A Remark on Iffy Oughts

Abstract

Every adequate semantics for conditionals and deontic ought must offer a solution to the miners paradox about conditional obligations. Kolodny and MacFarlane have recently argued that such a semantics must reject the validity of modus ponens. I demonstrate that rejecting the validity of modus ponens is inessential for an adequate solution to the paradox. 1 The Miners Paradox Every adequate semantics for conditionals and deontic ought must offer a solution to the miners paradox about conditional obligations. Kolodny and MacFarlane (2010) have recently argued that such a semantics must reject the validity of modus ponens. My goal in this paper is to demonstrate that rejecting modus ponens is inessential for solving the puzzle. I begin with a brief outline of Kolodny’s and MacFarlane’s analysis of the paradox and their reasons for rejecting modus ponens. Then I develop and defend a semantics for deontic conditionals that avoids the miners paradox while preserving the validity of modus ponens. The key observation of this paper is that Kolodny’s and MacFarlane’s case against modus ponens trades heavily on assumptions about logical consequence that their very own semantics shows are dubious. The validity of modus ponens may very well have its limits, but we have no compelling reason to think that it is invalid for deontic conditionals and certainly would need more than the miners paradox to show that it is. Here is the miners paradox. Ten miners are trapped either in shaft A or in shaft B, but we do not know which one. Water threatens to flood the shafts. We only have enough sandbags to block one shaft but not both. If one shaft is blocked, all of the water will go into the other shaft, killing every miner inside. If we block neither shaft, both will be partially flooded, killing one miner. Published in the Journal of Philosophy 109(7): 449–461, 2012 1I observe elsewhere that modus ponens is in fact a problematic rule of inference for socalled “Thomason conditionals” such as “If Mary is deceiving me, I’ll never believe it” (see Willer (2010) for discussion). My goal here is then not to offer an unqualified defense of modus ponens but rather to contribute to the (no less interesting) project of determining the exact scope of its validity.