terminologists get as close as possible to implementing a new proposal, reality always has the last word. Indeed, above all these orders and conditions, feasibility criteria, and recommendations made by language planning bodies, there is the life of words as a whole and that of each word in particular, and that often responds to forces beyond external control. In addition, there is the dynamic of languages and the freedom of choice of speakers, who do not necessarily heed plans addressed to them. 3.5–Standardization of neologisms Neology is an activity, that is, a process, a dynamic field, producing new lexical units in a language system, in a cultural body or a social group. This may be due to the desire to control an evolving world, where new objects or kinds of objects appear; or due to the desire to rename things that had already been given a name for reasons that are difficult to analyse, and which can account for totally subjective and collective phenomena such as snobbism, as well as the internal need to renew the lexical stock.—Rey (1988a)
Decisions taken by a standardization body about new proposals for designations to fill a gap in a special vocabulary or to replace an unsatisfactory designation are usually guided by prevailing political criteria (Consell Supervisor del Termcat 1990b). A standardization body considering acceptance of a neologism has the following options available: • A single proposal can be accepted as the standard form and all others are rejected. • More than one proposal can be accepted and the forms considered equivalents of one another. • More than one proposal can be accepted but one alternative is given preference. • A loan can be accepted without modifying any of its formal or functional aspects. • A loan can be accepted after formally or functionally adapting it to the receiving language. Proposals for designations can be rejected because none fully satisfies the standardization body or because more information is needed before a decision is taken. Finally, a form borrowed from another language or created using the resources available in the language itself can be accepted after modifying its spelling, its pronunciation, its grammatical category, its subcategory, etc.
210
terminology
There are three ways of adapting borrowings to a language: 1. changing one or more elements of the borrowed form: Examples: lasso (Spanish lazo); filibuster (Spanish filibustero); blitz (verb) (German noun Blitzkrieg) 2. deletion of an element Examples: Spanish autoedición, English desktop publishing; Catalan desconnectat, English off-line 3. addition of an element Examples: Spanish hoja de cálculo, English spreadsheet; Catalan base de dades, English database The parameters usually considered when adapting a loan are varied in nature. The new form should be readily acceptable to end-users and easily adapted to the language; examples: [in English]: yogurt, pajama, yoga, spaghetti. A neologism can also be based on the oral form if the written form carries relatively little weight. Terminological standardization bodies do not make arbitrary decisions but base themselves on general criteria for dealing with neologisms and apply them systematically. They therefore usually accept a new term because it is necessary, and widely used in the profession and therefore has already been accepted in its context. It may also have been recorded in general lexicographical works and disseminated by a language body. It may also be found to have a parallel form in other languages; examples: [in English] pied-à-terre, parka, hula. From the point of view of its form, they usually accept a neologism when it is well-formed and fits the etymology and derivational patterns of the language. Borrowed forms are usually accepted without adaptation when there is no satisfactory alternative and it proves difficult to adapt it to the language structure. Registered trademarks and toponyms or anthroponyms are usually kept in their original form. Preference is also given to international designations. 3.6–International criteria for creating terms Is the desire to standardize the undeniable destiny of terminology? It is undoubtedly so if we follow Wüster's idealist rationalism and objectivist mysticism. It is not so if terminology dares to question its governing principles, its opinions regarding general purpose language and the LSPs, its declared autonomy, its relationship to standardisation, its attitude towards neological activity. . . Questioning these matters is the aim of socioterminology, understood as the type of terminology that no longer works with labelling terms, with ‘‘objects to be pinned down,’’ but as a kind of terminology that, by accepting linguistic evolution, is based on the emergence and transfer of concepts and terms, with their continuous and sometimes contradictory transformations, and of specific activities.
terminology and standardization
211
Socioterminology deals with variation (as well as sociolinguistics and pragmatics), and tries to explain the differences found in the sciences, technical fields and industries. It then establishes three levels of analysis: that of terminography, that of ad-hoc terminology, and that of official intervention to face the challenge of computerisation; these levels of analysis are the tools that help to develop knowledge engineering (Gambier, 1991). Such a terminology would not only establish links with linguistics, the sociology of sciences, logic and philosophy . . . but also be affected by such contact in the effort to establish a theory. This, in turn, would lead us to question the currently dominant principles.—Gambier (1991a)
The analysis of terminological regularization in several different countries shows that there are two basic attitudes: a prescriptive attitude and a less forceful, less prescriptive attitude that recommends, rather than dictates, the use of a form. This attitude depends on the level of development of the subject field in question, on the formal distance between its linguistic code and those of the dominant loaning languages and on the social, political and linguistic conditions of the context. Highly industrialized societies approach standardization with attitudes and organizations that are different from those of societies undergoing industrialization. The reasons behind this difference lie in the degree of vitality of the language and the possibilities for new lexical items to be created spontaneously within the language itself. A society that produces terms can afford to wait and see how its language is evolving, because its own creative vitality already ensures continuity of the language. A society that must continuously import techniques, science and technology, and which is the recipient of knowledge created by others in other languages may want to control the entrance of adapted or direct borrowings if it wants to ensure that its own language is not overwhelmed by foreign structures. For this reason these societies exercise a stricter control over neologisms and establish principles to deal with them. Indo-European languages with terminological adaptation policies within an overall process of language standardization can rely on specific antecedents in the spontaneous terminology created in the dominant languages in technology (French or Italian for Romance languages; German and especially English for Germanic languages). They take the terms of the most representative languages as models to create their own terms, bearing in mind that, depending on the special field, they cannot stray too far from the languages used in international relations. Languages in other language families (Basque, Hebrew, Japanese, etc.), however, are quite different in this respect. They must propose their own models for formation and make a choice: either favour native solutions, which distance the language from the more international solutions, or make the solution more similar to that used by other dominant languages, which means the language does not turn to its own resources because they are considered useless for modern communication. After a balanced analysis of the pros and cons of each alternative, policy makers
212
terminology
for terminological neology have to choose which of these two options they favour. This decision is necessarily political in nature, and must conform to the proposals for action and strategies present in the language planning program the society has adopted. It is neither useful nor desirable for a society to adopt parochial solutions for specialized terminology, especially in some very internationalized fields, which would result in linguistic isolation, nor should a society depend excessively on structures borrowed mechanically from other languages, even when usage has practically made them legitimate. The latter approach in the long run leads to language replacement, at least for some contexts. TC 37 of ISO has provided guidelines that attempt to unify certain aspects of new terminology, regardless of the language. ISO Standard R704, approved in 1968 as Principles for Naming, sets forth principles that new units should adhere to and recommends favouring term formation processes that unify terminology in different languages. Although this recommendation may now seem obsolete, it is useful to list some of the principles and recommendations it establishes because they are still important and valid today: • The literal sense of a term should reflect the characteristics of the concept it names. • Complex terms can be considered to be abbreviated definitions. • The phonetic and written forms of terms should be suitable in the language they are to form a part of. • Terms should be as concise as possible without compromising comprehension. • New terms can be formed by abbreviating a simple word, the elements of a compound word, or phrase. • The abbreviated term can be either syllables or individual letters (particularly initials, e.g. PET, positron electronic tomography) of the words that form the base. • A new term should allow derivation. • The formation of complex terms in which the modified element does not correspond to the type of concept expressed by the complex term as a whole should be avoided. • New terms can by formed by altering the meaning of a term of the general language or of another special language, as long as the two special fields are quite different from one another in order to avoid ambiguity. • Many terms will inevitably have more than one meaning. Multiple meanings, however, only lead to confusion if: a) the concurrent meanings are similar to each other, and b) if the two terms often appear in the same texts. In these two cases
terminology and standardization
• •
•
•
213
different terms should be found to express the different senses of an ambiguous term. Synonymy is a strain on memory and gives the impression of confusion. Technical language should not favour variety but rather clarity. Most words that have an international written form derive from Latin or Greek. That is why they do not seem odd in the Romance languages or in English, which already contain a large number of words deriving from Greek and Latin. In other languages, such as the other Germanic languages or the Slavic languages, both the international words and the compound words based on native elements should be admitted as synonyms of one another. In these languages the native words are usually given official status and are thus found in laws, whereas the international words are better suited to communication with foreigners. When words or concepts are already rooted in use, they should not be changed unless there are important reasons to do so and there is some chance that the change will be accepted. Changes that do not gain wide acceptance lead to terminological divergence (i.e. new synonyms) and ambiguous terms. It is particularly difficult to achieve widespread acceptance of a change in meaning. In most cases a change of this nature leads to ambiguous terms. In this situation creating a new term to designate a new concept is recommended.
Chapter 7 Professional terminology: The role of terminologists in a language service
Language phenomena tend to evolve very slowly. At first glance, language seems static and unchanged day after day. Generally speaking, the linguistic situation of a country appears completely steady, forever fixed as a snapshot. On the one hand, this impression is partially due to the fact that the process of change is very slow; and, on the other, to the fact that it is quite difficult to see what is happening because we lack the parameters that would be useful in measuring the differences between two analyses. Nor is there any measure of language development, insofar as the intensity and the direction of the development are concerned. History shows us, however, that language, like all other living organisms, is constantly changing in a never-ending effort to adapt to the continuously evolving reality that it must convey.—Corbeil (1980)
The need to establish relations and to communicate with others that has led to modern economic growth and the appearance of international bodies has given rise to language issues that were unheard of in other times. For example, today we need to have constantly updated and exchangeable information and we must be able to express ourselves in several languages. The standardization of economic, scientific and cultural exchanges between organizations, with the subsequent need to standardize the formulae for designations, is also new. This situation has resulted in a strong demand for specialists in language able to respond to the multilingual requirements of industrialized societies. This situation is also the cause for the new technologies related to communication and information. The way work is organized has changed, as have the resources used to manage and address language issues. Given this new situation, non-dominant languages are attempting to survive amidst those languages considered dominant. The governments of countries with non-dominant languages pursue protectionist language policies in order to encourage the use of the minority language and attain its necessary modernization so that it can be used in all communicative situations. Complex organizations, public or private, need to take systematic linguistic
professional terminology
215
actions aimed both at codifying specific forms and diversifying language usage if they want to adapt themselves to situations requiring the use of more than one language. Only then can they ensure communication and set the status that corresponds to each language. In this context the idea of a language service arises. A language service is a working unit that organizes and resolves the technical aspects related to the language needs of organizations, without losing sight of the society in which these organizations exist.
1.–Linguistic needs and language planning The evolution of a language cannot be separated from the evolution of the society in which it is spoken. Political, economic, social and demographic changes that condition a society also condition the language, or languages, it uses. This evolution, however, is not a free, neutral process in which languages evolve independently but is beset by interferences that push the development of a language in one direction or another. Thus, the greater the economic and political power of a society, the more probable is it that its language will be dominant. For this reason, societies that are economically and technologically or politically dependent upon others need to take steps to protect their cultural identity by safeguarding their language rights. The regulatory measures for the individual and social language rights introduced by a community constitute the language policy of that community. Such measures affect the main language of the society and also other languages spoken in the community. Language standardization is a dynamic, gradual social process in which a language, starting from an unstable situation, achieves a stability which people perceive as natural. In this sense standardization processes are not natural, because they deal with political considerations, the appropriate legal framework, and with strategies for disseminating and carrying out the policy in education, communications media, etc. Language policies thus develop a series of strategic actions and intervening measures aimed at changing the natural evolution of a language. Successful language policies must be based on a consensus that allows the coherent implementation of the set of actions by the various agents taking part in the process and on clear objectives expressed in a standardization plan, i.e. a systematic strategy that includes the intervening measures and gives them a sense of unity. They must be rigorously planned and proceed within a legal framework that supports the intervention. They also need criteria for tracking the situation as it changes while the plan is applied. In addition, a standardization plan can only be successful if it is backed by the
216
terminology
political will and the necessary resources for carrying it out. In this respect it is vital for governmental organizations to be able to count on the collaboration of civic organizations and society as a whole. Without this support it is difficult to bring about a change that will be firmly anchored in society. Language services established as working units within organizations are important parts in the standardization process of a language. They help to rationalize the process and make it more effective. Generally speaking, language services are agents that favour language change in the area in which they act, and as such they should be a part of any coherent strategic planning. Language services can, in the initial stages of standardization of a language, be responsible for filling gaps produced by irregular developments which may be related to updating a language code, or to speakers’ lack of full linguistic competence. Once the language policy has been accepted, these services are easily reconverted and can take on the task of managing the information and communication needs of the organization, irrespective of the language used.
2.–Planning and language services The linguistic needs resulting from cultural and economic changes in developed societies are generally of three types. New communication needs arise due to the appearance of new opportunities for contacts inside and between societies. New needs for designations arise due to the rapid progress of science and technology and engineering, the microspecialization in many fields of knowledge and human activity and the creation of new interdisciplinary fields. New multilingual needs arise as all sorts of organizations act in the international arena. At the same time the need to defend communication in one’s native language becomes more widespread. In this regard, the result is an apparently contradictory trend for languages: movements towards monolingualism and multilingualism at the same time. There is a strong tendency toward the use of a single language, especially for certain communicative situations, yet with the development of political democracies, there is also a tendency to encourage multilingualism in which non-dominant languages pursue protectionist policies against the expansionist tendencies of the dominant languages. Given this panorama, governments are taking an increasingly active role in planning language use, even in the case of those languages that were until recently considered dominant. They have initiated strategic plans for standardizing language use and codifying certain forms of usage to ensure that their language can be used in all communicative contexts, no matter how advanced, specialized or highly formulaic.
professional terminology
217
Wanting to belong to the group of industrialized nations means that a country must be willing to accept that economic, professional, scientific, and cultural exchanges are increasingly necessary. Such exchanges allow us to maintain and disseminate "differences" yet at the same time favours the elimination of these "differences". It also assumes having enough updated information available so that it can be exchanged in equal conditions with the most advanced countries. To carry out this exchange of information, we need designations to express innovations, and they need to be formed in a way that is suited both to the language in which they are produced and to the equivalents in the other languages in which the exchange of information occurs. The key issue, then, is being able to communicate with diversity from one’s own identity. To deal appropriately with the language needs of organizations, it is essential to have professionals specifically trained in these areas. This is the purpose behind the language services of governmental, educational and professional organizations. These services deal with the linguistic and communicative aspects of their respective organizations and address their multilingual needs by giving priority to their own native language as the most natural vehicle for the usual communication of the organization and regulating the use of other languages in accordance with the characteristics of each communicative situation. These services, which act as organized groups within a larger body, deal with a wide variety of language questions such as the external image, signs, posters, etc.; internal signs, posters, bulletin boards, etc.; documentation for external use; documentation for internal use; external communication with users and clients; internal communication among the organization’s professionals. At the same time, these services also organize the language training of the organization’s staff so that all employees can correctly use the native and other languages of communication within a reasonable period of time.
3.–The technical tasks of language services Where language services have a key role in the standardization of a language, but where more than one language is involved, they share some features. On the one hand, they are designed to bring about a change in language use in the society as a whole and in the organization in particular, and on the other, they must be able to deal with all the organization’s communicative needs in all the languages involved. The tasks of language services can be divided into those related to codifying certain usages of the language and to those related to strengthening the use of the native language and regulating the use of the other languages that coexist in the society or in the organization.
218
terminology
Codifying certain registers of the language includes such jobs as: • translation of texts and documents in all the working languages of the organization; • drafting and correction of texts in several languages; • design of specific documents; • standardization of the names of positions, functions and other itemized lists specific to the organization; • review of the organization’s headquarters’ internal and external signs; • preparation of terminological glossaries and lists of phrases used in the organization’s special subject field; • providing answers to language, document and terminological questions raised by members of the organization or centre; • setting forth the reference linguistic criteria, etc. Strengthening the use of a language involves extending the knowledge of a language, e.g. the organization of language training for staff in accordance with the language-related tasks required by their jobs; and extending the use of language, e.g. proposals to stimulate use of a language, regulating language use, development of language profiles in jobs, dissemination of information about the language rights of individuals and the rights and obligations of organizations; designing and carrying out surveys of language needs, the state of knowledge and use of a language, the elements favouring change, etc. In societies with standardization plans for their native language, language services directed at changing the status of a language are indispensable. But even languages that enjoy fully standardized social status need professionals devoted to dealing with issues like the continuing modernization of their special vocabularies, the adaptation of their resources and models to technological innovations, the standardization of concurring forms in order to further communication without interference or ambiguity, etc. These languages, which are in a less difficult position than languages undergoing standardization, are used by bodies that need linguistic mediators to facilitate international communication in several languages. The creation of the language services in developed societies is fully justified for several reasons. The growth in multilingual and specialized communication is increasingly important in business and professional relations. Services are needed for technical translation, the teaching of general language and language for special purposes, technical interpretation, and technical writing. The need to ensure communication between languages implies a need to standardize formulae, units and models of communication and to establish reliable equivalents among languages.
professional terminology
219
Our need for information requires the creation of large data banks on various subjects. Languages play a key role in these data banks: on the one hand, they are the best way to access the information, and on the other, they constitute a valuable flow of information if they contain textual, grammatical, lexical, terminological, and bibliographical information. The widespread use of computers, particularly personal computers, and the growing automation of certain types of mechanical tasks that until recently were carried out by people involve language as an element of communication between people and computers, and in the future, between computers and computers. The importance of economic profitability and the commercialization of all goods has also turned language into a commodity. The language industries are devoted to developing, producing, and selling technology related to processing language such as software for correction, writing, translation, programmes for isolating and extracting terms, etc. The new communicative needs of society also create new needs in languagerelated fields. There are new professions concentrating on formal language processing and addressing new language-related issues, for example linguists specializing in computer science and technical writers. At the same time traditional language professionals must change their own working habits to adapt to using new technologies. The language service expert, also called a language mediator or communication mediator or, as coined by Engel & Picht (1990), an information manager, is one of the new professionals needed by our society. 4.–Language services and terminology Terminology plays an important role in any language service. The need for terminology is concentrated in three areas. First, in order to name the innovations that are constantly being developed an organization must have regularly updated terminology. Secondly, each organization has a number of language-related tasks that require highly specific documentation. Specialized terminology in this documentation, along with the phrasing and overall format of the document, is one of the most important elements involved. Thirdly, the language needs of most organizations are not limited to a single language, but often involve several languages whose use must be regulated. In multilingual situations, where different versions of the same document must be produced, translators need the assistance of terminologists for answering questions about specific cases of equivalents between languages. Multilingual terminology is fundamental for accurate translation of technical documents. All such organizations have two types of professionals involved with terminology: (a) the subject field specialists and translators and interpreters who assist
220
terminology
them in communication in foreign languages, and (b) terminologists, but also linguists, lexicographers, information scientists and language planners. For the former terminology is a tool for communication, for the latter it is the object of their work. Translators do not usually write terminologies in special subject fields but rely on the systematic work produced by terminologists; they consult terminologists on specific matters. A well-structured language service needs at least four types of specialists: • translators of technical texts • technical writers and editors • terminologists, who carry out systematic terminological research, write dictionaries, vocabularies and glossaries and find solutions to terminological queries raised by translators, technical writers and editors • information scientists, who are in charge of storing and disseminating information, and providing the other groups with the documents necessary for their jobs.
5.–The training of terminologists The teaching of terminology cannot be separated from the social context in which future terminologists have to work. Note what Maurais (1987) insightfully stated about the French Canadian context: The teaching of terminology must take into account the social context within which the future terminologist will work. What is this context? The best way to describe it is to recall the circumstances in which terminology appeared in Quebec. It was first conceived by translators at the time when the Canadian and Quebecois language policies were being defined. In my view, terminology is a discipline at the service of the language policy carried out in Quebec and is essential for the application of the policy on bilingualism undertaken by the Federal Government. The abovementioned situation results in the realisation that terminology was born within a bilingual context. The type of terminology practised here is mainly contrastive. (. . .) In Quebec, terminology also plays a role in the process of updating the lexicon, and it takes on two different aspects. First, there is lexical readjustment, that is, the Quebecois are now using French terminologies that already existed but which were unknown because the industrialization process had been carried out mainly in English. Second, there is lexical updating proper, which is an activity defined in terms of its relationship to translation.—Maurais (1987)
The training of professional terminologists should be organized in accordance with the specific needs of a particular society. In this regard we fully agree with the following basic principles proposed by Boulanger (1987) to set out an international agenda for training in terminology: • Real situations for training are not identical.
professional terminology
221
• The teaching of terminology and the training of terminologists derive from an ideology that is not only political but also scientific in nature. • Terminology involves both theory and practice, two inseparable sides to a single phenomenon. • Terminology is an organized discipline that cannot be improvised. • No one becomes a terminologist except by acquiring a background of knowledge and training in which theory and practice are intertwined. 5.1–Background assumptions Organizing the training of terminologists has to begin with a series of a priori conceptual distinctions which condition the design of a plan of study and without which such a plan loses sight of its real purpose. Some of these distinctions are: • Training is not the same as instructing, i.e. training in terminology presupposes providing a broad, well-thought out view of the subject. • Suitable training in terminology must necessarily include theoretical, methodological, and technical aspects, and practical experience in a professional setting. The relative weight of each component of the training will vary with the case. • A suitable plan of study must start from the characteristics of the social context in which it exists and cannot ignore the language or languages the professional is going to have to deal with. • A plan should clearly state what its aims are, who it addresses, and what requirements it has. The content of the plan of study can only be determined in function of these parameters. • The content of a training programme must be based on the activities its participants are likely to carry out. • A suitable plan must include information that covers present and future needs, and as such cannot ignore the technical resources used by the profession. • A plan should provide for separate specialisation for those interested in research, those interested in practical work, and those interested in training others. • Training in terminology can be realized by means of extensive or intensive courses, seminars, training periods at terminology centres, lecture series, etc. Each training situation requires the right kind of tuition. • Finally, a training programme operating within a language standardization context cannot ignore the social bases of terminology.
222
terminology
5.2–Training in terminology versus training terminologists Training people in terminology should not be confused with training terminologists. Although it is true that translators, documentation specialists and subject field specialists can be trained in terminology to intervene at a specific point in a research programme, terminologists should be the only professionals who actually write terminologies. They should be able to organize and carry out systematic monolingual, bilingual, or multilingual terminological work within a special subject field. To do this they must know how to deal with specific terminological tasks and how to supervise the terminological work of other terminologists. A programme of training terminologists should include elements from linguistics, particularly lexicology and lexical semantics, logic and classification theory, special subject fields, documentation, sociolinguistics, pragmatics and computing in addition to all the knowledge a specific social situation may require. I have argued for the terminologists' active role in the process of Gallicisation, so that terminologists become agents of linguistic change as opposed to being people who simply make lists of terms.—Maurais (1987)
In addition to the basic knowledge required, the training of terminologists can be implemented at a basic level, including the teaching of the principles and methods of terminology aimed at preparing terminologists-terminographers able to carry out applied research tasks; and at an advanced level, geared towards future terminology professors or directors of language planning projects or projects involving terminological activity. After reflection and consideration of all these provisos and guiding principles, a pragmatic, professionally oriented plan of study can be developed. The position of R. Kromp is illustrative in this respect: It is not the university's responsibility to train experienced professionals. First of all, terminologists must be trained in the rigourous techniques that they will have to use in their working environment. Training must be consolidated with practice and possibly even with one or more training courses. Above all, theoretical training must not be replaced by practical experience, which often is too general to be enriching.—Kromp (1987)
Beside the training of terminologists, there is also a need to consider the training of translators, interpreters, and technical writers; scientists and technicians; information and documentation specialists; language planners, and computer scientists specializing in artificial intelligence applied to language. Translators, interpreters, and technical writers need training in terminology to deal with the specific terminological problems that derive from working with more than one language or from special subject writing, or to work on multilingual terminological research.
professional terminology
223
Scientists and technicians learn terminology to enable them to participate as subject experts in standardization tasks and systematic research, and to deal with problems of designation that constantly arise in their fields. Information and documentation specialists need training in terminology to organize information into concepts so that it can be easily retrieved. They also take part in the writing of thesauri and classifications. Language planners take terminology programmes to intervene in the standardization and modernization of a language, to detect the terminological needs of the various social contexts and to implement the terminology they have helped to prepare. Finally, it is useful for computer programmers specializing in artificial intelligence to have training in terminology in order to construct expert systems for languagebased activities. To conclude, we would like to draw special attention to one of the ideas we have considered fundamental throughout this book — the need for terminologists to have a broad-based, solid background that allows them to reflect on their work and to act on their own, as well as specific training so that they can play an active role in the numerous research areas that all languages have to deal with. We firmly believe that terminologists are necessary for society, but they must not lose sight of their role nor of their objectives, as Corbeil aptly notes: The professional aspect of terminology should not be exaggerated. Terminology is primarily the business and responsibility of several groups of specialists. Terminologists, with their working methods and knowledge, are merely technical aids in a multidisciplinary field.—Corbeil (1988)
Notes
Chapter 1 1. E. Wüster defended his doctoral dissertation Internationale Sprachnormung in der Technik, besonders in der Elektrotechnik at the University of Vienna in 1931. The translation of his work into Russian is the starting point for the interest in terminology in technical domains and an indication of the increasing importance given to the standardization of terms. As Picht (1984) states, the Russian translation of Wüster’s work led the ISA (International Standardization Association) to establish its Technical Committee 37 for the purpose of unifying the methods and presentation of specialized terminologies. World War II stopped the work of TC 37 but it was resumed in the 1950s, thanks (again) to Wüster’s interest in the subject. 2. According to Rondeau (1983), the real father of terminology as a scientific discipline is Lotte. When Lotte was concerned with theoretical and methodological issues, Wüster was working on the processing of terminological data and did not begin to develop a general theory of terminology until the 1970s. 3. ISA (International Standardization Association), founded in 1926, was the first international standardization body and is the predecessor of ISO. 4. In 1971 this interest prompted Unesco to create Infoterm, the international center for documentation on terminology. 5. In 1969 Wüster published Die vier Dimensionen der Terminologierarbeit, in which he presents for the first time the four aspects of terminological work: the special subject field, the languages, the purpose and the degree of abstraction. His overall approach to theory, Einführung in die Allgemeine Terminologielehre und terminologische Lexikographie, only appeared posthumously, in 1979. 6. African languages have adopted a method for terminology that is halfway between compilation and creation. When they do not have written texts, terminologists base themselves on a two-fold survey that is their source for terminology: an ethnographic survey, to ensure the terminology respects the world view of African communities. See Halaoui (1990). 7. The various issues of the journal Termnet News on terminology in specific geographic areas can be consulted for information on terminological activity in different parts of the world.
226
terminology
8. The creation of an international Information center for terminology documentation by Unesco, Infoterm, has contributed much to the development of the discipline and has been a crucial factor in encouraging terminological activities around the world. Chapter 2 1. The study of the relations between special subject languages and general language has given rise to opposing views. Some say that general language and special languages constitute two autonomous sets in opposition to each another. A second view defends the autonomy of the two sets between which there is an intersecting relationship. The third position, which is the most widespread at present, considers that general language and special languages are two intersecting sets that, together, form the broader set of the language in its entirety; between these two subsets there are continuous exchanges in both directions. 2. To facilitate this moderate approach, ISO admits the original word in the language and the borrowing from another as synonyms, on condition that both are used. 3. In practice different languages, especially in fields related to the social sciences and to culture, business, and leisure, have views of the real world that partially differ from one another and therefore have conceptual systems that only partially coincide. 4. In theory terms have only one meaning. In practice this is true when terms are defined within a specific special subject field. 5. For a distinction between dictionaries, vocabularies and glossaries, see Boutin-Quesnel et al. (1990). 6. This opinion differs from the Sapir-Whorf hypothesis, which states that reality is only perceived through language. 7. And also of the morphemes in the case of morphologically complex words. In lexicology a morpheme is described as the ‘‘smallest unit of meaning.’’ 8. There are several types of concepts, according to the class of entity they represent: objects, qualities, actions, places, situations, relationships, etc. See Chapter 3, Section 2. 9. Felber (1985) states that concepts can represent an individual object or by abstraction a series of individual objects that share certain qualities. 10. Choosing the characteristic animal, for example, for the definition of dog excludes the characteristic human but not the characteristics carnivore or domestic. 11. The Universal Decimal Classification (UDC) is the best known hierarchical classification; libraries and documentation centres usually also have other, more precise classifications that allow for finer choices of information in the retrieval process. 12. The transfer of knowledge from specialists to students in a special field is an act of
notes
227
communication between two specialists with varying levels of expertise. Such a communicative act is basically aimed at increasing the decoder’s knowledge. 13. Sager (1990), discussing the model describing specialist communication, states the following: ‘‘In a model on specialist communication we assume the existence of at least two specialists in the same discipline, who are jointly involved in a particular situation where the sender is motivated to transmit a linguistic message which concerns the topic of his choice and which he expects a recipient to receive. We assume that the sender’s motivation arises from a need or desire to affect in some way the current state of knowledge of the recipient.’’ 14. We exclude administrative, legal, and management texts, which are written in highly structured formats and are used to establish different types of relationships between senders and recipients (e.g., applications, contracts, etc.) This type of text is discussed in Chapter Three, 1. 15. We distinguish between original terminology and adapted or created terminology. When a language names with its own resources a concept hitherto unknown to it, it tends to propose forms that are as transparent as possible; hence derivation, compounding and syntagmatic structures are the most used resources. Nevertheless, as new terms become more familiar, speakers tend to shorten them in discourse by using only a segment or by establishing an abbreviated form. 16. The following term banks are from this first stage: dicautom, machine-readable dictionary of the European Community of Coal and Steel for its translators. It dates from 1963 and is the basis for eurodicautom, the current plurilingual data bank of the EEC which was created in 1973. Banque de Terminologie du Québec [BTQ], created in 1973 by the Office de la Langue Française from an initial file in 1970, used to promote usage of French. termium, property of the Canadian government, built from an initial computerized file in 1974. team, data bank of the company Siemens in Munich, created in 1967-68 for translation purposes. termdok, created in Sweden from a 1968 system. lexis, of the German government, started in 1969. danterm, data bank of the School of Advanced Commercial Studies of Copenhagen, the result of a 1975 feasibility study. vinkik, data bank created in Moscow in 1979 containing the terminology from national and international technical standards. normaterm, data bank created in 1973 by the French Association for Standardization (afnor). It has French and English terminology for the ISO standards.
terminology
228
This information comes primarily from Felber (1984b), Rondeau (1983), Sager (1990), and the special issues of the journal TermNet News devoted to the countries and groups of countries that have terminology banks. 17. The information in French and English in the termium data bank is now accessible on CD-ROM. 18. On existing programs of computer-assisted work or automatic work for French see Les industries de la langue. Quebec: Conseil de la Langue Française, 1990, as well as many special issues of the journals Terminogramme, Meta and L’actualité terminologique that have been devoted to this topic. Also see Abbou et al. (1987). Chapter 3 1. The expressions proposed for neutralizing geographic, social, and historical differences, in order to ensure more effective communication in formal situations, are the standard variety of the language. Some linguists therefore consider it a functional variety of the language. 2. Several overlapping terms have been used to designate the concept of special language: special languages, specialized languages, and languages for specific purposes. Authors such as Kocourek (1982) or Rondeau (1983) consider the three to be synonyms. In contrast, Sager et al. (1980) explain that the term ‘‘languages for special purposes’’ (LSP) comes from language teaching as opposed to linguistics, and as such must be limited to usage in this context. The expression LSP is justified in language teaching due to the specific goals a language program must cover as a function of its addressees. —In the linguistics literature Sager et al. (1980) uses the term special languages, or, to be more precise, special subject languages: ‘‘In teaching English as a foreign language we can speak of English for general purposes (EGP) and English for special purposes (ESP), which represents a division of convenience for designing syllabuses and course outlines.’’ 3. This happens when we talk about the details of our job to friends in another profession, e.g. when we talk to a lawyer about a property issue, or when we talk to colleagues about the most recent football match. We ‘‘banalize’’ the specialized terminology of each subject area. Mass communications media particularly contributes to the banalization of specialized terminology. See Galisson (1978). 4. The classification complements Hoffmann’s and is based on: (1) the degree of abstraction of the subject; (2) the natural or artificial syntax used; (3) the type of special subject field; (4) the participants. 5. There are certain cases, especially when the defence of the rights of minority languages is concerned, in which the maintenance of the identity and reaffirmation of the language in formal contexts is as important as the form of the communication itself. 6. The mass media and wide availability of education have aided this dissemination
notes
229
of special subjects, and along with it that of the most commonly used terminology. —Terms like sonogram, CAT scan, and magnetic resonance imagery (or its abbreviation, MRI) are now part of the general terminology of speakers who had never heard of them twenty years ago. New terms are becoming a part of the general lexicon, depending on the development of technological progress and its commercial impact. 7. For example, variable ‘‘is’’ a term from mathematics ‘‘used’’ in statistics and economics; virus ‘‘is’’ a term from medicine ‘‘adapted to’’ computer science; cell, a term ‘‘from’’ cytology, is used in electronics: ’solar cell,’ ’photoelectric cell,’ ’voltage cell.’ 8. This would be the case for terms like vector (which belongs to both mathematics and physics) or tone (which can be defined from the standpoint of either music or acoustics). 9. Not all special subject fields exhibit these characteristics. There are more words based on Latin and Greek forms in the experimental sciences and mathematics than in engineering, for instance. See Martí (1991). 10. In this book we assume that derived words are formed from a lexical base that is a lexeme (currently or historically). If this lexeme is simple, it coincides with the root, to which inflectional affixes can be added. 11. This tendency is often violated because all languages can refer to the same reality in different ways, using different grammatical categories, e.g. operation (process noun) pressing (action noun) immunity (quality noun). 12. In this classification we only take into account terminological units that have been adopted to respond to a need for a designation in a special subject field, and do not consider the historical provenance of the language as a whole. Obviously, most lexical units in Romance languages come from Latin, and most in English have either a Germanic or neoclassical base. 13. Greek and Latin forms constitute one of the richest and most widely used sources for extracting and creating special language terms. For example: from Latin: adhesion, dissimilation, genocide, ferric, disjunction; From Greek: democracy, toxic, hematocrit, biosphere. 14. Xenisms are a separate class of borrowings, because they are designations in other languages that name idiosyncratic concepts in the language they belong to, e.g. ayatollah, gendarme, perestroika. When xenisms are used to designate a concept that also exists in the target language (which is often the case) they are borrowings. Native English speakers vary widely as to their pronunciation of borrowings. 15. It is important not to confuse phraseological units with what is often called collocation or context. A phraseological unit effectively constitutes the linguistic context of the term if it includes it in a specific text, but a context does not necessarily constitute a phraseological unit. We might also note that a context does not really correspond to a special subject field, whereas a phraseological unit is a unit in relation to a specific subject. See Picht (1987, 1991) and Kjaer (1990).
230
terminology
16. General lexicology treats this process of diversification in terms of multiple meaning and thus does not consider there is a new lexicographic unit. Rather, it treats the unit as a new meaning of an already existing word. In terminology, this is treated as a case of homonymy which leads to a new unit that is an original association of a designation and a concept that only has meaning in a special subject area. This view has recently been questioned by linguists working in socioterminology; see Gambier (1991a), Guespin (1991) and Gaudin (1991) 17. Derivational rules are very precise as to their environment. Each rule only applies to a single affix and is subject to a series of restrictions which affect both the base and the rule’s output. 18. This view leads to independent representations of the units in a dictionary. Each homonymous unit has its own entry. If terminologies are prepared for a specific area, each specialized dictionary will only contain the terms of the relevant discipline or domain. Chapter 4 1. Alongside translators who clearly specialize in translation there may also be translators with training in systematic terminology who in fact work like terminologists. 2. There are national and regional terminology and standardization bodies that also set standards for work and presentation of data, but in most cases they follow the overall layout of ISO standards. 3. This initial analysis allows the terminologist to find out whether the topic has been worked on previously and at what level, whether it is necessary to go forward with the search, if the work needs to be redirected, etc. 4. Consultations with subject field experts are very useful for resolving specific questions and in obtaining information. The answers they can give usually provide much information and consequently it is useful to transcribe them clearly and to identify the source. 5. In some areas in which terminological activity has become highly organized, access to term banks has superseded consultation of printed works. 6. Alphabetically arranged vocabularies and lexicons without definitions cannot be used as references on the contents of a subject field because they do not contain conceptual information; on the other hand, they can be used for subsequent verification of information. 7. Terms are extracted from a document because they are of interest for the terminology being prepared. The fragments of the document relating to a term are extracted because they may be of interest either for illustrating how a term is used in a text or because they can provide information about the term’s meaning. 8. We must differentiate between three types of search: monolingual, multilingual and monolingual with equivalents. Monolingual searches are usually done for lexicons of
notes
231
special subject fields and are often aimed at standardizing terminology in a language. —Multilingual searches are more often than not systematic. The end products of this kind of search are dictionaries or vocabularies with information about several languages. —In terminographic work, monolingual searches with equivalents are mainly carried out in only one language but it contains equivalents in others. Therefore, it is not the product of a complete multilingual search, but rather the result of a monolingual search to which a subsequent search for equivalents has been added. 9. An original text is preferred to one that is a translation from another language. For more information about criteria to evaluate the source documentation, see Rondeau (1983). 10. Examples of transparent units: flow chart, soldier of fortune, file server. Terms made up of morphemes (roots and affixes) are usually not transparent, e.g. informant, storage, extraction. 11. Multiword units in terminology are often used in an abbreviated form and over time this form may take the place of the phrase, e.g. magnetic resonance imaging: MRI central processing unit: CPU modulator/demodulator: modem facsimile: fax 12. The text in dictionaries is a mixed context, between a defining and a metalinguistic context. 13. Sometimes the definitions of terms come from extraction records, since many documents that have been used as extraction sources are informative manuals from the subject field and as such they include definitions or defining contexts. If the definition cannot be obtained during the extraction process, then specialized dictionaries can be consulted. Subject field specialists can be consulted as a last resort. 14. See ISO/R standard 639. 15. Works with a sociolinguistic factor that are designed to contribute to the standardization of the usage of a language usually include sociological information concerning the stage of standardization of terms (i.e. whether a term is standard, standardized by an authorized body, pending standardization or deprecated). Chapter 5 1. Term extraction systems analyze a text and provide a first list of units, with either a simple or complex structure, that might be terms. From this initial list the same system goes on to refine the list. 2. An indexed field is analyzed by the system when it is consulted. As a result, indexing (excluding grammatical function words) of all the words in the abstracts of the publications contained in a bibliographic database yields information about all the works that have a particular word in their description.
232
terminology
Chapter 6 1. Given the evolution of science and technology, standards have to be revised every five or six years to determine whether they are still acceptable and satisfactory. 2. The languages of the International Electrotechnical Vocabulary are English, French, Russian, German, Italian, Spanish, Dutch, Polish, and Swedish. 3. National standardization bodies are their country’s official representatives at the ISO and the IEC: for example, ANSI in the U.S., SCC in Canada, BS in the U.K., ON in Austria, GOST in the CIS, AFNOR in France, DIN in Germany, AENOR in Spain, etc. These bodies publish their official standards in the official languages of their country. See Frank (1981) for more information. 4. cenelec includes Austria, Belgium, Denmark, Finland, France, Germany, Ireland, Italy, the Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, and the United Kingdom. 5. Three types of terminology standards depend on the information included, and they are usually multilingual: standards with terms and definitions or explanations, standards with terms and illustrations, standards that only contain terms.
Bibliography* Actes du colloque Terminologie et technologies nouvelles. 1988. Quebec: Gouvernement du Québec. Adda, Rosine et al. 1979. ‘‘Néologie et lexicologie’’. Langue et langage. Paris: Larousse. Alber-Dewolf, Rosa. 1984. ‘‘Étude sur la création néonymique. Analyse comparée des procédés morphologiques et morphosyntaxiques de la formation des termes du domaine de la spectroscopie en anglais, en allemand, en français et en russe,’’ Thèses et Mémoires. Quebec: girsterm. Andersen, Aldean et al. 1987. ‘‘La création d’une banque de terminologie et perspectives d’avenir en terminotique’’. Meta 32, 2. 111–23. Arnzt, Reiner. 1986. ‘‘Réflexions méthodologiques sur l’harmonisation internationale des terminologies’’. Rondeau & Sager 1986. 219–25. Association cannadienne de normalisation. 1992a. Principes et méthodes de la terminologie. Norme CAN/CSA-Z780–92. Rexdale (Ont.): Association canadienne de normalisation. Association cannadienne de normalisation. 1992b. Terminologie-Vocabulaire. Norme CAN/CSA-Z781–92. Rexdale (Ont.): Association canadienne de normalisation. Association française de normalisation. 1993. Recueil de normes françaises: documentation. Paris: AFNOR. 2 vol. Auger, Pierre. 1982. ‘‘La problématique de l’aménagement terminologique au Québec’’. Terminogramme 13. 1–3. Auger, Pierre. 1984. ‘‘La Commission de terminologie de l’Office de la langue française et la normalisation terminologique’’. Terminogramme, 26–7. 9–12. Auger, Pierre. 1986a. ‘‘La francisation et la terminologie: l’aménagement terminologique’’. Rondeau & Sager 1986. Quebec: girsterm. 47–55. Auger, Pierre. 1986b. ‘‘Introduction’’. Répertoire des avis linguistiques et terminologiques. Vol. 1, mai 1979 - octobre 1985. Quebec: Gouvernement du Québec. 5–11. Auger, Pierre. 1988. ‘‘La terminologie au Quebec et dans le monde, de la naissance à la maturité’’. Actes du sixième colloque OLF-STQ de terminologie. L’ère nouvelle de la terminologie, 27–59. Quebec: Gouvernement du Québec. Auger, Pierre. 1989a. ‘‘La terminotique et les industries de la langue’’. Meta 34, 3. 450–6. Auger, Pierre. 1989b. ‘‘Informatique et terminologie: revue des technologies nouvelles’’, Meta 34, 3. 485–92.
—*This bibliography has been updated since the original version of this book was published in 1992 to aid those interested in further discussion.
234
terminology
Auger, Pierre, Rousseau, Louis-Jean & et al. 1978. Méthodologie de la recherche terminologique. Quebec: Éditeur officiel du Québec. Austin, John L. 1970. How to Do Things with Words. New York: Oxford University Press. Beacco, Jean-Claude, ed. 1992. ‘‘Ethnolinguistique de l’écrit’’. Langages 26, 105. Beaugrande, Robert de. 1987. ‘‘Special Purpose Language and Linguistic Theory’’. ALSED– LSP Newsletter 10, 2 (25). 2–10. Bédard, Claude. 1990. ‘‘Les produits ‘‘infolinguistiques’’: un tour d’horizon’’. Terminogramme 55. 7–9. Bédard, Édith & Jacques Maurais, eds. 1983. La norme linguistique. Quebec / Paris: Gouvernement du Québec/Le Robert. Béjoint, Henri. 1989. ‘‘À propos de la monosémie en terminologie’’. Meta 34, 3. 405–11. Bessé, Bruno de. 1991. ‘‘Le contexte terminographique’’, Meta 36, 1. 111–20. Bessé, Bruno de, ed. 1992. Terminologie et traduction, 2/3. Luxemburg: Commission des Communautés européennes. Bessé, Bruno de. 1994. ‘‘Contribution à la définition de la terminologie’’. Langues et sociétés en contact. Mélanges offerts à Jean-Claude Corbeil ed. by Pierre Martel & Jacques Maurais, 135–8. Canadiana Romanica, 8. Tübingen: Max Niemeyer. Blampain, Daniel, dir. 1994. ‘‘La traduction et l’interprétation dans la Belgique multilingue’’. Meta, 39, 1. Blanchon, Élisabeth. 1993. ‘‘Défi pour la terminologie’’. La Banque des mots, special issue. 88–91. Boulanger, Jean-Claude. 1981. ‘‘Bibliographie linguistique de la néologie 1960–80. 1. Études linguistiques’’. Études, recherches et documentation. Quebec: Éditeur officiel du Québec. Boulanger, Jean-Claude. 1984. ‘‘Quelques observations sur l’innovation lexicale spontanée et sur l’innovation lexicale planifiée’’. La Banque des mots 27. 3–29. Boulanger, Jean-Claude. 1986. ‘‘La stratégie pédagogique dans l’aménagement linguistique’’. Rondeau & Sager 1986. 56–65. Boulanger, Jean-Claude. 1988a. ‘‘Le syntagme en informatique: un projet de recherche’’. Terminogramme 46. 22–3. Boulanger, Jean-Claude. 1988b. ‘‘Les dictionnaires et la néologie: le point de vue du consommateur’’. Actes du colloque Terminologie et techonologies nouvelles. Quebec: Gouvernement du Québec. 291–318. Boulanger, Jean-Claude. 1989. ‘‘L’évolution du concept de NÉOLOGIE. De la linguistique aux industries de la langue’’. ed. by Caroline de Schaetzen. Terminologie diachronique. Actes du colloque 7 (Brussels, March 1988). Paris: Conseil international de la langue française. 193–211. Boulanger, Jean-Claude. 1990. ‘‘La création lexicale et la modernité’’. Le Langage et l’Homme 25, 4. 233–40. Boulanger, Jean-Claude & Pierre Auger. 1987. ‘‘Le formateur de terminologues: un globetrotteur’’. Table ronde sur l’enseignement de la terminologie à l’université: état de la question ed. by Jean Claude Boulanger & Ali Reguigui. Quebec: girsterm.
bibliography
235
Boutin-Quesnel, Rachel, Bélanger, Nycole et al. 1985. Vocabulaire systématique de la terminologie. Cahiers de l’Office de la langue française. Quebec: Gouvernement du Québec. Cabré, M. Teresa & Rigau, Gemma. 1985. Lexicologia i semàntica. Barcelona: Enciclopèdia Catalana. Cabré, M. Teresa. 1986. ‘‘Normes internacionals i metodologia de recerca en la moderna terminologia’’. Tradició i modernitat en l’establiment del llenguatge administratiu, 111–32. Barcelona: Publicacions de la Generalitat de Catalunya. Cabré, M. Teresa. 1987a. ‘‘Normalització terminològica’’. Actes de les II Jornades d’estudi de la llengua normativa, 49–79. Barcelona: Publicacions de l’Abadia de Montserrat. Cabré, M. Teresa. 1987b. ‘‘Les bases de dades terminològiques: especificitat i organització’’. Actas del II Congreso de lenguajes naturales y lenguajes formales. Barcelona: PPU. 47–64. Cabré, M. Teresa. 1987c. ‘‘Les necessitats terminològiques’’. COM ensenyar català als adults 13. 34–7. Barcelona. Cabré, M. Teresa. 1987d. ‘‘L’organisation terminologique de la langue catalane’’. Terminogramme 45. 3–5. Cabré, M. Teresa. 1987e. ‘‘El TERMCAT, Centre de terminologia del català’’. Llengua i Literatura 3. Barcelona. 647–50. Cabré, M. Teresa. 1988a. ‘‘Les bases terminològiques al servei dels professionals’’. NPQ (Notícies per a químics) 301, 7–9. Barcelona. Cabré, M. Teresa. 1988b. ‘‘Enseigner la terminologie, former des terminologues’’. Actes de première rencontre internationale sur l’enseignement de la terminologie. Geneva: Publications de l’Université de Genève. 214–22. Cabré, M. Teresa. 1989a. ‘‘La neologia efímera’’. Miscel·lània Joan Bastardas I, 37–58. Barcelona: Publicacions de l’Abadia de Montserat. Cabré, M. Teresa. 1989b. ‘‘La terminologie catalane: bilan des activités’’. Meta 34, 3. 544–51. Cabré, M. Teresa. 1989c. ‘‘Termcat, centre catalan de terminologie (Recherche et travail terminologique en Catalogne)’’. TermNet News 15. 40–3. Cabré, M. Teresa. 1989d. ‘‘Normes internacionals i metodologia terminològica’’. Llengua i Dret 11. 63–76. Barcelona. Cabré, M. Teresa. 1990a. ‘‘La néologie dans la presse catalane. Premières données d’un observatoire de néologie’’. La Banque des mots, special issue. 75–84. Cabré, M. Teresa. 1990b. ‘‘La lingüística aplicada, avui’’. Cabré & Payrató 1990. 11–30. Cabré, M. Teresa. 1990c. ‘‘Un projet sur la néologie de large diffusion’’. Colóquio de lexicologia e lexicografia. Actas. Lisbon: Universidade de Lisboa. 34–42. Cabré, M. Teresa. 1990d. ‘‘La formación terminológica’’. Actas del I Simposio Latinoamericano de Terminología, 1988. Caracas: Universidad Simón Bolívar. 55–83. Cabré, M. Teresa. 1990e. ‘‘Bancos de terminología: estado actual de BTERM’’. Actas del I Simposio Latinoamericano de Terminología, 1988. Caracas: Universidad Simón Bolívar. 37–54.
236
terminology
Cabré, M. Teresa. 1990f. ‘‘Neologia i text: producció i interpretació de noves paraules’’. Actes de les VII Jornades sobre l’expressió escrita: L’ampliació de vocabulari i la funció lèxica a l’escrit, 37–54. Barcelona: Universitat de Barcelona. Cabré, M. Teresa. 1990g. ‘‘Influència de l’organització dels serveis lingüístics canadencs a Catalunya’’. Barcelona: Llengua i Dret 14. 277–89. Cabré, M. Teresa. 1991a. ‘‘La neologia avui: naixement d’una disciplina’’. Límits 9. 47–64. Cabré, M. Teresa. 1991b. ‘‘Normalitzar la terminologia catalana’’. Miscel·lània d’Enric Casassas i Simó, 73–6. Barcelona: Publicacions de la Universitat Autònoma de Barcelona. Cabré, M. Teresa. 1991c. ‘‘Terminologie ou terminologies? Spécialité linguistique ou domaine interdisciplinaire?’’. Meta 36, 1. 55–63. Cabré, M. Teresa. 1991d. ‘‘Aspectes sobre la neologia: la neologia lèxica a través de dos diaris catalans’’. Miscel·lània Jordi Carbonell. Barcelona: Publicacions de l’Abadia de Montserrat. 359–76. Cabré, M. Teresa. 1991e. ‘‘Os servicios lingüísticos: definición e estructura’’. Cuadernos de lingua 3. 141–50. Cabré, M. Teresa & M. Lorente. 1991. Catàleg de diccionaris catalans (1940–88). Barcelona: Publicacions de la Universitat de Barcelona. Cabré, M. Teresa. 1992a. ‘‘Els serveis lingüístics: definició i estructura’’. Actes de les Primeres Jornades de serveis lingüístics, 19–28. Barcelona: Publicacions de la Generalitat de Catalunya. Cabré, M. Teresa. 1992b. ‘‘La formación de la terminología: uniformización y diversidad’’. Actas del Coloquio Iberoamericano sobre la enseñanza de la terminología, 103–34. Granada: Publicaciones de la Universidad de Granada. Cabré, M. Teresa. 1992c. ‘‘Servicios Lingüísticos y normalización’’. Sendebar 3. 87–105. Cabré, M. Teresa. 1993. ‘‘La terminologia, una qüestió indefugible de la normalització lingüística’’. II Congrès Internacional de llengua catalana. Àrea IV: Lingüística social, 331–8. Palma: Universitat de les Illes Balears. Cabré, M. Teresa. 1994a. A l’entorn de la paraula (I). Lexicologia general. Col·lecció lingüística catalana. Valéncia: Universitat de València. Cabré, M. Teresa. 1994b. A l’entorn de la paraula (II). Lexicologia catalana. Col·lecció lingüística catalana. Valéncia: Universitat de València. Cabré, M. Teresa. 1994c. ‘‘Autour de la norme lexicale’’. Langues et sociétés en contact. Mélanges offerts à Jean-Claude Corbeil ed. by Pierre Martel & Jacques Maurais, 351–61. Tübingen: Max Niemeyer. Cabré, M. Teresa. 1994d. ‘‘Terminologie et dictionnaires’’. Meta 39, 4. 589–97. Cabré, M. Teresa. 1995a. ‘‘On Diversity and Terminology’’. Terminology 2, 1. 1–17. Cabré, M. Teresa, ed. 1995b. ‘‘Lexicografia’’. Caplletra 17. Valencia: Publicacions de l’Abadia de Montserrat & Institut de Filologia Valenciana. Cabré, M. Teresa. Forthcoming. ‘‘Terminologia i diccionaris II’’. To appear in Miscel·lània Colon. Barcelona: Publicacions de l’Abadia de Montserrat. Cabré, M. Teresa. Forthcoming. ‘‘Sobre la diversidad y la terminología’’. To appear in Actas del III Simposio Iberoamericano de Terminología. Barcelona: Publicacions de la Universitat de Barcelona, PPU.
bibliography
237
Cabré, M. Teresa and Lluís Payrató, eds. 1990. Lingüística aplicada: noves perspectives, noves professions, noves orientacions. Barcelona: Publicacions de la Universitat de Barcelona-Caixa de Pensions. Candel, Danielle, Cormier, Monique C. & John Humbley, eds. 1994. ‘‘Hommage à Bernard Quemada. Termes et textes’’. Meta 39, 4. 535–826. Candel, Danielle & John Humbley. 1994. ‘‘Oralisation de sigles en aéronautique’’. LINX, Nanterre: Université de Paris X. 133–52. Cayer, Micheline. 1990. ‘‘Pour une meilleure compréhension de la méthode de recherche terminologique de l’Office de la langue française’’. Terminologies nouvelles 3. 72–85. Célestin, Tina, Godbout, Gilles & Pierrette Vachon-L’Hereux. 1984. Méthodologie de la recherche terminologique ponctuelle. Essai de définition. Quebec: Gouvernement du Québec. Centre d’études du lexique. 1990. ‘‘La définition’’. Langue et langage. Paris: Larousse. Chaumier, Jacques. 1988a. ‘‘Travail et méthodes du/de la documentaliste’’. Formation permanente en sciences humaines. Paris: Les Éditions ESF - Entreprise Moderne d’Édition et Librairies Techniques. Chaumier, Jacques. 1988b. Le traitement linguistique de l’information. Paris: Entreprise Moderne d’Édition. Chaffin, R., Herrmann, D. J. & M. Winston. 1988. ‘‘An Empirical Taxonomy of Part-whole Relations: Effects on Part-whole Relation Type on Relation Identification’’. Language and Cognitive Processes 3, 1. 17–48. Ciobanu, Georgeta. 1998. Elemente de terminologie. Editura Mirton Timisoara. Clas, André, dir. 1985. Guide de recherche en lexicographie et terminologie. Paris: Agence de coopération culturelle et technique. Clas, André. 1987. ‘‘Une matrice terminologique universelle: la brachygraphie gigogne’’. Meta 32, 3. 347–55. Clas, André & Paul Horguelin. 1991. Le français, langue des affaires. Montreal: McGrawHill. Collignon, Lucien & Michel Glatigny. 1978. Les dictionnaires. Initiation à la lexicographie. Paris: CEDIC. Colloque: Les stages en traduction et en terminologie. 1982. Quebec: girsterm. ‘‘Les commissions de terminologie’’ (1995). Terminometro, núm. hors-série. Paris: Union Latina. 9–10. Corbeil, Jean-Claude. 1980. L’aménagement linguistique du Québec. Col. Langue et société. Montréal: Guérin. Corbeil, Jean-Claude. 1983a. ‘‘Pédagogie de l’implantation de la terminologie. Synthèse présentée par Jean Claude Corbeil’’. Actes du quatrième colloque OLF-STQ de terminologie. Aménagement de la terminologie: diffusion et implantation, Quebec: Gouvernement du Québec. 179–83. Corbeil, Jean-Claude. 1983b. ‘‘Préface de la deuxième édition’’. Introduction à la terminologie by Guy Rondeau. Chicoutimi: Gaëtan Morin. 25–34. Corbeil, Jean-Claude. 1988. ‘‘Les terminologies devant Babel’’, Actes du colloque Terminologie et technologies nouvelles. Quebec: Gouvernement du Québec. 49–62.
238
terminology
Corbeil, Jean-Claude. 1990a. Les industries de la langue: un domaine à la recherche de luimême. Quebec: Gouvernement du Québec/Conseil de la langue française. Corbeil, Jean-Claude. 1990b. ‘‘Comment orienter l’usage d’une langue’’. Cabré & Payrató 1990. 79–85. Corbeil, Jean-Claude. 1991. ‘‘Terminologie et banques de données d’information scientifique et technique’’. Meta 36, 1. 128–34. Corbeil, Jean-Claude, ed. 1973. ‘‘Guide de travail en terminologie’’. Cahiers de l’Office de la langue française, 20. Quebec: Gouvernement du Québec. Corbin, Danielle. 1983. ‘‘Le monde étrange des dictionnaires (4): la créativité lexicale, le lexicographe et le linguiste’’. Lexique 2. 43–68. Corbin, Danielle. 1984. ‘‘Méthodes en morphologie dérivationnelle’’. Cahiers de lexicologie 45, 1984-I. 3–17. Corbin, Danielle. 1987. Morphologie dérivationnelle et structuration du lexique. Tübingen: Max Niemeyer. 2 vol. Corbin, Danielle & Pierre Corbin. 1980. ‘‘Le monde étrange des dictionnaires (1): Les ‘‘marques d’usage’’ dans le Micro Robert’’ Synonymies, Bulletin du Centre d’Analyse du duscours, 4. 237–324. Corbin, Danielle. 1984. ‘‘Le monde étrange des dictionnaires (5): DÉRIVER’’. Cahiers de lexicologie 45, 1984–II. 33–46. Corbin, Danielle. 1985. ‘‘Le monde étrange des dictionnaires (6): le commerce des mots’’. Lexique 3. 65–124. Cormier, Monique C. & Jacques Lethuillier, eds. 1991. ‘‘La terminologie dans le monde: orientations et recherches / Terminology in the World. Trends and Research’’. Meta 36, 1. Cormier, Monique C. & John Humbley, eds. 1993. ‘‘L’Europe au rythme de l’anglais’’. Circuit. Montreal: Ordre des traducteurs et interprètes agréés du Québec, 41. 2–17. Coseriu, Eugenio. 1973. Teoría del lenguaje y lingüística general. Cinco estudios. Madrid: Editorial Gredos. Cottez, Henri. 1989. Dictionnaire des structures du vocabulaire savant. Élements et modèles de formation. Series ‘‘Les usuels du Robert’’. Paris: Dictionnaires Le Robert. Couteaux, Paul-Marie. 1988. ‘‘Les effets de la compétition économique sur la situation du français’’. Actes du colloque Terminologie et techonologies nouvelles. Quebec: Gouvernement du Québec. 39–48. Cruse, D. A. 1986. Lexical Semantics. Cambridge: Cambridge University Press. Czap, Hans & Wolfang Nedobity, eds. 1990. TKE’90: Terminology and Knowledge Engineering. Proceedings. Second International Congress on Terminology and Knowledge Engineering. Frankfurt: Indeks Verlag, 2 vol. Dahlberg, Ingetraut. 1983. ‘‘Terminological Definitions: Characteristics and Demands’’. Duquet-Picard 1983. Quebec: girsterm. 15–34. Dahlberg, Ingetraut. 1993. ‘‘Faceted Classification and Terminology’’. TKE’93 Terminology and Knowledge Engineering. 225–34. Frankfurt-am-Main: Indeks Verlag. Diki-Kidiri, M., Joly, H. & C. Murcia. 1981. Guide de la néologie. Paris: Conseil international de la langue française. Dion, Louise. 1984a. ‘‘La notion en terminologie’’. Dion 1984b. 15–28.
bibliography
239
Dion, Louise, ed. 1984b. Travaux de terminologie, 3. Quebec: girsterm. Drodz, Lubomir. 1981. ‘‘Science terminologique: objet et méthode’’. Rondeau & Felber 1981. 115–31. Dubois, Claude & Jean Dubois. 1971. Introduction à la lexicographique: le dictionnaire. Paris: Larousse. Dubois, Jean. 1966. ‘‘Les problèmes du vocabulaire technique’’. Cahiers de lexicologie 9, 1966-II. 103–12. Dubuc, Robert. 1985. Manuel pratique de terminologie. 2nd edn. Brossard (Quebec): Linguatech. Dubuc, Robert. 1992. Manuel pratique de terminologie. 3rd edn. reviewed and updated. Brossard (Quebec): Linguatech. Duquet-Picard, Diane, ed. 1983. Problèmes de la définition et de la synonymie en terminologie. Actes du Colloque international de terminologie 7. (Université Laval, Quebec, 23–27 May 1982). Quebec: girsterm. Engel, Gert & Heribert Picht. 1990. ‘‘New Professional Profiles in Knowledge Engineering and Knowledge Transfer’’. Czap & Nedobity 1990 I. 47–71. Felber, Helmut. 1984a. Terminology Manual. Paris: Unesco, Infoterm. Felber, Helmut. 1984b. ‘‘L’activité du CT37 de l’ISO Terminologie (principes et coordination)’’’, Terminogramme 26–7. 6–8. Felber, Helmut. 1987. Manuel de terminologie. Paris: Unesco, Infoterm. Galinski, Christian. 1982. ‘‘Standardization in Terminology. An Overview’’. Terminology for the Eighties. Infoterm Series 7, 186–205. München: K.G. Saur Verlag. Galinski, Christian. 1990. ‘‘Recent Developments of Terminology. From the Theory of Terminology via Knowledge Theory to Terminological Knowledge Engineering’’. Cabré & Payrató 1990. 87–91. Galinski, Christian & Wolfang Nedobity. 1989. ‘‘International Terminology Standardization’’. Infoterm 1. 1–8. Galisson, Robert. 1978. Recherches de lexicologie descriptive. La banalisation lexicale, ‘‘Nathan-Université Series’’. Paris: Nathan. Gambier, Yves. 1991a. ‘‘Présupposés de la terminologie: vers une remise en cause’’. Cahiers de linguistique sociale 18. 31–58. Gambier, Yves. 1991b. ‘‘Travail et vocabulaires spécialisés: prolégomènes à une socioterminologie’’. Meta 36, 1. 8–15. Gaudin, François. 1991. ‘‘Terminologie et travail scientifique: mouvement des signes, mouvement des connaissances’’. Cahiers de linguistique sociale 18. 111–31. Gaudin, François. 1993. Pour une socioterminologie. Des problèmes sémantiques aux pratiques institutionnelles. Rouen: Université de Rouen. Gaudin, François, ed. 1995. ‘‘Usages sociaux des termes: théories et terrains’’. Meta 40, 2. 191–329. Gaudin, François & Allal Assal, eds. 1991. ‘‘Terminologie et sociolinguistique’’. Cahiers de linguistique sociale, 18. Rouen: Université de Rouen. Goffin, Roger. 1985. ‘‘La science terminologique’’. Terminologie et traduction 2. 9–29. Gouadec, Daniel. 1987. ‘‘Les horizons de la terminotique’’. Meta 32, 2. 130–8.
240
terminology
Gouadec, Daniel. 1990. Terminologie. Constitution des données, ‘‘afnor gestion’’. Paris: AFNOR. Gouadec, Daniel, dir. 1993. Terminologie et terminotique. Outils, modèles et méthodes. Actes de la première Université d’Autonme en Terminologie. (Rennes, September 1992). Paris: La Maison du dictionnaire. Gouadec, Daniel. 1994. Terminologie et phraséologie. Acteurs et aménageurs. Paris: La Maison du dictionnaire. Guespin, Louis. 1991. ‘‘La circulation terminologique et les rapports science, technique, production’’. Cahiers de linguistique sociale 18. 59–79. Guespin, Louis & Foued Laroussi. 1989. ‘‘Glotto-politique et standardisation terminologique’’. La Banque des mots, special issue. 5–21. Guilbert, Louis. 1971a. ‘‘La néologie scientifique et technique’’. La Banque des mots 1. 45–54. Guilbert, Louis. 1973. ‘‘La spécificité du terme scientifique et technique’’, Langue française 17. 5–17. Guilbert, Louis, ed. 1974. ‘‘La néologie lexicale’’. Langages 36. Guilbert, Louis. 1975. ‘‘La créativité lexicale’’. Langue et langage. Paris: Larousse. Guilbert, Louis & Jean Peytard, eds. 1973. ‘‘Les vocabulaires techniques et scientifiques’’. Langue française 17. Halaoui, Nazam. 1990. ‘‘Questions de méthode en terminologie des langues africaines’’. Terminologies nouvelles 3. 5–24. ‘‘Harmonisation des méthodes en terminologie’’. 1990. Terminologies nouvelles. Brussels: Réseau international de terminologie, 3. Hoffmann, Lothar. 1976. Kommunikationsmittel Fachsprache: Eine Einführung. Berlin: Akademie Verlag. Hoffmann, Lothar. 1979. ‘‘Towards a Theory of LSP. Elements of a Methodology of LSP Analysis’’. Fachsprache 1, 1–2. 12–7. Hoffmann, Lothar. 1984. ‘‘Seven Roads to LSP’’. Fachsprache 6, 1–2. 28–38. Hohnhold, Ingo. 1990. Übersetzungsorientierte Terminologiearbeit. Eine Grundlegung für Praktiker. Stuttgart: InTra. Humbley, John. 1987. ‘‘L’emprunt sémantique dans la terminologie de l’informatique’’. Meta 32, 3. 321–5. ‘‘Les industries de la langue’’. 1990. Terminogramme 55. ISO 1087 (1990): Terminology—Vocabulary = Terminologie—Vocabulaire, Geneva, International Organization for Standardization, 17 pp. ISO 1951 (1973): Lexicographical Symbols, particularly for use in Classified Defining Vocabularies, Geneva, International Organization for Standardization, 30 pp. ISO 6156 (1987): Magnetic Tape Exchange Format for Terminological/Lexicographical Records (MATER), Geneva, International Organization for Standardization, 25 pp. ISO 639 (1988): Codes for the Representation of Names of Languages, Geneva, International Organization for Standardization, 17 pp. ISO 704 (1987): Principles and Methods of Terminology, Geneva, International Organization for Standardization, 16 pp.
bibliography
241
ISO/R 860 (1968): Unification internacionale des notions et des termes, Geneva, International Organization for Standardization, 12 pp. Jakobson, Roman. 1963. Essais de linguistique générale, translated and prefaced by Nicolas Ruwet. Paris: Éditions de Minuit. Kerpan, Nada. 1980. ‘‘La documentation et les services linguistiques’’. Meta 25, 1. 7–10. Kjaer, Anne Lise. 1990. ‘‘Phraseology Research —State-of-the-Art. Methods of Describing Word Combinations in Language for Specific Purposes’’. IITF Journal 1, 1–2. 3–20. Kocourek, Rostislav. 1982. La langue française de la technique et de la science. Wiesbaden: Brandstetter. Kocourek, Rostislav. 1991a. La langue française de la technique et de la science. Vers une linguistique de la langue savante. Wiesbaden: Brandstetter. Kocourek, Rostislav. 1991b. ‘‘La nature du français technique et scientifique’’. In Kucera, A. Clas & J. Baudot. Dictionnaire compact des sciences et des techniques. Volume II. Alemand-français. Wiesbaden: Brandstetter. Kocourek, Rostislav. 1991c. ‘‘Textes et termes’’. Meta 36, 1. 71–6. Kocourek, Rostislav, Monique C. Cormier, & Lise Lapierre, eds. 1994. ‘‘Langue et terminologie spécialisée. Specialized Language and Terminology’’. Initiales. Halifax: Université Dalhousie, 14. Kromp, Richard. 1987. ‘‘Formation des terminologues et besoins de l’entreprise’’. Table ronde sur l’enseignement de la terminologie à l’université: état de la question ed. by JeanClaude Boulanger & Ali Reguigui. Quebec: girsterm. 111–8. Lapalme, Guy. 1988. ‘‘Les systèmes experts en terminologie’’. Terminogramme 46. 14–5. Lerat, Pierre. 1988. ‘‘Terminologie et sémantique descriptive’’. La Banque des mots, special issue. 11–30. Lerat, Pierre. 1989. ‘‘Les fondements théoriques de la terminologie’’. La Banque des mots, special issue. 51–62. Lerat, Pierre. 1990a. ‘‘L’hyperonymie dans la structuration des terminologies’’. Langages 98. 79–86. Lerat, Pierre. 1990b. ‘‘L’analyse morphologique des termes nouveaux’’. La Banque des mots, special issue. 23–34. Lerat, Pierre. 1995. Les langues spécialisées, col. ‘‘Linguistique nouvelle’’. Paris: PUF. ‘‘Les logiciels de terminologie’’. 1991. La Banque des mots, special issue. Lyons, John. 1977. Semantics, Volumes I and II. Cambridge: Cambridge University Press. Manu, Adrian. 1984. ‘‘Les organismes internationaux de normalisation terminologique’’. Terminogramme 26–7. 3–6. Martel, Pierre & Jacques Maurais, eds. 1994. Langues et sociétés en contact. Mélanges offerts à Jean-Claude Corbeil. Tübingen: Max Niemeyer. Maurais, Jacques. 1987. ‘‘La place de l’enseignement de la terminologie dans l’aménagement linguistique du Québec’’. Table ronde sur l’enseignement de la terminologie à l’université: état de la question ed. by Jean-Claude Boulanger & Ali Reguigui. Quebec: girsterm. 139–45. Melby, Alan. 1991. ‘‘Des causes et des effets de l’asymétrie partielle des réseaux sémantiques liés aux langues naturelles’’. Cahiers de lexicologie 58. 6–43.
242
terminology
Mounin, Georges. 1979. ‘‘La linguistique comme science auxiliaire dans les disciplines juridiques’’. Meta 24, 1. 9–17. ‘‘La normalisation terminologique’’. 1984. Terminogramme 26–7. Odgen, Charles K. & I. A. Richards. 1946. The Meaning of Meaning: a study of the influence of language upon thought and of science of symbolism. New York: Harcourt, Brace and World. Office de la langue française. 1980. Énoncé d’une politique relative à l’emprunt de formes linguistiques étrangères. Quebec: Gouvernement du Québec. Office de la langue française. 1985. Énoncé d’une politique linguistique relative aux québécismes. Quebec: Gouvernement du Québec. Office de la langue française. 1989. Document synthèse sur la méthode de recherche terminologique systématique de l’Office de la langue française pour le séminarie du Réseau international de néologie et de terminologie. Quebec: Office de la langue fançaise, Direction des productions linguistiques et terminologiques. Otman, Gabriel. 1989. ‘‘Terminologie et intelligence artificielle’’. La Banque des mots, special issue. 63–95. Otman, Gabriel. 1991. ‘‘Des ambitions et des performances d’un système de dépouillement terminologique assisté par ordinateur’’. La Banque des mots, special issue. 59–96. Pagès, Alain. 1984. ‘‘L’apparition néologique’’. La Banque des mots 28. 131–40. Paradís, Claude & Pierre Auger. 1987. ‘‘La terminotique ou la terminologie à l’ère de l’informatique’’. Meta 32, 2. 102–10. Pavel, Silvia. 1987. ‘‘La terminologie de l’avenir: un dialogue Homo Sentiens—Machina Sapiens’’. Meta 32, 2. 124–9. Perron, Jean. 1988. ‘‘Le dépouillement terminologique assisté par ordinateur’’. Terminogramme 46. 24–6 & 29–31. ‘‘Phraséologie. Actes du séminaire international (Hull, mai 1993)’’. 1993. Terminologies nouvelles, Brussels: Réseau international de néologie et de terminologie, 10. Picht, Heribert. 1984. ‘‘Breve historia y situación actual de la teoría, la investigación y las prácticas terminológicas’’. Actas del primer seminario nacional de terminología (Caracas, 11–15 April 1983). Caracas: Universidad Simón Bolívar. 24–32. Picht, Heribert. 1987. ‘‘Terms and their LSP Environment — LSP Phraseology’’. Meta 32, 2. 149–55. Picht, Heribert & Jennifer Draskau. 1985. Terminology: An Introduction. Guildford: The University of Surrey. Plante, Pierre & Lucie Dumas. 1988. ‘‘Le dépistage automatique des termes’’. Terminogramme 46. 26–8. Plénat, Marc, ed. 1994. ‘‘Les sigles’’. LINX, 30. Quemada, Bernard. 1978. ‘‘Technique et langage’’. Histoire des techniques ed. by Bertrand Gille. ‘‘La Pléiade’’, 1146–240. Paris: Gallimard. Quemada, Bernard. 1987. ‘‘Notes sur lexicographie et dictionnairique’’. Cahiers de lexicologie 51, 1987-II. 229–42. Quist, Christian. 1984. ‘‘Characteristics of Special Subject Languages in the Technical Language Group’’, ALSED–LSP Newsletter 7, 1. 229–42.
bibliography
243
Régent, Odile. 1992. ‘‘Pratiques de communication en médecine: contextes anglais et français’’, Ethnolinguistique de l’écrit. Langages 26, 105. 66–75. Rey, Alain. 1965. ‘‘À propos de la définition lexicographique’’. Cahiers de lexicologie 6, 1965-II. 67–80. Rey, Alain. 1976. ‘‘Néologisme: un pseudo-concept?’’. Cahiers de lexicologie 28, 1. 3–17. Rey, Alain. 1977. Le lexique: images et modèles, du dictionnaire à la lexicologie. Paris: Armand Colin. Rey, Alain. 1985. ‘‘La néologie: un problème de création, de diffusion et d’acceptation’’. Actes du colloque national sur les services linguistiques (Ottawa, October 1984). Ottawa: Ministre des Approvisionnements et Services Canada. 231–56. Rey, Alain. 1988a. ‘‘Dictionnaire et néologie’’. Actes du colloque Terminologie et technologies nouvelles. Quebec: Gouvernement du Québec. 279–89. Rey, Alain. 1988b. ‘‘Les fonctions de la terminologie: du social au théorique’’. Actes du sixième colloque OLF-STQ de terminologie. L’ère nouvelle de la terminologie. 87–108. Quebec: Gouvernement du Québec. Rey, Alain. 1988c. ‘‘Terminologie et lexicographie’’. Parallèles 10. 27–35. Rey, Alain. 1993. La terminologie: noms et notions. ‘‘Que sais-je?’’, 1780. Paris: PUF. Rey, Alain. 1995. Essays on Terminology, translated and edited by Juan C. Sager. Amsterdam & Philadelphia: John Benjamins. Rondeau, Guy & Helmut Felber, ed. 1981. Textes choisis de terminologie. I Fondements théoriques de la terminologie. Quebec: girsterm. Rondeau, Guy & Juan C. Sager, eds. 1986. Termia 84. Terminologie et coopération internationale. Quebec: girsterm. Rousseau, Louis-Jean. 1988. ‘‘Néologie et problèmes de normalisation’’. Actes du colloque Terminologie et technologies nouvelles. Quebec: Gouvernement du Québec. 333–48. Sager, Juan Carlos. 1988. ‘‘The Status of Terminology as an Independent Discipline’’. Parallèles. Cahiers de l’École de Traduction et d’Interprétation de l’Université de Genève 10. 21–3. Sager, Juan Carlos. 1990. A Practical Course in Terminology Processing. Amsterdam & Philadelphia: John Benjamins. [2nd edn. 1997]. Sager, Juan Carlos, Dungworth, David & Peter F. McDonald. 1980. English Special Languages. Principles and Practice in Science and Technology. Wiesbaden: Brandstetter. [2nd edn. 1997]. Sager, Juan Carlos & R. L. Johnson. 1980. ‘‘Standardization of Terminology in a Model of Communication’’. International Journal of the Sociology of Language 23. 81–104. Schaetzen, Caroline de, ed. 1989. Terminologie diachronique. Actes du colloque (Brussels, 25–26 March 1988). Paris: Conseil international de la langue française. Spillner, Bernd. 1992. ‘‘Textes médicaus français et allemands. Contribution à une comparaison interlinguale et interculturelle’’. Langages 26, 105. 42–65. Strehlow, R. A., ed. 1988. Standardization of Technical Terminology: Principles and Practices. Second Volume. Philadelphia: ASTM. Table ronde sur les problèmes de découpage du terme. 1979. Quebec: Éditeur officiel du Québec.
244
terminology
Tackels, Stéphane. 1990. Typographie et terminologie. Guide de présentation des travaux terminologiques. Quebec: Gouvernement du Québec. ‘‘La terminologie en France’’. 1995. Terminometro, numéro hors série. Paris: Union Latina. Temmerman, R., Feke, S. & L. Luyten. 1990. Terminologie een methode. Inleiding tot theorie en praktijk van systematische terminografie. Amsterdam: Acco Leuven. Terminologie et enseignement des langues. Actes du colloque international organisé par l’Association Européenne des Linguistes et des Professeurs de Langues (AELPL) (31 January–1 February 1991), Cergy-Pontoise. 1991. ‘‘Paroles & Actes’’. Paris: LA TILV ed. ‘‘Terminologie & informatique’’. 1988. Terminogramme 46. ‘‘Terminologie et linguistique de spécialité. Études de vocabulaires et textes spécialisés’’. 1994. ALFA 7/8, 1994–5. Halifax: Université Dalhousie. Trudel, Michèle. 1991. ‘‘L’indexation multiple dans la Banque de terminologie du Québec’’. Terminogramme 61. 4–8. Varantola, Krista. 1986. ‘‘Special Language and General Language: Linguistic and Didactic Aspects’’. ALSED–LSP Newsletter 9, 2 (23). 10–9. Wijnands, Paul. 1985. ‘‘Pour une redéfinition du néologisme lexicographique’’ La Banque des mots 29. 13–34. Wüster, Eugen. 1955. Bibliography of Monolingual Scientific and Technical Glossaries/ Bibliographie de vocabulaires scientifiques et techniques monolingues. Paris: Unesco, 2 vols. Wüster, Eugen. 1968. The Machine Tool: An Interlingual Dictionary of Basic Concepts. London: Technical Press. Wüster, Eugen. 1974. The Road to Infoterm. Infoterm Series 1. Pullach/Munich: Verlag Dokumentation Saur K.G. Wüster, Eugen. 1976. ‘‘La théorie générale de la terminologie—un domaine interdisciplinaire impliquant la linguistique, la logique, l’ontologie, l’informatique et les sciences des objets’’. Essai de définition de la terminologie. Actes du colloque international de terminologie. Quebec, Manoir du Lac Delage du 5 au 8 octobre 1975. Quebec: L’Éditeur officiel du Québec. 49–57. Wüster, Eugen. 1981. ‘‘L’étude scientifique générale de la terminologie, zone frontalière entre la linguistique, la logique, l’ontologie, l’informatique et les sciences des choses’’. Rondeau & Felber 1981. 55–114. Dictionaries and texts referred to in the body of the book Encyclopaedia Britannica, 15th edn. 1998. Chicago: Encyclopaedia Britannica. Glazier, Stephen. 1992. Random House Word Menu. New York: Random House, Inc. Newton, Isaac. 1990. Axioms or Laws of Motion. Reprinted in Encyclopaedia Britannica 1989. Yearbook of Science and the Future. Chicago: Encyclopaedia Britannica. Quirk, Randolph and Gabriele Stein. 1990. English in Use. London: Longman. The American Heritage Dictionary of the English Language. 1992. Boston: Houghton Mifflin Cop. Webster’s Third New International Dictionary of English. 1961. Springfield, Mass: MerriamWebster.
Index abbreviation–34, 73, 86, 87, 110, 142, 149, 156, 200 acronym–61, 86, 92, 93, 207 affix–34, 73, 84, 85, 89, 92 antonym–137, 144 artificial intelligence–44, 52–54, 161, 162, 166, 167, 173, 175, 192, 193, 222, 223 artificial language–59–61 banalization–63 blending–89 borrowing–4, 88–90, 94, 158, 201, 206–8, 210, 211 classes of concepts–88, 135 clipping–87, 92, 93, 110, 207 cognition–39, 41, 42, 44 compounding–92, 94 computerized terminology–160–93, 204 concept–33, 34, 37–8, 40–5, 52, 81, 87–8, 95ff, 116, 167, 173, 194 conceptual class–87, 88 conceptual field–43, 99, 100, 103, 104, 106, 133, 144 conceptual relationship–101, 103 conceptual structure–43, 51, 104, 112, 126, 132, 133, 135, 162 conceptual system–8, 18, 34, 38, 44, 88, 104, 108, 117, 133 conceptual unit–40 consultation service–159 correspondence record–121, 127, 151 cross-reference–108, 124, 139, 142, 144, 146, 147, 149, 150, 165 database–4, 23, 24, 51–5, 117–20, 127, 129, 151, 153, 155, 162–6, 170–6, 181, 184, 186–93, 202 data bank–6, 14, 22, 24, 48, 51, 54, 156, 158–60, 163–5, 168–93, 219 defining context–138, 139, 147 definition–31, 95–8, 104–9, 141
deprecated term–145, 149 derivation–82, 88, 89, 92–4, 200, 207, 210, 212 descriptive work–132 descriptor–9, 50, 51, 107, 135, 141, 160, 165, 172 designation–38, 40, 82–3, 92–5, 116, 194 dialect–29, 31, 32, 57, 61, 64, 65, 76, 77, 88, 110, 142, 154, 206 dictionary–31, 36–8, 54, 55, 107, 108, 110, 148, 168, 177 dictionary entry–8, 80, 108, 147, 204 discourse–27, 33, 36, 37, 56–8, 63, 64, 67, 71, 77, 82, 83, 87, 91, 111–13, 156, 201, 206 documentation–6, 44, 50–4, 66, 117, 123, 124, 133–5, 146, 150, 151, 155, 156, 159, 165, 168, 173, 177, 201, 217, 219, 222, 231 document bank–165 document corpus–121 equivalence–48 expert–6, 18, 34, 36–8, 45, 46, 55, 62, 64, 65, 67, 70, 76, 77, 88, 95, 111, 114, 118, 121, 133, 136, 137, 147, 148, 150–1, 156–9, 168, 169, 173, 180, 201, 219, 223 extraction–117, 121–3, 127, 130, 133–9, 141, 145, 151, 161, 165, 173, 181, 184, 192 file–53, 54, 118, 145, 147, 151, 153, 165, 166, 168, 170, 185–6, 189, 190 terminological file–54, 168 general language–7, 34, 36, 40, 45, 46, 49, 51, 56, 58–85, 89, 90, 93, 108, 111–14, 119, 120, 137, 140, 194, 200, 205–8, 212, 218 general language vocabulary–40, 84, 114, 200 generic concept–100 glossary–107, 115, 147, 149 homonymy–40, 61, 108, 111, 144, 149, 194, 200
246
index
index–51, 147, 149 indexing–50, 51, 161, 165, 171 initialism–34, 86, 92, 93, 110, 156, 207 interdisciplinary–7, 25–55, 216 international standard–89, 116, 142, 197, 198, 199 internationalization–1, 23, 34 international committee–116, 198 ISO–19, 95, 104, 107, 116, 119, 145, 195, 197, 198, 201, 202, 212 knowledge acquisition–44, 134, 167 language language planning–3, 5, 6, 14, 16, 17, 19, 21, 48–51, 201, 204, 208, 209, 212, 215–17, 222 language service–22, 214–23 language standardization–3–4, 5, 10, 13, 16, 19, 20, 21, 49, 132, 176, 199, 208, 211, 215, 216, 217, 221, 223 lexeme–31, 37, 56, 73, 91–3 lexical component–30, 82, 94 lexicalization–93, 207 lexicography–7, 9, 28, 30–2, 34, 37–8, 99, 108, 111, 115, 129, 140, 163, 205, 208 lexicology–8–10, 29–31, 32–7, 115, 222 lexicon–4, 7, 11, 15, 19, 20, 26, 28, 29, 32–5, 42, 43, 46, 49, 50, 62, 69, 80, 81, 83–5, 87, 92, 93, 109, 111, 114, 115, 119, 148, 149, 176, 204, 220 linguistics–7–10, 13, 20, 22, 25–9, 32–4, 36, 52, 54, 61–2, 83, 103, 111, 112, 150, 160, 161, 166, 167, 169, 208, 211, 222 loan translation–94, 201, 207 logical relationship–8–9, 44, 100–3 methodology–1, 2, 5, 8, 13, 17, 18, 28, 32, 35, 37, 118, 136, 148, 150, 161–4, 208 mini-search–152 monolingualism–3, 216 monolingual record–124, 151 monolingual search–129, 130–51 morpheme–56, 82–5, 90, 92, 94, 101 morphology–32, 46, 62, 82, 84–5, 204 multidimensional nature–4, 45, 58, 205 multilingual search–129, 151–2 neoclassical form–84, 88, 89, 92 neologism–50, 116, 119, 120, 145, 150, 167, 204–11 neology–18, 130, 203–13 nomenclature–59, 110, 177
nominalization–73, 88 onomasiological approach–8, 38, 44, 162 ontological relationship–38, 44, 45, 100, 103 phoneme–56, 84 phrase–27, 56, 57, 73, 75, 85–7, 90–2, 137, 161, 173, 177, 180, 206, 207, 212, 218 phraseology–79, 80, 82, 91 polysemy–40, 61, 108, 109, 111, 194, 195 pragmatics–27, 36, 46, 68, 70, 112, 211, 222 prefix–84, 92, 94, 207 prescriptive work–10, 33, 38, 132, 142, 144, 147, 150, 178, 196, 211 protocol for use–121 query–127–9, 152–9, 171, 175, 176, 186–9, 191 query record–121, 127, 129, 153, 155 record–54, 93, 107, 120, 121–9, 136–40, 142, 144–6, 151, 157, 159, 165, 168–72, 181, 184–6, 188–9 reference bibliography–133 reference documentation–123, 134–5 register (see record) representation–40–2, 46, 53, 75, 83, 104, 121, 127, 133, 140, 145, 146, 149, 150, 161, 164, 167, 168, 170, 173, 175, 181, 196, 202 search ad-hoc search–121, 127, 129, 152–9 systematic search–121, 129–52 semantics–8–10, 26, 32, 39, 51, 154, 222 semasiological approach–8, 38 source–31, 116–17, 124, 140–1, 181, 185 special language–5, 7, 11, 13, 17, 23, 45, 48, 56–80, 84, 87, 91, 110, 120, 124, 169, 173, 200, 204, 206, 208, 212 special language vocabulary–115, 177, 208 special subject special subject field–36–7, 65ff, 80, 220–2 special subject subfield–80, 140, 156, 197 special subject text–65 standard–5, 10, 17, 63, 110, 116, 132, 139, 145, 151, 154–7, 170, 177, 180, 191, 195, 196, 200, 201, 209 standardization (see also language standardization)–2, 5, 11, 13, 16, 18, 20, 21, 32, 33–4, 37, 38, 43, 49, 51, 83, 104, 120, 151, 156, 157, 194–213, 214, 215, 218
index standardization plan–49, 215, 218 standardization process–3, 4, 11, 17, 176, 194, 201, 215, 216 subject field (see also special subject field)–140, 147, 148, 156–8, 162, 166, 172, 178, 185, 201 subsystem–7, 29, 39, 42, 56, 66, 81, 92 suffix–84, 92, 94, 188, 207 superordinate concept–68, 144 symbol–8, 34, 41, 42, 50, 59, 71, 73, 81, 87, 101, 110, 120, 142, 164, 167, 196 synonymy–108–10, 142, 194, 200, 206, 213 technical committee–14, 119, 198, 201, 202 term extraction–117, 133, 135, 136–7, 181, 192, 231 terminography–18, 19, 37, 38, 104, 115–59, 165, 201, 211 terminographer–7–8, 12, 115, 116, 119, 222 terminological data bank–156, 175–193 terminological file–54 terminological unit–8, 25, 37, 38, 80–114
247
terminological record–116, 121, 123–7, 136, 139, 140, 141, 144–5, 146, 147, 149, 151, 165, 184 terminologist–6ff, 18ff, 34, 38, 41ff, 51, 54, 94–5, 111ff, 133, 147, 152ff, 159–66, 172, 173, 201, 209, 214–23 thesaurus–9, 31, 51, 55, 135, 172 translation–12, 13–14, 16, 17, 22, 32, 47–8, 50, 54, 55, 94, 108, 115, 160, 161, 168, 174, 176, 192, 201, 207, 218, 219 truncation–86, 92–3 unit of knowledge–43 univocal nature–108 variation–76 vocabulary–4, 13, 17, 22, 32, 40, 43, 64, 70, 73, 76, 84, 114, 119, 148, 149, 200, 202, 209 working methods–5, 6, 9, 12, 53, 112, 129–59, 163, 223 work schedule–134, 136, 169 work station–54, 160, 168, 174 working document–146
Author Index Andersen, Aldean–193 Auger, Pierre–5, 12, 15, 19, 49, 50, 130, 160, 161, 173, 199 Austin, John L.–79
Kjaer, Anne Lise–229 n. 15 Kocourek, Rostislav–59, 59, 61, 62, 66, 69, 75, 228 n. 2 Kromp, Richard–222
Beaugrande, Robert de–61, 62, 68, 70 Boulanger, Jean-Claude–24, 173, 203, 204, 220 Boutin-Quesnel, Rachel–151, 226
Lapalme, Guy–55
Cabré, M. Teresa–32, 36, 112, 118, 119, 148 Cayer, Micheline–19, 22 Célestin, Tina–152 Corbeil, Jean-Claude–12, 173, 214, 223 Coseriu, Eugenio–88 Diki-Kidiri, M.–208 Draskau, Jennifer–45, 62, 64, 65, 67, 78, 129 Drodz, Lubomir–13 Dubuc, Robert–9, 14, 18, 116
Manu, Adrian–195 Maurais, Jacques–15, 220, 222 Melby, Alan–169 Murcia, C.–208 Nedobity, Wolfang–202 Otman, Gabriel–162 Pavel, Silvia–175 Payrató, Lluís–112 Picht, Heribert–45, 62, 64, 65, 67, 78, 129, 219, 225 n. 1, 229 n. 15 Quemada, Bernard–61, 62
Engel, Gert–219 Felber, Helmut–19, 39, 178, 180, 194, 226 n. 9, 228 n. 16 Galinski, Christian–52, 202 Galisson, Robert–63, 228 n. 3 Gambier, Yves–112, 211, 230 n. 16 Gaudin, François–230 n. 16 Godbout, Gilles–152 Goffin, Roger–11, 28, 173 Gouadec, Daniel–80, 168, 176 Guespin, Louis–195, 199, 200, 201, 230 n. 16 Guilbert, Louis–11, 204 Halaoui, Nazam–17, 115, 225 n. 6 Hoffmann, Lothar–61, 228 n. 4 Jakobson, Roman–46, 74, 75 Joly, H.–208
Rey, Alain–1, 15, 61, 112, 205, 209 Richards, I. A.–40 Rondeau, Guy–5, 6, 24, 50, 61, 68, 70, 176, 178, 180, 181, 196, 204, 206, 225 n. 2, 228 n. 16, 228 n. 2, 231 n. 9 Rousseau, Louis-Jean–130 Sager, Juan Carlos–6, 9, 10, 24, 42, 43, 46, 56, 62, 64, 70, 163, 164, 165, 176, 177, 178, 180, 181, 227 n. 19, 228 n. 2, 228 n. 16 Tackels, Stéphane–147 Vachon-L'Hereux, Pierrette –152 Varantola, Krista–46, 67, 71, 76 Wüster, Eugen–2, 5, 7, 8, 9, 12, 13, 16, 18, 25, 33, 34, 40, 52, 53, 81, 198, 210, 225 n. 1, 225 n. 2, 225 n. 5
In the series TERMINOLOGY AND LEXICOGRAPHY RESEARCH AND PRACTICE (TLRP) the following titles have been published thus far or are scheduled for publication: 1. CABRÉ, M. Teresa: Terminology. Theory, methods and applications. 1999. 2. ANTIA, Bassey Edem: Terminology and Language Planning. An alternative framework of practice and discourse. 2000. 3. TEMMERMAN, Rita: Towards New Ways of Terminology Description. The sociocognitive approach. 2000. 4. SAGER, Juan C.: Essays on Definition. 2000.