"Internal Reasons" Philosophy and Phenomenological Research, LV, (1995) 109-131, by Michael Smith (Australian National University),
Reviewed by Lee Overton (Wake Forest University)
f, as Michael Smith holds, there is "an analytic connection between the desirability of an agent's acting in a certain way in certain circumstances and her having a desire to act in that way in those circumstances if she were fully rational," (109) and if, as Smith holds, talk of desirability can be replaced by talk of having a reason, then there is a certain "internalism requirement" on reasons.
Two construals of this requirement emerge from the literature, both characterizing the information sought from one's fully rational self. One holds that you seek practical advice for someone in your current psychological state. The other holds that you seek to know, by example, what your better self (who does not have your current psychological state) would do in your situation, for you intend to copy him. These are, respectively, the "advice" and "example" construals of the internalism requirement, and there is a straighforward argument for why the advice construal has to be right. My better self is calm and cool, even in situations that are tense for me. What he would do, I may be incapable of doing, given how uptight I am. So it would be foolish to use him as an example. Better to ask him what I should do, given my hang-ups. Better, that is, to ask him what he would do if he were me, rather than what I would do if I were him.
Smith's primary concern in this paper is to show that it is only if we think of reasons as being non-relative and think of reasons on the 'advice' model, "that we can properly account for the normative significance of reason claims". (125) So if it can now be shown that the "non-relativity" of reasons is presupposed by our ordinary concept of a reason, then we are well on our way to vindicating the view that there is a fundamental irrationality in one's believing one ought to do thus and such while failing to desire accordingly. So, taking on Bernard Williams, with whom he shares a fondness for reasons internalism, Smith rejects Williams' view that what is a reason for one person could fail to be a reason for another. Although both agree that our ordinary concept of a reason is to be cashed out in terms of where good reasoning would take you, Smith thinks in addition that ordinary thought presupposes that practical reason would bring us all, ultimately, to the same place, despite our having different motivational sets at the start.
It is a commonplace that better imaginative acquaintance can result in a change in desire. But this is nothing, argues Smith, compared to the changes in desire that may occur through a Rawlsian process aimed at "a more coherent and unified desiderative profile and evaluative outlook". (114) He writes: "We can ask ourselves whether we wouldn't get a more systematically justifiable set of desires by adding to this whole host of specific and general desires another general desire, or a more general desire still, a desire that, in turn, justifies the more specific desires that we have. And an answer might be that we would. If the new set of desires--the set we imagine ourselves having if we add a more general desire to the more specific desires we in fact have--exhibits more in the way of coherence and unity, then we may properly think that the new imaginary set of desires is rationally preferable to the old." (114-115)
Reason might therefore lead us to adopt beliefs about desires that we would have but don't yet have. And if we see that in a more coherent and unified state, we would have such and such desire, then by God we ought to have those desires--for we would be in a more rational state if we did. A rational person's desires would change at this point. "As this procedure of systematic justification continues we can therefore well imagine wholesale shifts in our desiderative profile. Systematic reasoning creates new underived desires and destroys old. Since each such change seems rationally required, the new desiderative profile will seem not just different from the old, but better; more rational." (116)
'State x is a more unified state than state y' does not imply that from y a move to x is rationally required. There are surely many states that will serve as well as x. But, argues Smith, it is "part of our task, in trying to come up with a systematically justifiable set of desires, to come up with the same set of desires as our fellow rational creatures would come up with if they set themselves the same task." (118) Now, perhaps Smith thinks that it is only in the context of a community each member of which trusts the others' judgments to be, for the most part, as reliably expressive of his values as are his own judgments, that there can be such a thing as rational desire-revision. Thus: if something is a reason, it follows that it is our reason.
At best, this undermines the possibility of individual relative reasons. Group relative reasons seem viable still. But even this is conceding to Smith too much. For his argument is better seen as against individual relative moral reasons or values than against individual relative reasons. And in that case, the initial substitution of 'has a reason to act' for 'is desirable to act' comes into question. Why can't I have reasons that are relative to me? There is no obvious reason why the process of desire revision could not leave some of my arbitrary desires untouched. Why should I care about unifying and systematically justifying all of my desires? Surely I don't aspire to be just like everybody else as regards my desire-identity!
And, finally, there is another straightforward worry: even despite failure to show the non-relativity of reasons, part of what is unnerving about Smith's argument is the unabashed faith in conceptual analysis that leads him to say, in parting, "it of course remains an open possibility that there are no internal reasons--and hence there are no reasons for action at all." (130) Surely what Smith calls an "apriori presupposition" is better understood as a starting assumption of persons considering, together, whether there is a reason to act--an assumption that may be revised in the face of conflicting evidence.