Abstract: There are two definitions of 'pragmatic encroachment' that are widely accepted in the literature on encroachment. Unfortunately, both definitions are inadequate. If we accept these definitions, then the claim that knowledge is subject to pragmatic encroachment ends up being entailed by a certain contextualist view of knowledge ascriptions, which I call shiftable contextualism. At the start of this talk, I will explain how shiftable contextualists endorse the letter of the claim that knowledge is subject to pragmatic encroachment, while denying the intended substance of this claim.

In the rest of the talk, I defend shiftable contextualism, pointing out several virtues of this view that have so far gone unnoticed. Shiftable contextualism successfully generates bound readings of certain knowledge ascriptions, where rival views fail to generate these readings. Shiftable contextualism can account for many of the ordinary language judgments that motivate pragmatic encroachment, while at the same time avoiding counterintuitive verdicts often used to argue against encroachment. Shiftable contextualists also vindicate the theoretical principles that motivate pragmatic encroachment, such as the principle that rational action must be based on knowledge. With the right sort of contextualist view at hand, we can reject the thesis that knowledge is subject to pragmatic encroachment, while retaining the attractive consequences and applications of this thesis. In short, we can do without encroachment.