Daniel Tiskin

This reply to Oleg Domanov’s target paper is not concerned with the technicalities of the proposed approach. Rather, I discuss the fruitfulness of the underlying ideas in dealing with Quine’s famous “double vision” scenario, for which the approach is designed. I point out some key ingredients of Domanov’s proposal: (a) context dependence of propositional attitude ascription (and ascribability); (b) replacement of individuals with finer-grained entities for reference and quantification, such as Kaplan’s “vivid names”, Frege and Yalcin’s senses or Percus and Sauerland’s concept generators; and (c) using the apparatus of cross-identification functions. I show that those ingredients were already present in a body of work preceding the target paper. On the other hand, there are known problems related to the fact that sometimes the choice of the pertinent mode of presentation depends on the choices associated with quantifiers higher in the syntactic tree. No account based on manipulations with the global context, such as Domanov’s in its current form, can handle them.

As Maier’s aim is to extend the notion of unreliable narration onto film, this reply focuses on the consequences of the difference between textual and filmic narration. Textual fiction imitates, or at least uses the resources typical of, a true textual description of events, which is itself highly conventional in that it uses arbitrary linguistic signs and chooses to describe those properties of objects and events that matter to the author, leaving the remainder unspecified. On the contrary, filmic narration imitates the perception of real events of which the watcher is supposed to be witness. Even if the arrangement of frames is conventional, the content of a particular frame is presented to the observer as if the latter happened to be at the scene, thus in the totality of its detail; and the connection between the object filmed and its depiction in film is causal rather than conventional. Moreover, it is natural even for non-fictional texts to describe the scene in some rhetorically plausible order, whereas a real-life scene presented to our sight by pure chance need not follow any coherent plot.

Intentional Identity, introduced by Geach :627–632, 1967), refers to pairs of attitude reports where a pronoun embedded into the second report is anaphoric on a quantifier embedded into the first one. In the Geach sentence the antecedent carries no commitment to the existence of witches, and moreover the sentence does not require that Nob should know anything about Hob or Hob’s mental state. This fact has given rise to the conviction, almost universally shared, that in Intentional Identity reports the anaphoric pronoun cannot be D-type, i.e. that it cannot borrow its reference and descriptive content directly from its antecedent.We show that the perceived non-committing truth conditions can be derived via a D-type analysis of pronouns, which are taken to be syntactically complex. The crucial ingredient of the proposal is that the predicate within a pronoun in Intentional Identity ascriptions receives a “non-specific transparent” reading ; Schwager ), so the second attitude holder is required to know Hob’s thoughts no more than Ralph is required to know Ortcutt’s name in the famous scenario due to Quine :177–187, 1956).

Many theories of the de dicto / de re ambiguity for quantifiers and descriptions follow the tradition started by Kaplan and Lewis in that they make use of notions that are epistemic in nature, such as the notion of acquaintance. This may create the impression that the question about de re in affitude report semantics should always be resolved by looking at the reported affitude; if the latter qualifies as de re according to some epistemological criteria, then also the affitude report may be true de re. The present paper aims to provide an argument to the contrary. The argument proceeds in several steps. First, we point out the existence of a wide range of de re readings (mostly already known from the literature), some of which do not target referential or quantificational expressions at all. Second, we show that the existing analyses either give wrong predictions for such cases, or are incomplete, or at least inelegant. Third, we offer a new analysis (which, as it turns out, has predecessors not united by any single tradition) whose main ingredient is the observation that the right predictions result from modifying the Context Set of the conversation in certain intuitive ways before the affitude report is added to the Common Ground. This is the semantic contribution of the paper. The philosophical upshot is that the reasons for an affitude report being de re at least in some cases have little to do with the grounds on which the corresponding mental state - the affitude itself - qualifies as de re. We also speculate as to how the proposed analysis, which divorces de re'ness from epistemic rapport, may be extended onto more traditional cases.

The comment is focused on two aspects of the commented paper, viz. the choice of the appropriate auxiliary formalism for the analysis of Russell's joke about the yacht, and the more general issue of what sort of arguments are acceptable in the logical investigation of natural language semantics. As for technical issues, we suggest that an extensional language should be used which contains variables over possible worlds and over degrees and show how Russell's joke could be accounted for in such a formalism; in addition, some remarks concerning Borisov's technical work are provided. At the methodological side, we argue against purely intuitive considerations regarding the presence or absence of this or that component in the semantics of a given lexical item. Real evidence as to which sentences containing that word are acceptable and what they mean should be used instead.