Online research in philosophy

Since Kaplan’s "Demonstratives", it has become a common-place to distinguish between the character and content of an expression, where the content of an expression is what it contributes to "what is said" by sentences containing that expression, and the character gives a rule for determining, in a context, the content of an expression. A tacit assumption of theories of character has been that character is autonomous from content – that semantic evaluation starts with character, adds context, and then derives content. One consequence of this autonomy thesis is that the rules for character can contain no variables bound by content-level operators elsewhere in the sentence. Tacit appeal to this consequence features essentially both in Jason Stanley’s recent argument that all contextual ambiguity must be linked to "elements in the actual syntactic structure of the sentence uttered" in his "Context and Logical Form" and in my arguments against character-based theories of complex demonstratives in my "Complex Demonstratives". However, I argue here that the autonomy thesis is unmotivated, and show that we can separate Kaplan’s notion of character into two independent components: an aspect of meaning which is context-sensitive, and an aspect of meaning that is exempted from scopal interactions with other operators. The resulting semantic framework allows constructions similar to Kaplan’s rejected notion of "monsters begat by elegance", but which are both more empirically adequate and more theoretically versatile. Having made the distinction between context-sensitivity and autonomy from scopal interaction, I show how it allows binding into the character of expressions and hence undermines the immediate success of both Stanley’s argument and my former argument against character-based theories of complex demonstratives, and discuss briefly the prospects for reinstating modified versions of those arguments. Finally, I show how that same distinction allows a defusing of Kripke’s modal argument against a descriptive theory of names. Once autonomy from semantic interaction is separated from context-sensitivity, the first of those two alone can be used to capture the modal rigidity of proper names..

Both context relativists and circumstance-of-evaluation relativists agree that the traditional semantic interpretation of some sentence-types fails to deliver the adequate truth-conditions for the corresponding tokens. But while the context relativists argue that the truth-conditions of each token depend on its context of utterance—each token being thus associated with a distinct intension—circumstance-of-evaluation relativists preserve a unique intension for all the tokens by placing circumstances of evaluations under the influence of a certain ‘point of view’. The main difference between the two approaches is that only the former can operate locally. It is shown that, for this reason, circumstance-of-evaluation relativism makes erroneous semantic predictions about (relative) gradable adjectives.

This paper argues in favor of a treatment of discourse about fiction in terms of operators on character, that is, Kaplanesque ‘monsters’. The first three sections criticize the traditional analysis of ‘according to the fiction’ as an intensional operator, and the approach to fictional discourse grounded on the notion of contextual shifts. The final sections explain how an analysis in terms of monsters yields the correct readings for a variety of examples involving modal and temporal indexicals.

Context Shifting Arguments (CSA) ask us to consider two utterances of an unambiguous, non-vague, non-elliptic sentence S. If the consensus intuition is that what’s said, or expressed or the truth-conditions, and so possibly the truthvalues, of these utterances differ, then CSA concludes S is context sensitive. Consider, for example, simultaneous utterances of ‘I am wearing a hat’, one by Stephen, one by Jason. Intuitively, these utterances can vary in truth-value contingent upon who is speaking the sentence, while holding hat-wearing constant, and so they express distinct propositions and differ in their truth conditions. Since these differences are not the result of ambiguity (lexical or structural), vagueness, conversational implicature, or syntactic ellipsis, we have pretty strong evidence that ‘I am wearing a hat’ is context sensitive.

This paper concerns the central method of generating evidence in support of contextualist theories, what we call context shifting experiments. We begin by explaining the standard design of context shifting experiments, which are used in both quantitative surveys and more traditional thought experiments to show how context affects the content of natural language expressions. We discuss some recent experimental studies that have tried and failed to find evidence that confirms contextualist predictions about the results of context shifting experiments, and consider the criticisms of those studies made by DeRose (forthcoming). We show that DeRose's criticisms are incomplete, and we argue that the design of context shifting experiments he proposes is itself subject to some of the same problems as the studies he criticizes. We propose a refined approach to the design of context shifting experiments that addresses these problems and which allows us to to investigate the effect of context on both positive and negative sentences. This aspect of our design allows us to control for several forms of bias, including a particular form of "truth bias" that favors positive over negative sentences. We then deploy our improved design in an experiment that tests a large number of scenarios involving different types of expressions of interest to contextualists, including "know" and color adjectives like "green". Our experiment (i) reveals an effect of changing contexts on the evaluation of uses of the sentences that we examined, thereby overturning the absence of results reported in previous experimental studies (so-called null results) and (ii) reveals previously unnoticed distinctions between the strength of the contextual effects we observed for scenarios involving knowledge ascriptions and for scenarios concerning color and other miscellaneous scenarios.

Kaplan says that monsters violate Principle 2 of his theory. Principle 2 is that indexicals, pure and demonstrative alike, are directly referential. In providing this explanation of there being no monsters, Kaplan feels his theory has an advantage over double-indexing theories like Kamp’s or Segerberg’s (or Stalnaker’s), which either embrace monsters or avoid them only by ad hoc stipulation, in the sharp conceptual distinction it draws between circumstances of evaluation and contexts of utterance. We shall argue that Kaplan’s prohibition is also essentially stipulative, and that it is too general. The main difference between ourselves and Kaplan is that the basic carriers of a truth-value is a sentence-in-a-context; our account is utterance-based.

Keith DeRose has argued that context shifting experiments should be designed in a specific way in order to accommodate what he calls a `truth/falsity asymmetry'. I explain and critique DeRose's reasons for proposing this modification to contextualist methodology, drawing on recent experimental studies of DeRose's bank cases as well as experimental findings about the verification of affirmative and negative statements. While DeRose's arguments for his particular modification to contextualist methodology fail, the lesson of his proposal is that there is good reason to pay close attention to subtle aspects of the design of context shifting experiments.

Kaplan claims in Demonstratives that no operator may manipulate the context of evaluation of natural language indexicals. We show that this is not so. In fact, attitude reports always manipulate a context parameter (or, rather, a context variable). This is shown by (i) the existence of De Se readings of attitude reports in English (which Kaplan has no account for), and (ii) the existence of a variety of indexicals across languages whose point of evaluation can be shifted, but only in attitude reports. We develop an alternative account within an extensional framework with overt quantification over times, worlds and contexts. Various typological facts are discussed, esp. the distinction between English, Amharic and Ewe pronouns, and that between English and Russian tenses.

Semantic theories that violate semantic innocence, i.e. require reference-shifts when terms are embedded in ‘that’ clauses and the like, are often challenged by producing sentences where an anaphoric expression, while not itself embedded in a context in which reference shifts, is anaphoric on an antecedent expression that is embedded in such a context. This, in conjunction with a widely accepted principle concerning unproblematic anaphora, is used to show that such reference shifting has absurd consequences. We show that it is the (...) widely accepted principle concerning anaphora that is to be blamed for these consequences, and not the supposed sin of reference shifting.

Kaplan (1989) insists that natural languages do not contain displacing devices which operate on character---such displacing devices are called monsters. This thesis has recently faced various empirical challenges (e.g. Schlenker 2003 and Anand & Nevins 2004). In this note, the thesis is challenged on grounds of a more theoretical nature. It is argued that the standard compositional semantics of variable binding employs monstrous operations. As a dramatic first example, Kaplan's formal language LD ("Logic of Demonstratives") is shown to contain monsters. For similar reasons, the orthodox lambda-calculus-based semantics for variable binding is argued to be monstrous. This technical point promises to provide some far-reaching implications for our understanding of semantic theory and content. The theoretical upshot of the discussion is at least threefold: (i) the Kaplanian thesis that "directly referential" terms are not shiftable/bindable is unmotivated, (ii) since monsters operate on something distinct from the assertoric content of their operands, we must distinguish ingredient sense from assertoric content (cf. Dummett 1973, Evans 1979 and Stanley 2002), and (iii) since the case of variable binding provides a paradigm of semantic shift that differs from the other types, it is plausible to think that indexicals---which are standardly treated by means of the assignment function---might undergo the same kind of shift.