LINGUIST List 29.4136

Wed Oct 24 2018

Review: Pragmatics; Semantics: Lasersohn (2018)

Editor for this issue: Jeremy Coburn <jecoburnlinguistlist.org>



Date: 15-May-2018
From: Enrico Torre <contactenricotorre.com>
Subject: Subjectivity and Perspective in Truth-Theoretic Semantics
E-mail this message to a friend

Discuss this message


Book announced at https://linguistlist.org/issues/29/29-807.html

AUTHOR: Peter Lasersohn
TITLE: Subjectivity and Perspective in Truth-Theoretic Semantics
PUBLISHER: Oxford University Press
YEAR: 2018

REVIEWER: Enrico Torre, Independent Researcher

SUMMARY

In his monograph ‘Subjectivity and perspective in truth-theoretic semantics’, Peter Lasersohn pursues the following objectives (p. xiii): 1) demonstrating that a truth-conditional theory can account for the meanings of sentences which concern matters of opinion; 2) arguing for the use of a model of semantic contents as determining extensions only relative to parameters left indeterminate in the context of utterance; 3) motivating the plausibility of a relativized conception of truth; 4) drawing out the implications of this strategy for the analysis of a range of specific linguistic phenomena.

The book is divided into eleven chapters. Chapter 1 defines sentences concerning matters of opinion as those whose contradiction can result in ‘faultless disagreement’ (Kölbel 2004) between two speakers, i.e. a kind of disagreement which does not imply a factual mistake on either part (e.g., ‘Licorice is tasty’). However, Lasersohn observes that sentences expressing matters of opinion are often used as if they had truth values, just like those expressing matters of fact (e.g., ‘This costs $5’). In order to preserve the distinction between the two different kinds of sentences, the author proposes the adoption of a system containing different notions of truth, including a monadic concept of truth which applies only to the contents of sentences which deal with matters of fact and, and a relativized notion of truth which applies to both types of sentences. Following Kaplan (1989), the author distinguishes a concept of meaning, labeled ‘content’ at an intermediate level between the semantic content which does not vary with pragmatic context (‘character’, in Lasersohn’s terms) and the traditional notion of ‘denotation’.

Chapter 2 reviews the following alternatives solutions: analyses which treat personal tastes as involving some kind of indexical reference; analyses which consider these sentences as involving a hidden argument to predicates like ‘tasty’, which is treated as a quantified variable; analyses of claims like ‘Licorice is tasty’ as simple one-places predicates whose truth or falseness is inaccessible, so that people have to rely on their own judgments; claims that taste sentences do not possess truth values at all; claims that disagreement between two people asserting and denying that licorice is tasty are to be explained at a metalinguistic (e.g., different meanings of the word ‘tasty’ for different speakers) or metacontextual levels (e.g., the use of the sentence may convey information about the context). The author dismisses all solutions on the grounds that some of them fail to grasp the intuitive contradiction between two people uttering sentences like “Licorice is tasty” and “Licorice is not tasty” while others require additional conditions which are not necessary for the contradiction to arise.

In the following two chapters, Lasersohn outlines a formal grammar for a fragment of the English language to be used to the relativist treatment of taste sentences in Chapter 5. Chapter 3 provides the syntactic and semantic provisions necessary to distinguish the interpretation exhibited by predicates of personal taste from that exhibited by indexical expressions. Special attention is dedicated to noun phrases, in particular to pronouns and their use as bound variables. Then, a treatment of relevant phenomena like negation and intensionality is also provided. Chapter 4 expands the grammar provided in the previous chapter to the analysis of verb tense and expressions of spatial location and orientation.

Chapter 5 expands the grammar to treat some simple cases of personal taste sentences as expressing contents whose truth value can vary from person to person. Importantly, the author introduces the ‘judge’ parameter, which assigns denotations to contents while being neutral to the determination of content. The judge parameter is a feature of the context of assessment (i.e., the situation in which a truth value is ‘judged’) rather than the context of use (i.e., the situation in which an expression is used). As a consequence, the truth value of taste sentences depends on the value of a parameter which is left indeterminate by the context of use, and this indeterminacy gives the effect that these sentences are only subjectively true or false. Once the indeterminacy is resolved, a truth value may be assigned on an objective basis. In Lasersohn’s words, “to say that ‘Elderberries are tasty’ is true relative to a context of assessment with John as the judge is to say that it is objectively true that elderberries are tasty for John” (p. 94).

Chapter 6 considers several possible alternative analyses which assume the presence of ‘hidden’ and ‘disguised’ indexical elements. The author outlines various ways of including them into his grammar: phonological reduction, pro-drop, syntactic ellipsis, constructional indexicality and sublexical indexicality, arguing that none of these solutions is viable. Indeed, appealing to hidden or disguised indexicality is of little help in explaining why one person saying, “Licorice is tasty” and another one saying, “Licorice is not tasty” are intuitively understood to be contradicting each other. Likewise, accounting for the intuition of contradiction by claiming that each speaker asserts the negation of the content asserted by the other speaker, it is impossible to explain the intuition of interpersonal variation in truth value.

Chapter 7 pairs Lasersohn’s semantic theory with a pragmatic theory that explains how the context of assessment provides specific values for their parameters and how people assess the truth values of each other’s assertions. The author argues that a system of pragmatic norms establishes default values, but in appropriate circumstances, a sentence use may be intentionally assessed relative to non-normal parameter values. Normally, the truth or falsity of a sentence content will be assessed relative to the assessor’s current perspective (‘autocentric stance’). However, a person may perform a truth assessment from another perspective than one’s own (‘exocentric stance’) to serve a specific purpose. For instance, if Mary tells John that the merry-go-round is fun, and John is trying to decide whether to buy a ticket for his son Bill, John will consider if what Mary said is likely to be true relative to Bill, rather than himself. A sentence may also be considered in the abstract, without performing a truth assessment (‘acentric stance’).

In Chapter 8, the author extends his semantic and pragmatic theory to deal with sentences ascribing mental attitudes. This is accomplished by treating attitude predicates as creating intensional contexts. The bearer of a mental attitude involving truth assessment is expected to adopt a stance in performing that assessment, which is supposed to be somehow reflected in the semantics of attitude predicates. The adoption of this perspective is illustrated through the analysis of attitudes, grammatical constructions, and their intersection, which are formalized and discussed, with a particular attention reserved to pro-dropping strategies. In the end of the chapter, the author addresses and refutes non-indexical contextualism, on the grounds that it forces a distinction between what we accept as true and what we accept someone as having truthfully said.

Chapter 9 considers the notion of relative truth as applied to the theory of speech acts, in particular, the act of assertion. First of all, the author shows that it is possible to make sense of the idea of asserting a content which is true in relation to certain perspectives but not others. This is argued both from a normative point of view (i.e., the perspective of Grice’s Maxims of Conversation) and a contextual point of view (which includes the ‘common ground’ between all the participants in a conversation). Then, the author enquires into why people insist on assertions with regard to matters of taste even in the face of opposition, despite the fact that there are no facts which can resolve the issue, arguing that the answer is to be found in terms of social advantage for one’s own tastes. At the end of the chapter, the author shifts his attention to questions, which differ from assertions in being by default exocentric.

Chapter 10 deals with the theoretical possibility that certain sentence contents may vary in truth value from perspective to perspective but, at the same time, have an objective truth value which cannot be relativized. More informally, sentence contents which can be seen as neither entirely matter of opinion nor entirely matter of fact. The author identifies and discusses several types of sentences which can be analyzed in this way: aesthetic judgments and refinement of taste, contingent futures, epistemic modality, and also cases of sentences unrelated to taste (scalar cut-offs, sufficiency, derogation). All of these are discussed in some detail and with the aid of a range of examples.

Finally, Chapter 11 enquires into the notion of truth itself. The author approaches the question by telling an evolutionary fable, arguing that truth is an idealization of the more basic one of ‘reliability’: a representation of the environment is reliable when it works for the relevant participant in a representation process. For instance, John’s communicating that a wolf is nearby will be received by Mary, who will imagine herself in John’s position, identify the relevant mental representation, and assess the utterance for reliability. According to the author, truth differs from reliability in three aspects: 1) it abstracts away from practical consequences of representation use; 2) reliability is scalar whereas truth is bivalent; 3) a representation can be reliable for certain purposes but not others, whereas truth is not purpose-sensitive. The argument is then developed and formalized

EVALUATION

Overall, this monograph represents a welcome contribution to the study of sentences concerning matters of opinion within the framework of logical semantics. As a result, it will be particularly appealing to specialists in formal linguistics and, more generally, scholars who are interested in the study of linguistic expressions of subjectivity. As such, it fits in very well in the series ‘Oxford Studies in Semantics and Pragmatics’, which include other authoritative studies such as Elbourne (2013) and Chierchia (2013). Lasersohn’s book is well-written, its background and aims are made clear in the beginning, and each chapter has a specific function. Although the volume as a whole coheres, most chapters may also stand as self-contained contributions. This is no surprise, given that the book presents the result of a research program whose ad interim results were published in several journal articles over the last fifteen years. The objectives of the book are listed at the very beginning and consistently pursued over the following chapters.

Lasersohn’s ingenious truth-conditional treatment of sentences which intuitively concern matters of opinion is backed by convincing arguments, which provide solid support in favor of the adequacy of logical semantics to account for this kind of sentence. Indeed, by means of a rigorous, yet relatively flexible formal grammar, the author is able to provide an explicit account of a range of possible linguistic expressions dealing with taste and other matters of opinion. In this respect, a the adoption of a grammar which allows multiple notions of truth, including relativistic notions of truth (which apply to all examples) and absolute notions of truth (restricted to sentences about matters of fact), plays a key role, enabling the author to demonstrate the virtues of the notion of relativized truth on both formal and theoretical grounds. At the same time, the author’s concise but thorough comparison between his own approach to subjectivity and analyses which were previously provided in the literature is appropriate, spelling out some subtle yet crucial differences which may not be immediately identified (especially by non-specialists).

The author’s choice to write a simplified grammar for a fragment of the English language rather than a wholly artificial logical language is to be praised since it is helpful for the reader, especially if s/he is not familiar with a high level of formalization. Likewise, Lasersohn’s selection of the examples to be illustrated throughout the book is very sensible. Indeed, the sentence contents taken into consideration are relatively simple and their discussion is clear; as a result, the task for the reader is again substantially simplified. Given that the range of linguistic phenomena covered in the book is relatively wide, the adoption of these strategies are very important to help the reader follow the development of the argument.

For the reasons mentioned above, my evaluation of Lasersohn’s monograph is overall positive. Indeed, this book represents a fine contribution to the study of (apparently) problematic sentences in truth-conditional semantics and it will certainly spark a lively debate in the formal linguistic circles. However, I will now focus on what I consider to be the basic weaknesses of this book. Although the book is a valuable contribution, the achievement of its objectives is partly held back by the very perspective adopted by the author. Indeed, Lasersohn’s truth-conditional treatment of sentences concerning matters of opinion is mainly directed to an audience of ‘insiders’ (i.e., formal linguists and analytical philosophers); as such, it adopts a style which may seem irksome for semanticists who do not share the same background.

Although this is a legitimate choice, this reviewer cannot help seeing this as a missed opportunity because, in principle, the book might interest a wider audience. As a matter of fact, an account of sentences concerning matters of opinion in truth-conditional terms may be of considerable interest for advocates of different approaches. For instance, the adoption of an intermediate level of meaning between ‘character’ and ‘denotation’ may be of interest to proponents of (European) structural semantics, who normally distinguish ‘signification’ from ‘denotation’. Likewise, not all advocates of functional/cognitive linguistics are happy with the abrogation of the distinction between semantics and pragmatics, which often implies a reduction of the former to the latter; on the contrary, some of them are happy to admit of the use of truth conditions, in particular to deal with assertions (on this point, see Itkonen 2018). As a consequence, adopting a less technical, more accessible point of view may have helped broadening the debate, allowing a discussion of the potential of truth-conditional semantics to reveal some points of contact among different, but not necessarily watertight theoretical paradigms. To this end, keeping the level of formalization to a minimum in favor of a more narrative style may remarkably help scholars from a plurality of backgrounds to engage in the debate. The full version of the formal grammar could still have been included as an appendix.

At the end of this review, I will just outline a couple of (in a sense, complementary) issues with the very structure of the book. On the one hand, the book does not have a concluding chapter. Of course, this is no fatal problem, but the book would have benefitted from the presence of a general conclusion providing a summary of how the objectives listed in the preface were achieved, what solution were adopted for each of the issues encountered in the previous chapters, and the directions which may be taken in the near future. Conversely, it is less than clear how the last chapter fits in with the rest of the monograph. Indeed, the question of why a language might employ expressions whose interpretations depend on context is surely important and Lasersohn’s proposed explanation is intriguing, but this issue seems to be collocated at a different, more general theoretical level than those dealt with in the previous chapters. As such, it would appear more consistent to either present this argument at the beginning of the book or omit it altogether.

REFERENCES

Chierchia, Gennaro. 2013. Logic in Grammar: Polarity, Free Choice, and Intervention. Oxford: Oxford University Press.

Elbourne, Paul. 2013. Definite Descriptions. Oxford: Oxford University Press.

Itkonen, Esa. 2018. In Defense of Traditonal Semantics (and Against All-Out Psychologism). http://users.utu.fi/eitkonen/In%20defense%20of%20traditional%20semantics.pdf (04 May, 2018)

Kaplan, David. 1989. Demonstratives. In J. Almog, J. Perry, and H.K. Wettstein (eds.), Themes from Kaplan. Oxford: Oxford University Press. 481-563.

Kölbel, Max. 2004. Faultless Disagreement. Proceedings of the Aristotelian Society 104. 53-73.


ABOUT THE REVIEWER

Enrico Torre holds a PhD in Linguistics from Lancaster University, UK. His research interests include English linguistics, theories of language and the philosophy of linguistics. He is currently investigating the notions of analogy, pattern, and family resemblance in the history of linguistics. Moreover, he is exploring the connections between contemporary linguistic theories and the structuralist tradition. In the recent past, he has analysed the patterns of use of Italian idioms.



Page Updated: 24-Oct-2018