Printer Friendly

G. Preyer and G. Peter, eds.: Context-Sensitivity and Semantic Minimalism.

G. Preyer and G. Peter, eds.

Context-Sensitivity and Semantic Minimalism.

New York: Oxford University Press 2007.

Pp. 363.

US$150.00 (cloth ISBN-13: 978-0-19-921332-0); US$49.95 (paper ISBN-13: 978-0-19-921331-3).

The present volume is a timely collection, as both context sensitivity and semantic minimalism are hot topics in current philosophy of language and linguistics, especially in the border dispute between semantics and pragmatics. However, all but the most advanced reader will likely be a bit lost with this volume due to the lack of any editorial guidance. Readers are basically left to fend for themselves in making sense of the disputes (terminological and substantive) in the fifteen chapters comprising this book. Most of the papers take Cappelen and Lepore Insensitive Semantics (Blackwell 2005), as their point of departure (pro or con), and Part 1.1 of Pagin and Pelletier (below) is a useful summary. There are seven chapters under the title T. The Defense of Moderate Contextualisin' (with contributions from P Pagin and J. Pelletier, K. Taylor, K. Kepa and J. Perry, I. Maitra, S-J. Leslie, E. Corazza and J. Docik, E. Camp); six chapters under the title 'II. On Critiques of Semantic Minimalism' (with contributions from J. Atlas, J. MacFarlane, L. Clapp, R. Elugardo, E Robins, H. Jackman), ending with a section entitled 'Back to Semantic Minimalism' with a single article by E. Borg. There are no introductions accompanying Parts 1 or 2, so the reader does not know what holds the chapters in these parts together, nor what the difference between a defense of contextualism and a critique of minimalism might be. Here is a bit or orientation.

Lots of things go by the label 'context sensitivity' (or, better, 'knowledge-of-context' sensitivity). Perry usefully distinguishes three roles for context (cf. the Introduction to Recanati, in Perspectivai Thought, Oxford 2007, for an elaboration of Perry's taxonomy). First, the pre-semantic role of context is to fix the language and operative meaning of the expression uttered. Second, the semantic role of context is to determine extension, given meaning (or 'character'). This can be done in two ways: i) allowing context to help fix the content (e.g., proposition) expressed, ii) allowing context to contribute to the circumstance of evaluation (for Kaplan typically a world-time pair), which then fixes extension (e.g. truth-value). Or both, depending on the case. For instance, 'The universe is expanding' uttered at time t might express an eternal content true at any time in case the universe is expanding at t. Or it might express a neutral content, true in circumstance t if the world is expanding at t, but possibly false in circumstances when it is not. There is mischief to be found in having the word 'context' label these two sensitivities; a generic 'situation-sensitivity' with two species, 'context' and 'circumstance' sensitivity, would help. Third, the post-semantic role of context is to provide implicit or 'unarticulated' information carried by the utterance, and perhaps such things as implicatures, nonliterality and indirection. One dispute regarding context sensitivity has to do with whether implicit information is best captured at the level of content (most 'contextualists'), circumstance of evaluation (Lewis, MacFarlane), or both (Recanati, op. cit.).

Lots of things go by the name of 'semantic minimalism'. The basic idea common to all forms is that the semantic content of an utterance should be as 'close' as possible to the (compositional) semantics of the sentence uttered. There are four favored ways of getting from linguistic meaning to proposition(s) expressed: i) by 'modulating' the meaning of overt constituents, ii) by 'saturating' overt constituents, such as indexicals, iii) by adding 'unarticulated' constituents by 'completing' a propositional function or radical, or iv) by 'expanding' one proposition into a more appropriate one. Some authors, e.g., Cappelen & Lepore, countenance only a small set of con text sensitive items ('contextuais' and 'indexicals'), but allow the fixation of reference of these terms to appeal to speakers' intentions. Others, e.g., Borg, subscribing to modularity, abjure appeals to speakers' intentions in favor of 'individual concepts' of the referents. Finally, some authors (Bach, Recanati) embrace the role of unarticulated constituents in the content expressed; others (Cappelen & Lepore) resist it, or limit it to existential quantification (e.g., Borg's 'liberal' truth conditions), or restrict its scope by finding evidence that it is really not as 'unarticulated' as supposed (e.g., Stanley). Those are a lot of issues (and there are more)--too many to cover here--so I will say a few words about the chapters that open and close the volume and about a chapter each from the book's two main parts.

Cappelen's introduction to the volume, 'Semantics and Pragmatics: Some Central Issues', is very useful in orienting the reader to issues surrounding the phenomena of context-sensitivity (if not to the other articles in the volume). He sets out the problem of contextual variation and illustrates it with a selection of examples from the literature. He then illustrates how three explanatory strategies might apply. For example, a semantic explanation of variability of content would be the standard account of indexicals such as T and 'now'. A pragmatic account of variability of content might appeal to i) mechanisms of conversational implicature, or ii) variation in speech act contents due to a) the lack of a proposition ('Kiara has had enough'), or b) speech act multiplicity, or c) unarticulated constituents. An index explanation would not apply to traditional contents, but to distributions of truth value, per MacFarlane (below). Cappelen lists six challenges that the field must meet to make progress, including: developing new diagnostic procedures, clarifying the metaphysics of propositions, relating semantic to speech act content (what is said-asserted), understand contextual mechanisms, accounting for shared content, and fitting in compositionality. He ends by provocatively announcing that there is no semantics-pragmatics distinction and that looking for it is a 'waste of time'.

Korta and Perry's chapter ('Radical Minimalism, Moderate Contextualism') follows Kaplan in holding that linguistic meaning is a feature of linguistic types and determines the 'semantic contributions' to the proposition expressed, whereas 'locutionary content' is a feature of linguistic tokens and is rooted in intuitions about what is said. They favor, along with Cappelen and Lepore and others, a minimalist semantics (lexical meaning plus composition), and hold that pragmatics contributes to what is said, and that there are multiple things said on most occasions of utterance. They think, contra Cappelen and Lepore and others, that the concept of what is said is theoretically useful, but they make no mention of the experimental failures to find such a stable pretheoretic concept. They see an informationally incremental series of propositions expressed, and the utterance-bound proposition has a distinctive role to play by quantifying over the contextual and the intentional--it can be computed from linguistic information alone.

Semantic minimalists, such as Cappelen and Lepore, hold that the sentence 'Jane is tall' can expresses many propositions, such as that she is tall for a seven year old or tall for a basketball player. But if it does it will always also express the proposition P: that Jane is (just plain) tall. MacFarlane ('Semantic Minimalism, Nonindexical Contextualisin') doubts that there is such a proposition (the 'intension problem'--what would a world be like for P to be true?). He argues that P is entailed by any proposition ascribing height to Jane, even P*: 'Jane is (just plain) short'. So if Jane has any height she is (just plain) tall and short. Not good. MacFarlane's solution, which he calls 'nonindexical contextualism' is to incorporate a 'counts as' parameter in the circumstance of evaluation and let the evaluation of the (just plain) proposition vary from circumstance to circumstance. Thus, in a 'basketball player' circumstance P might be evaluated false, but in a 'seven year olds' circumstance is might be evaluated true; hence, context sensitivity without traditional content variability. One major problem to be faced is managing the 'counts as' parameter; is there one "-valued parameter or "-parameters? How are they identified? And are they identified in advance, or only in the circumstance? In any event, much more work needs to be done here.

According to Borg ('Minimalism vs Contextualism in Semantics'), minimalists thinks of contents as the literal meaning of sentences and the result of little or no pragmatic processes, whereas contextualists hold that pragmatic effects are endemic in literal content, and defend their position with three types of argument: i) context shifting ('Jill is tali'), ii) incompleteness ('Steel isn't strong enough'), and iii) inappropriateness ('There is nothing to eat'), where the proposition semantically expressed is not contextually appropriate. In response, Borg's minimalists hold four theses: i) every indexical-free sentence expresses a proposition, ii) only obviously context sensitive expressions such as indexicals require contextual input, iii) semantic content is not the content of the speech act being performed, and iv) there is always a syntactic trigger for contextual contributions to content. On Borg's view it is not the number of context-sensitive items that mark the minimalist-contextualist divide, as Cappelen and Lepore might have it, but rather the type of content and the mechanism that delivers it.

Robert M. Hamish

University of Arizona
COPYRIGHT 2009 Academic Printing and Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Hamish, Robert M.
Publication:Philosophy in Review
Article Type:Book review
Date:Oct 1, 2009
Previous Article:Catherine Mills: The Philosophy of Agamben.
Next Article:Yuriko Saito: Everyday Aesthetics.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters