Terry Horgan
Publications
Decision Theory (and related topics)
Counterfactuals and Newcomb's Problem, Journal of Philosophy (1981),331-56. Reprinted in The Philosopher's Annual, 1981, and in R. Campbell and L. Sowden (eds.), Paradoxes of Rationality and Cooperation (U. of British Columbia Press, 1985).
I argue that one-box reasoning and two-box reasoning in Newcomb's problem both employ counterfactual conditionals essentially, and that the crux of the problem concerns the question of how best to resolve the vagueness of the relevant counterfactuals. If the appropriate resolution is what David Lewis calls the "standard" resolution, then two-box reasoning prevails. But I argue that a nonstandard resolution is appropriate, for puposes of practical decision making, and hence that one-box reasoning prevails. I then discuss the implications for decision theory.
Newcomb's Problem: A Stalemate, in R. Campbell and
L. Sowden, eds., Paradoxes of
Rationality and Cooperation (
Let's Make a Deal, Philosophical Papers 24 (1995), 209-22.
Paul Moser and D. Hudson Mulder have recently argued, on the basis of a decision problem known as the Monty Hall Problem, that it is sometimes rational to do in an isolated individual case what one knows in advance would not be rational to do over a long run of cases, all of which are just like the given case in all relevant respects. I show, to the contrary, that the single-case probabilities in the Monty Hall Problem coincide with the statistical probabilities, and more generally, that epistemic single-case probabilities necessarily coincide with the corresponding epistemic statistical probabilities.
The
Two-Envelope Paradox, Nonstandard Expected Utility, and the Intensionality of
Probability, Nous 34 (2000),
578-602.
I argue that an adequate
diagnosis of the reasoning that generates the two-envelope paradox reveals two
important morals: (1) epistemic probability is intensional (i.e.,
substitutivity salva veritate fails for sentential contexts governed by the
probability operator), and (2) this intensionality generates hitherto
unnoticed, nonstandard, forms of expected utility. The flaw in the paradoxical
reasoning, I claim, is that it involves a form of nonstandard expected utility
whose maximization is not rationally appropriate in context. I propose a
general normative principle to govern the rationally appropriate maximization
of various kinds of nonstandard expected utility in various decision situations.
The
Two-Envelope Paradox and the Foundations of Rational Decision Theory, in B. Brogaard and B. Smith, eds., Rationality and Irrationality: Proceedings
of the 23rd International Wittgenstein Symposium. öbv
& hpt (2001), 172-91. [HTML]
Sleeping
Beauty Awakened: New Odds at the Dawn of the New Day, (Forthcomming in Analysis).
[HTML]
