Joshua May's excellent book in moral psychology is a welcome corrective to the way in which psychological and neuroscientific findings have propagated into moral philosophy. It's not a corrective to the inclusion of the empirical sciences in moral philosophy: May is clearly happy to engage with the empirical literature. He's a philosopher who understands better than most how to read it, and how not to read it. And he is clearly skeptical about the philosophical presentations of some recent psychological literature.
The book is a largely empirical argument that, first, the empirical findings that the sentiments drive moral decision-making in fact show much less than that, when read carefully (or sometimes simply when read). Second, there is good empirical evidence that reason plays a crucial role in moral decision-making. May therefore proposes that we recognize the prominent place that reason plays in moral decision-making.
May's opponent for much of the beginning of the book is the sentimentalist, one who holds that moral judgments are the expression of non-rational motivational states: emotions, feelings, mere desires, etc. He steps through some of the findings that have been taken as evidence for the sentimentalist position – we are moved by intuitions, emotions, disgust, etc. – and shows convincingly that these findings do not show that such sentiments alone determine our moral judgments. Often they show nothing at all about moral judgments in general, and occasionally they show that some moral judgments can be somewhat amplified, though not changed, by such sentiments.
If the sentiments are not sufficient to explain moral judgments, then morality must also need reason – at least on the assumption that reason and the sentiments are the two major categories of mental phenomena that could explain moral judgments. This overview of how the empirical challenges to the primacy of reason in moral judgments have failed is excellent and enough on its own to make the book worth reading.
May does more than argue for the importance of reason by process of elimination, though. He also argues that we have independent evidence that reason plays a central role in morality. One primary argument for this relies on the importance of rationalization in our moral lives; or, to be more specific, on the importance of what May calls “ante hoc rationalization.” Post hoc rationalization recruits reasons that support what one has already done or decided. Ante hoc rationalization, by contrast, recruits reasons that support what one wants or intends to do. The fact that we engage in ante hoc rationalization demonstrates, for May, that reason is so important to us that we'll engage in reasoning before acting even when the reasoning is mere rationalization.
The distinction between ante hoc and post hoc rationalization is interesting, though it's worth considering what argumentative weight rationalization can bear. What both forms of rationalization have in common is that the rationalizer marshals reasons even though the conclusion is already established. If I'm trying to decide which job candidate to hire by looking at how well their demonstrated skills fit what the current job needs, then I'm reasoning, not rationalizing: the conclusion isn't established until I've gone through the various candidates. If I've instead already decided to hire one candidate and I'm then looking at her skills to construct a narrative of how those particular skills fit the job, then I'm rationalizing, or holding fixed the decision I intend to make and looking for reasons in support of it. If I rationalize after I've hired her, then it's post hoc; if I haven't hired her yet, then it's ante hoc rationalization.
Particular cases of rationalization often won't separate cleanly into ante hoc and post hoc. This is because we can arrive at practical conclusions in fits and starts, deciding temporarily or for the sake of argument, then revising and rethinking our decisions. Further, it will be hard to distinguish ante hoc rationalization from genuine deliberation. This is because practical deliberation is rarely settled by finding a sufficient reason on one side. If none of the qualities of the job candidates suffice to make some one candidate better than the others, I still have to hire someone. One perfectly ordinary way to decide is to imagine, in turn, hiring each candidate and then to think about how well I could defend that decision, i.e., how well I can rationalize the decision once I've made it. Good reasons and arguments can take a while to appear, and holding a conclusion fixed while looking for reasons in support of it like this might yield, over time, some strong reasons that would not have been initially obvious. But holding onto a conclusion to see what reasons can support it can be indistinguishable from rationalization. Such is the complicated nature of deliberation.
May need not disagree. My muddying the distinctions between rationalization and deliberation likely helps his general claim that reason plays a central role in morality. If I rationalize my decision to hire someone instead of simply hiring her without citing any reasons, this is evidence that I at least recognize the importance of reason in making the decision. Otherwise, I would have just hired her, reasons be damned.
Nevertheless, despite the evidence behind May's provocative claim, the role of rationalization in our decision-making still might be fairly limited. May makes a good case that we at least pay homage to reasons even when we're not reasoning. But paying homage doesn't mean that we actually recognize a role for reason. We also pay homage to reason by using the structure of reasoning to express what look like sentiments, as when we offer justifications like “I just wanted to,” “I was bored,” or “I just felt like it.” Those placeholder justifications have the form of reasoning, but would count as genuine reasoning only to someone who has an exceptionally broad idea of what reasoning is. (If the point isn't obvious in the everyday examples, consider this example of “reasoning”: “I did it because colorless green ideas sleep furiously.”) So does paying homage to reasons mean that we actually take reason to play a role, or does it just reflect a way of organizing or expressing our thoughts?
Of course, we sometimes don't even use the structure or language of reasoning: we fail to justify our own actions at all, and even fail to bring them up for evaluation.
So, the fact that we do rationalize needn't show that we take reason to play a central role in moral motivation, let alone show that reason in fact does play a role in moral motivation. What May identifies might instead be something aspirational about our motivation, something that adults work to develop and maintain. Morality doesn't require reason, that is, but perhaps we – deliberating adults, here and now, who live with others who also care about morality, etc. – do.
May does provide other evidence in support of the claim that reason plays a central role in morality. But the strength of his book is that it engages so well with the empirical literature, and it's hard to see how one could get to a stronger conclusion that reason is necessary or central to morality from the empirical evidence that reason plays some role.
The reason is that the empirical evidence underdetermines any theoretical account of moral motivation. There are many ways of dividing up the mental world, and mental states and phenomena can be distinguished or unified differently, depending on our theoretical goals. Anger, indignation, disgust, and contempt are, for example, all alike in being unpleasant. At higher levels of abstraction, those mental states relevant to moral motivation end up classified as sentiments (or desires, or passions), together with pleasure, satisfaction, pride, and ecstasy. Other states and processes, mutatis mutandis, can be finely individuated or abstractly lumped together as the work of reason.
Sometimes it's clear how to make such classifications, but not always, and definitions of such abstract terms are contestable. If, for example, desires are defined as any motivational state, then desires are analytically necessary to any motivation; if a desire is defined somewhat more narrowly as “a goal-directed, conative, motivation-encompassing state with some content, so it's a state whose function is to bring about what's desired” (p. 189, italics added), then some motivational states will not be included. For example, toddlers regularly build, destroy, rebuild, redestroy, etc. the same structure of blocks. (Adults do similar things, but in much more complicated ways.) What goal do they have? What are they trying to bring about? Even if they have no goal or nothing they are trying to bring about, it seems equally unclear that they have no desires. Similar questions could be asked for a number of more complicated adult motivational states.
Or consider reasons: is my belief that something is just or good a reason that motivates, a complicated desire, both, or neither? One's answer will reveal the person's definition of “reason” or “desire” as much as anything deeper about our moral motivational structure.
Now, May is arguing in opposition to those who think morality can be reduced entirely to sentiments. And, unless one's definition of “sentiment” is implausibly expansive, he's surely right. So, while May himself uses a fairly expansive notion of reason, he is nevertheless arguing in favor of a pretty ordinary view, namely that morality depends at least in part on reason, and against the counterintuitive view that morality is primarily or entirely motivated by our sentiments. The view he ends up defending is therefore fairly conventional.
Not only does he defend a fairly conventional view, but May even buffers the reader against any future evidence she may encounter that the sentiments move us morally. May (following Joshua May and Victor Kumar, “Moral Reasoning and Emotion,” in Aaron Zimmerman, Karen Jones, and Mark Timmons (eds.), The Routledge Handbook on Moral Epistemology (New York: Routledge, 2019), pp. 139-56) proposes a paradoxical observation: the “debunker's dilemma.” This dilemma holds that psychological (or other empirical) findings either are general enough to be relevant to many of our moral views and judgments, or they genuinely challenge (or debunk) some of those moral views and judgments, but not both. If there is evidence that, say, disgust affects our moral judgments, then it will turn out either that disgust affects many of our moral judgments, but in a way we would find unsurprising and unobjectionable; or that disgust affects our moral judgments in a disturbing but very minor way, but not both. This is a gripping conjecture, and May provides example after example in which it's true. But it's also a conjecture that cries out for a systematic proof or for a counterexample.
Maybe a proof for this would come from the fact that psychology just can't be all that surprising in the morally relevant aspects of psychology because we're too familiar with them to be systematically wrong in any way that matters.
Of course, May doesn't need any particular proof in order to maintain his observation that psychology's challenges are either serious or general, not both. And perhaps his own positive account of the role of reason in morality could even support this claim from a different direction. I've focused here only on the aspects of his account that respond to the sentimentalist, not his own positive account, so I won't assess now how well his own account could finish up this argument.
What May does need in a positive account is to move beyond what we already know, ordinarily, to be true about the relationship of reason and sentiments. The sentimentalist's claims were surprising, and May has now reassured us that, as we originally suspected, reason plays a role. And he's given us reason to hold onto that ordinary view despite future evidence against it. This is certainly progress if we thought the sentimentalist had the weight of empirical evidence on their side. Following May's lead, perhaps we can further remind ourselves why it's okay to have confidence in our ordinary views that, of the disparate things that fall under the broad umbrella of “reason,” some seem to motivate us. But perhaps May has also given us confidence to wonder, given the advances in empirical psychology, what we get out of the increasingly outdated reason/sentiments dichotomy that he has wonderfully illustrated to be at least as distracting as illuminating.
