(1) According to a great many views of intention, rationally intending action brings with it a belief condition; you must believe (or fail to disbelieve) that you will so act. According to the Lockean picture of belief and credence, believing some proposition is a matter of having sufficiently high credence in it. I argue these two positions, although independently plausible and popular, are jointly inconsistent. Roughly, the problem derives from cases where one intends to act but has low credence they will do so. Since Lockeans hold high credence makes for belief (and low credence does not) this will typically violate any belief condition placed upon intention. I fill out the sketch just given, consider objections and close the discussion by arguing that this is a reason to drop both Lockeanism and its justificatory analogue.

(2) There is a relatively intuitive notion of ‘activity’ according to which it is an instance of agency, somehow comprised of individual actions. One example of this is joint action. When a set of individual people perform a set of individual actions in a certain way, they all take part in the same activity. In the literature on joint action, there is more or less consensus that intentional joint action must be caused by some kind of joint intention. If one accepts the Davidsonian orthodoxy that for something to be an action (intentional or not) it must be caused by an intention, a certain conclusion is tempting: For something to be a joint action (intentional or not) it must be caused by a joint intention. Unfortunately, there are simple counterexamples to this claim. If that is so, we need a different way to explain what it is for sets of actions to form activities or joint activities. I develop an austere view according to which a set of individual actions form an activity just in case they depend on each other for their relation of part to whole. One virtue of this view is that facts about which activities are performed do not depend on facts about personal or collective identity. This, I suggest, has wide-ranging normative implications.