ABSTRACT
According to the resemblance account of ‘what it's like’ and similar constructions, a sentence such as ‘there is something it's like to have a toothache’ means ‘there is something having a toothache resembles’. This account has proved controversial in the literature; some writers endorse it, many reject it. We show that this conflict is illusory. Drawing on the semantics of intensional transitive verbs, we show that there are two versions of the resemblance account, depending on whether ‘resembles’ is construed notionally or relationally. While well-known criticisms of the resemblance account undermine its relational version, they do not touch its notional version. On the contrary, the notional version is equivalent to various accounts usually interpreted as rivals to resemblance. We end by noting that this resolution of the controversy (a) explains why ‘like’, which is a comparative, appears in a construction that concerns the properties of events and (b) removes any pressure to suppose that ‘like’ is ambiguous between a comparative and a non-comparative sense.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 See also Lewis (Citation1988), Hellie (Citation2004), Stoljar (Citation2016), among many others.
2 Occurrences of ‘something’ in intensional positions are what Friederike Moltmann (Citation1997, Citation2003, Citation2008) calls special quantifiers, which differ from ordinary objectual quantifiers in that they do not replace expressions whose function is to refer – they are ‘non-nominal quantifiers’.
3 Here we paraphrase Zimmermann slightly, for readability. His analysis is presented formally by Meier (Citation2009) as follows: ‘John resembles a pig’ is true iff there is a possibly complex property such that
1. | , and | ||||
2. | For all y, if y is a prototypical representative of pigs, then . |
On analyses of this sort, John can resemble a pig even if there are no pigs; all that is required is that he has certain properties. Meier then goes on to develop this analysis to accommodate the fact that ‘resembles’ is not only an intensional transitive verb, but also a multidimensional comparative. On her view, ‘resembles’ is a multidimensional intensional comparative.
4 For discussion of this point, see Stoljar (Citation2016).
5 The traditional statement of this view comes from Montague (Citation1974), who derived the relational reading of sentences involving ITVs using a special rule of quantifying in, together with a type-shift. Montague's view has been widely, although not universally, adopted. Proponents of similar analyses include Quine (Citation1956), Zimmermann (Citation1993, Citation2006), and Richard (Citation2000), among many others. Fodor (Citation1970) criticizes scopal analyses of the different readings of attitude verbs on the grounds that scope cannot generate all of the available readings. But even those who do not think that the notional/relational distinction is a scopal one accept that the distinction does not arise from a lexical ambiguity in the verb – verbs such as ‘seek’, ‘need’, and ‘owe’ are not lexically ambiguous. Our point in the text is that the same is true for ‘resemble’.