The Tyranny of the Subjunctive

David J. Chalmers

Much of contemporary philosophy of language, philosophy of mind, and metaphysics rests on Kripke's analysis of necessity in Naming and Necessity. Kripke's analysis itself rests largely on claims about what is true of certain possible worlds. And these claims are themselves grounded in intuitions about certain subjunctive conditionals. I argue that it is possible to use indicative conditionals in this role instead, yielding a different sort of evaluation of truth in possible worlds (epistemic evaluation), and a correspondingly different analysis of necessity, one on which necessity is tied much more closely to epistemic notions. Further, I argue that there is no substantive reason to favor subjunctive conditionals over indicative conditionals in the analysis, so no reason to favor Kripke's evaluation of truth in a possible world or his analysis of necessity. This unargued bias in favor of subjunctive conditionals has affected almost everything in philosophy (e.g. direct reference, externalism, essentialism) over the last thirty years. I will speculate on how different philosophy might look (e.g. Fregeanism, internalism, anti-essentialism) if Kripke* had used indicative conditionals instead.