As we discussed in “New Compatibilism” among the main attractions of new compatibilism are its foundation in the seemingly obvious fact that we act for reasons, as is evidenced by the efficacy of our daily practice of giving and asking for reasons, and the idea that our daily practices of responsibility (including the assumption that at least some people sometimes do things for which they are not responsible and our habit to hold people by default responsible for what they do) can be legitimized on the basis of this fact, without further appeal to obscure notions such as the ability to do otherwise. In this section we will argue that the BCN-findings discussed in “BCN Findings” cast doubts on this latter claim.
The BCN-evidence indicates that many actions for which the actor can give reasons are automatic responses to external stimuli, many of which are not recognized by the actor. These findings have led many scientists to the conclusion that, normally, people do not act for reasons. We do not think that this conclusion follows. Saying that a certain act was performed for a particular reason does not necessarily mean that that act resulted in one way or another from a conscious process of reasoning. There might be ways to accommodate the BCN-finding that normally people act automatically while maintaining that people normally act for reasons.
Yet, we do think that these findings spell trouble for the new compatibilist, for, if these findings are true, new compatibilists must find a way to view automatic actions as actions for a reason if they are to avoid the conclusion that acting for reasons is exceptional. Because new compatibilists are also committed to the thesis that the ability to act for reasons distinguishes actions for which the actor is responsible from action for which she is not, it follows that new compatibilists must come up with an account of what distinguishes automatic actions for a reason from automatic actions that were not for a reason. It is far from obvious that such a distinction can be made without introducing obscurities such as the ability to do otherwise.
This challenge is aggravated by the BCN-thesis that giving reasons is more a matter of interpretation than of recollection. As we explained in “New Compatibilism” the project of new compatibilism is to legitimate our common practices of responsibility independent of the issue whether someone ever could have acted otherwise. In practice, the issue of personal responsibility typically arises when someone behaves immorally without an adequate excuse or exemption. In such cases of wrongdoing we assume that there is a distinction between cases in which wrongdoers are responsible for doing wrong, and those in which they are not. We shall call the first type of cases ‘blameful wrongdoing’ and the latter ‘blameless wrongdoing’.
According to new compatibilism, the distinction between blameful and blameless wrongdoing in our daily practice is made and should be made based on the ability to act for reasons. However, someone who acts immorally without an adequate excuse or exemption in fact ignores the reasons or moral justifications for not acting in such a way. Clearly, someone who steals without an adequate excuse fails to respond to the reasons not to steal. That is, cases of wrongdoing by their very nature disclose a failure to respond adequately to reasons (see , Chapter 3). This means that the distinction between blameful and blameless wrongdoing cannot be viewed as a distinction between wrongdoings in which one did respond to reasons and wrongdoings in which one did not; it must be seen as a distinction between being able to respond to reasons and not being able to do so. It follows that the new compatibilist is committed to the view that in cases of blameful wrongdoing a person did not respond to the reasons there were, although she could have done so. This sounds astonishing like saying that she would have been able to do otherwise!
At this point, new compatibilists might want to point out that our every day practices provide evidence for the view that in some cases of wrongdoing wrong doers are capable of responding to reasons although in fact they did not do so. The kleptomaniac is deemed unaccountable because she communicates that she feels awful about her behavior and freely commits herself to treatment (showing she is not able to respond to reasons). The thieving student is deemed accountable either because her remorse shows her to be aware of having made a wrong choice (showing that she is able to respond to reasons); or, alternatively, because her anger at us, shows her unwillingness to accept our norms and values, which we firmly believe to be valid (showing that she responds to reasons, albeit to reasons we do not accept). So although we have a problem to explain how it is possible that people are sometimes able to respond to reasons to which they did not respond, we might be sure that this sometimes is the case.
However, if the BCN-sciences are right that reasons are inferred after the fact, it is far from clear that this practice provides the evidence needed by the new compatibilist. For the BCN-view suggests that differences between the self-reports of the thieving student and the kleptomaniac might result only from differences in their ability or willingness to account for their failure to respond to the reasons there are, rather than from differences in their capability to act for reasons.
The problem is not that people can make mistakes in their self-reports or lie about their motivations. The problem is that if giving reasons is a matter of post hoc interpretation, the alleged fact that we can make a distinction between blameful and blameless wrongdoers on the basis of the reasons they provide, does not provide evidence for the view that there ever are wrongdoers who are able to respond to reasons although they do not do so. It might only indicate that some people are better than others in rationalizing their failure to respond to the reasons there are. The thieving student might be just as incapable to act for reasons as the kleptomaniac although she thinks otherwise (and due to her rationalizing talents is able to convince others of it too).
Let us stress that we do not think that the BCN-sciences have shown that it is impossible for someone not to respond to reasons while being able to do so, or that the BCN-sciences show that our everyday ways of judging whether some wrongdoer can be held responsible for what she did are illegitimate. Our point is rather that in the light of the BCN-sciences it is unclear what it would mean to fail to respond to reasons while being capable of doing so and hence that it is unclear how on the new compatibilist view someone could ever be responsible for doing wrong. The new compatibilist needs an account of what it means to be able to respond to reasons while in fact failing to do so, an argument based on this account that shows that it is a real possibility that someone who did not respond to reasons was nevertheless able to respond to those reasons, and an argument that this account legitimates our common practices of deciding about the blamefulness of wrongdoings.
Fischer & Ravizza ( Chapter 3] provide an account that seems tailor made to solve this problem. Put very simply, the idea is that a person who in a certain case did not respond to certain reasons is nevertheless able to respond to those reasons if there is a scenario available in which she would have responded differently.Footnote 13 Suppose someone stole a wallet. If she would not have stolen it when a police officer was standing next to her, this would show that she was nevertheless able to respond to the reasons not to steal. As Fischer and Ravizza put it: “we believe that, if an agent’s mechanism reacts to some incentive ... this shows that the mechanism can react to any incentive” [9: 73].
At first sight this is an interesting and appealing idea. However, the argument is flawed. It does not help to distinguish between the thieving student and the kleptomaniac. For, a kleptomaniac, too, would be able not to steal under conditions such as having a police officer stand beside her. However, most of us do not need a police officer to prevent us from stealing. Therefore, the very fact that someone would steal means that she does not respond to the reasons to which most of us would respond i.e. internal reasons. Why would the fact that a person reacts to other reasons (the presence of a police officer) indicate she was nevertheless able to react to the reasons to which she did not respond? Fischer and Ravizza [9: 74] suggest that this inference is justified to the extent that the decision mechanism is the same in both situations. Yet, this response only displaces the problem, for it gives rise to the question of how to decide whether the same mechanism is in play. How can we be sure that this is the case and on what basis do we infer it?
The assumption that the ability to react to one reason provides evidence that one is able to react to all the reasons one recognizes is seriously challenged by situationist psychologists who have investigated the influence of features that we do not notice or believe to be relevant in explaining behavior [e.g. 33]. The general view of this paradigm is that our behaviors and actions are better explained in terms of the particularities of the situation that often go unnoticed or are deemed irrelevant than in terms of internal character traits, virtues, and morality. If it really is the case that the particularities of the situation strongly influence our ability to respond to reasons, our ability to respond to reasons in situations in which we respond to reasons, does not indicate that we were able to respond to reasons in situations in which we do not do so. So in the light of the BCN-sciences it is not clear how, on a new compatibilist account, it would ever be possible to do something wrong while being responsible for it.
The new compatibilist is not only committed to explain how it is ever possible that a wrongdoer was able to respond to reasons, although in fact she did not do so. The new compatibilist should also explain why in cases of wrongdoing it is reasonable to assume by default that the wrongdoer was able to respond to reasons. After all, in our daily practices we hold wrongdoers by default responsible for what they did. However, if situationist psychologists are right about the influence of the situation on what we do and decide, this practice must be reconsidered. If the particularities of the situation rather than differences in agential decision procedures determine whether or not someone acts morally it seems not reasonable to hold the agent responsible for acting wrong. As Doris observes, the challenge of situationism is that the sensible default attitude would be: “[...] a general agnosticism about responsibility attribution, since we could never confidently rule out the presence of competence defeaters” [33: 138].
This is a problem for new compatibilism, which argues that the assumption that we are reasons-responsive beings is metaphysically modest. Therefore, new compatibilism needs our reasons-responsiveness to be an unproblematic, easy to observe aspect of our everyday lives.Footnote 14 When the reasons we provide for acting immorally can be taken at face value, we can infer that at least sometimes some of us act immorally and are to blame for it e.g. the thieving student, but not the kleptomaniac. The BCN-sciences challenge this assumption.
So, the BCN-sciences pose at least three problems to the new compatibilist. First, the new compatibilist should find a way to accommodate the insight that most of our everyday life is determined by automatic processes triggered by external cues without introducing obscurities. Second, in the light of the BCN-thesis that giving reasons is a matter of post hoc interpretation, the new compatibilist cannot point to our every day practice to determine whether someone was responsible for doing wrong on the basis of the reasons they give to rebut the challenge that the new compatibilists’ idea that some wrongdoers are able to respond to reasons to which they do not respond is as obscure as the libertarians’ idea that those wrongdoers were able to do otherwise. Third, the new compatibilist should answer the challenge that everyone’s ability to act for reasons is heavily compromised by the influence of the situation.
Before concluding, let us emphasize that in our view these challenges are problems that should be addressed rather than findings that undermine new compatibilism. We ourselves are not convinced of the new compatibilist view in all its aspects. However, we do believe that it correctly points to the importance of our abilities to respond to reasons as a key to understanding our moral practices. For that reason, we believe that our understanding of free will and responsibility will be improved if we take BCN-findings concerning these abilities into account. We suspect that, in doing so, some version of incompatibilist free will will reappear, but this does not alter our view that our common point of focus should be the efficacy of reasons, moral deliberation, and our general reasons-responsiveness.