Despite their differences, the epistemic and the pragmatic creation myths rest on the common assumption that intentions have a teleofunction, some ultimate purpose they are designed to serve. Velleman thinks it is more plausible that their existence is an accident, that is to say, that they are the byproduct of some more general endowments of human nature. In other words, Velleman is tempted to think of the human will as, in Gould & Lewontin’s phrase (1979), a spandrel, a feature formed not by design but as an accidental byproduct of some other designed feature or features. This leads him to be skeptical about both teleological myths. In telling his own creation myth, Velleman pursues two aims. His first aim is to show that the assumption behind the two teleological myths can be dispensed with. His second aim is to show that the accident that led to the emergence of the human will more closely approximates the epistemic than the pragmatic creation myth.
Velleman’s own account of intentions characterizes them as an agent’s commitment to the truth of some act-description of his or her forthcoming behavior that reliably causes this act-description to come true. He argues that this account of intentions “posits nothing more than the predictable consequences of two motivational states whose utility in the design of a creature is far more general than that of the human will” (Velleman 2007, p. 211). In other words, the human will is a spandrel, a feature arising from the accidental confluence of two designed features. What are these two features? The first, according to Velleman, is curiosity, defined as the creature’s drive to understand what goes on in its environment. The second is self-awareness, through which the creature realizes that it is part of its environment and that its own behavior is part of what goes on in this environment. Self-awareness thus allows a creature to acquire an objective conception of itself. A creature that is both curious and self-aware will in turn be driven to understand its own behavior, that is, to understand “how the egocentrically conceived world of doing things is connected to the objectively conceived world of things understood” (Velleman 2007, p. 211). In understanding this, it will have acquired the capacity for intentions.
We can now see why Velleman thinks his own creation myth has more affinities with the epistemic than with the pragmatic creation myth. Curiosity is an epistemic drive and self-awareness is an epistemic capacity. As their byproduct, the capacity for intentions inherits this essential epistemic dimension. We can also understand why he means his own myth as an antidote to the methodological assumption inherent to the idea that intentions serve a specific teleofunction. Curiosity and self-awareness are, Velleman claims, designed for far more general purposes than that of the human will.
I think, however, that this is also where the creation myth told by Velleman reaches its limits. Important questions are left unanswered: What are these more general purposes served by curiosity and self-awareness? What good is curiosity? What good is self-awareness? Unless he is willing to consider the will as a spandrel of spandrels, Velleman owes us answers to these questions. From an evolutionary point of view, it is unclear what benefits knowledge of their environment and knowledge of themselves could confer on creatures endowed with curiosity and self-awareness unless this knowledge found some behavioral expression. It isn't too difficult to see how a better understanding of their environment can promote more effective behavior, enhance the satisfaction of desires and needs, and ultimately have a differential impact on reproductive success in creatures endowed with curiosity. One should note, however, that pushing Velleman’s story one step further in his direction has the effect of undermining his claim that his own myth has strong affinities with the epistemic creation myth for it suggests that the epistemic function of curiosity is ancillary to its pragmatic function, rather than the reverse.
It is less obvious how we should answer the question what good is self-awareness, what purposes it is designed for. My aim in the next two sections will be to remove two obstacles that prevent us from looking in the right direction for an answer to this question. The first obstacle lies in the fact that philosophers have tended to neglect an important pragmatic function of intentions. Thus, Velleman notes, rightly in my view, that Bratman’s account of the pragmatic functions of intentions leaves many present-directed intentions without a purpose. However, rather than looking for some further pragmatic purpose intentions may serve, beyond scheduling deliberation and enhancing action coordination over time, Velleman turns his attention to epistemic functions. I will argue in section 4 that they both neglect a further important pragmatic function of intentions, namely their role in the online monitoring and control of action. The second obstacle lies in the fact that one central feature that makes us human, our deep sociality, is either ignored or at best a peripheral concern in philosophical accounts of intentions. Of course, I am not denying the obvious: many philosophers, and Bratman prominently among them, have explored joint agency and collective intentionality. Typically, however, their focus has been on whether or not joint agency should be seen as continuous with individual agency and thus on whether or not the conceptual framework developed to account for individual intentions could be fruitfully extended to shared intentions.[1] Rarely if ever, however, do they consider the possibility that shared intentions may shed light on some of the features and functions of individual intentions. In section 5, I will argue that the control and monitoring function of intentions plays a crucial role in contexts of joint action. I will further argue that this function might indeed be the primary function of intentions and might have become established because of the role it serves in solving the coordination problems that arise in joint action and because of the benefit thus conferred on creatures capable of solving these coordination problems.