Kasher, Asa, ed. (1998) Pragmatics: Critical Concepts, vol. III: Indexicals and Reference. Routledge, v+217pp, hardback ISBN 0-415-11734-8 (for the series), Routledge Critical Concepts.
Michiel Leezenberg, Department of Philosophy, Faculty of Humanities University of Amsterdam
This is the fifth of a projected set of six reviews of this anthology, one for each volume. Previous reviews can be found at: http://linguistlist.org/issues/12/12-2673.html (Volume I) http://linguistlist.org/issues/12/12-2154.html (Volume IV) http://linguistlist.org/issues/12/12-2917.html (Volume V) http://linguistlist.org/issues/12/12-2221.html (Volume VI)
For the original announcement of the six volume set, see http://linguistlist.org/issues/9/9-518.html
This review in part continues where the reviewer's discussion of Volume I ended; for general comments on Kasher's collection as a whole, the reader may check that review.
The present volume covers the articles numbered 39 through 47 of Kasher's collection. The selections are preceded by a preface that, this time around, provides little more than the keywords of the papers included and some bio- and bibliographical data and trivia. It features no discussion whatsoever of the origin and status of the notions of indexicality and deixis, even though these two related notions have arguably moved from the status of a philosophical problem to that of a major defining topic of contemporary pragmatics (see e.g. Levinson 1983: ch. 2).
39, K. Donnellan, 'Reference and definite description', introduces the now-famous distinction between the referential and the attributive use of definite descriptions. On its referential use, a description is employed to pick out a specific individual (rather than whoever fits the description), regardless of whether the description actually applies to that person. Thus, in "Who's the man drinking a Martini?", the description may be successfully used to pick out, or refer to, that specific person, even if he in fact turns out to be drinking water rather than a Martini. Such referentially used descriptions, Donnellan argues, come close to being purely referential expressions that establish their reference without an intermediate descriptive content or meaning: successful reference to the right person may be achieved even if the person referred to is not really drinking any Martini at all, and thus does not fit the description.
Donnellan's paper has been instrumental in the development of what has been called 'California semantics', or 'direct reference theory' (cf. Kaplan 1977), according to which the determination of the extension or reference of an expression is not mediated by a Fregean sense or Carnapian intension, but rather involves the referent itself.
40. R. Montague, 'Pragmatics', a highly technical paper, tries to implement Carnap's view of pragmatics (see volume I) in the shape of an indexical semantics that uses the tools of formal logic, especially intensional type logic. For the average reader, it will be difficult if not impossible to follow the argument of this paper without a semantics textbook at hand. Montague's work has, of course, been seminal for the development of model-theoretic semantics; but his formulation of pragmatics as an extension of this approach to context-dependence, in which the 'pragmatic' context of utterance and the 'semantic' circumstance of evaluation are conflated into a single 'index' for the interpretation of a sentence, has not been the predominant one in later years.
Rather, following Kamp (1971) and especially Kaplan (1977, 1979), nowadays a strict distinction is usually made between what has been called 'character' as a function from contexts to contents, and 'content' as a function from circumstances of evaluation to extensions. This view, in which character strictly precedes content, matches Stalnaker's approach to pragmatics as the study of contextual factors that determine propositions as opposed to semantics as the study of the propositions themselves; both in the Kamp-Kaplan and in the Stalnakerian formulation, these ideas have been widely influential in formal semantics and pragmatics. Surprisingly, however, neither Kamp nor Kaplan is included in this volume, even though their work lies at the basis of most articles that have been included. Neither is Stalnaker's 1978 paper, 'Assertion' (reprinted in volume IV), referred to here, despite its obvious relevance.
41. H. Wettstein, 'How to bridge the gap between meaning and reference', following in Kaplan's footsteps, distinguishes pure indexicals, like 'I' and 'here', from demonstratives like 'this' and 'that'. The former establish their reference in virtue of wholly linguistic rules, whereas the latter involve much more complex factors, such as acts of demonstration like pointing. Wettstein further argues that the reference of an indexical is determined by the features of communicational interaction that make the reference available (i.e., by some of the social and institutional aspects of language), rather than a causal connection between the utterance and the referent or by the speaker's intending to refer to a particular thing, as competing theories claim. He thus opens the way to an approach to indexicality that pays more systematic attention to the interactional aspects of indexicality; but in the present volume, this is no more than a suggestion.
42 S. Davis, 'Linguistic semantics, philosophical semantics, and pragmatics', distinguishes two semantic traditions, the one ('philosophical semantics') associating with a sentence an abstract object like a meaning, proposition, or thought (in the Fregean, non- mentalist sense), the other ('linguistic semantics') associating with it a specifically mentally object like a belief. In sentences containing indexicals, the Fregean thought it expresses is not identical to its linguistic, mentalistic meaning. Pragmatics, Davis concludes, should investigate the human capacity to link these two.
43 and 44. J. Hankamer & I. Sag, 'Deep and surface anaphora' and 'Toward a theory of anaphoric processing', argue, first, that syntactically controlled intersentential anaphor poses a problem for the view that discourse structure is merely a pragmatic entity defined on the meanings of its constituent sentences, which are held to be recoverable on a sentence-by-sentence basis. Consequently, they argue, there must be such a thing as the syntax of discourse. Their initial discussion, which is couched in the terminology of the 1970s versions of generative grammar, is elaborated in the later 'Towards a theory of anaphora' (first published in 1984). Here, they argue that phenomena of anaphora and ellipsis cannot be understood in terms of purely linguistic objects, but require reference to mental representations or models.
45 R. Jackendoff, 'Pragmatic anaphors and categories of concepts' (a small fragment from his 1987 book 'Consciousness and the computational mind'), teases out some of the implications of Hankamer & Sag's findings, linking claims regarding linguistic anaphora to questions of conceptual structure.
Strangely, the three papers by Hankamer & Sag and Jackendoff do not really deal with indexicality at all, but rather with anaphora; and though there undoubtedly are interesting relations between the two (witness, for example, the fact that expressions like 'that' may be used both demonstratively and anaphorically), these notions really should be kept analytically distinct, or at the very least, their interrelationship should be the subject of more explicit discussion. Moreover, these selections largely fail to address the (rather problematic) status of pragmatic phenomena like indeixicality in such mentalistic approaches to language. Their inclusion in this volume, and even in the collection as a whole, thus remains unaccounted for.
46. The inclusion of the next article, G. Nunberg, 'Indexicality and deixis', by contrast, is rather less problematic. In a characteristically dense and rewarding argument, Nunberg further refines the picture established by Kaplan and his followers, by distinguishing between deixis and indexicality, i.e., context-dependent items whose interpretation is claimed to be determined, respectively, by salience and other intention-driven factors, or by purely linguistic factors). Nunberg argues that indexicals are rather more complex than the proponents of direct reference theory like Kaplan allow for: first, they may provide rather more than an indication of how the indexical term is related to the utterance, as they may contain descriptive information about the referent's gender, number, etc. Second, indexicals may allow for what Nunberg calls 'deferred reference', like referring to a man while pointing to the flower he has been holding. Kaplan and others treat such cases as derivative or 'deviant', but Nunberg makes a case for their relevance to drawing the border between semantics and pragmatics. Numerous asides make his article a rich source of insights and observations.
47 A. Kronfeld, 'Reference and computation', by contrast, is once again a rather odd selection in this volume. It mainly discusses the question of how reference is established in dialogue; or rather, how speaker and hearer mutually come to understand the subject of conversation, and how to get a computer to achieve this kind of understanding. This is a topic of obvious pragmatic interest, involving questions of mutual knowledge, collaborative planning, and the recognition of intentions; but it is not an obviously relevant selection in a volume that deals with indexicality and reference. Discussions of mutual understanding and the establishment of topics of conversation would equally well, or actually better, in place in the parts dealing with, e.g., implicature (part 7) or talk in interaction (part 10). Moreover, Kronfeld's contribution clearly belongs to the interface between semantics, pragmatics and computational linguistics; and this raises the question why this fragment is included in this volume rather than in volume VI, which specifically deals with the interface between pragmatics and other disciplines.
It will be clear from the above that Kasher's selection is rather idiosyncratic at several junctures. Thus, it juxtaposes the concepts of indexicality and reference, which (in a pragmatic context at least) could equally well have been included in the section on speech act theory (witness chapter 4 of Searle's 'Speech acts', appropriately titled 'Reference as a speech act') or presupposition (witness Strawson's classical paper 'On referring'). At the same time, part of the selections are decidedly conceptualist in character, and consequently do not say much of substance about either indixicality or reference.
In part, however, the editorial idiosyncracies are certainly justifiable: thus, Kasher skips the earliest philosophical discussions of the phenomenon of indexicality, and the question of whether it can be eliminated from logically perfect languages -- a question that occupied the minds of, among others, Peirce, Russell, and Reichenbach. Moreover, the selections that have been included only discuss indexicality in the restricted, philosophy-inspired sense of reference to an identifiable feature or parameter of the context, like speaker, time, or place. That is, they treat contextual parameters as individual objects of some sort referred to. The question of how indexical expressions may refer to groups of people (as in 'we') or to properties (as in 'thus') does not loom large in such research.
Even less represented are discussions of more linguistic and social science inspired approaches to deixis (e.g., tense or temporal deixis), let alone 'social deixis', like the use of more or less polite pronouns (e.g., French 'tu-vous') or the obligatory use of 'politeness particles' as in Japanese. The general phenomenon of politeness is arguably a kind of context-dependence, and even less controversially a core area of pragmatic research, around which a huge literature has developed. It is, however, entirely absent from the present volume, and receives an all too brief discussion in the very last two papers included in volume VI of the collection: two older (though admittedly canonical) contributions by Leech and Brown & Levinson (1987). Once again, a more explicit motivation of the particular choices and orderings made in this set would have been welcome here. But even given the editor's apparently self-imposed thematic restrictions, the failure to include any work by Kaplan, which is perhaps the most influential approach available, and which is referred to by several of the selections that are included, is really indefensible.
In conclusion, to judge from the contents of volumes I and III, and from the list of selections included in the other four volumes, Kasher's pragmatics anthology presents an impressive array of classical papers and different approaches and perspectives. But it can hardly lay claim to the title of an almost complete, or even a relatively representative, selection of the most fruitful, relevant and influential writings in and around theoretical pragmatics up until the early 1990s. Given the sheer scope and bulk of Kasher's anthology, this is a bit of a disappointment.
REFERENCES Brown, P., & Levinson, S. (1987) Politeness, Cambridge University Press Kamp, Hans (1971) Formal properties of 'Now'. Theoria
Kaplan, David (1989 [1977]) Demonstratives. In Almog, J., a.o. (eds.), Themes from Kaplan, Oxford University Press
Kaplan, David (1979), On the Logic of Demonstratives, Journal of Philosophical Logic
Levinson, S.C., (1983) Pragmatics, Cambridge University Press
ABOUT THE REVIEWER Michiel Leezenberg teaches Philosophy of Science at the University of Amsterdam. His research interests include the semantics-pragmatics interface, the foundations of the social sciences and the history and ethnography of linguistic thought. Among his recent publications are Contexts of Metaphor (Elsevier Science 2001) and a Dutch-language textbook on Philosophy of Science for the Humanities, co-authored with Gerard de Vries (Amsterdam University Press 2001).
|