This is a review of David Chalmers's /Constructing the World/. The short, short version: there are issues, but you should definitely read the book.

In Meaning and Necessity (1947/1950), Carnap advances an intensional semantic framework on which modal claims are true in virtue of semantical rules alone, and so are a priori. In 'Empiricism, Semantics, and Ontology' (1950), Carnap advances an epistemic-ontological framework on which metaphysical claims are either trivial or meaningless, since lacking any means of substantive confirmation. Carnap carried out these projects two decades before Kripke influentially argued, in Naming and Necessity (1972/1980), that some modal claims are true a posteriori. How should (...) a neo-Carnapian respond to Kripke's results? Some (notably, Chalmers and Jackson, in their 2001) have suggested that an extension of intensional semantics along lines of "epistemic two-dimensionalism" can accommodate Kripke's results while largely preserving commitment to the semantics-based a priority of modal claims. Here we consider how best to implement this suggestion, and how the resulting semantics fits with Carnap's second project. We find that the most promising (and most Carnapian!) post-Kripke version of Carnap’s semantics---abductive two-dimensionalism---presupposes an epistemology which undermines Carnap's metaphysical anti-realism. (shrink)

A number of philosophers—among them Joseph Levine, David Chalmers, Frank Jackson and Jaegwon Kim—have claimed that there are conceptual grounds sufficient for ruling out the possibility of a reductive explanation of phenomenal consciousness. Their claim assumes a functional model of reduction (regarded by Kim as an alternative to the traditional Nagelian model) which requires an a priori entailment from the facts in the reduction base to the phenomena to be explained. The aim of this paper is to show that this (...) is an unreasonable requirement—a requirement that no reductive explanation in science should be expected to satisfy. I argue that the functional model is not substantively different from the Nagelian model properly understood, and that the question whether consciousness is reductively explainable—in a sense involving property identifications or in some weaker sense compatible with Nagelian reduction—is a fundamentally empirical question, not one that can be settled on conceptual grounds alone. 1. Introduction2. Kim's critique of the Nagelian model of reduction3. The functional model of reduction4. Is consciousness reducible?5. Psychophysical reduction: concluding remarks. (shrink)

This paper compares and contrasts two conceivability arguments: the zombie argument (ZA) against physicalism, and the perfect actor argument (AA) against behaviourism. I start the paper by assuming that the arguments are of the same kind, and that AA is sound. On the basis of these two assumptions I criticize the most common philosophical suggestions in the literature today about what is wrong with ZA, and what is right in it. I end the paper by suggesting that the comparison between (...) the two arguments makes plausible an epistemic response to ZA according to which the conceivability of a zombie - that is, someone identical to me in all physical respects but different in some phenomenal respect - is being confused with the conceivability of something else, i.e. someone identical to me in physical respects that are epistemically available, but different in some phenomenal respect. (shrink)

Type-B materialists (to use David Chalmers's jargon) claim that though zombies are conceivable, they are not metaphysically possible. This article calls this position regarding the relation between metaphysical and epistemic modality “modal autonomism,” as opposed to the “modal rationalism” endorsed by David Chalmers and Frank Jackson, who insist on a deep link between the two forms of modality. This article argues that the defense of modal rationalism presented in Chalmers and Jackson (2001) begs the question against the type-B materialist/modal autonomist. (...) The argument proceeds as follows. Modal rationalists claim that for all nonphenomenal macro properties, the appropriate supervenience conditional is both necessary and a priori. Hence, type-B materialists must engage in special pleading when they claim that the relevant supervenience conditional for phenomenal properties, expressing the supervenience of the phenomenal on the physical, is necessary but not a priori. However, what Chalmers and Jackson demonstrate, if anything, is that the conditional that includes all the microphysical plus the phenomenal in the antecedent, and nonphenomenal macro facts (such as facts about water and other natural kinds, among other things) in the consequent, is a priori. The question arises why, since facts about water and the like do not metaphysically supervene on the phenomenal facts, is it appropriate to include the phenomenal facts in the antecedent of the relevant supervenience conditional. This article argues for the following claims: First, that it's crucial to the general semantic framework Chalmers and Jackson defend that they do include the phenomenal facts in the supervenience conditional; without them, the conditional would not be a priori. Second, that the only way to argue from the a priori character of these conditionals to the applicability of modal rationalism to the nonphenomenal cases is to rely either on modal rationalism itself or on the denial of type-B materialism. Obviously, in the context of this argument, either way would beg the question. (shrink)