Frank Jackson's (1998) supervenience argument against non-naturalism fails, because its premises entail that non-naturalism is false even under the supposition that the normative fails
to supervene on the natural. But non-naturalism is clearly true
under such a supposition. Jackson's premises are too strong, and I develop an under-explored approach on which non-naturalists can reject the conjunction of these premises.
Expressivism, even of a quasi-realist variety, has metaphysical commitments that differ from realism, since it has different consequences for the naturalness or eliteness of the normative.
The theoretical benefits of analyzing the modal operators ☐ and ♢ in quantificational terms have been
assumed to come with an ontological cost. The cost is an ontology of “possible worlds”, which may be either the concrete worlds of David Lewis, or else some kind of abstract entity. In either case, quantifying over these entities is supposed to be necessary for formulating adequate modal analyses of semantic content and supervenience, and for capturing intuitive claims about cross-world resemblance and duplication. I show how, with two independently motivated resources, we can reject the assumption that the benefits of quantificational analyses require this ontology. The resources in question are primitive second-order quantifiers, which bind variables in predicate-position and have no analysis in terms of first-order
quantifiers, and a hyperintensional connective like 'in virtue of'. With these resources, we can formulate
quantificational analyses of ☐ and ♢ which avoid commitment to an ontology of possible worlds.
Much of contemporary experimental philosophy involves taking surveys of 'folk' concepts. The results of these surveys are often claimed to be surprising, and treated as evidence that the relevant folk intuitions cannot be predicted from the 'armchair'. We conducted an experiment to test these claims, and found that a solid majority of philosophers could predict even results that were claimed to be surprising in the literature. We discuss some methodological implications as well as some possible explanations for the common surprisingness claims.
Some expressivists (most notably, Simon Blackburn) have claimed that expressivists can adopt a piecemeal strategy
to show that all of the sentences realists accept are consistent with expressivism. I argue that the project cannot be carried out in the way Blackburn describes. This is because Blackburn's claims about the meaning of the realist's sentences commit him to claims about the meaning of the parts of these sentences— commitments the realist does not share.
Ethical Vagueness and Practical Reasoning
Sometimes it is vague what one ethically ought to do. The question I ask in this paper is whether there is a sense of 'ought' (perhaps an ought of practical rationality) which determinately yields verdicts about what to do in some of these borderline ethical cases. My answer is that there is, but what one practically ought to do depends on the theory of vagueness one adopts. I outline an approach available to the Epistemicist which takes as central the notion of expected moral value, and argue that standard linguistic approaches to vagueness cannot yield the same practical verdicts. I close by outlining how these differences might be seen as an instance of a widespread issue in metaethics, which involves the conflict between pre-theoretic intuitions about ethics and the overall best theory of the subject matter.
Luck: Evolutionary and Epistemic
In an epistemic framework that is knowledge–first (holding knowledge to be the primary aim of our epistemic activity) and safety–theoretic (holding that a safety condition on knowledge captures the way in which knowledge requires non–lucky true belief), an interesting evolutionary argument against the claim that moral beliefs can constitute knowledge goes as follows. Since the evolutionary pressures that shaped our moral belief–forming capacities didn't select for particular beliefs because these beliefs captured the truth about morality, we would be lucky in a sense if they did select for true beliefs. This is a kind of evolutionary luck. If evolutionary luck is a species of epistemic luck, then the relevant moral beliefs, even if true, are not knowledge. I argue that when epistemic luck is understood in terms of the safety condition, this argument in its general form is unsound. I close by outlining how a luck–based argument like this relies on the details of the relevant empirical science, and how instances that refer to specific evolutionary processes (and not evolutionary processes in general) may yet be sound.
Primitive Relative Fundamentality
The notion of metaphysical fundamentality is typically explained as an absolute notion, distinguishing what is (fully) fundamental from what is not. But the theoretical role for fundamentality of grounding reference, laws, etc. requires a relative or comparative notion of fundamentality. I explore the options for reconciling these two aspects of fundamentality, and endorse a solution which takes relative fundamentality as primitive.
Natural and Unnatural Relations (with David Manley)
Many philosophers hold that there is some kind of connection between natural properties (in David Lewis's sense) and properties that are resemblance-conferring. We ask how a connection of this kind might be preserved in the case of highly natural relations.
Reference Magnetism as a Solution to the Normative Twin Earth Challenge (with Tristram McPherson)
This paper explains how reference magnetism, when grounded in metaphysical fundamentality, provides a solution to the Moral Twin Earth argument from Horgan and Timmons (1992).