In his 2009 book Willing, Wanting, Waiting, he argues that choices are the way we form intentions and they have three features

(i) choice is an action;

(ii) choice is not determined by one's prior beliefs and desires;

(iii) once the question of what to do has arisen, choice is typically both necessary and sufficient for moving to action.

It is argued that choice is needed because of agents' inabilities to arrive at judgements about what is best. Nevertheless, choice differs from random picking: in choosing, agents frequently (though not always) deploy abilities that enable them to make good choices. In such cases, judgements about what is best will frequently follow the choice.

If a necessary condition on one's free will is that one be able to maintain one's resolutions, then awareness of the effort expended can be seen as a further source of the experience of freedom. This interpretation explains, and is in turn supported by, some studies showing that belief in determinism tends to undermine moral motivation. Determinism can be easily (but wrongly) interpreted as showing that effort will be futile.

Some recent studies have suggested that belief in determinism tends to undermine moral motivation: subjects who are given determinist texts to read become more likely to cheat or to go in for vindictive behaviour. One possible explanation is that people are natural incompatibilists, so that convincing them of determinism undermines their belief that they are morally responsible.

I suggest a different explanation, and in doing so try to shed some light on the phenomenology of free will. I contend that one aspect of the phenomenology is our impression that maintaining a resolution requires effort — an impression well supported by a range of psychological data. Determinism can easily be interpreted as showing that such effort will be futile: in effect determinism is conflated with fatalism, in a way that is reminiscent of the Lazy argument used against the Stoics. If this interpretation is right, it explains how belief in determinism undermines moral motivation without needing to attribute sophisticated incompatibilist beliefs to subjects; it works by undermining subjects’ self-efficacy. It also provides indirect support for the contention that the awareness of the exertion of effort is one of the sources of the phenomenology of free will.

I suggest that choice is an important factor in our experience of free will. When we focus on an action, we inevitably raise the question of what to do: of whether to perform it, or some other action. That is what happens when we focus on our experience of free will with respect to some action. So whilst it is surely wrong to think that only chosen actions are free, it is understandable that choice will loom large in any discussion of free will Equally understandable is how the knowledge that our experience gives us could be mistaken for knowledge of a grander metaphysical claim. Our experience tells us that our choice is not determined by our beliefs and desires, or by any other psychological states — intentions, emotions etc. — to which we have access. Those could be the same, and yet we could choose differently. From there it is easy to move to the thought that we could be just the same in our entirety, and yet we could choose differently: that the world is indeterministic. That I think is one of the pressures towards libertarianism. It is not the only one: others, needing different responses, come from considerations of moral responsibility. But it is, I think, the most immediate. I hope I have gone a fair way to blocking it.