Abstract
In a recent article, Lars Bo Gundersen has suggested a new semantics for counterfactuals (Pacific Philosophical Quarterly 85: 1-20, 2004). The new semantics gives up on the idea that the actual world must be at the center of world orderings. Instead, statistically normal worlds are at the center. Part of the motivation for this idea is that a very influential account of knowledge, so-called ‘tracking’ accounts, seem right, but suffer serious problems with more traditional semantics for counterfactuals. Gundersen's semantics, if correct, helps tracking accounts with certain problems, but leads to new problems that seem intractable. So while Gundersen's semantics might be correct, those who endorse tracking accounts of knowledge should be loath to endorse it.
© Philosophia Press 2005