My conclusion should not be a surprising one. I think that alief governs all sorts of belief-discordant behavior—the cases with which I began the paper, and the ones that I have presented along the way. But if alief drives behavior in belief-discordant cases, it is likely that it drives behavior in belief-concordant cases as well. Belief plays an important role in the ultimate regulation of behavior. But it plays a far smaller role in moment-by-moment management than philosophical tradition has tended to stress.

The aim of this article is to expand the diet of examples considered in philosophical discussions of imagination and pretense, and to offer some preliminary observations about what we might learn about the nature of imagination as a result. The article presents a number of cases involving 'imaginative contagion': cases where merely imagining or pretending that P has effects that we would expect only perceiving or believing that P to have. Examples are offered that involve visual imagery, motor imagery, fictional emotions, and social priming. It is suggested that imaginative contagion is a more prevalent phenomenon than has typically been recognized.

Contemplating imaginary scenarios that evoke certain sorts of quasi‐sensory intuitions may bring us to new beliefs about contingent features of the natural world. These beliefs may be produced quasi‐observationally; the presence of a mental image may play a crucial cognitive role in the formation of the belief in question. And this albeit fallible quasi‐observational belief‐forming mechanism may, in certain contexts, be sufficiently reliable to count as a source of justification. This sheds light on the central puzzle surrounding scientific thought experiment, which is how contemplation of an imaginary scenario can lead to new knowledge about contingent features of the natural world.

By the age of two, children are able to engage in highly elaborate games of symbolic pretense, in which objects and actions in the actual world are taken to stand for objects and actions in a realm of make-believe. These games of pretense are marked by the presence of two central features, which I will call quarantining and mirroring (see also Leslie 1987; Perner 1991). Quarantining is manifest to the extent that events within the pretense-episode are taken to have effects only within that pretense-episode (e.g. the child does not expect that ‘spilling’ ( pretend) ‘tea’1 will result in the table really being wet), or more generally, to the extent that proto-beliefs and proto-attitudes concerning the pretended state of affairs are not treated as beliefs and attitudes relevant to guiding action in the actual world. Mirroring is manifest to the extent that features of the imaginary situation that have not been explicitly stipulated are derivable via features of their real-world analogues (e.g. the child does expect that if she up-ends the teapot above the table, then the table will become wet in the pretense), or, more generally to the extent that imaginative content is taken to be governed by the same sorts of restrictions that govern believed content.

Through careful analysis of a specific example, Parfit’s ‘fission argument’ for the unimportance of personal identity, I argue that our judgements concerning imaginary scenarios are likely to be unreliable when the scenarios involve disruptions of certain contingent correlations. Parfit’s argument depends on our hypothesizing away a number of facts which play a central role in our understanding and employment of the very concept under investigation; as a result, it fails to establish what Parfit claims, namely, that identity is not what matters. I argue that Parfit’s conclusion can be blocked without denying that he has presented an imaginary case where prudential concern would be rational in the absence of identity. My analysis depends on the recognition that the features that explain or justify a relation may be distinct from the features that underpin it as necessary conditions.

What I call the puzzle of imaginative resistance is the puzzle of explaining our comparative difficulty in imagining fictional worlds that we take to be morally deviant. I argue: (1) that imaginative resistance is a widespread feature of our engagement with fiction; (2) that the primary source of imaginative resistance is not (as previous commentators have suggested) our inability to imagine morally (and otherwise) deviant situations, but our unwillingness to do so; and that (3) this unwillingness is explicable if imagining involves something in between belief and mere supposition.

Ever since Locke (and particularly in the last 50 years or so) the philosophical literature on personal identity has centred on arguments of a certain type. These arguments use an assumed convergence of response to purely imaginary cases to defend revisionary conclusions about common-sense beliefs concerning the nature or importance of personal identity. So, for instance, one is asked to contemplate a case in which A's brain is transplanted into B's body, or a case in which some of C's memories are implanted in D's brain, or a case in which information about the arrangement of the molecules which compose E is used to create an exact replica of E at another point in space-time.
Thinking about these cases is supposed to help us tease apart the relative roles played by features that coincide in all (or almost all) actual cases, but which seem to be conceptually distinguishable. So, for instance, even though we can ordinarily assume that the beliefs, desires, memories, etc. which are associated with a given body will not come to be associated with another body, it does not seem to be in principle impossible that such a state of affairs should come about. Indeed, it seems that we can describe a mechanism by which such a situation might come about: for instance, A's brain (and with it A's beliefs, desires and memories) might be transplanted into B's body. And since the scenario described strikes us as something of which we can make sense, it seems we can make judgments of fact or value about which of the two factors really matters in making A who she is. We might ask, for instance, whether it would be true to say that A had survived in a body that used to belong to B, or whether it would be right to punish the B-bodied human being for A's actions, or whether if we were A before the intended operation, we ought to worry about what would be happening to the B-bodied person afterwards. And on the basis of these judgments about what we would say in the imaginary case, we can return to the actual case having learned something about which features are essential and which accidental to our judgments concerning the nature or value of personal identity. My goal in this paper is to suggest reasons for thinking that this methodology may be less reliable than its proponents take it to be, for interesting and systematic reasons.