on pejoratives cannot be conducted without analyzing the occurrences of these terms. I hope that this will not disturb or offend my readers and that the theoretical contribution will compensate for any discomfort.
NOTE
1 1. Copp (2001, 2009) did propose an analysis of moral terms based on conventional implicatures. However, he characterizes conventional implicatures in such a way that they are very different from Grice’s and Potts’ and they are thus labeled ‘conventional simplicature.’
THE PRESUPPOSITIONAL ACCOUNT OF HYBRID EVALUATIVES
In this section I develop a unified account of two classes of terms that have provoked a lively debate in linguistics, philosophy of language, ethics and metaethics: slurs and thick terms. These expressions seem to have a hybrid nature, as they carry at once descriptive and evaluative contents. The label ‘hybrid evaluative,’ introduced in Cepollaro and Stojanovic (2016), works as an umbrella term that covers both slurs and thick terms (and possibly other expressions). The core tenet of this unified account is the following: both classes of terms carry descriptive content at the level of truth conditions, but they presuppose an evaluative content, i.e. they trigger an evaluative presupposition. Thus, for example, the adjective ‘lewd’ has roughly the same truth-conditional content as ‘sexually explicit beyond conventional boundaries,’ but it also triggers the presupposition that things that are sexually explicit beyond conventional boundaries are bad because of being so. Along similar lines, I analyze a slur like ‘wop’ as having the same truth conditions as ‘Italian’ but triggering at the same time the presupposition that Italians are bad because of being Italian.
According to this view, slurs are very similar to other evaluatives: so-called objectionable thick terms. Many authors (Eklund 2011, 2013, Harcourt and Thomas 2013, Kyle 2013, Väyrynen 2013) distinguish between unobjectionable and objectionable thick terms: the former convey an evaluation that speakers are willing to accept as appropriate or warranted; the latter convey an evaluation that speakers are not willing to share. Whether the HE-evaluation is shared/sharable is completely orthogonal to whether it is negative or positive: a speaker might approve of the negative evaluation associated with ‘selfish’ and reject the one associated with ‘lewd’; she may share the positive evaluation conveyed by ‘courageous’ and reject the one associated with ‘chaste.’ These two features, polarity—the HE-evaluation being positive (P) or negative (N)—and objectionability—the HE-evaluation being warranted (W) or unwarranted (U)—give rise to four combinations. While thick terms seem to admit all four combinations (PW ‘generous,’ PU ‘chaste,’ NW ‘brutal,’ NU ‘lewd’), slurs tend to instantiate NU only, i.e. they typically convey negative and objectionable evaluations. Unlike the polarity of the evaluation, the objectionable/unobjectionable distinction is not lexically encoded, but depends on the sets of values endorsed by speakers.
The first chapter of part I introduces the class of hybrid evaluatives (HEs): I settle some preliminary issues about the definition of ‘hybrid evaluative’ and discuss the presuppositional behavior of these expressions. In chapter 2, I focus on the semantics of HEs. chapter 3 is dedicated to the conversational dynamics which slurs and thick terms give rise to, by focusing on the ways in which people can react to them. In chapter 4, I defend the two main tenets of my proposal—(i) the uniformity claim: slurs and thick terms should be analyzed along the same lines, and (ii) the presuppositionality claim: slurs and thick terms trigger evaluative presuppositions—against potential objections. In chapter 5, I consider non-standard uses of HEs, including the reclamation of slurs and the variability of thick terms.
A New Class
1.1 WHAT COUNTS AS A HYBRID EVALUATIVE
‘Evaluative,’ ‘expressive,’ ‘normative,’ ‘pejorative,’ ‘slur,’ ‘thick,’ and ‘thin’ are widespread labels in ethics, metaethics, philosophy of language and linguistics, and yet there is not much consensus about how to understand these notions. In this section I put forward a criterion to distinguish HEs from descriptive terms that can be used evaluatively. I elaborate on the controversial notion of ‘group’ involved in the definition of slurs (section 1.1.1) and discuss the problematic thin-thick dichotomy by apppealing to the distinction between at-issue and not-at-issue content, while acknowledging that a clear-cut divide is hard to obtain (section 1.1.2). I briefly discuss some proposals that take into consideration the idea that slurs and thick terms can be analyzed along the same lines (section 1.1.3).
In the literature on thick terms, the paradigmatic examples are expressions such as ‘lewd,’ ‘generous,’ ‘courageous,’ ‘chaste,’ and ‘cruel.’ Thick terms are studied in metaethics (Hare 1963, Williams 1985, Blackburn 1992, Gibbard 1992, among others) and are typically believed to combine descriptive and evaluative meanings. The former can usually be paraphrased: the descriptive content of ‘lewd’ is something like ‘sexually explicit beyond conventional boundaries’; ‘generous’ descriptively corresponds to ‘willing to give without expectation of compensation,’ ‘chaste’ to ‘abstaining from sexual intercourse,’ and so on. Moreover, thick terms also convey (or express, depending on what theory one favors) a value judgment about such an object or individual. Thus, in calling a person ‘lewd,’ speakers usually imply something negative about the person, while in judging an act as ‘generous,’ they typically convey a positive evaluation of the act at stake. Scholars investigate how the use of a thick term is associated with value judgments, but they disagree about what semantic and pragmatic machinery makes it possible for them to express or convey evaluations.
In recent years, philosophers and linguists have focused their attention on another class of terms that appear to mix description with evaluation, namely, slurs. Slurs are derogatory terms targeting individuals and groups on the basis of their belonging to a certain category, such as sexual orientation (terms such as ‘faggot,’ targeting homosexual men; or ‘dyke,’ targeting homosexual women, etc.), nationality (‘wop,’ targeting Italians; ‘boche’ targeting Germans, etc.); ethnic origins (‘chink’ targeting Asian people, especially Chinese, etc.) and so on and so forth. Once more, these terms raise interrogatives as to whether they lexically encode evaluation, whether these expressions allow non-offensive uses (Brontsema 2004; Bianchi 2014a; Croom 2014, Cepollaro 2017a, Ritchie 2017), whether they can be mentioned without provoking offense (Hornsby 2001; Anderson and Lepore 2013a, 2013b), whether they constitute a legitimate and uniform lexical category (Croom 2011, Nunberg 2018), and so on.
It is not always clear how to determine whether a certain lexical
