Do all our reasons depend on our desires?
Let’s imagine I borrow money from my neighbour, and I genuinely believe that it is right to repay borrowed money (hereafter, to G). I have, however, utterly no desire to do so. Do I have a reason to G? I shall set out how Reasons Internalism (hereafter RI) requires the presence of a desire for there to be a reason to G and how this is not required by Reasons Externalism (RE). I shall explore possible avenues for RE and conclude that it seems possible that desires are not required for all reasons.
Terminology
First, let’s get clear on some terminology. By ‘reasons’ we are interested in normative / justifying reasons, not merely explanatory / motivating reasons. The latter explain why an agent did (or was motivated to do) an action, whereas the former outline considerations in favour of the agent doing the act. These can come apart: if I dance every morning in the mistaken belief that it will make my hair curl, there is an explanatory reason but not a normative reason for my dancing.
Next, by ‘desires’, I mean any of a range of possible subjective states, such as desire, aspirations, intentions, etc… When I talk of ‘motivation’, this does not require an agent to be fully moved to action but to have at least some motivation to do so.
Finally, following Williams, I will for simplicity talk about internal and external reasons, although the original paper was about reasons statements.
Reasons Internalism
Following Williams, RI holds that having a reason to G necessitates having, or being able to acquire, a corresponding motivation. In turn, the Humean Theory of Motivation holds that motivation requires both a belief and a desire. Thus it follows that to have a reason to G I must hold both a suitable belief and also a desire that is appropriately related to G. Williams refines this to claim I must either have an element (such as a desire) in my subjective motivation set (S) that would be served by my G’ing, or that such a relationship or motivation can be established through rational deliberation from an existing element in S. Williams focuses on the latter and thus we will take RE to be the denial that a reason requires such motivation as can be established through rational deliberation from an existing element of S; in other words, it holds that a desire is not necessary for a reason.
Examples and the challenge for RI
Some examples will make this clearer. Williams’s own example is of someone who requires medication but who has no current desire to preserve his health and, further, is such that no motivation could be established despite any rational deliberation from an existing element in his S. According to RI, he has no reason to take the medicine. This view would be rejected by most REs, who would claim that health is sufficiently important to be protected no matter the inner state of the agent. Back to our original example, since I have no desire to G I have, according to RI, no reason to G. Again, many REs would disagree, and this leads to a compelling criticism of RI: it seems problematic that moral reasons, such as to keep a promise, are contingent on the subjective desires of the agent. It feels intuitive that at least some reasons should exist independently from the agent.
The internalist rejection of external reasons
An internalist of Williams’s school would deny the possibility of such external reasons, for the following reasons. In some cases, an agent who is not already motivated comes later to do the thing in question. Clearly, then, he must have acquired a new belief about the external reasons statement that he previously rejected. The claim is that since there was no original motivation, the agent had nothing to deliberate from to acquire the new motivation and belief, and given the Humean claim that reason alone cannot motivate, this possibility is rejected. Thus, RE is unable to explain the motivation of the agent, which RI insists is a requirement of reasons. Yet, as we have seen, REs are keen to preserve the possibility of external reasons that are not based on desires. How, therefore, might they respond?
Externalist response – substantive deliberation
One possibility would be to accept that rational deliberation is required for the having of a reason, but to deny that this need be procedural deliberation from an existing motivation. It could be noted that the procedural deliberation requirement takes no account of what the existing motivation actually is, it just requires that one exists which, it might be argued, is an unsteady footing for a reason, particularly moral ones An alternative might be to instead require some kind of substantive deliberation: the agent must care about certain substantive things and be able rationalise from that to the motivation (Parfit 1997). My concern with this, however, is that given that Williams remains vague about the contents of a motivation set, more work needs to be done to see whether substantive concerns can truly be differentiated from the contents of S.
Externalist response - conversion
Another possible response for the Externalist is to reject rational deliberation as a requirement for a new motivation. A possibility here might be a conversion of some kind (McDowell 1995), perhaps similar to a religious conversion or falling in love and suddenly acquiring a totally new perspective on the world without having deliberated. In our case, we might perhaps imagine that I awake from a dream in which I suddenly ‘saw the importance of paying my debts, and find I am now motivated to do so.
An Internalist might object to this on the basis that it does not meet the explanatory role of reasons that we have seen is a requirement. However, we have already noted that Williams does not examine the nature of any existing element in the agent’s S before holding it as a candidate to be used in rational deliberation. If conversion is a possibility, it seems feasible that any existing element in the agent’s S might itself have come from a previous conversion, and this would therefore be problematic for the RI account as well.
Externalist response - dual reasons
A further avenue the RE might explore is the possibility of having both internal and external reasons concurrently. As Parfit tells us, having an external reason to do G need not mean I have no internal reason to G. Say I have no desire at all to repay the money I borrowed: any moral reason for me to do so is external. However, I do have a desire to retain the friendship with my neighbour and thus have an internal reason to repay the money. This feels a plausible suggestion, not least because it seems to match how we feel in situations where we feel pulled in two directions, thinking that when a situation is looked at from one angle we have no desire or motivation to act but from another we do. Indeed, I feel this might better capture William’s Owen’s change of heart about the military better than his acquiring a new belief: an internal reason (of not upsetting family?) won out over his lack of military desire.
Conclusion
I both find dual reasons and conversion to be promising avenues for the existence of external reasons, and thus I tentatively conclude that not all reasons need depend on desires.
