Wordtrade LogoWordtrade.com
Philosophy

 

Review Essays of Academic, Professional & Technical Books in the Humanities & Sciences

 

Views into the Chinese Room: New Essays on Searle and Artificial Intelligence by John Preston, Mark Bishop (Oxford University Press) The most famous challenge to computational cognitive science and artificial intelligence is the philosopher John Searle's "Chinese Room" argument. Searle argued that, although machines can be devised to respond to input with the same output as would a mind, machines--unlike minds--lack understanding of the symbols they process. 19 essays by leading scientists and philosophers assess, renew, and respond to his challenge.

In the mid-1970s one of the USA's best-known philanthropic organizations, the Alfred P. Sloan Foundation, invested substantial funds in a programme designed to stimulate progress in a burgeoning cross-disciplinary study of the nature and workings of the mind: `cognitive science'. Although, with hindsight, it can be traced back to the 1950s, cognitive science came to public recognition (and was dubbed by the psychologist Christopher Longuet-Higgins) only in the early 1970s. It comprises a constellation of disciplines (the core members being psychology, linguistics, artificial intelligence, and neuroscience) which currently attempts to explain cognitive phenomena (thinking, reasoning, intel­ligence, perception, learning, understanding, belief, knowledge, memory, etc..) on the basis of hypotheses about the kinds of information-processing which support them. Motivated and underpinned by a certain philosophical perspec­tive, the constellation subsequently broadened to include parts of or approaches to related fields like anthropology, archaeology, and sociology.

The University of California at Berkeley was one of the main beneficiaries of the Sloan Foundation's programme, as part of which prominent researchers were funded to travel around the country, lecturing at universities. How one of these researchers, a philosopher from UC Berkeley, came to be thought of as supplying the best-developed and most pointed threat to a core component of cognitive science is the story we have to tell. Although there is still tremendous controversy over its success, there is some consensus over the import of this `Chinese Room argument' (CRA), which John Searle first published in a paper entitled `Minds, Brains, and Programs'. The argument turns on an easily understood thought-experiment which mobilizes readily available intuitions. If sound, it undermines the official self-image of artificial intelligence (AI), one of the supposed foundations of much contemporary cognitive science. It may well also be contemporary philosophy's best-known argument. 

Views into the Chinese Room was conceived as a forum in which to provide opportunities to restate the original argument and envisaged responses, to develop those responses, and to indicate lines of argument which Searle did not anticipate. Several contributors have honed and developed their arguments by reference to each other's contributions. Searle was not asked, or`given the opportunity here, to respond to the chapters in this volume. But although it's not supposed to contain the last word on the debate, the volume does not simply take stock. Rather, it attempts to latch onto a new phase of the debate, in which detailed analysis and unpicking of the arguments pro and con predominates over the original `replies' which Searle himself enumerated.

In preparing the volume, the editors became more aware than ever of a sort of consensus among cognitive scientists to the effect that the CRA is, and has been shown to be, bankrupt. Despite the fact that several notable `names' within the philosophy of mind agree with Searle, it's true that the negative consensus among computationalists has become, if anything, even more solid. Some prominent philosophers of mind declined to contribute on the grounds that the project would give further exposure to a woefully flawed piece of philo­sophizing. Even some who have contributed to the volume think of the CRA not just as flawed, but as pernicious and wholly undeserving of its fame.

Despite this consensus it is notable, however, that there is (still) little agree­ment about exactly how the argument goes wrong, or about what should be the exact response on behalf of computational cognitive science and Strong AI. We should probably find it extraordinary how much opinions can differ, and how wide the variety of topics which can be raised by, a scenario as apparently simple as the Chinese Room. But Searle's thought-experiment is a microcosm of much contemporary philosophy and cognitive science. Its importance, both philosophically and practically (in its impact on the self-image of current and proposed cognitive science research programmes) has ensured that it has been widely attacked (and defended), with almost religious fervour. It raises a host of issues about mind and mentality, language, meaning and understanding, intentionality, computers, cyborgs, and our self-conception. It can also be used to raise large methodological questions about how cognitive science should be done (computationalism versus `cognitive neuroscience', versus some more person-centred alternative?), as well as about what philosophy should be ('scientific'? or `analytic'? or perhaps `phenomenological'?). At the very least, it forces those involved in contemporary cognitive science into clarifying exactly what general theoretical theses they want to defend.

So far we've looked in an informal way at the Chinese Room scenario-the thought-experiment-and some of the conclusions Searle draws from it. As we shall see, Searle sets out the underlying argument against Strong Al and computa­tionalism in different ways on different occasions. However, he never explicitly presents it as a piece of reasoning about the thought-experimental scenario. If it were presented thus, its premises would presumably be:
1. The person in the room has access only to the formal, syntactic features of
the symbols he or she is presented with.
2. To understand the Chinese input, the person in the room would need access to the semantic features of those input symbols.
3. No set of formal or syntactical principles is sufficient for understanding.

But what exactly is this argument's conclusion? If the conclusion pertains only to the person in the room (if, for example, it's simply that the person in the room doesn't understand the Chinese input), then it's relevant to Strong Al only if that view makes a claim about the analogous part of a suitably programmed computer. It's often argued that Strong Al makes no such claim.

Exactly how is the Room supposed to be analogous to a computer? Searle says that when ensconced in the Chinese Room he `simply behaves] like a computer', is `simply an instantiation of the computer program', that he is the computer and that he has `everything that artificial intelligence can put into me by way of a program' (Searle 1980a: 418 (B69-70)).16 The English-language rules constitute the computer program, and the first two batches of symbols the database to which the program has access.

However, some commentators, such as Ned Block and John Haugeland, urge that the person in the room is analogous not to the computer, as Searle usually claims, but only to its central processing unit (CPU), the executive part of the computer which controls and coordinates everything else happening in the machine. Haugeland, for example, argues here that Searle fails to apply his own proposed criterion for adequate theories of mind, asking himself only what it would be like to be part of the Chinese-understanding system, rather than the system itself. These commentators then remind us that Strong Al's claim is not about the CPU, but about the computer as a whole, the entire system. However if, as Copeland points out in what he calls the `logical reply', the Chinese Room Argument is supposed to pertain to the system as a whole, then although germane, it isn't watertight. It would then be of the form: `No amount of symbol manipulation on the person's part will enable him to understand the Chinese input, therefore no amount of such manipulation will enable the wider system of which he is a part to understand that input'.

Since the conclusion is about a system that isn't even referred to in the premises, this Chinese Room Argument (as it stands) must be logically invalid. It commits, as Haugeland puts it, a part-whole fallacy.

As I mentioned, Searle never presents the Chinese Room Argument in this way. In the abstract of his original 1980 article, he set it out as a derivation from axioms, thus:
(1) Intentionality in human beings (and animals) is a product of causal features of the brain.
(2) Instantiating a computer program is never by itself a sufficient condition of intentionality, therefore
(3) The explanation of how the brain produces intentionality cannot be that it does so by instantiating a computer program.
(1) is supposed to entail:
(4) Any mechanism capable of producing intentionality must have causal powers equal to those of the brain.

(In more recent work, Searle explicitly says `threshold causal powers', since the brain may have more than is necessary to produce mentality. And from (2) and (4) is supposed to follow:
(5i Any attempt literally to create intentionality artificially (strong AI) could not succeed just by designing programs but would have to duplicate the causal powers of the human brain. (Searle 1980a: 417, not reproduced in Boden)

The central argument of the original article, Searle tells us, is directed towards establishing axiom (2).

Whenever he presents what he thinks of as the underlying CRA's abstract logical structure, however, or the point of the Chinese Room scenario, Searle always does so as follows:

  1. Programs are purely formal (syntactical).

  2. Minds (human ones, at least) have semantics, mental (i.e.. semantic) contents.

  3. Syntax by itself is neither the same as, nor sufficient for, semantic content.

Therefore,

   4. Programs by themselves are not constitutive of nor sufficient for minds.

It's noteworthy that in such presentations of what we might call (following Searle, and Hauser 1997) `the Brutally Simple argument', the Chinese Room scenario is said merely to illustrate or remind us of the truth of premise (3), rather than to constitute the argument against Strong Al. That so many comment­ators took the scenario to constitute (i.e.. exhaust) the argument is why Searle felt able to say, in 1987:

[I]n all of the vast amount of literature that has grown up around the Chinese room argument, I cannot see that any of my critics have ever faced up to the sheer logical structure of the argument. Which of its axioms do they wish to deny? Which steps in the derivation do they wish to challenge?

Whether treated as the argument underlying the Chinese Room scenario, as a streamlined reformulation of that argument, or as a separate though related piece of reasoning, the Brutally Simple argument does attract its own

commentators. Dennett, perhaps the most determined critic of the Chinese Room Argument, has explicitly denied all three of its premises. In this volume, Copeland, Haugeland, and Hauser concern themselves expli­citly with the Brutally Simple argument. Haugeland, for example, seeks to show that serious Al, while not committed to denying Searle's logical truth (that syntax is not sufficient for semantics), can respond to the CRA by denying that computer programs are purely syntactical. To do so, he outlines the con­ceptual foundations of Al in a way that takes account of the causal powers of programs and data. 

Computers and Translation: A Translator's Guide edited by H. L. Somers (Benjamins Translation Library, 35: John Benjamins Publishing) This book is, broadly speaking, and as the title suggests, about computers and translators. It is not, however, a Computer Science book, nor does it have much to say about Translation Theory. Rather it is a book for translators and other profes­sional linguists (technical writers, bilingual secretaries, language teachers even), which aims at clarifying, explaining and exemplifying the impact that computers have had and are having on their profession. It is about Machine Translation (MT), but it is also about Computer-Aided (or -Assisted) Translation (CAT), computer-based resources for translators, the past, present and future of translation and the computer.

Actually, there is a healthy discussion in the field just now about the appropri­ateness or otherwise of terms like the ones just used. The most widespread term, "Machine Translation", is felt by many to be misleading (who calls a computer a "machine" these days?) and unhelpful. But no really good alternative has presented itself. Terms like "translation technology" or "translation software" are perhaps more helpful in indicating that we are talking about computers, the latter term emphasising that we are more interested in computer programs than computer hardware as such. Replacing the word "translation" by something like "translator's" helps to take the focus away from translation as the end product and towards translation as a process' carried out by a human (the translator) using various tools, among which we are interested in only those that have something to do with computers.

The first seven chapters look at various uses to which a translator might put the computer while the second half of the book focuses more on MT. In Chapter 2 we describe the development of the ideas behind the translator's workstation, and look at some of the computer-based tools that can be made easily available to translators, with a special focus in Chapter 3 on one of these tools, the translation memory. Chapter 4 concerns the special place of terminology in the CAT scenario. Translators have always been aware of the need to access technical vocabulary and be sure that the terms chosen are correct and appropriate. As Lynne Bowker describes, computers can play a particularly useful role in this question, as term banks and other sources of terminology are available in various forms, both on-line and in the form of machine-readable dictionaries and thesauri.

A relatively new translation activity that has emerged in recent years goes under the name of software localization. In the early days of computers, most software (and hardware) that was produced was biased towards (American) English-speak­ing users. It has now been recognised that products aimed at a global market must be customized for local aspects of that global market. Software localization involves translating documentation, including on-line help files, but also often involves customizing the software itself, inasmuch as it contains language (for example, how to translate Press Y for Yes into French). In Chapter 5, Bert Esselink condenses some of the ideas from his comprehensive book on the subject, to give an indication of the problems involved, and some of the tools available to assist the translator in this specific task.

In today's commercially-oriented world, much translation work is motivated by commercial considerations. Socio-economic factors thus influence the develop­ment of MT and CAT systems, and it is the major European languages (English,

French, Spanish, German, Italian, Portuguese, Russian) plus Japanese, Chinese, Korean and to a certain extent Arabic that have received attention from the developers. Bad luck if you work into (or out of) any of the several thousand other languages of the world. In Chapter 6 we look at the case of CAT and minority languages — an ironic term when one considers that the list of under-resourced languages includes several of the world's top 20 most spoken languages (Hindi, Bengali, MalaylIndonesian, Urdu, Punjabi, Telegu, Tamil, Marathi, Cantonese). We will consider what the prospects are for translators working in these languages (and other languages more reasonably described as "minority") and what kinds of computer-based tools and resources could be made available.

The next chapter looks at the place of computers in the academic world of translator training. Sara Laviosa considers the use of the computer in Translation Studies: in particular, this chapter looks at how computer-based corpora — collec­tions of translated text — can be used to study trends in translation practice.

The remaining chapters focus more closely on MT. In Chapter 8, Doug Arnold explains why translation is hard for a computer. Readers of this book will have views on what aspects of translation are hard for humans, but Arnold points out that some aspects of language understanding in the first place, and then the render­ing of what has been understood in a foreign language in the second place, present difficulties to computers which, after all, are basically sophisticated adding ma-chines. At least some of the difficulty is down to the nature of language itself, and in Chapter 9, Paul Bennett describes how the scientific study of language — linguistics — can help to provide some solutions to the problems.

The next three chapters focus on MT from the commercial point of view. In Chapter 10, John Hutchins, MT's unofficial historian and archivist-in-chief, details the current state of the art in commercially available MT and CAT software. Chapter 11 presents the developer's point of view. Co-authored by Laurie Gerber, formerly one of Systran's senior linguists, and Scott Bennett who, at the time of writing was a senior member of Logos's development team, and before that had helped oversee the productization of the Metal MT system by Siemens. In Chap-ter 12, Jin Yang and Elke Lange report on Systran's intriguing experiment in which they have made their MT system freely available on the World Wide Web. This contribution explains why the company is happy to see their product freely used, and reports on a period of close monitoring of the web-site, and users' feedback and opinions.

John White's chapter on how to evaluate MT will be essential reading for anyone thinking of MT or CAT tools as a solution to their translation needs, whether they be an individual freelancer, a small translation company or part of the translation department of a large company. White gives a practical and historical overview of what to evaluate, how to evaluate it, and, above all, some of the pitfalls to avoid.

The next three chapters address aspects of the practical use of MT. In Chapters 14 and 15 we look at two strategies for getting the best out of MT: Eric Nyberg, Terako Mitamura and Wolf Huijsen describe their approach to controlled lan­guage, explaining the basic idea behind the concept, and how it can be implemented within a translation scenario. An important feature of the controlled-language approach is the need to gain acceptance of the underlying idea from the authors of the texts to be translated, and to overcome the negative preconceptions of loss of author's creativity and the resulting sterility of textual form that the term controlled language inevitably invokes. While the controlled-language approach restricts the syntax and vocabulary of the texts to be translated in a pre- and proscriptive way, the sublanguage approach takes advantage of naturally occurring restrictions and preferences in the style and vocabulary of the texts. In Chapter 15 we look at the classical example of a successful sublanguage MT system — the Canadian Meteo system — and consider whether this was a "one-hit wonder" or whether the success of this experiment points the way for future MT success stories.

In Chapter 16, Jeffrey Allen looks at the question of revising MT output, usually termed post-editing to distinguish it from the parallel task of revision often performed on human translations. Allen brings out some of the differences in these two tasks, and outlines some strategies and techniques to make the task easier and more efficient.

In the final chapter shows the use of MT and CAT tools in the teaching of translation, both to trainee translators, and to language students in general.

Philosophy and Linguistics edited by Kumiko Murasugi and Robert Stainton offers ten new essays in areas of overlap between philosophy and theoretical linguistics. Specific areas include semantics, philosophy of language, and philosophy of linguistics, with papers on topics as diverse as the language of perception, compositionality, situations, moral competence, and the notion of simplicity in generative grammar. Of special interest is a section devoted to mixed quotation.

The papers are by top scholars in the field: the contributors are Herman Cappelen, Susan Dwyer, Reinaldo Elugardo, James Higginbotham, Ernie Lepore, Peter Ludlow, Adele Mercier, Stephen Neale, Paul Pietroski, Francois Recanati, Robert Stainton and Zoltan Gendler Szabo.

 At the core of the essays in Philosophy and Linguistics are questions of logical form: a joint linguistic and philosophical investigation into the syntax and semantics of various natural language constructions. The papers by Higginbotham and Neale are concerned with the logical form of perceptual reports and plural definite descriptions, respectively. The project of uncovering logical form is supplemented by philosophical reflection on the limits of semantics and on the methodology of linguistic research. The former are represented here by a paper by Mercier on the problem of naming and a paper by Recanati on how to accommodate context-dependence within semantics. The latter are represented by Dwyer's paper on a possible application of Chomsky's innatism in moral psychology and a paper by Ludlow on how to evaluate simplicity-claims in linguistics.

The last section of the volume contains a case study. In the first article, Cappelen and LePore offer a paratactic analysis of sentences containing mixed quotation, like `Alice said that life is "difficult" to understand: Their account is an elegant combination of slightly modified versions of the Davidsonean semantics for direct and indirect quotation. Since-as the authors argue-the rivals of the paratactic account have no straightforward way of achieving a similar theoretical unity, the paper can be seen as a new argument for a Davidsonean semantics for propositional attitudes. The replies by Elugardo, Pietroski and Stainton contain detailed criticism of the proposal and alternatives for the syntax, semantics and pragmatics of mixed quotation.

The exchange about mixed quotation illustrates two of the most striking features of many contemporary debates in philosophy of language. First, the focus is a specific linguistic phenomenon, but the analysis is used in support of general theses about the interpretation of natural languages. Second, argument and criticism are sensitive to detailed linguistic considerations. The discussion about Cappelen and LePore's proposal brings together questions about logical form, the limits of semantics and methodology, thereby exemplifying the current interplay between philosophy and linguistics.

The collection is sure to be of interest to scholars whose work bridges both linguistics and philosophy, as well as to some philosophers of mind.

 Kumiko Murasugi received a Ph.D. in linguistics from the Department of Linguistics and Philosophy at MIT. She is currently Adjunct Research Professor in the School of Linguistics and Applied Language Studies at Carleton University in Ottawa.

Robert Stainton holds a Ph.D. in philosophy from the Department of Linguistics and Philosophy at MIT. He is currently Associate Professor at Carleton University in Ottawa. His previous publications include Philosophical Perspectives on Language, and numerous articles on philosophy and linguistics.

Excerpt:

Perhaps the most exciting thing about recent philosophy of language is is openness towards genuine linguistic problems. There was a time when philosophers could write about semantics with no regard to what linguists would say about the expressions they were considering. They had the excuse that questions about exactly how natural languages manage to express what they do had not yet been investigated in formal detail. When raising questions about the truth-conditions of a certain type of sentence, they could simply suggest a formula in Logish4 (a mixture of first-order logical symbolism and simplified English), replace the original sentence with it and discuss this `translation' instead. Philosophers of language these days ace required to call their translations `logical forms'. The shift is not merely terminological. Philosophers today are supposed to think of these formulae as-perhaps somewhat simplified-representations of the real syntactic and semantic complexity of natural language sentences, and they are supposed to show--or at least indicate--how these formulae are derivable from acceptable rules and principles of syntax and semantics.

The fate of Russell's theory can again be used to illustrate the reorientation of philosophy of language in recent years. Most contemporary philosophers of language tend to bracket Russell's reservations about the subtleties of `the' and take his theory of description to be directly about the interpretation of natural language sentences. According to this line, definite descriptions can be identified with a certain type of quantificational device. What the Russellian quantificational form of a sentence containing a definite description yields is neither a characterization of a use of the sentence (as Strawson and with him the ordinary language philosophers maintained), nor a useful proxy for the sentence (as Quine and the ideal language tradition have argued), but a perspicuous presentation of its exact truth-conditions. Some semanticists go even further. They argue that for the Russellian account to be acceptable, it is not enough that all sentences containing singular definite descriptions have Russellian truth-conditions. The quantifier postulated by the Russellian must have syntactic reality: it must appear on the level of the logical form. Its presence there must be syntactically testable; it must be subject to the usual syntactic operations that quantifier phrases can undergo.

Understanding Russell in this way puts new burdens on a Russellian. How does the Russellian analysis fare when one considers sentences with multiple quantifiers. modal operators, propositional attitude verbs? Can one expand the Russellian account to cover plural descriptions, generic descriptions or descriptions containing mass nouns? What is the relationship between definite descriptions and anaphoric pronouns that can often replace them within complex sentences? Such problems would not exist for either Strawson or Quine. Strawson could argue that the definite article is not employed in a singularly referring manner in most of these complicated cases, so the Russellian theory narrowly construed says nothing about the truth-conditions of uses of such sentences. Quine could argue that there is no obvious need for incorporating all of this complexity into the regimented language of science. The very point of philosophical analysis for him is the elimination of some of the more obscure features of ordinary language. But if one is interested in assigning truth-­conditions to English sentences, one cannot bypass the complications. Their presence is a fact of language to be accounted for by the semanticist.

Should we think of the current linguistic focus of much of philosophy of language as a sign of `naturalizing' another branch of traditional philosophy? Some do. But it is not obvious that the new interest in logical forma concern for harmonizing one's views on reference, meaning and use with the results of linguistics-will yield solutions to most philosophical problems about the nature of language. After all, as linguistics matures, linguists increasingly turn their backs on the foundational issues of their discipline.

If they defer to philosophers who in turn defer to linguists in these matters, we can hardly expect progress. One might hope, more modestly but perhaps still in harmony with some version of naturalism, that the results of linguistics will sharpen our understanding of the old problems and that they will provide new and interesting constraints on philosophical reflection on them.

The Structure of Modern English: A Linguistic Introduction is an extensive overture to formal aspects of modern English structure, including: phonology, morphology, lexical and sentence semantics, syntax, and pragmatics.

This text is for advanced undergraduate and graduate students interested in contemporary English, especially for those primary area of interest is English as a second language, primary or secondary-school education, English stylistics, theoretical and applied linguistics, or speech pathology. Focus is exclusively on English data, providing an empirical explication of the structure of the language, rather than exploring theoretical questions or theory for theory's sake. The text does use linguistic theory but presupposes little or no background in linguistics or any particular linguistic predilection.

Accompanying the text is a pedagogically useful CD-ROM that is a complete interactive workbook with numerous self-testing exercises. Additionally, the CD presents suggestions for pedagogical applications of the material in the textbook in a teacher’s section.

English Sentence Analysis: An Introductory Course is a formal 10-week course for students of English as a second language. It is designed for students of linguistics and other formal language study with an analytical orientation The CD-ROM provides much background information and drills for study. For example study a chapter from the book. Exercises in each chapter and answers in the Key to Exercises let you check your understanding. Then confirm your understanding with the chapter exercises in the Practice Program on your PC. The Practice Program keeps your score on screen and analyzes your results. If you score under 80% the program presents you with a new set of exercises focusing on your problem areas. Want to know why your answer is correct or incorrect? You can go to Feedback. Still not clear? The Theory Module will lead you through. Have fun. This text is designed for self-study and class use.

Sign Language in Indo-Pakistan: A Description of a Signed Language by Ulrike Zeshan provides the first formal description of sign language used by the deaf in India and Pakistan. It pioneers cross-cultural sign language and systematical analyzes the individual signs as well as showing the principles of discourse organization. Highly original and important work.

 The Organizing Property of Communication by Francois Cooren

Cooren's remarkable book is destined to become one of the keystone texts of an emerging new school of theorization and research: the investigation of organization as a phenomenon of language. While the study of both organization and language have a long tradition, the links between them have been tenuous, to the point that it could often be said that they hardly existed at all. In the literature on organization the theme of language figures very little, and language study, in turn, has tended to be indifferent to issues of organization. Cooren's work, along with that of his colleagues (many of them associated with the communication program at the University of Montreal), is concerned to construct a causeway between the two disciplines - or, to use the French expression, a passerelle. To understand the full implication of Cooren's book it may be useful, in a preface, to fill in some of the background to his thinking on this topic, along with that of his colleagues.

It is of course almost a truism to say that organization is a constructed reality, and that the uses of language are fundamental to such a construction. Berger and Luckmann's The Social Construction of Reality is, after all, now approaching the status of a classic. But to say that organization is a "construction" of language-in-use, while plausible, tends to leave the impression that its existence is entirely contingent on intersubjective process. In such a view, organization exists only in the consciousness of its members. Cooren's position could hardly be more opposed to this idea. In his book, he sets out to demonstrate the contrary: that organization is real –inter-objective, not inter-subjective.

The field of organizational communication, that forms the backdrop to Cooren's exposition, emerged`as an academic domain in the 1960s. It brought together two traditions of research, the social psychology of work, and speech communication. Social psychology is an established discipline everywhere. Speech communication, on the other hand, is a field of study unique to the American educational system. Its origin is in rhetoric and elocution and it has thus served to marry practical concerns about effective expression in organizational contexts with a continuing theoretical interest in the principles of rhetoric and argumentation. Over the years, organizational communication scholars have explored phenomena of human interaction in organizational contexts, including questions of effective supervision, socialization of new members, leadership, negotiation of hierarchical status, conflict resolution and organizational `climate.' Theirs is an eclectic field that has freely borrowed from sociology concepts such as discursive formation (Foucault) and structuration (Giddens), among others. One issue that has been missing from this literature, however, is the one this book addresses, namely the ontology of organization: what an organization is. Cooren's work, which is primarily a study in speech act theory and its possible interpretations, and thus addressed most evidently to students of language and discourse, should nevertheless prove to be of great interest to organizational scholars.

In taking up the problem he has set himself, Cooren draws on a formidable arsenal of resources, including speech act theory and its interpretations, Greimasian narrative theory, rhetorical theory, and the translation theory of Latour and his associates. These different disciplines might seem, superficially, to be quite distinct from each other, but Cooren's contribution, and one of the reasons why this is an important book, is in delving into the foundations of each of them to reveal their deeper connections in a basic underlying logic of communication. What emerges from his analysis is the centrality of the organizing function of language. But Cooren's interpretation of language function leads us to a reexamination, and ultimately the rejection, of the standard concept of the relationship of language and communication: language as medium of expression. He argues for a much more powerful role of language in human affairs: language no longer as constructed, but language as constructor. Language, he contends, has agency embedded within it, and can thus`act through the mediation of the people who speak it. It is an argument, in other words, for looking at language differently, as interobjectivity. It is because of the agencies of language that organization is enabled to stabilize over time and be extended in space.

A central element in the development of this alternative perspective on both communication and organization is Derrida's critique of Austin's initial formulation of speech act theory. Cooren elaborates on Derrida's objection to Austin's assumed primacy of the speech situation and in doing so he invites us to set aside the psychological conditions that may be attached to the accomplishment of acting in and through speech and instead to look at the materialization of speech in text, even when the text is inscribed in no more substantial a medium than the human voice. Taking his cue from Latour, Cooren argues that what acts is not one person on another, with language having no more significant a role to play than that of medium, but a hybrid actant embodied in textual form. In the end, after all, it is the text that acts for the speaker. Because the power to act is embedded in the text it may become detached from its site of occurrence and, once multiplied and amplified, take on the role of autonomous actor. It is because of the enablement of text, Cooren proposes, that organization can transcend the strictly local context of an embodied interaction to assume the extensions of modern corporate hegemony and big government.

Cooren's book, I would suggest, is a `must read' for scholars in both language study and organizational science. It represents a genuinely new approach to some old questions, and it does so without ever abandoning the highest standards of academic scholarship. The book is meticulously written, cogent and profound. It is destined, I believe, to become one of those works that marks a transition in our way of thinking about our fields.

James R. Taylor, Ph.D.Emeritus Professor of Communication Universite de Montreal

Pragmatic Meaning and Cognition by Sophia S.A Marmaridou provides a good overview of philosophical and cognitive approaches to language use and meaning. A synthesis of such approaches leads to a dynamic concept of pragmatic meaning which is on the one hand grounded in cognition and motivated by linguistic and cultural convention and, on the other, creates a framework for studying the interactive and social dimensions of the development of meaning in linguistic communication. Through an experientialist approach based on connectionist models, the author shows that by internalizing pragmatic meaning people become social agents who reproduce, challenge or change their social parameters during interaction. Pragmatic Meaning and Cognitionis suitable as a course book in Pragmatics and Semantics and of interest to those concerned with cognitive models and dynamic and social aspects of linguistic communication.

 Pragmatic Markers and Propositional Attitude edited by Gisle Andersen and Thorstein Fretheim

The present collection of papers grew out of a panel on Particles and Propositional Attitude at the 6th International Pragmatics Conference, Reims, July 19-24 1998.

We find it necessary to comment briefly on our choice of the collocation 'pragmatic marker' in the title Pragmatic Markers and Propositional Attitude. Most contributors to this volume refer to their objects of study not as 'markers' but as 'particles'. While we are aware that the term 'particle' has been variably used with reference to a morphologically rather disparate set of linguistic expressions, including grammaticalized phrases as well as monomorphemic words, we judged the even broader term 'marker' to be more adequate, considering the wide range of linguistic phenomena to which the present collection of articles is devoted. It should also be borne in mind that some of the particles explored here are claimed to interact with and receive support from other linguistic devices, like sentence type or specific prosodic properties of the verbal stimulus, which could be called `markers' but which are certainly not `particles'.

The modifier 'pragmatic' is potentially more controversial than the head noun `marker'. Although it is true that the kind of meaning encoded by what the editors of this volume refer to as pragmatic markers frequently does not affect the truth conditions of the proposition expressed by the utterance, we do not want to leave the readers with the impression that we equate 'pragmatic' content and 'non-truth-conditional' content. We consider a study of the meaning of a given linguistic item in a given utterance to belong to the domain of pragmatics, because part of the utterance meaning of the item can only be derived as a result of the addressee's extra-linguistic inferential processing of the stimulus containing it. The lexically encoded meaning of the markers examined in this book generally underdetermines the contribution of those markers to the overall meaning communicated by the utterances in which they occur. The semantic meaning of a marker equals its encoded meaning, but its encoded meaning only represents a very useful (and occasionally quite necessary, so it seems) constraint on the kinds of pragmatic, or extra-linguistic inferences that the addressee processing an utterance will draw in his effort to comprehend the message communicated. Pragmatically derived meaning affects the hearer's recovery of explicitly communicated assumptions and implicitly communicated assumptions alike. `Explicit' is not to be equated with `semantic', and `implicit' is not to be equated with `pragmatic'. Following Relevance Theory, the editors of this volume tie the semantics/pragmatics distinction to two fundamentally different sorts of cognitive process in utterance interpretation, namely the distinction between linguistic (semantic) decoding and extra-linguistic (pragmatic) inference.

In addition to semantically encoded concepts there are also semantically encoded instructions for the hearer to follow in order to derive intended cognitive effects, including implicitly communicated ones; in the relevance-theoretic terminology there is a procedural semantics alongside a conceptual semantics.

No article in the present collection disputes the view of the semantics/pragmatics dichotomy adumbrated here, and most authors are seen to explicitly share it with us.

It is far from accidental that we did not let the term `discourse markers' appear in the title of this book instead of `pragmatic markers'. For Bruce Fraser, `discourse markers' constitute a subtype of pragmatic markers, specifically "an expression which signals the relationship of the basic message to the foregoing discourse". The connection between the notion `discourse marker' and textual functions is highly salient in Deborah Schiffrin's account, for example, where she defines discourse markers as sequentially dependent elements that bracket units of talk. Indeed, 'discourse markers' is a term which has come to be associated predominantly with discourse analysis and such markers are assumed to play a key role in establishing coherence relations in discourse. The emphasis on corpus-based data is a salient feature of discourse analysis and is therefore also seen to prevail in studies which are professedly concerned with the function of so-called discourse markers. Some of the papers included in this volume do rely extensively on the use of corpora, but many of them rely mainly, or even exclusively, on invented examples based on native user competence. More importantly, however, the linguistic function of the majority of markers examined in this volume has rarely been associated with the label `discourse marker', and the same goes for most of the theoretical issues raised by the authors. By choosing the term `pragmatic marker' instead of `discourse marker' we believe we do not run the risk of evoking unintended connotations.

 The goal of Bridging and Relevance by Tomoko Matsui is to help fill a major gap in our current knowledge of verbal understanding. Often, in interpreting an utterance, the hearer has to choose between a range of possible interpretations: the question that interests me is how that choice is made. What makes this question particularly interesting is that wrong choices lead to misunderstanding; by examining the mechanisms by which choices are made to show not only how misunderstandings arise, but how they may be reduced.

The focus of  this research is reference assignment, where the hearer must decide which referent the speaker had in mind. Reference assignment has been approached from many different angles: by linguists, logicians, philosophers, psychologists and workers in the field of artificial intelligence. All are agreed that contextual or background assumptions play a crucial role, but there is little agreement about how they are selected, and, once selected, how exactly they are used. Matsui argues, following recent work in relevance theory, that the role of contextual assumptions in reference assignment is identical to their role in other aspects of utterance interpretation. Thus, a detailed investigation of reference assignment may provide a solution to a wide range of further problems that arise in verbal understanding.

Within the domain of reference assignment, Matsui focuses attention on the phenomenon of `bridging reference'. on which there is a substantial literature in linguistics, psycholinguistics and artificial intelligence. This is illustrated in (1) and (2):

(1) I went into the room, The window was open.
(2) John went walking at noon. The park was beautiful.

What interests Matsui is the interpretation of the referential expressions `the window' in (1) and `the park' in (2). In (1), although it is not explicitly stated in the first clause that the room had a window, the natural assumption is that it did, and that the window referred to in the second clause is the one in the room. Similarly, in (2), although it is not explicitly stated in the first clause that John went walking in a park, the natural assumption is that he did, and that the park referred to in the second clause is the one where John went walking. Matsui is interested in what makes these the natural assumptions, and what happens in more complicated cases of the same general type.

The term `bridging' was introduced by the psycholinguist Herb Clark to describe the process by which the existence of a referent which has not itself been explicitly mentioned is inferred from something which is explicitly mentioned - in (1) from the mention of a room, and in (2) from the mention of the fact that John went walking. By `inference' here, Matsui mean a non-demonstrative inference process of hypothesis formation and evaluation. In (1), for instance, it is reasonable to suppose that, in interpreting the referential expression `the window', the first hypothesis to come to the hearer's mind will be that the window in question is in the room just mentioned. In more complex cases, there may be several possible hypotheses, and the hearer will have to choose between them. Matsui investigates what determines the order in which hypotheses occur to the hearer, and how the choice between them is made.

In this book, Matsui draws the usual distinction between semantics (the theory of sentence meaning) and pragmatics (the theory of how utterances are interpreted in context). Pragmatic interpretation involves not only a knowledge of sentence meanings, but also the ability to draw inferences based on contextual assumptions, or knowledge of the world. Bridging, which involves going beyond what is explicitly stated in the utterance, is clearly a pragmatic rather than a strictly semantic phenomenon, and must be investigated by primarily pragmatic means. Matsui work provides some important clarification to how meaning is made.

 Essays on Definition by Juan C. Sager.

Definition occupies a central place in all sciences and is a fundamental tool in logic, philosophy of ideas and semantics; each of these areas view the activity of defining from a different angle. In everyday communication, the result of this activity, represented in the various forms of definition considered appropriate for their objective, is also essential for establishing relationships between things and ideas and their names for which purpose they are collected in glossaries, dictionaries and other reference tools. Given these various viewpoints on what is involved in defining and the wide range of uses of definitions, there is, under­standably, a great diversity of interpretations of what is meant by defining and its product, the definition. There is, thus, no general agreement about what a definition is, what knowledge it represents and conveys and what quality criteria it must satisfy.

Defining and definitions play an essential role in terminology and lexicogra­phy because they are the conventional means for establishing the meaning of lexical items, or, expressed differently, for connecting the concept with the word or term that represents it. This crucial role, which lies at the root of language as a symbolic system, is, however, viewed very differently by lexicographers and terminologists. To judge by the many treatises on the subject, not only are there different methods of defining and different resulting types of definition, but even the limited objectives of lexicographical and terminological definitions themselves seem far from clear. It would therefore appear to be useful to step back from the specific concerns of lexicography and terminology and consider the nature and functions of defining and definitions in those sciences which have discussed this topic for the last two millennia.

The present collection of essays on definitions from Plato and Aristotle to modern times, assembles interesting, sometimes less widely known and, perhaps, unexpected texts. They examine the subject from the point of view of philos­ophy, which is essential for a theory of terminology seeking to establish the relationship between concepts and terms. These essays deal mainly with theoretical issues but they also consider the practice of defining and therefore serve as back­ground to all manner of studies in terminology. In addition they form a useful complement to the better known discussions of definitions in lexicography.

Talking at Cross-Purposes. The Dynamics of Miscommunication by Angeliki Tzanne

In my Introduction, I stated that my main aim was to show that the occurrence of miscommunication in discourse can be accounted for satisfactorily only when examined in relation to the dynamics of social interaction. Additionally, I set out to show that misunderstandings should be analysed not in terms of isolated turns or stages, but as a process of meaning breaking and negotiating in the course of an encounter. Finally, my intention was to investigate the relation between the creation and development of misunderstandings and the participants' considerations for their face in social interaction. As part of this last aim, Tzanne sets out to examine the role of miscommunication in dynamic aspects of interaction, such as challenging the authority of figures which are considered by the society to be invested with power in a given situation.

With respect to the overall aim of the book, Tzanne’s data provide evidence to support the claim that, in general terms, the creation and development of miscommunication in discourse relate closely to the dynamic nature in which talk-in-interaction proceeds. This claim is based on the following findings.

(1) The creation of misunderstandings relates to the dynamic way in which conversation develops by constructing its own interpretative context turn by turn. The tendency participants display to interpret a linguistic item on the basis of directly preceding discourse leads to misunderstandings if the item in question is meant as a shift from the frame established by previous discourse in the exchange.

(2) The creation of misunderstandings relates to the dynamic way in which social interaction proceeds as a succession of frames (situated activities) and participants' roles. Miscommunication arises when interlocutors fail to realize an intended shift of a co-participant's role or a change of the hitherto operating activity.

(3) The development of the misunderstandings examined in this work yields different combinations of reparative steps. Each reparative step is constrained by steps taken previously to it and at the same time contributes to the construction of the context on whose basis other steps will be taken.

These findings lead to the overall conclusion that miscommunication does not consist of a set of isolated turns or stages, but that it constitutes a process of interrelated steps, shaped dynamically as a series of participants' choices in the course of an exchange. This implies that misunderstanding is the result of a joint effort to communicate made by speaker and hearer, and that, as such, it should not be considered to be the hearer's `failure to understand correctly' the speaker's meaning. What Tzanne has attempted to show in this work, is that no one participant should be thought of as being responsible for the creation of a given misunderstanding, and that, consequently, `blame' which has been assigned explicitly or implicitly to hearers by previous studies should be dissociated from the occurrence of communicative breakdowns in talk-in-interaction.

With respect to the development of misunderstandings, my data show that participants' choices of reparative steps are affected primarily by people's need to ensure a smooth and unchanging flow of the discourse situation in which they are engaged, and by considerations for their own face as well as for that of their interlocutors. The variety of repair strategies used in misunderstandings reveals that in repair, participants are faced with two conflicting goals, restoring successful communication and maintaining their own and others' positive public image (face) in the encounter. As regards the relation between miscommunication and participants' face considerations in interaction, my findings lead to the conclusion that the amount of attention paid to the issue of saving face while repairing a misunderstanding depends on the power relations holding between the participants involved, with powerful speakers being the ones who show less regard for face wants. Repair strategies which attempt to reconcile these contrasting needs, together with misunderstandings which remain unresolved as a result of participants' attending exclusively to face wants, indicate that, overall, for interlocutors, face considerations appear to be more important than the need to communicate successfully in interaction.

An interesting finding that emerges from Tzanne’s data concerning face considerations and miscommunication is that the notion of face appears to be related not only to the stage of repair but also to the creation of misunderstandings. In particular, Tzanne found that face considerations not only motivate interlocutors' choices of repair strategies, but they also make people create misunderstandings intentionally as a way of preserving or enhancing their positive public image (face) in an encounter. Affecting misunderstanding can function as a tease in a conversation between friends, or be used as a face-saving strategy by an interlocutor whose positive image is seriously threatened in a given encounter.

But the most interesting use to which intentional misunderstandings are put is when they are employed by the less powerful participants in an asymmetrical encounter as a means of attacking the face of the powerful figure in this encounter. This renders miscommunication a complex facework strategy, with the aid of which the people who affect misunderstanding manage to undermine the face of their interlocutor without seriously threatening their own. By contrast, as we have seen, there are cases where this kind of facework results in enhancement rather than in loss of the hearer's face. That misunderstandings are used as a facework strategy used to challenge the source of authority in an encounter suggests that miscommunication is closely related to the constant negotiation and change of participants' goals, roles and relations in social interaction.

In addition to yielding interesting findings concerning miscommunication per se, the present study of the mechanisms of misunderstanding has general implications concerning discourse processing in social encounters. These implications relate mainly to the process of meaning-making in the course of an unfolding conversation and to the influence of roles and activities on interpretation.

 Beyond Physicalism by Daniel D. Hutto

Unlike standard attempts to address the so-called `hard problem' of consciousness, which assume our understanding of consciousness is unproblematic, this book begins by focusing on phenomenology and is devoted to clarifying the relations between intentionality, propositional content and experience. In particular, it argues that the subjectivity of experience cannot be understood in representationalist terms. This is important, for it is because many philosophers fail to come to terms with subjectivity that they are at a loss to provide a convincing solution to the mind-body problem. In this light the metaphysical problem is revealed to be a product ofthe misguided attempt to incorporate consciousness within an object-based schema, inspired by physicalism. A similar problem arises in the interpretation of quantum mechanics and this gives us further reason to look beyond physicalism, in matters metaphysical. Thus the virtues of absolute idealism are re­examined, as are the wider consequences of adopting its understanding of truth within the philosophy of science. This book complements the arguments and investigations of The Presence of Mind, and develops insights first introduced in Current Issues in Idealism.

 Language and Linguisticality in Gadamer's Hermeneutics edited by Hans Georg Gadamer and Lawrence K. Schmidt

Excerpt from Introduction by Lawrence K. Schmidt

Language lies at the heart of philosophical hermeneutics and is a central theme in contemporary philosophy. To honor Hans-Georg Gadamer's centennial, we decided to revisit the question of language forty years after the publication of Truth and Method. In that work language is the guide for envisioning a hermeneutic ontology. Gadamer initiates these considerations by arguing that the hermeneutic object and the execution and fulfillment of the hermeneutic event of understanding are determined by linguisticality. In "Europa and die Oikoumene"  Gadamer notes that he chose the term Sprachlichkeit (linguisticality) to prevent one from erroneously concluding that there is no common basis in the development of different languages. This common root is an "impulse towards the word" in the ontological structure of human being in the world. In Truth and Method Gadamer writes: "The linguisticality of our experience of the world is prior to everything that is known as an entity and is spoken of. As such it grounds the universality of hermeneutics, so "Being that can be understood is language.”

Yet, as Gadamer relates to Grondin, the third part of Truth and Method is to some extent an initial investigation.' After the publication of Truth and Method several critical questions concerning the concept of language were raised. In the last decades Gadamer has hinted that he hoped to publish a book about language to present his further thinking concerning this central topic; however, as yet we have only indications in several of his essays.

To indicate some of Gadamer's later thinking on language, this volume begins with two recent essays. In "The Boundaries of Language" (1985) Gadamer clarifies his discussion with Habermas, among others, concerning what does and does not count as language. Returning to Aristotle's definition of humans in the Politics, Gadamer remarks that understanding the beneficial distinguishes human language from animal communication. To recognize the beneficial implies a distancing from one's own position and a sense of the temporal. Yet such distancing can only occur in human interaction, in our being with one another in the world. So the basic structure of language is mutual agreement (Ubereinkommen) emerging from a conversation.

If this is language, what is not? Gadamer first considers laughing, again following Aristotle. Laughing is not pre-­linguistic, but is similar to language in that it requires such a distancing relation to oneself in a social context. This type of communication is termed by Gadamer the co-linguistic. Looking at the emergence of language in children, Gadamer discovers the fundamental, transitional element of communication to be a playing with one another and not the mere repeating of sounds. In playing, one is taken up into the game and so begins to distance oneself from one's own particular stance. The prosody of speaking plays a developmental role and indicates that "the pre-­linguistic is always already, in a certain sense, underway towards the linguistic".

In his conclusion the important dimension of the trans- linguistic is broached. What enters languages is surrounded by what has not been said. All languages may be expanded to include what has not yet been thought and expressed. The conversation that we are, is an infinite task. So the trans-linguistic indicates the universality of the hermeneutic dimension. The conversation is always underway to find the correct word.

"Towards a Phenomenology of Ritual and Language" may be an initial sketch of part of Gadamer's planned work on language. It is a long essay explicating four major themes in Gadamer's consideration of language. First, language is concealed from our thinking even though it has moved to the center of philosophic discourse. It is hidden since one has understood language as the naming of an object already known. In adopting the methodology of modern natural sciences, one has overlooked the central connection of language and thinking. In following Heidegger's thinking, Gadamer adopted as his theme this question of the close relation between word and concept, language and thinking. He finds there is no first word, a first naming of an object known, but rather that "Language exists, however, only in the with-one-another of the conversation". The study of language as conversation must include a rhetoric based upon the subject matter (Sache), where one examines how one is lead by and to the subject matter in answering the question of the conversation. Speaking occurs in the finitude of human being and so implies an unending quest for the correct word.

The second section considers the root of linguisticality in human rituals. Using the terms, together-with (Mitsamt) and with-one-another (Miteinander), Gadamer identifies the difference between animal communication and human language. Although a mixture of the together-with and with-one-another, humans alone have the freedom to distance themselves from the constraints of nature. The fundamental form of such distancing is questioning. A question implies a direction of meaning. This is the root of all pointing and naming and not an already understood objectification. Gadamer notes that his previous discussions of language and linguisticality did not consider enough the life-world context of speaking and language. The communality of the conversation has its roots in the communality of the ritual. In the ritual one acts not individually, or even with one another, but as a whole. One distances oneself from oneself in being taken up into the whole with others, as in play. Language, as the development of mutual agreement, emerges from the roots of ritual. Such communality is contrasted to Nietzsche's will to power and indicates Gadamer's difference to Derrida.

Thirdly, Gadamer examines how literature, a work of art in language, comes from this life-world ground of language in rituals. The connection is the telling of sagas or myths. Here one discovers the transition from speaking to writing. The storytelling art emerges from a ritual context of the constitution of mutual identification. Gadamer refers to the famous scene in the Odyssey where all are gathered together and hear the singer's tale of the Trojan horse. The telling of myths held the community together, could overcome differences, and created the bond of mutual agreement. The written form was long prepared for in variations of storytelling, and the written was at first read aloud. New possibilities occurred in the written work of art. Its task was to create the same sort of involvement in the reader as the listener experienced in the presence of the storyteller. The reader was to be taken up into the story, losing himself or herself in the saying of the story.

In the final and longest section, Gadamer develops the connection between the life-world basis of language in ritual and the conceptuality of thinking and philosophic language. Language is not merely the vocalization of thought, but in different languages we discover, Gadamer says quoting Kleist, "the refinement of thoughts through speaking" (GW 8, 427). Each cultural tradition expresses its way of thinking in the language that carries this tradition. Referring to Heidegger, Gadamer recounts the concealment of Greek concepts through their translation into Latin and the solidification of this way of thinking in the philosophic concepts of Western philosophy. To understand the development of concepts from living language Gadamer examines the relationship between rituals and philosophic language. We are returned to the with-one-another of live conversation. In such a conversation, following Plato, thinking should be understood as the reawakening recollection of what is meant in conceptual differentiations. What is thought and said in such a conversation is too narrowly construed if it is thought to defend a final foundation for a concept. One needs to consider the priority of language games in the move towards conceptuality. Gadamer concentrates on the rationality of practical reason to demonstrate the growth of the rational from its roots in the lived with one another. The discussion of rituals enables us to understand the rational and normative as what is found to be appropriate in "the many inter-weavings of ethos and logos through which the human is integrated into the unpredictable play of the world" (GW 8, 438). What is correct in a particular concept is not determined by a correspondence to given rules, but by being the proper application of the common, lived experience of that language community.

Our discussion of language and linguistically in Gadamer's later thought begins with Jean Grondin's discussion of "Play, Festival, and Ritual in Gadamer." In reviewing these three central images Grondin elaborates upon a theme he finds running through Gadamer's thought, the immemorial. The concept of play demonstrates the limits of the objectifiable and the phenomenon of being taken up into an artwork, and for us the sense of being taken up into the language of the conversation. In the image of the festival the accomplishment and temporality of understanding is discovered in the repetition of the celebration. One is taken up into the whole of the festival, and although there may be variations in its execution and different persons participating, it is nevertheless the same festival. So the "festival celebrates the enduring in the perishing." Rituals help constitute a tradition and one's participation in that tradition. The concept of the ritual illustrates aspects of a tradition that cannot be objectively known but that still affect us more than we know. Grondin finds that the concept of ritual more clearly presents what Gadamer meant by tradition in Truth and Method. It points to the immemorial as what lies hidden from reason, is not ever completely graspable in concepts, but makes reason itself possible.

In "On the Hermeneutic Understanding of Language: Word, Conversation, and Subject Matter," Istvdn Feher carefully examines the concept of linguisticality in Gadamer's thought. After differentiating Gadamer's turn towards language from the analytic "linguistic turn," he examines how language functions as the universal medium of understanding. What can be understood is language, and the human universe is thereby linguistically constituted, since there is no place outside from which one can view the world. The basis for this is discovered in the linguisticality of experience itself. Following Heidegger, we first hear not sense data but a linguistically structured something, the crackling fire. Feher then considers how language and the subject matter under discussion belong together. What is understood is the speculative self-presentation of the subject matter in language. This view is contrasted to the view of linguistic philosophy. The subject matter that speculatively appears in language belongs also to the community whose conversation constitutes that language. Further, this appearing brings truth to the word. So the truth of a subject matter is embodied in the word of language as the conversation that finds the right word.

John Sallis, in "The Hermeneutics of Translation," explores one of Gadamer's central images for understanding in Truth and Method, the act of translating. Gadamer's discussion of translation centers on the particular case of translating between two languages. Sallis notes that Gadamer follows Locke's determination of translation as the transposition of meaning from one language to another, but disagrees that this transposition preserves a correspondence of meaning. For Gadamer, all translation is interpretation. Quoting Gadamer in "Lesen ist wie Obersetzen" that every translation is like a betrayal, Sallis concentrates not on the necessary loss, but on those rare situations where one could speak, with Gadamer, of a gain. Through an examination of two cases in Schlegel's translation of Shakespeare, Sallis uncovers complications that interrupt "the would-be pure circulation" of meaning at the root of the concept of translation. One case "is a certain non-reciprocity," where the translation contains more than the original and so prevents the appropriateness of a counter-translation without loss. The second case "is what might be called over-translation," where a term is translated throughout a text with a meaning that it attains only at the end of that discourse. The generative determination in the original is lost in the translation. Although there are indications in Truth and Method that begin to question the classical idea of translation as transposing meaning, it is really in the previously mentioned essay that Gadamer exerts "a stronger thrust . . . toward unsettling the classical determination." Gadamer worries that in the case of poetry, even if a conceptual transposition were correct, the translation would fail if it did not transpose the prosody from the inner voice of the poet to the inner ear of the reader. Sallis closes by hinting how this might allow another reading of the Critias, where the force of words and not their meaning is noticed in reference to speaking about translation.

Nicholas Davey's contribution, "On the Other Side of Writing: Thoughts on Gadamer's Notion of Schriftlichkeit," raises the question of how to understand Gadamer's concept of textuality, especially in the sense of the art of writing. To accomplish this he juxtaposes two theses in Gadamer's discussion of writing. The first is Gadamer's contention that the written word, especially in the eminent text, is inexhaustible. In the process of understanding a text, there is not one meaning established, for example by the author's intention, according to which any interpretation may be judged. The ideality of the word transports the written beyond the contingencies of its act of being written down. A correct interpretation hears some of what was being said, but, in Davey's words, "the full meaning is always with held." The second thesis is that the written is an alienated form of speech, so that for the written to be understood it must be brought into the living voice. The symbols of writing relinquish aspects of the spoken, so that "the speculative dimension of listening" that hears the unsaid in the saying can be lost, if the reader is unable to bring the text to speak. To bring to speak, Davey notes, does not mean just to read a text out loud, but to be taken up by the text's saying so that the text effects "a fusion or, rather, a hermeneutic transposition of horizons." The juxtaposition of these is to expose the inner and outer aspects of the activity of writing and not to devalue the act of writing. Davey argues for a hermeneutics of the practice of writing by way of an examination of Hegel and a dialectic of the without and the within of language. Language is without in being independent of any subjectivity, in its ideality. It is within in shaping one's historically effected consciousness. The resolution is found in the speculative event of understanding. Davey discovers such a speculative event not only in listening to what a text that speaks says, but also in the practice of writing itself.

P. Christopher Smith, in "Plato's Khora as a Linguistic Index of Groundlessness," conducts a careful reading of part of the Timaeus "against the grain" in order to elaborate a tension between two forms of language. One is the muthos of Homeric storytelling, a temporal sequencing of events in audible speech, correlated with Gadamer's concept of linguisticality. The second is logos as the demonstration for onlookers in writing, a spatial disposition into static classes, correlated to textuality (Schriftlichkeit). To clarify these two forms of language and establish the context in which the khora, the groundless, is first spoken of, Smith notes that at the beginning of Critias's telling of the tale of Atlantis, there is a discussion of writing (21e-23c). Here we are told that the temporal sequence in a tale is inexact and must be evaluated against the logical sequence of written records. When Timaeus begins, his distinction (28a) between the unchanging and changing mirrors the distinction between what is best written about and what is best told. Smith examines how Timaeus's exposition shifts between these two forms of language as it progresses. Timaeus, the mathematician, wants to present all in the logical, written order but reaches a specific limit in the move to spatialize the temporal and visualize the acoustical, "of which `he khora' is the linguistic index." In distinction to Gadamer, Smith argues that the "`unaccustomed argument"' of 48d-a is different from that announced at 53c, which is more logical. Here, in the discussion of the khora, there is a type of thing that "eludes our insight, and it can only be told about in a muthos or story." The problem is how one is to speak of this "wherefrom" that is named the khora. This requires a type of "illegitimate thinking." Unlike Gadamer's constructive reading, Smith, in his deconstructive reading, emphasizes the danger of things lapsing into the indeterminacy of the khora. Here Timaeus can only continue by speaking mythically. So the khora functions as a linguistic index of groundlessness, and Timaeus finds himself pushed back "into the narrative and temporal origins of language from which his spatialized optical thought is but a derivative abstraction."

In "Participation and Ritual: Dewey and Gadamer on Language," Lawrence Schmidt examines the similarities between each philosopher's understanding of language. The participatory nature of language in Dewey is examined by raising a question concerning the discovery of significance in a series of events that establishes the use of tools. Schmidt worries that such linguistic understanding would appear to be possible without the participation of others, contrary to Dewey's thesis. After briefly noting two other discussions of this situation, he turns to Gadamer's discussion of the coemergence of language and experience. Linguisticality is examined as the basis for Gadamer's universal claim: being that can be understood is language. So for Gadamer any understanding of the significance of a series of events would already involve language. It is then asked whether this includes too much as language. Gadamer's discussion of the boundaries of language is examined to differentiate between human language and animal communication. The communal nature of language is rooted in Gadamer's discussion of rituals, which mirrors Dewey's sense of participation.

In "A Written History of Effects: From Concept to Application," Ben Vedder is not worried about the emergence of concepts from the linguisticality of the life-world, but about how one should approach the interpretation of texts thus developed. He begins with an examination of Gadamer's descriptive concept of the effect of history or effective history. From Gadamer's discussion of the conditions that allow for understanding texts, Vedder examines the requirements for textual research within these conditions, what he calls "historically effected research." The aim is to discover how the content of a text effects another text. Due to the history of effect at work in the interpreter's own consciousness, there cannot be an ontologically neutral position from which interpretation can occur; rather, there is a necessary ambiguity involved in determining how a text has been effected by other texts. The meaning of a text can be approached by means of all the other "texts which are produced by the power of expression of the `original text'." And we also know that such historical critical research is itself historically conditioned by the effect of history on the interpreters. So although no research can claim to have the answer, and one knows other interpretations will follow, "the point is to empirically trace effects in such a way that relations between texts become visible." Vedder next incorporates the recipient, arguing the recipient is necessary for the effect of history to occur, for the text's matter to be able to effect. So research on the history of effects must also consider the social context that influences reception, for example, institutions that canonize texts. Because the reception of a text, or part of it, effects another text, research on effects becomes intertextual research. But such research is more than the mere tracing of textual relations, it needs to investigate the "history of the truth of a text-by studying the radiance of its power of expression."

This collection on language and linguisticality in philosophical hermeneutics closes with a discussion of some practical implications of the human way of being that is distinguished by linguisticality. In "The Enigma of Health: Gadamer at Century's End," Fred Dallmayr examines Gadamer's reflections in The Enigma of Health on the art of healing in modern western society. Modern medicine is in a precarious position between two ways of dealing with the world. One is the modern scientific way that calculates, controls, and is "capable of replacing the natural by the artificial." The other concerns the "exercise and cultivation of practical judgment." Health is unique since it cannot be technically engineered. Although disease may be combated, the physician must aid nature in its restorative process. "Human health participates in the general selfmaintenance and self-balancing of nature at large," which Gadamer finds expressed in the Phaedrus. Dallmayr then demonstrates how the conclusions drawn from health reflect general themes in Gadamer's thinking as a whole, discussing Gadamer's treatment of judgment in Truth and Method, the mimetic relation in art, and nature as energeia. After noting the affinities of Gadamer's thought to Heidegger's and Adorno's, Dallmayr concludes by developing the broader social and political consequences of this discussion. Human freedom needs to be defended from technical control. We must learn to again listen to the language of things and not dominate them through the rationality of science. He notes that, without explicit reference, Gadamer's thoughts on "non-mastery and natural recovery" are similar to ideas in Buddhism and Taoism.

Headline 3

insert content here