Judges, experiencers, and taste

ABSTRACT

This paper reviews the claim that certain predicates, including what are called predicates of personal taste, have a sometimes-hidden element for a judge or experiencer. This claim was advanced in my own earlier work, as well as a number of other papers. My main goal here is to review some of the arguments in favor of this claim, and along the way, to present some of my earlier unpublished work on the matter. In much of the earlier literature, this claim was part of a debate between relativists, contextualists, and others about the semantics of ‘subjective’ or ‘perspectival’ predicates. I shall argue here that these issues are independent. Whether we opt for experiencer or judge parameters is independent of whether we prefer relativist semantics to any other kind.

KEYWORDS:

• Predicates of personal taste
• judge
• experiencer
• contextual parameters
• semantics

Acknowledgments

This paper grew out of a presentation given at the conference on Pluralism, Relativism, and Skepticism organized by the Middle East Society for Analytic Philosophy in 2019. Thanks especially to Sherif Gamal Salem, both for organizing the conference and encouraging me to write this paper. Thanks also to Markus Kneer and an anonymous referee for comments on an earlier draft. I have borrowed heavily from some unpublished work of my own. For that work, thanks to Kent Bach, Paul Elbourne, Jeff King, and Adam Sennet for many conversations about the ideas I have been developing. Earlier versions of what became a long unpublished paper I have borrowed from were presented at the Workshop on Context and Intentions at the Center for the Study of Mind in Nature, University of Oslo, February 2009; the Workshop on Relativism at the Institute of Philosophical Research, National Autonomous University of Mexico, February 2009; the Conference on the Contextualist Challenge in the Philosophy of Language, Queens University, Ontario, September 2009; the University of Buenos Aires, April 2013; and the Kline workshop on Context-Sensitivity in Language, University of Missouri, November 2016. Thanks to all the participants at those events, and especially Emma Borg, Herman Cappelen, Lenny Clapp, Chris Gauker, Claire Horsik, Peter Lasersohn, and Alex Radulescu for helpful comments and discussions. Special thanks to Maite Ezcurdia, my commentator at the UNAM workshop, for giving such insightful comments under less than ideal circumstances.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1 Though epistemic modals have been central to the relativism debate, I shall for the most part ignore them here. They bring with them their own complications, which are beyond the scope of this paper. See the survey by Egan (2011) for many references.

2 To my knowledge, the name comes from Kölbel (2002), though for taste predicates, it was highlighted by Lasersohn (2005). Early work of MacFarlane (e.g. MacFarlane, 2003) focused on different phenomena but was certainly along similar lines. MacFarlane takes up taste predicates in MacFarlane (2014).

3 Relativism is of course, a long-standing philosophical issue, going back to pre-Socratic times. See the overview by Baghramian and Carter (2021) for more background. The focus here is on a form of relativism that emerged around the beginning of the 21st century, focusing on philosophy of language, semantics, metaphysics, and related areas. One striking thing about this revival of relativism is that it took place among people who were inclined to dismiss many forms of relativism as a mistake. Many of us thought of relativism as what we sometimes derisively called ‘freshman relativism’: the tendency of our introductory students to simply say things are ‘true for you but not me’ and that ‘everything is relative’. There was also not in this group much enthusiasm for more sophisticated ideas we might find about ‘conceptual schemes’ from such authors as Kuhn (1962) or Putnam (1988) or Strawson (1959). The revival offered a form of relativism that seemed to those of us skeptical of the idea to be level-headed, well-argued, and right or wrong, a really interesting new development. Pioneering works of Egan, Hawthorne, and Weatherson (2005) and Lasersohn (2005) and MacFarlane (2003) took somewhat different approaches to the issue, but came to related conclusions.

4 Lasersohn (2017) presents a different formulation, but as he stressed in his earlier work, part of the idea was to present a conservative departure from familiar semantic machinery. I shall follow suit. Relativism has many different forms. Important work from Egan (2007), Kölbel (2002), MacFarlane (2005, 2014), Recanati (2007), and Richard (2004) all offer different forms. There is some useful discussion of the varieties of relativism in Kölbel (2003), MacFarlane (2009), and Weatherson (2009). A strongly anti-relativist position is offered by Cappelen and Hawthorne (2009). For our purposes, the semantically conservative model of Lasersohn (2005) will be helpful.

5 The issue is somewhat more complicated for Lewis himself.

6 This is one of the leading approaches to all gradable adjectives, including tall, rich, etc. See Barker (2002), Bartsch and Vennemann (1972), Bierwisch (1989), Cresswell (1977), Heim (1985), Kennedy (1997, 2007), Rett (2015), and von Stechow (1984). Alternative views are defended by Burnett (2014) and Klein (1980).

7 See Glanzberg (2007) for a more extended presentation of this view.

8 See the discussion in Kaplan (1989a, 1989b), Lewis (1970, 1980), and Stalnaker (1998). One further point. I happen to disagree with the Kaplan-inspired view that indices include times (Glanzberg, 2011; Glanzberg and King, 2020; King, 2003). As the Kaplan framework is familiar, I have followed it here in spite of that.

9 A good discussion of some of these issues is from Huvenes (2012).

10 As I leave the main relativism/contextualism debate, let me mention one more issue. There is a debate over the status of retraction and the nature of disagreement. See López de Sa (2015), MacFarlane (2007), Marques (2018), Marques and García-Carpintero (2014), Zakkou (2019), and the survey by Zeman (2017). For interesting empirical discussion, see Kneer (2021a).

11 Well, I did in Glanzberg (2009/2016).

12 I am here mashing together Lasersohn and MacFarlane. MacFarlane’s view has a number of very different features, to which I am not doing justice. But I think one of the lessons from MacFarlane can be put this way.

13 Very common does not mean universally agreed upon. For an overview of the issues about ellipsis, see Merchant (2019).

14 For more on psych adjectives, see Bennis (2000) and Landau (1999, 2006, 2010b). For a more relativist take on these issues, see Lasersohn (2008).

15 There are some complex syntactic questions about the nature of the implicit argument, that relate to the structure of the passive itself. We will not be concerned with that structure, but see Baker, Johnson, and Roberts (1989) and Jaeggli (1986).

16 For overview of the debate, see Bhatt and Pancheva (2006), Jones (1991), and Landau (2000). Landau concludes that the argument fails in this case, but that it is much better off in the case of experiencers. Control has been a central topic in syntax for many years, notably since Chomsky (1965) at least. But at the same time, it remains a difficult, contentious, and perhaps not fully understood phenomenon. I shall not try to survey the many issues that make this so. See Landau (2013) for a comprehensive overview.

17 In the case of ‘optional’ oblique arguments, like instrument arguments, psycholinguistic results appear quite complex. See, for instance, Rissman, Rawlins, and Landau (2015).

18 These observations are picked up by the more extensive discussions of personal taste in Moltmann (2010), Schaffer (2011), and Snyder (2013).

19 A related construction for adjectives like stupid is discussed by Barker (2002), focusing mostly on semantic properties. But this sort of evaluative adjective behaves differently from predicates of personal taste, and has a different argument structure. For more discussion of the syntax, see Bennis (2000, 2004), Cinque (1990a), Landau (1999, 2006), and Stowell (2004).

20 As I mentioned, Landau also concludes that the case of agentless passives is not obligatory control, though Higginbotham (1997) defends the control argument for the existence of the agent argument. See also Landau (2010b).

21 This is not to claim that we have the simple argument that PRO is always controlled by a syntactically represented controller. That is a contentious claim. As I have already noted, not all control structures pattern with obligatory control. Moreover, there are a number of examples in the literature of ‘pragmatic control’, where no controller seems to be available. These typically involve discourse effects, like:

(i) John agreed to kill Mary. But he instantly felt some hesitation. To kill her would leave poor little Billy without a mother.

(This example is from Peter Lasersohn (p.c.), but related ones can be found in Bresnan (1982).) As Higginbotham (1997) notes, the judgments on these sorts of cases are rather delicate. But more importantly, the cases of obligatory control, where discourse effects are not present, give us a strong enough pattern to provide some substantial evidence. Of course, this makes the argument for the existence of the parameter indirect, built from a number of syntactic patterns all of which make more sense if we posit the implicit argument.

22 For more discussion, see Cinque (1990b) and Hicks (2009). Though relatively little is known about the properties of this particular construction, see Lasnik and Fiengo (1974) for some interesting observations.

23 Schaffer (2011) and Snyder (2013) come to similar conclusions.