In ‘Kripke on epistemic and metaphysical possibility: two routes to the necessary a posteriori’, Scott Soames identifies two arguments for the existence of necessary a posteriori truths in Naming and Necessity (NN). He argues that Kripke’s second argument relies on either of two principles, each of which leads to contradiction. He also claims that it has led to ‘two-dimensionalist’ approaches to the necessary a posteriori which are fundamentally at odds with the insights about meaning and modality expressed in NN. I argue that the alleged second argument is not in NN. I identify the mistakes that lead to Soames’ misinterpretation.

There has been a groundswell of interest in the account of modality that Kant sets forth in his pre-Critical Only Possible Argument. Andrew Chignell's reconstruction of Kant's theistic argument in terms of what he calls ‘real harmony’ has a prima facie advantage in that it appears to be able to block the plurality objection (namely, that even if every modal fact presupposes some ground, this does not entail that all modal facts share the same ground). I argue that it is both textually and philosophically problematic to interpret Kant's argument in terms of real harmony. Then, I set forth an alternative response to the plurality objection which does not require the adoption of the problematic notion of real harmony. Instead, I argue that the objection can be overcome by observing that the argument seeks to ground modal facts as a totality and that, according to Kant, such relations can be accounted for only by their schematization in a single intellect.