It is tempting to appeal to the concept of sensitivity or responsiveness to reasons to account for the difference between some scenarios where agents act freely and some where they don’t. Some actual-sequence theorists have offered views of freedom and responsibility that draw on this important concept (Fischer and Ravizza, McKenna). According to reasons-sensitivity views (henceforth RS views), the difference between addicts and non-addicts, for example, is that addicts are not sensitive to reasons when they take the relevant drug, but non-addicts are. Roughly, the thought is that there is an appropriately wide range of reasons to refrain from taking the drug that the non-addict is sensitive to, and the addict is not. These reasons are, in fact, absent but are such that, were they to be present, the non-addict would (in normal circumstances) respond to them. For example, if it had been apparent to the non-addict that a good friend needed his help at the time, he would have helped his friend instead of taking the drug, but the addict would not have. According to an RS view, this (at least partly) explains why the non-addict is free and the addict is not.
Many important questions arise for RS views given their appeal to the concept of ranges of reasons that are “appropriately wide.” As it turns out, it is quite tricky to give an adequate account of when a range of reasons is appropriately wide. Even addicts may be sensitive to some reasons to refrain from taking the drug (for example, an addict could be disposed to refrain from taking the drug if he had reason to believe that the drug would kill him; still, we wouldn’t think that this is enough for him to act freely when he takes the drug). Also, what seems to make the relevant range of reasons “appropriately wide” is, arguably, not just the number of reasons that agents are sensitive to, but also the type of reasons involved, and the fact that reasons can come in patterns that make some sort of overall rational sense. (Fischer and Ravizza, and also McKenna, discuss these issues and develop RS views that accommodate these important considerations.)
In this post I want to discuss a different, and I think underappreciated, issue that arises for RS views concerning the relevant ranges of reasons.
Let’s start by comparing two acts:
A1: Ingesting the contents of a can of poison
A2: Drinking a glass of water
RS views would say that I only do these acts freely if I am suitably sensitive to reasons when I perform them, in other words, if I am sensitive to the relevant range of reasons to refrain from performing those acts. Now, according to some natural way of identifying and counting reasons, there are more reasons one can think of to refrain from doing A1 than there are to refrain from doing A2. So it seems that an RS view shouldn’t require that the relevant ranges of reasons be equally wide in both cases. Instead, it seems that it should require a wider sensitivity range in the case of A1 than in the case of A2.
In other words, presumably, what determines whether the relevant range of reasons is appropriately wide is not just the absolute number of reasons the agent is sensitive to (and the type of reasons, and whether the reasons fit together in the right kind of way) but, also, the ratio or proportion of those reasons to the total number of reasons that there are or could be, given the kind of act that we’re dealing with. Here’s one possible way of formulating this claim:
Flexibility: One of the things that determine whether the relevant range of reasons is appropriately wide is the particular type of act in question.
The “flexibility” arises from the fact that different types of acts will call for different required widths for the relevant ranges of reasons, so there won’t be a single required range (not even a vaguely specified one) for all kinds of acts.
I suspect that recognizing Flexibility would help RS views avoid some potential objections concerning the relevant ranges of reasons. For, once Flexibility is recognized, there is no longer the pressure to identify a single sensitivity range (however vague) that would plausibly work for acts of all types (a task that seems daunting). This is an important advantage that Flexibility provides. In addition, I believe that Flexibility results in another significant advantage for RS views.
Wolf, and more recently Nelkin, have suggested that the freedom required for responsibility consists in being able to do the right thing for the right reasons. This results in an asymmetrical view of freedom, one according to which blameworthiness requires the ability to do otherwise but praiseworthiness doesn’t. (This is because one can do the right thing for the right reasons even if it was the only thing one could have done; in contrast, if blameworthiness for an act requires being able to do the right thing, for the right reasons, then being blameworthy requires being able to do otherwise.) In this sense, blameworthiness requires more than praiseworthiness, and this is a result of the types of acts and the types of reasons involved in each case.
As an actual-sequence theorist, I don’t believe that blameworthiness requires the ability to do otherwise. Still, there is something about the Wolf-Nelkin view, and, in particular, about the claim that there is some asymmetry between blameworthiness and praiseworthiness that is grounded in the types of acts and reasons involved in each case, that rings true to me. And I think Flexibility can capture this.
Imagine that there is an act so good and so obviously rational that there could apparently be no reasons (moral or non-moral in nature) to refrain from performing it. (The assumption that there is, in fact, such an act is not essential for the asymmetry to arise, but it helps illustrate how the asymmetry arises, by means of illustrating how it would arise in the limiting case.) Consider, for example, the act of pressing a button to prevent the world’s imminent destruction, or to prevent the horrible torture of your infant child, or something along these lines. Assume there is no possible reason to refrain from performing that act. In that case the reasons-sensitivity requirement would be met trivially: if there are no possible reasons to refrain from performing that act, then the relevant range of reasons to refrain that one must be sensitive to in order to be appropriately sensitive to reasons is zero. Perhaps in those cases being “sensitive to reasons” is just a matter of acting on the basis of the relevant reasons to act in that way, and reasons to refrain don’t play a role.
More generally, other things being equal (in particular, assuming the range of non-moral reasons is the same), there are more reasons to refrain from performing a morally bad act than there are to refrain from performing a morally good act. If so, Flexibility yields an asymmetry between blameworthiness and praiseworthiness: it entails that, other things being equal, the range of reasons that one needs to be sensitive to in order to be blameworthy is larger than the range of reasons that one needs to be sensitive to in order to be praiseworthy. In this sense, one needs to be sensitive to more reasons to be blameworthy than to be praiseworthy, and thus it’s (thankfully!) “easier” to be praiseworthy than it is to be blameworthy.
Note that this is not the Wolf-Nelkin asymmetry. But the claim is that it captures what “rings true” about their asymmetry, from the perspective of an actual-sequence view of freedom. We can make sense of the existence of an asymmetry between blameworthiness and praiseworthiness without committing ourselves to the ability-to-do-otherwise requirement (for neither praiseworthiness nor blameworthiness).
How does all this sound?