Mental agency has seen a resurgence of interest from philosophers in recent years, after a prolonged period of neglect. Wayne Wu’s excellent book contributes to cementing the status of this topic as a target of philosophical study in its own right, independently of bodily agency. It offers fresh and well worked-out perspectives, in both substance and method, on what is involved in the mind’s making movements. Part I of the book, comprising the first two chapters, integrates and reworks previously published pieces by Wu on the structure of action, highlighting the central place he finds within the structure of action for the notion of bias. Part II (Chs. 3 & 4) spells out and defends a novel theory of intention as ‘practical memory’. And Part III (Chs. 5–7) draws out some of the theoretical implications of the book’s general outlook on the nature of mental agency for specific case studies, including biases in attending, inference, and the introspection of perceptual experience.
Alongside its substantive theoretical contributions, a couple of which I discuss briefly below, the book is noteworthy for its methodological innovation. Going beyond empirically informed philosophy, Wu is explicit about his aspiration to offer ‘a biology of agency, writ large where philosophy plays a part’ (p. 1; emphasis in the original). This intended methodological orientation, or reorientation, is worth bearing in mind when reading the book, though it should not and does not detract from engaging with it as philosophy.
A principal focal point for Wu’s theorizing, in this book as well as elsewhere, is attention. In fact, few can claim to have done more than him in recent decades to revive philosophical fascination with attention – a fascination which, as is so often the case when the pendulum of philosophical fashion swings in one direction, has waned in the aftermath of seminal discussions by the likes of William James (Reference James1950) and F.H. Bradley (Reference Bradley1886) around the turn of the previous century. Today, thanks in no small part to Wu’s own work, the level of interest from philosophers in attention better matches the steady stream coming from cognitive scientists, driving home the centrality of this mental process and the need to give it better philosophical grounding. I personally share Wu’s conviction that attention is an utterly central psychological process for theorizing about the nature of mental agency and the mind more broadly (as I argue elsewhere; see Levy, Reference Levy2019, and Levy, Reference Levy, Brent and Miracchi2023). Nevertheless, at certain points in the book I found myself wondering if Wu isn’t asking attention to do too much work.
One place where this query comes up is a distinction drawn in Chapter 2 between attending-in-action and attending-as-action. The former for Wu is a component of (all) action, whereas the latter is a kind of action (e.g., looking or listening). As he puts it, ‘every action involves attention though not every action is an attending’ (p. 64). It is the former, constituent sense of the term, that Wu is primarily interested in, recruiting it to explain the nature of mental action within a broadly ‘causalist’ template. It is this component that constitutes the agent’s ‘guiding a response’ (p. 62) when acting, thereby also serving to exclude causal deviance (§2.10; see also Wu, Reference Wu2016). The agent from Chisholm’s (Reference Chisholm and Lehrer1966) famous example of deviance resolves to drive over to his uncle’s house and murder him, but is so unnerved by this decision that on his way there he accidentally knocks over a pedestrian who turns out to be his uncle. Had the nephew guided his response by attending to his target, he would have recognized his victim and intentionally run him over. Attention would have eliminated deviance.
I have doubts about Wu’s appeal to attention to solve the problem of causal deviance for causalism about intentional action. For one thing, attention does not guarantee recognition: Chisholm’s nephew may have failed to notice that his victim is in fact his uncle even while looking straight at him. Further, the proposed solution does not cover cases of so-called ‘secondary deviance’, where a response may be fully attentive and still deviantly bring about its effect – as with Davidson’s example of an agent who (attentively, we may assume) shoots at someone and misses, ‘but the shot stampedes a herd of wild pigs that trample the intended victim to death’ (Davidson, Reference Davidson and Honderich1973, pp. 152–3; for other doubts about Wu’s treatment of deviance, see Shepherd, Reference Shepherd2021, Ch. 2). Even more fundamentally, however, I am not confident that we have been given sufficiently good reason to accept Wu’s very distinction between attention-as-component and attention-as-action. Thinking of ‘intention’ – another of Wu’s primary theoretical targets here – Elisabeth Anscombe memorably warned at the outset of her classic monograph on this topic that ‘[w]here we are tempted to speak of “different senses” of a word which is clearly not equivocal, we may infer that we are in fact pretty much in the dark about the character of the concept which it represents’ (Reference Anscombe1963, p. 1). There is a sense of ‘attending’, as when we look at or listen to or touch O, that seems clearly to convey an action performed on O. If Wu’s (intricate, sophisticated) case for thinking of attention as a constituent of action trades on a different sense of a term that does not seem equivocal, this leaves in place a residual theoretical tension.
As noted, intention, like attention, is itself one of the focal points of “Movements of the Mind”. The book contains a striking account of the nature of intention as ‘memory for action’. On Wu’s gloss, ‘intending is the action of practical remembering’ (p. 8). (Interestingly, the suggestion that intending is in fact a form of acting provides an otherwise unlikely point of convergence with the views of Michael Thompson (Reference Thompson2008, Part II). But why think of intention as a form of memory at all – a claim which, as Wu readily acknowledges, would strike many philosophers as bizarre? The central argument for this idea is an exemplar of the book’s general methodology, mentioned above. Wu identifies in cognitive science a sustained project of investigating a system that is functionally parallel to intention. That system (or perhaps better, systems) is working memory, a ubiquitous and expansive construct thought to retain and regulate short-term memory stores required in real-time for performing various actions (Baddeley, Reference Baddeley1992). Wu’s claim here does not seem to be reductive; intention is not thought to consist in working memory. Rather, ‘working memory provides the algorithmic basis of mnemonic action in the interaction of subpersonal systems’ (p. 106). Put differently, working memory implements or realizes, at the subpersonal level, the subject-level attitude of intending.
If the functionality at issue were claimed to be exhausted by remembering – by the retaining and retrieving of action-relevant information in the service of action – the parallel between working memory and intention would be ruled out for omitting the executive or practical dimension of intending. But Wu invokes an executive function, viz. the so-called ‘central executive’ – a component or concomitant system of working memory tasked with regulating ‘attention in keeping processing on task-relevant memoranda’ (p. 100). The function of working memory in regulating specifically how the agent’s attention is allocated combines nicely with Wu’s proposal, discussed above, that all mental action (and perhaps all action as such) essentially involves the agent directing her attention (a process that Wu additionally conceptualizes, at various points here as well as in other work, as a matter of solving a ‘selection problem’ of coupling input to response (§1.3).
What emerges is a highly elegant framework that neatly slots personal level attitudes at the upper floor directly above lower-level implementing mechanisms. In fact, the picture may be altogether too neat and elegant. Consider: why should we believe that there exists such a clean divide that allows for Wu’s one-to-one mapping between the personal and subpersonal levels of explanation in the psychology of action? Why think of working memory as mirroring rather than partly involved in implementing intentions? One way to make the question a bit more concrete is to ask what grounds there are, independently of the hypothesized parallel with working memory, for thinking of intending as a form of remembering at all. Wu considers and rejects an objection to this identification, namely that memory and intention have mismatched directions of fit, pointing out that the claim is ‘contentious’ from the point of view of the science (p. 109). In a more positive vein, he notes that in deciding, i.e. forming an intention, we retain ‘decision’s content to serve future behavior… [In this way,] intention has the functional role of memory for work’ (93). But the storing of intentions to be carried out at some future time seems perfectly consistent (as Wu himself may well concede) with thinking that in order to function properly, intention relies on, rather than being constituted by, memory. As Wu notes, ‘[w]orking memory was introduced into psychology to theorize about retaining plans and intentions in controlling behavior’ (p. 94). Arguably, allegiance with the science of intention-implementation has here crowded out philosophical motivation for the thesis of intending as remembering.
That this book provokes such substantive and methodological queries as raised above (and many more besides) is resounding testament to its intellectually stimulating import. It should be read by anyone with an interest in studying mental agency in a way that remains faithful to the rich science on this topic. And, equally, it should be read by anyone who feels that they can fruitfully study mental agency while essentially ignoring the science, so that they may get a sense of what they have been missing.
