Cognitive penetration and taste predicates: making an exception to the rule

The relevance of cognitive penetration has been pointed out concerning three fields within philosophy: philosophy of science, philosophy of mind, and epistemology. This paper argues that this phenomenon is also relevant to the philosophy of language. First, I will defend that there are situations where ethical, social, or cultural rules can affect our taste perceptions. This influence can cause speakers to utter conflicting contents that lead them to disagree and, subsequently, to negotiate the circumstances of application of the taste predicates they have used to describe or express their taste perceptions. Then, to account for the proper dynamics of these cases, I will develop a theoretical framework build upon two elements: the Lewisian idea of the score of a conversation (Lewis, 1979), and Richard’s (2008) taxonomy of the different attitudes speakers can have in taste disagreements. In a nutshell, I will argue that speakers can accommodate these conflicting contents as exceptions to the rule that determines the circumstances of application of taste predicates.Keywords: Cognitive penetration, Common ground, Circumstances of application, Accommodation, Exceptions, Score of the conversation, Taste predicates.

The Projection Problem for Predicates of Taste

Utterances of simple sentences containing taste predicates (e.g. "delicious", "fun", "frightening") typically imply that the speaker has had a particular sort of first-hand experience with the object of predication. For example, an utterance of "The carrot cake is delicious" would typically imply that the speaker had actually tasted the cake in question, and is not, for example, merely basing her judgment on the testimony of others. According to one approach, this 'acquaintance inference' is essentially an implicature, one generated by the Maxim of Quality together with a certain principle concerning the epistemology of taste (Ninan 2014). We first discuss some problems for this approach, problems that arise in connection with disjunction and generalized quantifiers. Then, after stating a conjecture concerning which operators 'obviate' the acquaintance inference and which do not, we build on Anand and Korotkova 2018 and Willer and Kennedy Forthcoming by developing a theory that treats the acquaintance requirement as a presupposition, albeit one that can be obviated by certain operators.

Judgement, taste and closely related Germanic languages

We will look at a treatment of the semantics of taste predicates using TTR (Type Theory with Records). The central idea is that we take the notion of judgement from type theory as basic and derive a notion of truth from that, rather than starting from a semantics based on a notion of truth and trying to modify it to include a notion of judgement. Our analysis involves two types of propositions: Austinian propositions, whose components include a situation and a type, and a subtype of Austinian propositions called subjective Austinian propositions, whose components in addition include an agent who makes the judgement that the situation is of the type. We will argue that attitude verbs can select either for propositions in general (subjective or objective) or for subjective propositions, but that there is no type of objective propositions which can be selected for. We will discuss some apparent counterexamples to this from Germanic languages and argue that there is a phenomenon akin to switch reference in certain attitude predicates when their complement involves a subjective proposition.

Disagreement about Taste as Disagreement about the Discourse: Problems and Limitations

In the present paper I present the metalinguistic solutions to the ‘lost disagreement’ problem proposed (independently) Sundell and Plunkett [2013] and Barker [2012]. I argue that metalinguistic negotiations about taste, even though successful in explaining the intuition of disagreement in a vast number of cases, are not an accurate solution to the disagreement problem in contextualism when it comes to the most paradigmatic case of “tasty”. I also argue against the account of faultless disagreement explained via vagueness of taste predicates [Barker, 2012]. I believe that the notion of faultlessness employed in the discussion of vagueness [Wright, 1994] is a different notion than the one employed in the discussion of taste discourse [Kölbel, 2003].

Gradable adjectives and disagreement about personal taste

Contextualism and Relativism offer competing semantic accounts of personal taste predicates. I argue in this paper that Michael Glanzberg?s defense of contextualism from one relativist argument-the Lost Disagreement Argument-is not successful. I show that Glanzberg?s scalar analysis of the adjectives from which personal taste predicates are built fails to capture the characteristic subjectivity of these predicates. I propose an alternative analysis according to whicheach personal taste adjective denotes multiple functions from a set of objects to an ordered scale of measurement of the appropriate dimensional property. This analysis succeeds where Glanzberg?s fails and it favors a relativist treatment of personal taste predicates.

Taste predicates and the acquaintance inference

Simple sentences containing predicates like "tasty<br /> and "beautiful" typically suggest that the speaker has first-hand knowledge of the item being evaluated. I consider two explanations of this "acquaintance inference": a presuppositional approach, and a pragmatic-epistemic approach. The presuppositional approach has a number of virtues, but runs into trouble because the acquaintance inference has a very different projection pattern from that of standard presuppositions. The pragmatic-epistemic approach accounts for the main data discussed in the paper, but faces challenges of its own.

The dynamics of subjectivity

I adapt the dynamic framework for vagueness of Barker 2002 to the analysis of subjective taste predicates. I argue, following Kennedy 2013, that there are two qualitatively distinct types of subjectivity in natural language, which I call mapping subjectivity and (vague) standards subjectivity, and that the matrix predicate *find* is sensitive to the distinction between them. Novel to the present analysis is the proposal that *find* requires not just a complement that supports mapping subjectivity, but also a context that supports nonvacuous entailments about those scalar mappings. I part ways from Kennedy and from Barker 2013 in treating mapping subjectivity as a fact of the world, unassimilable to the metalinguistic variety of subjectivity associated with vague standards.

The dynamics of subjectivity

I adapt the dynamic framework for vagueness of Barker 2002 to the analysis of subjective taste predicates. I argue, following Kennedy 2013, that there are two qualitatively distinct types of subjectivity in natural language, which I call mapping subjectivity and (vague) standards subjectivity, and that the matrix predicate *find* is sensitive to the distinction between them. Novel to the present analysis is the proposal that *find* requires not just a complement that supports mapping subjectivity, but also a context that supports nonvacuous entailments about those scalar mappings. I part ways from Kennedy and from Barker 2013 in treating mapping subjectivity as a fact of the world, unassimilable to the metalinguistic variety of subjectivity associated with vague standards.