On Truth-Conditions forIf(but Not Quite OnlyIf)

Abstract

What we want to be true about ordinary indicative conditionals seems to be more than we can possibly get: there just seems to be no good way to assign truth-conditions to ordinary indicative conditionals. Some take this argument as reason to make our wantings more modest. Others take it to show that indicative conditionals don't have truth-conditions in the first place. But we have overlooked two possibilities for assigning truth-conditions to indicatives. What's more, those possibilities deliver what we want and turn out to be equivalent.

A Uniform Theory of Conditionals

“…In this paper I raise three problems for this attempt at uniformity and propose a new uniform theory that solves them. This improved theory builds on the analysis of indicative conditionals developed in Starr (2013a) and Gillies (2009), which already has important advantages over Stalnaker's (1975). I will also show that the compositional explausible analysis of conditionals.…”

“…I will begin by evaluating whether or not Stalnaker's proposal can provide a satisfactory answer to these questions. After arguing that it cannot, I will propose a semantics for ¡ and combine it with the semantics for indicatives offered in Starr (2013a) and Gillies (2009) to yield a uniform analysis of conditionals. For simplicity, ¡ will be modeled as a specialized sentential operator.…”

“…These difficulties are solved by the new analysis proposed in §3. That analysis takes the dynamic semantics for indicative conditionals advanced in Gillies (2009) and Starr (2013a) and transforms it into a uniform theory of conditionals that explains the basic contrasts between the two varieties by paralleling the extension/revision contrast. On this theory, the antecedents of indicative and subjunctive conditionals propose different hypothetical changes to the information at stake in a conversation or inquiry.…”

Parsing and presuppositions in the calculation of local contexts

“…However, attention to the presuppositional component of the meaning of a conditional does point the way towards a promising solution. Conditionals are commonly thought to presuppose that their antecedent is compatible with the context set (see Stalnaker 1975, von Fintel 1998, Gillies 2009. This is very natural to implement on the strict conditional approach, according to which conditionals essentially have the meaning of a universal quantifier over the context set, since universal quantifiers are generally taken to presuppose that their restrictor has nonempty intersection with their domain (see Heim & Kratzer 1998 and citations therein).…”

Lifting conditionals to inquisitive semantics

“…In many-though not all-of these approaches, a conditional sentence is taken to express a proposition. Without any pretense of exhaustiveness, let me mention a few accounts of this kind: the material account, familiar from classical logic; strict accounts (e.g., Warmbrōd 1981;Gillies 2009); the selection function account (Stalnaker 1968); the variably strict account (Lewis 1973); the restrictor account (Lewis 1975;Kratzer 1986); premise semantics (Kratzer 1981); and causal accounts (Schulz 2011;Kaufmann 2013). 1 All these accounts work under the assumption that the antecedent and the consequent of a conditional express a proposition, and describe how these propositions are used to determine the proposition expressed by the conditional.…”