On Being a Random Sample
At times we are inclined to reason as though we are random samples from a larger group of individuals. Sometimes this involves using a fact about the group to draw conclusions about ourselves. (I call this reasoning 'inwards.') And sometimes it involves reasoning the other way around—starting with ourselves and working our way out to the group or the world as a whole. (This is reasoning 'outwards'.) A principled rule for how de se evidence should affect de dicto credences, and vice versa, would help us solve several puzzles in confirmation theory. (Among these are Sleeping Beauty, Doomsday, and the question whether the fine-tuning of the universe counts as evidence for the existence of many universes.) This paper looks at three competing rules for reasoning inwards and outwards, each of which replaces the standard conditioning rule for updating on evidence.
God and the Bayesian Conception of Evidence
Evidential arguments for and against the existence of God ask us to do something we are not very good at: set aside our knowledge of some very salient facts in order to reconstruct the hypothetical probability of those facts given competing hypotheses. There may be no alternative, but this process is beset with the danger of cognitive bias and should be undertaken with vigilance. It can help to honestly ask oneself whether one would have been willing to accept the opposite outcome as evidence in the other direction.
Fine Tuning and Confirmation Theory: Some Complications
This paper was given as a critique of the probabilistic fine tuning argument at the St Thomas Summer Seminar in 2015. I point out some problems with standard fine tuning arguments and draw a more general conclusion about how evidence is treated in debates about the existence of God. Some sections of these notes overlap with "God and the Bayesian Conception of Evidence", but there is much more in-depth discussion of the Fine-Tuning argument.
Moral Realism and Semantic Plasticity
Work in progess.
Are moral terms semantically plastic—that is, would very slight changes in our patterns of use have shifted their meanings? This is a delicate question for moral realists. A 'yes' answer seems to conflict with the sorts of intuitions that support realism; but a 'no' answer seems to require a semantics that involves hefty metaphysical commitments. This tension can be illustrated by thinking about how standard accounts of vagueness can be applied to the case of moral terms, and also by considering how realists should respond to the Moral Twin Earth problem. I argue that moral realists can accept the semantic plasticity of moral expressions while accounting for contrary intuitions in a way that is nearly cost-free.
The reference book
In this book, we rethink the semantic phenomenon of reference and the cognitive phenomenon of singular thought. Rejecting any special relation of causal or epistemic acquaintance, we explore a semantic account that unifies definite and indefinite descriptions with names and demonstratives. On our account, all four types of expression are specific existentials, each with its own presuppositional profile. We argue that many of the phenomena associated with reference are due to the covert use of singular restrictions on the quantifier domain.
- Kent Bach, 'Consulting The Reference Book'
- Michael Devitt, 'Lest Auld Acquaintance Be Forgot'
- Genoveva Martí, 'For the Disunity of Semantics'
- James McGilvray, 'Review of The Reference Book'
And our response from that issue:
Keeping Up Appearances: A Reducer's Guide
forthcoming in Journal of Philosophy
Metaphysicians with reductive theories of reality like to say how those theories account for ordinary usage and belief. A typical strategy is to offer theoretical sentences, often called ‘paraphrases’, to serve in place of various sentences that occur in ordinary talk. But how should we measure success in this endeavor? Those of us who undertake it usually have a vague set of theoretical desiderata in mind, but we rarely discuss them in detail. My purpose in this paper is to say exactly what they are, and why.
Dispositions Without Teleology
Oxford Studies in Metaphysics, vol 10
We argue against accounting for dispositions (and of the progressive aspect) in terms of a fundamentally teleological metaphysics, and we defend our previous conditional account from some novel objections. (Coauthored with Ryan Wasserman)
Dispositions, Conditionals, and Counterexamples
Our earlier paper about dispositions in Mind elicited three response papers: one by Daniel Bonevac, Josh Dever, and David Sosa; one by Sungho Choi, and one by Barbara Vetter. In this paper, we respond to our critics, focusing on the role of centering and of counterexamples in refuting conditional analyses of dispositions. (Coauthored with Ryan Wasserman)
The Folk Probably Do Think What You Think They Think
Much of experimental philosophy consists of surveying 'folk' intuitions about philosophically relevant issues. Are the results of these surveys evidence that the relevant folk intuitions cannot be predicted from the ‘armchair’? We found that a solid majority of philosophers could predict even results claimed to be 'surprising'. But, we argue, this does not mean that such experiments have no role at all in philosophy. (Co-authored with Billy Dunaway & Anna Edmonds)
Safety, Content, Apriority, Self-Knowledge
I motivate a revised version of safety and then use it to (i) challenge traditional conceptions of apriority, (ii) refute ‘strong privileged access’, and (iii) resolve a well-known puzzle about externalism and self-knowledge by showing that it can be treated as any other closure puzzle. Along the way I illustrate why sensitivity is of little use when it comes to certain kinds of closure puzzle.
When Best Theories Go Bad
It is common for contemporary metaphysical realists to adopt Quine’s criterion of ontological commitment while at the same time repudiating his ontological pragmatism. This paper argues that the resulting approach to meta-ontology is unstable. In particular, if we are metaphysical realists, it may be best to repudiate some of the ontological commitments incurred by our best first-order theories. This version fixes typos in the published version.
Dispositionality: Beyond the Biconditionals
Suppose dispositions bear a distinctive connection to counterfactual facts, perhaps one that could be enshrined in a variation on the well-worn schema ‘Necessarily, x is disposed to ϕ in ψ iff x would ϕ in ψ’. Could we exploit this connection to provide an account of what it is to be a disposition? This paper is about four views of dispositionality that attempt to do so.
A Gradable Approach to Dispositions
We argue that previous theories of the relationship between dispositions and conditionals are unable to account for the fact that dispositions come in degrees. We also propose a fix for this problem which also avoids the familiar problems of finks and masks. (Coauthored with Ryan Wasserman.)
On Linking Dispositions With Conditionals
Analyses of dispositional ascriptions in terms of conditional statements famously confront the problems of finks and masks. We argue that even conditional analyses of dispositions tailored to avoid finks and masks, face a battery of new difficulties: (i) Achilles' heels, (ii) accidental closeness, (iii) comparatives, (iv) explaining context sensitivity, and (v) absent stimulus conditions. We conclude by offering a proposal that avoids all seven problems. (Coauthored with Ryan Wasserman.)
Properties and Resemblance Classes
I examine the competing merits of resemblance-class theories of properties, arguing that the ‘companionship’ and ‘imperfect community’ problems are not avoided by appealing to classes of tropes instead of objects. If I am right, trope theory loses one of its primary selling points, and resemblance nominalists of either type must appeal to certain troublesome remedies.
A Guided Tour of Metametaphysics
in Metametaphysics (OUP, 2009).
Critical study of Mumford’s Dispositions
review of Correia's Existential Dependence and Cognate Notions