Reasons and causes both function as constituents of explanations, but of different sorts of explanations. Let’s start with the idea that explanations are answers to why-questions (seems reasonable enough that we can just assume that it’s correct for now). If that’s the case, what sorts of why-questions are answered with reasons? An obvious starting point seems to be a question of the form, “Why did you do X?” In this case, a reason is given as a first personal report of some sort of mental state or phenomenon led to the performance of X. Davidson cashes this out as what he calls a primary reason: the pair constituted by a pro-attitude, e.g., a desire for Y, and a belief, e.g., that doing X is a necessary step in achieving Y. In nearly all actual cases, of course, one would only respond with a partial explanation stating either the desire or the belief while believing the other to be assumed or understood. Let’s call this sort of explanation a rationalization (or, maybe, a justification).
The goal of a rationalization is to make sense of one’s action. It places the action in the space of reasons and offers a justification for acting in the way that one has. One asks, “Why did you do X?” when one does not understand or cannot make sense of your reasons for acting, i.e., when one cannot place herself in your shoes and discover why you did what you did. A rationalization, then, makes explicit your reasoning. Or, at least, it tries to…often times, as Davidson points out, we don’t act for the reason we think or claim to act for. In making explicit your reasons, the motivations and beliefs that led you to act in a certain way, i.e., that you took to justify your action, are now open to examination by the community and can be called into question, i.e., you can be forced to defend them.
Causes, on the other hand, function as constituents not of rationalizations but of, you guessed it, causal explanations (perhaps we could say scientific explanations, but then we would have to construe science very broadly, or perhaps we could say physical explanations, but then we would beg the question with regards to limiting the scope of causal explanation to the physical, i.e., non-rational, realm). We’ve spent a long time cashing this out in class, so I won’t get into details here, but causes seem be parts of answers to questions like, “Why did Y happen?” or “Why did X do Y?” They also seem to be open to counterfactual questions of the form, “If Z hadn’t happened, would X have still Y’d?” Causal explanations seems to fulfill a need for understanding of interactions between objects (or absences of interactions, in some cases).
Now, if you buy any of this, it seems that rationalizations and causal explanations fulfill our desires for different kinds of understanding, i.e., they answer different varieties of why-questions that are asked with different goals in mind. A complication arises when one asks a why-question of the form, “Why did person-X do Y?” This strikes us as an ambiguous question because it could either 1) be asking a third party to offer a rationalization of person-X’s action or 2) be requesting a causal explanation of person-X’s action. One form of causal explanation that can be offered might be neurological. One could point to a stimulus that caused a certain set of neurons to fire, that caused other neurons to fire…until finally the neurons fired that caused the action. Thus, the answer might be something like, “Person-X did Y because neurons A, B and C fired in sequence.”
Take notice in the last sentence of the “because” locution, and then consider an answer to the same question in the form of a rationalization. One might answer, “Person-X did Y because she believed doing Y would bring about Z, and Z would give her pleasure.” This rationalizes person-X’s Y-ing, it makes it reasonable by making explicit one set of possible reasons for her acting and opening them to critique. However, the ambiguity of the question and structural similarity of the answers (the mirroring use of “because”) could quite easily lead one to think that the rationalization offered was meant to be a causal explanation of person-X’s action, i.e., something of the form, “Person-X was caused to do Y by her mental state (which is cashed out in terms of primary reasons).” Indeed, Davidson thinks this is the only way to make sense of this use of because. We must, he argues, treat reasons as causes lest human actions be uncaused. Davidson argues that the having of a primary reason is being in a certain mental state. This state can be described variously in terms of rationalization (the having of reasons) or in terms of neural activity. Thus, the stating of reasons is simply a redescription of the neural event that is the cause of the action, though, he admits, it’s certainly far from clear that the connection between neural events and reasons is a knowable and he’s not sure how to cash it out.
Davidson’s story, I think, complicates matters far too much. We need only recognize that human actions are caused and that the causal story to be told is one about stimuli, neuron firings and actions, in order to satisfy ourselves that human action is not uncaused. Reasons need not come into this causal explanation at all. However, reasons do play an important role in our attempts to justify our actions, and the explanations offered via the giving of reasons are the response to a very different sort of why-question than the explanations offered via the pointing out of causes.
One final point might be helpful in thinking this through. Consider the case of a zombie who acts just like a human being but who lacks consciousness. We could quite reasonably give a causal explanation of the zombie’s actions, but it seems nonsensical to ask for the zombie’s reasons. In the case of the zombie, unlike that of the full human, one could not re-describe the neural events as the having of reasons. I think this could be cashed out to show that Davidson’s thought that reasons are causes in some way misses the mark.
