- Chapter 1: Many Sorts of Naturalism
- Chapter 2: Organic Naturalism
- Chapter 3: Practical Primates
- Chapter 4: What We Are
- Chapter 5: Practical Reasoning
- Chapter 6: Natural Reasoning
- Chapter 7: Conclusion
Chapter 5: Practical Reasoning
There could be no reasons unless a rational animal has a general conception of its own good, and thus a general sense of how to live. (Jennifer Frey, The Will to Do Good, 79.)
“How should one live?” This question is central to neo-Aristotelian writers such as Bernard Williams, Philippa Foot, Alasdair MacIntyre, John McDowell, and others. The question is so important, I think, for at least four reasons. First, the question implies that the questioner is aware of a dichotomy or distinction between the way one is in fact living and the way one might live. As a matter of fact, every capable adult is already living in a particular way. I take it for granted that most people learn to live in a particular way from their culture and family of origin, while also trying to satisfy more or less their own idiosyncratic preferences. But a normal part of human life is pausing to reflect on one’s own motives, methods, means, and ends. A crisis can trigger such reflection: what is wrong with my way of life, my values, and/or my choices? And exposure to other people – be they friends, fictional characters, or historical figures – who seem extraordinarily happy can trigger such reflection: what do they know that I do not? What are they doing that I am not?
Secondly, the “how should one live?” question assumes that there are good human lives and bad human lives. I hope that it is uncontroversial to point out that some of the members of our race are fools. (I leave it to the reader to supply illustrations.) If there are ways one definitely should not live, then there is at least a way or set of ways one should live. Even if it is difficult to answer the question of how one should live, we should not be fully skeptical that there is an answer (or a set of answers).
Thirdly, the question implies that the questioner is at the age of reflection. Young children do not wonder how to live. And, according to my account, practical reasoning is an essential part of one’s maturation from a child to a practically wise human being.
Fourthly, the question calls for a certain kind of answer, namely, a practically reasonable answer. Recall Jay Wallace’s general definition of practical reasoning as “the general human capacity for resolving, through reflection, the question of what one is to do.”^[@seppracticalreason] Although sometimes we reflexively act without thinking, and other times contemplate without acting, (“four and four makes eight”), it seems obvious, on the face, that deliberation and resolute action are not like this. One resolves what to do by considering practical reasons. When a child asks a “how?” question about, say, how to open a jar, we offer a practical instruction: hold the base tightly, grip the lid and twist to the left. As adults, we ask “how?” questions about large, multifaceted projects: How to manage a company merger? How to save for retirement? How to raise a child? The “instructions” for such answers will be complex. The “how should I live?” question is simply our most complex long-term project. The answer or answers cannot be an overly vague resolution (e.g., “help to improve the world”), nor mere specific platitudes (e.g., “do no harm”). Rather, a good answer will distinguish between overall good ways and overall bad ways to live and include a set of practical reasons, some general enough to give a trajectory to one’s whole life, and others specific enough to provide guidance through the day-to-day matters of human life.
In short, an answer to the “how should one live?” question requires practical wisdom. Practical wisdom is unique among virtues in several ways. First, it is perhaps the one clearly non-optional virtue. Everyone has the obligation to become practically wise, regardless of circumstances, social roles, aptitudes, cultures, and so on. The universality of the obligation arises from the mere fact that one is a practical, rational primate. Secondly, practical wisdom is also unique in that it enables one to acquire other virtues, such as courage or moderation, by providing its possessor with the insight and moral skill to develop specific good habits in the varied circumstances of normal life. Thirdly, practical wisdom is recursive: the practically wise person is the most well-equipped to root out folly and become more practically wise.
The neo-Aristotelian framework for doing ethics views ethical reasoning as a holistic process that must be sensitive to the whole range of practical reasons. According to such thinkers, there can be no adequate theory of ethics without a theory of practical rationality. According to the arguments of the last chapter, virtues are traits that enable one to live a distinctly human life and that partly constitute that life. In this chapter, I shall argue that the practically wise person is engaged in “mapping the landscape of value”^[@seppracticalreason, section 6] – that is, developing the knowledge and good intentions needed to pursue what is truly worthwhile and avoid what may seem worthwhile but is actually worthless. If successful, I shall be lending support to the age-old view that the skill of engaging in practical reasoning – reliably and successfully – is the virtue of practical wisdom. The practically wise person is one who knows the answer or answers, if there are any such answers. The one who answers this question poorly lives foolishly and, ipso facto, badly. He acts on bad reasons and fails to act on good reasons. The one who answers it well lives wisely, and ipso facto, well. Hence, it is essential to virtue that one be practically wise. Or so I shall argue.
Section 1 breaks ground on this complex matter through a sustained discussion of John McDowell’s “Virtue and Reason.” I offer a qualified defense of his thesis that virtue is a form of practical knowledge, including an initial perceptual sensitivity to the salient facts of a situation with the skill to do what is required by those facts.
Section 2 highlights an ambiguity in McDowell’s contrast between ‘moral’ and ‘practical’ reasons. He confuses the genus ‘practical reasoning’ for one species of ‘moral reasoning’ – that is, reasoning about one’s obligations to others. I attempt to remedy this confusion by putting in historical context the relationship between ‘moral’ and ‘practical’ reasons. McDowell confuses two frameworks for approaching ethics: the ‘quandary frame’ and the ‘character frame.’
Section 3 offers a more coherent alternative. It reprises the argument that human beings are practical reasoning animals by placing our distinctive activity in context of the general inclination of all living things to their own life and health. In this light, practical reasoning is a necessarily substantive form of reasoning about ends, rather than a merely instrumental one about means, because in order to have any reasons at all one must have a first principle of practical reason, namely, a general evaluative conception of what is to be pursued and hence how to live.
Section 4 addresses some serious objections to my way of framing ethical reasoning. For example, how, exactly, is a rational calculative process central to moral virtue? Here, I deal with three objections that challenge the notion that successful practical reasoning is essential to human virtue.
Virtue as Practical Reasoning
John McDowell’s “Virtue and Reason” argues, among other things, that virtue is a particular kind of practical knowledge. Practical reasoning is both a rational process and also an initial, perceptual sensitivity that makes visible to us practical reasons. Even though he allows that practical reasons are ultimately intersubjective features of our social world, he argues that they are no more and no less objective than theoretical reasons. In this section, I trace his discussion in some detail, including his statements of various objections and responses to them.
What kind of knowledge is virtue, according to McDowell? It is a practical and dispositional what to do. It is not simply propositional. Rather, it is a non-codifiable perceptual sensitivity to salient facts along with a disposition that leads the virtuous knower to act properly – so long as no countervailing psychological factors interfere. Some objections to his thesis will be addressed as we proceed.
How does it make sense to conceive of virtue as practical knowledge? Consider a platitudinous value such as kindness. Suppose kindness is really a virtue. What does it mean to predicate kindness of someone? We cannot ascribe a virtue to someone who acts kindly once or twice, or who does so (even consistently) by pure luck. Justifying the ascription of a virtue requires that a person “has a reliable sensitivity to a certain sort of requirement which situations impose on behavior” and such “deliverances of a reliable sensitivity are cases of knowledge.”^[@mcdowell1979virtue 332] McDowell is gesturing toward three or four plausible criteria for the ascription of a virtue: reliability means the kind person must be regularly or habitually disposed to kind thoughts, feelings, and behaviors; sensitivity means that the kind person demonstrates an alertness to the fact that a friend is in need, a child is sad, an elderly parent is lonely, etc.; practical knowledge means the kind person knows what to do in such situations; and intentional behavior means that the person correctly feels the imposition to avoid cruel and indifferent behavior and to act on what the situation requires.
McDowell has made it plausible that sensitivity to reasons to behave a particular way is at least necessary for virtue. But is it sufficient? He offers two answers. The first answer is that the presence of a virtue in someone exhaustively explains her behavior. For example, when the kind person sees that a situation requires kindness, that “requirement imposed by the situation” must “exhaust his reason for acting as he does.”[@mcdowell1979virtue 332] An ulterior interest (say, in a mercenary reward) would disqualify the action as an example of kindness. The kind person’s action is explained by the simple fact that it would be a kind action.
Now, the kindness is not the only reason for action. There are many reasons for action and many situations where no single overriding reason is obvious. Rather, the question of what to do seems to generalize into a question of what is good or advisable, all things considered. McDowell concedes the point. He illustrates it with the example of a parent who is overly indulgent to a child out of kindness. Certainly, the parent is sensitive to what kindness requires but not sensitive enough to fairness or to considerations of the child’s health, and so on.
To accommodate this observation, McDowell generalizes this point to encompass all of virtue:
Thus the particular virtues are not a batch of independent sensitivities. Rather, we use the concepts of the particular virtues to mark similarities and dissimilarities among the manifestations of a single sensitivity which is what virtue, in general, is: an ability to recognize requirements which situations impose on one’s behavior. It is a single complex sensitivity of the sort which we are aiming to instill when we aim to inculcate a moral outlook.^[@mcdowell1979virtue 333]
McDowell is saying that if the kind person’s behavior arises from a response to the salient facts he is sensitive to, then the virtuous person’s behavior in general is explained by just the fact that it is virtuous. The virtuous person’s behavior, then, arises from a general sensitivity to what situations require. If virtue is a “single complex sensitivity” that constitutes an entire “a moral outlook,” then virtue seems to be not just a perceptual capacity to notice what is required but also a metacognitive capacity to reflect upon, rank, and order, the various requirements imposed by a situation before acting accordingly.
I have a complaint about McDowell’s clarification here, which I shall explain below. In brief, it seems wrong to call the single sensitivity “virtue” when it includes considerations that do not seem intuitively moral at all, such as prudential considerations. For now, I must examine McDowell’s response to the non-cognitivist critic who challenges the notion that practical reasoning can, by itself, motivate one to action.
Reason, Practice, and Motivation
The first challenge to his own thesis that McDowell addresses comes from moral anti-realism, specifically, expressivism. Expressivists are among the chief contemporary proponents of an alternative, Humean, model of practical reasoning which denies that practical reason is “a capacity for reflection about an objective body of normative truths regarding action.”^[@seppracticalreason, section 2. Wallace cites Parfit (2011) and Scanlon (2014).]
Thus far, it is fairly clear that I have been assuming a kind of realism. While defending the assumption would take us too far afield, I should point out that it is not viciously circular. Most of us have no pre-analytic objection to the seeming fact that some reasons for acting are good reasons and others bad. Some brute norms (such that it is wrong to torture animals, or that one is not to use ineffective means to achieve one’s ends) have a quasi-analytic force to them. Realism about practical reasons is what Nagel calls a “defeasible presumption.”^[@nagel1989view 143.] Even anti-realism’s most sophisticated advocates concede the that realism is the default view. Mackie admits that moral thought and language assumes objectivity, for the notion of objective value has “a firm basis in ordinary thought, and even in the meanings of moral terms.”1 Gibbard goes so far as to suggest that Platonism about reasons is common sense.2
Nevertheless, anti-realism has a serious challenge to the defeasible presumption. Subjectivism is motivated by considering a problem about the status of practical reasons within (a broadly-construed) naturalism. The anti-realist worries that the “defeasible presumption” lying at the center of “the main tradition of European moral philosophy” commits one to non-natural norms and a corresponding non-naturalistic human capacity to intuit them. Philosophers such as Gibbard insists: “Nothing in a plausible, naturalistic picture of our place in the universe requires … non-natural facts and these powers of non-sensory apprehension.”^[@gibbard1992wise 154]
The anti-realist alternatives aim either to debunk the objective purport of moral reasoning or to reclaim it within the confines of a respectable naturalism.
The Humean model of practical reasoning asserts that “cognition and volition are distinct.”^[@mcdowell1979virtue 335] Practical reasons cannot motivate, at least not by themselves.^[He says: “Reason is the discovery of truth and falsehood.” (Treatise of Human Nature, Part I.1.)] If this were so, moral reasoning could not satisfy the “practical requirement” – it could neither move us to action nor explain why we acted. Indeed, a large part of the appeal of expressivism is that it can satisfy the practical dimension of practical reason (though at the cost of the rational dimension).
Hence, the non-cognitivist critic would be quick to respond to McDowell with a counterexample of two persons in the same situation who are sensitive to an identical range of reasons for action but respond differently. If such a situation were to obtain, it would disconfirm McDowell’s thesis that virtue is practical knowledge.
The expressivist has a neat explanation of reasoning, action, and motivation. If reasons cannot motivate by themselves, then practical reasoners act when reasons co-exist with a conative mental state (such as a desire, interest, or attraction).3 Practical reasoners do not simply enjoy a “single complex sensitivity” to what situations require. Instead, the cognitive bit judges an object, while the conative state provides the movement toward the object. For example, one is aware that one’s friend is in trouble and that the friend is able to be comforted (the cognitive bit) and a desire (or motivation or inclination or settled passion) for helping one’s friend (the non-cognitive bit). The expressivist would say that surely these two together – and neither in isolation – explains the behavior.
This challenge presents a pair of twin challenges: is virtue-knowledge practical – and if so, wouldn’t it be impossible for an agent to perceive what a situation requires and still do wrong? Secondly, is virtue-knowledge rational – and if so, mustn’t it be codifiable and consistent? The very notion of a unitary “practical reasoning” is a paradox.
Is Practical Reasoning Practical?
McDowell’s response to the expressivist critic is this: one must already be sensitive to a particular range of requirements for action in order to even notice the salient facts (e.g., that one’s friend is in trouble). It is quite plausible to interpret the difference between the vicious and virtuous person as lying not just in their psychological reactions to what they notice about the world but in the noticing itself.[See also @little1995seeing] The morally calloused person does not notice the fact that his actions are causing others pain. Better, the morally calloused person does not notice the fact as morally salient.
This response from McDowell is not conclusive, but it is a good start. It highlights, but does not alleviate, the deep disagreement between the Humean and the Aristotelian camps. He concedes the conditional that if two people are identically sensitive to a morally salient fact but act differently then virtue cannot simply be a sensitivity. But, for McDowell, one person’s modus ponens is another’s modus tollens. So if virtue is to be identified with a single complex sensitivity, then a supposed situation in which two persons perceive a situation and its practical requirements identically but act differently cannot obtain.^[@mcdowell1979virtue 333] Is there any way to bridge the divide without begging the question in either direction? McDowell suggests we look to Aristotle.
Aristotle allowed that sometimes the “appreciation of what [a virtuous person] observes is clouded, or unfocused, by the impact of a desire to do otherwise.”[@mcdowell1979virtue 334] It is possible that a person correctly perceives what a situation requires (and hence has the relevant virtue) but fails to act correctly due to interference from other psychological factors. Desires, fears, etc. might cause a “distortion in one’s appreciation” of the relevant reasons.^[@mcdowell1979virtue 334]
This Aristotelian reply is also not conclusive. McDowell cites an objection from Donald Davidson to the effect that a person might fail to perform the resultant right action even without such clouded appreciation. McDowell concedes. But Davidson’s move changes the subject slightly, from virtue and vice to continence and incontinence. For Aristotle, continence (or self-control) is not a virtue. If one can only do the right thing by gritting one’s teeth and bearing it, one has not fully attained the relevant virtue. Continence is still comparatively better than incontinence – but not as good as virtue.
The continent person is able to perform the right action because he recognizes it as right, despite countervailing pressures (from desires, say) to do the wrong action. Since the possession of a full virtue includes possession of the proper motivation as well, continence is only needed in the absence of a fully developed virtue. Put differently, the virtuous person is not just one who “balances” reasons to φ against countervailing reasons to π. The virtuous person is the one for whom simply identifying the appropriate reasons to act silences countervailing reasons. For example, in this situation, courage requires that I run into danger. The virtuous persons acknowledges the danger (and feels rightly apprehensive) but also sees that courageous action in the face of this danger is required; the latter perception, according to McDowell, “silences” other pressures.^[@mcdowell1979virtue 335] The merely continent person has to “weigh” reasons; the virtuous person fluently acts on the best reason.
In my view, McDowell’s reply to Davidson’s objection is not quite adequate. Fully explaining the common occurrence that we judge what is to be done but fail to do it would require a lengthier discussion. I have argued in chapter 4 that some virtues are excellent rational practices. The fullness of virtue is not merely the sensitivity to what is required – which an incontinent person might have – but also a well-ordered psychology (including emotions) and a set of rational habits that empower the agent to follow through on doing what is required. Unlike the continent person, the virtuous agent has overcome the psychological factors or other factors that cloud the appreciation of what is required. And unlike the incontinent person, the virtuous agent has cleared away other factors that interrupt the execution of the thing to do. Once the fullness of a virtue is attained, the possessor does not need to stop and “weigh.” He sees what is required and acts. (I shall comment a bit more on moral motivation below.)
Is Practical Reasoning Rational?
McDowell’s case that practical knowledge can motivate the virtuous person requires addressing twin challenges. We have addressed one side of the paradox which challenges the practicality of virtue-knowledge. The other side of the paradox challenged its rational credentials. Pretty clearly, the paradigmatic case of knowledge is theoretical knowledge, i.e., knowledge that p. Such knowledge is categorical, propositional, and codifiable into a deductive logical system. McDowell’s critic then poses the following argument: knowledge is codifiable. However, virtue-knowledge is practical knowledge or ‘knowing-what-to-do’, which is not codifiable. Therefore, virtue must not be knowledge.
The error in this objection, McDowell thinks, is not an error in moral theory but a “deep-rooted prejudice” that rationality is a rule-following procedure. If rationality is a rule-following procedure, then it follows that either practical rationality and morality are likewise rule-following procedures or that practical rationality and morality are not, ultimately, sufficiently rational. Some Humean philosophers (but not necessarily Hume) think that morality is a non-rational domain of sentiments, desires, commitments, approvals, and so on. Other Kantian philosophers (but not necessarily Kant) think that morality is a rational domain and hence must be a matter of identifying first principles and applying them to particular situations. What both parties share is a belief that “rationality must be explicable in terms of being guided by a formulable universal principle.”[@mcdowell1979virtue 337. MacIntyre, similarly, denies the assumption that normative ethical rules can be derived from universal ethical principles the way we “apply” universal logical truths to particular logical conclusions via a middle term. Cf. @macintyre1984does.] This common belief McDowell wishes to refute.
McDowell’s argument here (drawing on Wittgenstein and Kripke[@kripke1982wittgenstein]) is that even apparently obvious cases where the rational thing to do seems to require following an objective rule turn out to be cases of a much messier process in which there is no such objective rule we can appeal to. For example, take the objective rule of extending a series of numbers two at a time. Suppose Smith instructs Jones to “add 2” to a number and continue applying the rule indefinitely. We are confident (as is Smith) that Jones will “churn out the appropriate behavior with the sort of reliability which a physical mechanism, say a piece of clockwork, might have.” We tend to expect that Jones will produce “2, 4, 6, 8,” etc. McDowell thinks this confidence is based on postulating a “psychological mechanism, underlying his behavior, by an inference analogous to that whereby one might hypothesize a physical structure underlying the observable motions of some inanimate object.”[@mcdowell1979virtue 337] The postulate of a psychological mechanism is mistaken because, as it turns, out, the rule being followed is not so simple as “add 2 indefinitely.” It is logically possible that Jones interpreted Smith’s instruction as a different rule that happened to produce the same result. The attempt to stamp out this possibility by adding new meta-rules or sub-rules iterates the problem. It is still logically possible that Jones follows a different meta-rule or sub-rule that happens to produce the same result. Wittgenstein’s conclusion is that even apparently simple rules, successfully followed, cannot be exhaustively described.
McDowell’s conclusion is that the true “ground and nature of our confidence” is our participation with Jones in a common form of life. What is a ‘form of life?’ This is a term of art, also drawn from Wittgenstein and quoted with approval from Stanley Cavell. It refers to the shared result of acculturation or formation. For example, how do we learn reliably to use words and expressions in our native language? There is no clear mechanistic process that explains exactly when a child learns to make exclamations – such as a pained “ow!” or an excited “ooh!”. There is no clear mechanistic process by which we learn when to laugh at jokes or when to cry in pity. Instead of a mechanistic process, McDowell suggests that children learn words and behaviors by “bildung” or formation. The result is a “congruence of subjectivities”.^[@mcdowell1979virtue 339] Jones is able to follow Smith’s rule (and we are confident that we know what his instruction meant) even though it is not stated exhaustively because we all share a common practice of, say, adding or producing a series of numbers. More generally, all of our shared rationality is not grounded in “external” objective rules but in a shared form of life.
It is disconcerting to many to consider that nothing keeps rationality “on the rails” but a congruence of subjectivities. McDowell admits this is a disconcerting hypothesis; it induces “vertigo.” But, he says, our response to such vertigo should not be to embrace a “consoling myth”. That “consoling myth” consists of two notions: (a) that rational rule-following is enabled by a psychological mechanism that guarantees consistency; and (b) that there exist objective facts of the matter over and above the congruence of subjectivities. If we abandon these two notions and embrace the model of deductive rationality as grounded only in our intersubjective form of life, then the corresponding model of practical rationality will become tenable.
I think McDowell concedes too much here, as I shall explain below. Nevertheless, my purpose here is to agree with McDowell that both forms of rationality – the practical and the theoretical – are on a par. They stand or fall together. Either they are both intersubjective or both objective. Practical reasoning may be relatively less codifiable than theoretical reasoning, but each is equally a form of knowledge.
McDowell asks a related query: what, if anything, guarantees that the moral person’s behavior is intelligibly the same from case to case? If moral knowledge were formulable as a universal principle, then it would be consistent from case to case and situation to situation; but if, as McDowell has been arguing, both deductive reasoning and practical reasoning are not merely rule-following mechanisms, then how do we explain the virtuous person’s reliably correct behavior? His answer invokes Aristotle’s notion of a practical syllogism. According to McDowell, the ‘practical syllogism’ takes the following shape:
- X is good to do, desirable, worthwhile, etc. (E.g., it is good to instantiate justice in the classroom).
- Z would be X. (E.g., giving everyone a chance to re-take a quiz that was unavailable due to technical problems would instantiate justice in my classroom.)
- Therefore, Z would be good to do, desirable, worthwhile, etc.
On the strictly deductive logical model, the role of the major premise is to provide solid universal ethical principles from which it is possible to derive a codified set of particular moral duties. McDowell resists this model. The strictly non-cognitivist alternative is that there must be no universal ethical principles at all – only universal psychological states, such as consistent desires, plans, values, or norms. McDowell also resists this model. Instead, the role of the major premise is to articulate a “certain conception of how to live… [namely] the virtuous person’s conception of the sort of life a human being should lead.”^[@mcdowell1979virtue 343. Emphasis added.] What kind of life should a human being lead? The answer “cannot be definitively written down.”^[@mcdowell1979virtue 343]
If the kind of conception of a good life that the virtuous person has is approximate and non-codifiable, it becomes hard to see why we are bothering to fit moral reasoning into a syllogistic pattern at all. McDowell’s response is that understanding virtue-knowledge within a practical syllogism does a good job of providing a plausible explanation of moral motivation (reasons one might act in some way) and moral behavior (reasons one acted that way). To paraphrase McDowell: “Explanations of judgments about what to do are also explanations of actions.”^[@mcdowell1979virtue 342. Verbatim: “The explanations, so far treated as explanations of judgments about what to do, are equally explanations of actions.”] I can explain your behavior by understanding that you were concerned for your friend’s welfare and so offered to help. Likewise, you can explain your decision to help simply by citing the fact that your friend was in need. So the general structure of the practical syllogism is useful.
What’s more, McDowell concedes that there is a kind of circularity to his account: “the rationality of virtue… is not demonstrable from an external standpoint.”[@mcdowell1979virtue 346.] And: “Any attempt to capture it in words will recapitulate the character of the teaching whereby it might be instilled: generalizations will be approximate at best…”^[@mcdowell1979virtue 343] The virtuous person’s conception of how to live is itself conditioned by the moral outlook. That conception of how to live, in turn, conditions what particular saliences are noticed (what minor premises) and generates practical conclusions about what is to be done. The upshot of the combination of non-codifiability with a practical syllogistic form is that the virtuous person takes for a rule of life some conception of how to live but that this conception is part of what it means to be a virtuous person – and thus ensues a certain circularity.
McDowell bites the bullet on the incorrigible intersubjectivity of theoretical and practical reasoning. I think he does so because he fails to grasp Foot’s insight that objective, natural, normative facts are able to “keep us on the rails.” I am not motivated to think this out of a desire to be “consoled.” In any case, the presence or absence of a frightening vertigo in the arguer is irrelevant to the argument. The notion that all practical and deductive reasoning is ultimately answerable to the world is a more adequate explanation. As I have argued in chapter 2, both scientific reasoning and ethical reasoning can conform or fail to conform to the relevant range of normative facts. I shall criticize McDowell’s intersubjective notion a bit more in the next chapter.
In sum, McDowell thinks virtue is a kind of practical knowledge. That is, virtue is a sensitivity to salient facts that call for a particular response and that intrinsically motivates the virtuous person that response – absent interfering passions. The hypothetical counterexample presented by his Humean critic is one wherein two agents are “sensitive to” or “notice” identical reasons for action but do not act identically. McDowell’s response is that while noticing a requirement for action is necessarily motivating to some extent, other psychological factors may interfere with the resulting correct action. Furthermore, the kind of “knowledge” that virtue amounts to is uncodifiable, but that does no harm to the account. Virtue-knowledge is rather a broad conception of how to live and a series of specific sensitivities to a range of specific practical reasons. Practical reasoning is consistent, moreover, but not by being “objective” (in the sense that even McDowell admits would be desirable) but by being rooted in our communal form of life – precisely the same way in which logical reasoning is. Both are “intersubjective” and rooted in our form of life, but both are as objective as need be.
Moral and Practical Reasoning
While I shall discuss what I think McDowell gets wrong below, on my view, he gets this much right: practical reasoning is indeed by definition a form of reasoning. It is like theoretical reasoning in that it is normative.
Broadly, we can say that theoretical reasoning is a process by which I aim to determine what to believe – to answer the question “What should I believe?” When I assess evidence for and against some proposition p, I am looking for reasons to believe p is true or false. The successful conclusion of a rational argument is the judgment that p or not-p. (Or I may not have enough evidence to judge either way, in which case I may withhold judgment.) Similarly, when I consider a scientific hypothesis, I suppose that p and then conduct an experiment that will reveal reasons that confirm or disconfirm the supposition. To fail to believe p upon coming to know good evidence for it, or to believe p in spite of good evidence against it, is to make an intellectual error. If q entails p and I already know and affirm that q, then I ought to affirm that p. Similarly, if some reason to π entails a decisive reason to φ, and I already know and am committed to π, then I ought to φ.
So far as we know, all theoretical reasoners are also practical reasoners. We can imagine creatures such as angels, Artificial Intelligences, and intelligent aliens who might think without acting; but so far as we know, to be a reasoner at all is to be responsive to what Sellars called the “space of reasons,” including both practical and theoretical reasons. This consideration is part of the reason why, in chapter 3, I insisted that practical reasoning, and not abstract theoretical reasoning, defines human nature. If this is right, then the burden of proof lies with those who would artificially separate the knowing and the practicing.
That said, my complaint against McDowell’s account is that he confuses moral and practical reasons. Suppose Jane can pretty well diagnose a car engine by listening to the way it whines, hums, and clicks. When all John hears is noise, Jane is “sensitive to a range of requirements for action” and knows what to do (say it needs a new timing belt). By McDowell’s lights, she has a practical disposition which is virtue. It strains common sense to call any and all such sensitivities “virtues.”
Even if we introduce a requirement that practical knowledge must be concerned with requirements pertaining to other people, similar analogies arise in other contexts. For example, an American football kicker is sensitive to the salient facts of what is required to score a field goal, which will help his team to win. And a general contractor is sensitive to the salient facts of what is required to build a structure up to code, and so on, which will help protect the safety of whoever ends up living in the structure. Both of these practical skills help others but neither seems to amount to moral virtue. Even though a contractor willfully who fails to build a structure safely might be morally and legally liable for any subsequent injuries, something strikes us as odd about classifying the skill as a virtue.
There is a second, related complaint. McDowell admits that one might potentially need to rank, order, and weigh a dozen different kinds of reasons (kindness, fairness, appropriateness, prudence, etc.) before one resolved what to do. He seems to switch from talking about moral reasons to talking about any practical reason without any mention of the switch. By failing to render a clear distinction between moral and other practical reasons, I believe McDowell falls prey to a habitual way of framing moral discussions that is a subtle mistake.
The habitual way of framing moral discussions we may call the “quandary frame,” borrowing the term from a classic article by Edmund Pincoff. Pincoff contrasts “quandary ethics” with another way of framing ethical discussions which he calls “character” ethics. On this frame, ‘moral’ considerations contrast with prudence and any other kind of practical consideration. ‘Moral’ considerations most commonly refer to “other-regarding” considerations (opposed to self-regarding ones), altruistic (as opposed to egoistic), considerations of benevolence (as opposed to selfishness), or conscience (as opposed to self-love).^[The relation between virtue and happiness or self-regard is an expansive one: For fuller treatments, one might begin with: @macintyre1967egoism. @slote1995agent. @annas2009egoism. @huang2010self. @bloomfield2012eudaimonia.]
The contrast between moral and all other practical reasons gives rise to a distinctive way of approaching ethics. Pincoff writes of quandary ethics:
The business of ethics is to clarify and solve “problems,” i.e. situations in which it is difficult to know what one should do; that the ultimate beneficiary of ethical analysis is the person who, in one of these situations, seeks rational ground for the decision he must make; that ethics is therefore primarily concerned to find such grounds, often conceived of as moral rules and the principles from which they can be derived; and that meta-ethics consists in the analysis of the terms, claims, and arguments which come into play in moral disputation, deliberation, and justification in problematic contexts.^[@pincoffs1971quandary 552. MacIntyre offers a criticism along similar lines to Pincoffs in his “Does Applied Ethics Rest on a Mistake?”]
According to Philippa Foot, the quandary frame is the way most modern philosophers approach ethics. She says:
Many if not most moral philosophers in modern times see their subject as having to do exclusively with relations between individuals or between an individual and society, and so with such things as obligations, duties, and charitable acts… ‘moral’ and ‘prudential’ considerations [are] contrasted in a way that was alien to Plato or Aristotle.^[@foot2001natural 68]
Relatedly, Martha Nussbaum says:
This question [of how ‘moral’ ends figure among other ends] is posed in a characteristically modern way, presupposing a distinction between the moral and the non-moral that is not drawn, as such, by the Greek thinkers. But if one objects to that characterization, one can rephrase it: for example, What role does concern for others for their own sake play in here scheme of ends? What role does political justice play in her scheme of ends? And so forth.”[@nussbaum1999virtue 174]
Kant and Hume agree on the quandary frame, despite their significant substantive disagreements. They both present morality as a kind of crisis strategy. On any given normal day, agents are free to pursue their own self-interested inclinations – get a good job, save for retirement, eat healthy foods, exercise, make friends, and so on – so long as they commit no wrong. So long as life presents no moral dilemmas, moral reasoning is idle.
The alternative type of ethics is what Pincoff calls “character” ethics (of which I take neo-Aristotelian virtue ethics to be a token). Such ethics is focused on the long-term project of living well by executing worthwhile goals in every day life. Aristotle is the premier example of a character ethicist because he thought of ethics as a branch of the whole practical enterprise:
…[ethics and politics] is a very wide-ranging subject having to do generally with the planning of human life so that it could be lived as well as possible. Moral problems are given their due but are by no means stage-centre. The question is not so much how we should resolve perplexities as how we should live.^[@pincoffs1971quandary 553-4]
The Greek way of framing moral questions viewed all practical ends as ‘moral.’ MacIntyre provides the clearest summary of the older use of ‘moral’:
‘Moral’ is the etymological descendant of ‘moralis’. But ‘moralis’, like its Greek predecessor ethikos – Cicero invented ‘moralis’ to translate the Greek word in the De Fato – means ‘pertaining to character’ where a man’s character is nothing other than his set dispositions to behave systematically in one way rather than another, to lead on particular kind of life… The early uses of ‘moral’ did not contrast with ‘prudential’ or ‘self interested’” nor with ‘legal or ‘religious’… The word to which it is closest in meaning is perhaps most simply ‘practical’.^[@macintyre1984after 38]
MacIntyre’s point is not merely etymological; it is conceptual. When quandary ethicists conceive of ‘moral reasons’ as a special overriding type of practical reason concerned with duties to others (contrasted with self-regarding prudential reasons), they fall under the illusion that moral reasons may not be practical and that practical reasons may not be moral. By contrast, the character ethicist views life as presenting the variety of possible ends that could clash or harmonize that all need to be accounted for.
It is helpful to observe that, at some point in the history of western moral philosophy, the topic of the “moral” began to separate off from the broader topic of the practical. Foot cites Mill as an early proponent of the distinction:
J. S. Mill, for instance, expresses this modern point of view quite explicitly, saying in his essay On Liberty that ‘A person who shows rashness, obstinacy, self-conceit . . . who cannot restrain himself from harmful indulgences’ shows faults (Mill calls them ‘self-regarding faults’) which ‘are not properly immoralities’ and while they ‘may be proofs of any amount of folly . . . are only a subject of moral reprobation when they involve a breach of duty to others, for whose sake the individual is bound to have care for himself’.^[@foot2001natural 68]
Mill distinguishes folly from immorality by treating folly as a failure to provide goods for oneself. He treats imprudence as “bad” but not morally bad.
While I don’t intend to suggest that there is something automatically laudable about the older Aristotelian emphasis, my contention is that the modern emphasis on “relations between individuals or between an individual and society” fails to capture much of what is interesting about the “how should one live?” question. The modern distinction obscures the real ethical situation.
To return to McDowell, I can now put my complaint in clearer relief: is he a quandary ethicist or character ethicist? In my view, McDowell’s view represents a mixture (indeed, a confusion) of the two. Like the character ethicist, he emphasizes the “how should I live?” question and invokes practical knowledge as an important part of the answer. However, like the quandary ethicist, he represents moral considerations pertaining to the rights, obligations, or duties to others (such as kindness) as a special, perhaps overriding, kind of reason. He does not seem to notice that broadening the virtuous person’s perceptual sensitivity to what any situation requires renders his account ambiguous. Are moral reasons one type of practical reason, or can any practical reason count as a “moral” reason (broadly construed).^[@foot2002virtues chapter 13, “Are Moral Reasons Overriding?”; Cf. also @mcdowell1978moral]
Practical Reasoning as Pursuing the Human Good
The remedy for this confusion is to return to and defend a more consistent account of practical reasons. Happily, this account will reinforce what we have argued above about the natural normativity in the human life form and all organic life. This section builds on the work of Philippa Foot and on Jennifer Frey’s recent discussions of Anscombe and Aquinas.4
On the Aristotelian account, as developed by Aquinas, practical reasoning is by definition an end-oriented activity that aims at the perceived good of one’s form of life. The primary question is not “why should one respond to moral reasons instead of prudential ones?” but “why do we act at all?” and “how can we act well?” Asking this question, and answering it, is a practically rational activity that defines the human life form. Certainly, as Foot says, some practical reasons have to do with “obligations, duties, and charitable acts” to others; but others pertain to what is required for oneself and even for third-person objects such as the environment, possessions, and perhaps even abstract objects.
Considered thus broadly, the normativity of practical reasoning is clear: some reasons for acting are good while others are bad. Errors of morality, then, belong to a wider class of practical errors. As Foot says: “I want to show that judgments usually considered to be the special subject of moral philosophy should really be seen as belonging to a wider class of evaluations of conduct with which they share a common conceptual structure.” [@foot2001natural 66-67] On this frame, any reason to φ or not to φ is a practical reason, and successfully sorting through all such reasons is a virtue, namely practical wisdom. Unsuccessfully doing so is the vice of imprudence or practical folly, which inhibits one’s ability to live a human life.
Defending the Aristotelian account requires us to revisit in more detail some of what was argued above in chapter 2. Recall the observation that all organisms act toward ends, with or without reflection. Frey summarizes Aquinas in this way:
All living things are a self-sustaining system of powers that functions to bring the living thing into being and to sustain its being. The movement of any part of a living thing, at any particular moment, is necessarily explained by reference to the movement of the whole thing towards a single end: the coming to be, maintenance, or reproduction of that very form of life.^[@frey 68]
As I argued above, all living things exhibit teleological movement. In proper circumstances, they grow into maturity, which is the exemplification of their form of life. This form of life is what Aquinas calls a thing’s “nature”: wolf hunts in packs by nature, trees extend roots into the ground by nature, reptiles warm themselves in the sun by nature, and so on.
The sunflower has no consciousness with which to incline toward sunlight. ‘Things are specified by their power.’ When it comes to higher organisms, insects and mammals and so on, organisms have “appetite.” They demonstrate the capacity to sense and to move consciously toward or away from certain objects: The antelope pursues healthy grass and flees a lion. The animal can only experience what is good or bad for it as a particular object.
While natural norms are features of all living beings, human beings are distinct in also being aware of such norms. Humans grow, reproduce, and enjoy conscious experiences like other animals and also know that they do so. Obviously, plants and animals do not “naturally incline” toward their good by reflecting or choosing it. Frey points out:
Aquinas would agree with us that it is a category mistake to say that a sunflower wants to grow towards the light, if by this we mean that the flower somehow registers a positive feeling or has an inner impression towards the light, which “causes” it to move toward the light. The plant does not apprehend or desire anything; thus Aquinas is very careful to say that it does not have a power of appetite. In fact, Aquinas is at pains to note that a plant has no window onto the world at all – it just has conditions in which it characteristically comes into being, maintains, and reproduces itself.^[@frey 69-70]
Lower organisms naturally incline toward their own good. Higher organisms perceive objects but do not perceive them as falling under universal categories. By contrast, a human being can recognize universals. Human beings are specified by their “power” – their capacity to engage in cognitive and deliberative activities. While animals can not only sense but perceive, humans have the capacity of “intellection” – the power of abstracting formal properties from what is perceived. An animal can sense an informed, organized object; an animal can be affected by the object. But the human animal can acquire information from the organized object. Animals may perceive something as dangerous or as desirable. Human beings perceive that the dangerous thing is a predator or the desirable thing is food.^[@haldane1996coming]
The extra ability to perceive under universal categories brings with it the human capacity for taking up natural inclinations or aversions in a deliberative act. Natural inclinations may be underwritten or overridden. Confronted with a delicious and healthy salad sitting on someone else’s plate, I recognize it as not mine and hence choose not to reach for it. Confronted with a lion in a zoo, I choose not to flee, for I recognize it as not dangerous. Frey summarizes:
Rational animals, like any animal, have a natural inclination towards their good as a whole, and like lower animals this power is actualized through their apprehension of things in the world. But Aquinas argues that a rational animal relates to the world through the application of universal concepts, and thus it is inclined to pursue or avoid things under an intellectual, universal apprehension of them. Thus, Aquinas says that the will is inclined towards its objects under the formality of the “universal good,” rather than the particular good.^[@frey 75]
By the same token, human beings are capable of an extra ability to err. The conclusion that all living things move toward their own natural ends is compatible with the biological judgment that some specimens are defective, just as it is compatible with the ethical judgment that some agents – such as Dostoevsky’s Underground Man – are practically irrational in failing to pursue their own natural ends. Human beings are supposed to practically reason well. When they do not, the defect that arises is more than merely animal. Any animals might be afflicted by sickness or injury; only human animals can inflict themselves with new injuries and even new illnesses.
We have been speaking of the human capacity for recognizing and pursuing particular ends as good. As we saw in chapter 4, a full conception of virtue demands that we expand our scope to include the whole of life, the conception of our human good that constitutes the answer to the “how to live?” question. McDowell gets this part right in his discussion of the practical syllogism. Every rational practice is undertaken in pursuit of some particular end in context of a total conception of what is good in general. Frey continues:
Consequently, we can say that rational animals have an understanding of different levels of ends, and at least a vague sense of how they are supposed to hang together as a whole. This conception of how it all hangs together is what Aquinas calls the ultimate end – a rational animal’s general, conceptual understanding of how to live or go on. Aquinas thinks that any sane, mature adult will necessarily have cobbled together some such conception. Aquinas calls this conception “the universal good”, and he argues that it is the will’s proper object. Everything that is willed is willed under this rational aspect of good, as to be pursued because in accord with my general conception of the good. In fact, Aquinas thinks there could be no reasons unless a rational animal has a general conception of its own good, and thus a general sense of how to live. [@frey 78-9, italics in original.]
Frey’s argument here is that the question of ‘how to live’ is a question about my good as a human being; answering that question requires the human activity of practically reasoning. And since every “sane, mature adult” engages in this activity, every sane mature adult has a general notion about the answer. The crucial insight is that without such a general notion, we would not engage in rational action at all. Frey continues:
No human action is intelligible without attributing to the agent herself some conception of this end, no matter how inarticulate, unsystematic, or unreflective it might be. Aquinas takes it for granted that in coming to be a human being – i.e., being raised in a community of other human beings, coming into the possession of concepts, a language, and coming to have a world – one comes into some such conception, and thus comes to act voluntarily.^[@frey 87]
Human beings act. And all intelligible actions are undertaken in pursuit of some end. Therefore, all intelligible actions of humans are undertaken in pursuit of some end. This conclusion can accommodate the commonsense observation that not every move we make counts as an intelligible action. Aquinas makes a helpful distinction between the “actions of a human” and “human actions.” An action of a human can be any motion of a human body: mumbling while asleep, scratching an itch, or idly tapping a foot. A human action is by definition an action in pursuit of a goal which is perceived as a good. A human action is an action such as running a race, starting a conversation, or tapping out a message in Morse Code. These actions are both intelligible and intentional in a way other animal actions are not. With this distinction in mind, we can state an important clarification. It is not the case that a human being without any practical reasons would perform immoral deeds; a human being without any practical reason would not do anything at all. He might move about, driven by instinct or fear or desire, but such a person would not engage in any human actions. Like Melville’s Bartleby the Scrivener (from the short story of that title), the person who does not engage in practical reasoning or identify any practical reasons would simply waste away and die.^[@melville1966bartleby]
Aquinas’ distinction between human actions and the actions of a human can also go to explain some compulsive or addictive behaviors. Unfortunate people in the overwhelming grip of, say, a heroine addiction, might be injecting themselves with the drug against their own evaluations of what should be done. However, in extreme cases of addiction, the action hardly falls under the description of a human action. Heroine is so highly addictive that one or two uses can create a dependency that lasts a lifetime. The addict’s initial decision to use the drug can still fall under the description of a human action, perhaps aimed at some perceived good such as pleasure or joining in a social group. But just as one is free to choose to slide into a muddy hole in the ground but not necessarily free to climb back out, one is free to use a habit-forming drug but not necessarily free to stop feeling the overwhelming compulsion, even though one might wish never to use again.
If all action aims at some good, then where does the process begin? How can one pursue ends before actualizing the natural ability to practically reason? We can again compare practical reasoning with demonstrative or theoretical reasoning. Aquinas puts the comparison this way:
…as “being” is the first thing that falls under the apprehension simply, so “good” is the first thing that falls under the apprehension of the practical reason, which is directed to action: since every agent acts for an end under the aspect of good. Consequently the first principle of practical reason is one founded on the notion of good, viz. that “good is that which all things seek after.” Hence this is the first precept of practical reason, that “good is to be done and pursued, and evil is to be avoided.^[@aquinas IIa. Q.94. Art. 2]
Aquinas points out that the first thing human beings apprehend as theoretical reasoners is simply “existence” or “being” – infants perceive that some things are there and others not there. They eventually come to perceive objects as objects, as individual objects, and to name and categorize them with language acquired in a social setting. Likewise, the first thing human beings apprehend as practical reasoners is simply the “good” or “desirable.” The use of ‘good’ here, it bears repeating, is not a special moral sense of good, but simply means ‘desirable’ or ‘to be pursued.’ An entity is ‘good’ when it is considered as an object of inclination. Hence, infants perceive that some things are to be pursued and others avoided. To be theoretically rational is to judge a proposition p as true or false, as best one can, according to the rational assessment of the reasons for affirming or denying p. Similarly, to be practically rational is to judge a practical reason φ to be pursued or avoided, in accord with the rational assessment of the reasons for pursuing or avoiding φ. Without a general principle in either case, practical reasoning and rational practice are unintelligible.
Given this basic and abstract formulation of the structure of practical reasoning, we can further specify good ends. Just as the basic structure of reasoning begins with the apprehension of being in general and grows to include apprehension of particular beings, concepts, and categories, practical reason begins with the apprehension of good in general and then determines particular goods.
Practical reason is the movement of thought towards, rather than away from, material particulars…. practical reasoning is a movement from general knowledge of what is good and how to live, towards the production of the kind of life that is essentially characterized by such knowledge. When it is done well, what is understood is the same as what is produced: human form or human life.^[@frey 2]
Such basic goods are apprehended as contributing to a distinctively human life form.
For practical reason, the starting points are the most primitive human goods that the will is naturally inclined to seek: life, knowledge, family, friendship, play, political community, and so on. These are the ends that all human beings want for their own sake, as intrinsically valuable to them. And they want these things in a rational way – viz., because they have a conceptual apprehension that they are constitutive of their general good.^[@frey 88]
Having said this, we should make two clarifications. First, I think Frey is asserting a generic truth when she says “these are the ends that all human beings want”; the truth admits of exceptions. Whatever the causes of psychopathy, some people seem insensitive to the obvious draw of natural ends. Such people don’t want knowledge, don’t care to have friends, don’t like to play, detach from their families, and in some cases show careless disregard for life – both their own and that of others. The important point is not that “primitive human goods” are pursued by all without exception – though indeed they are pursued by the vast statistical majority. The important point is that without some notion of primitive human goods, we could not identify disorders like psychopathy. Social behavior is not merely statistically normal but normative.
Secondly, some readers might object that this account equates “pursuing the good” with “pursuing the human good,” including such “primitive goods” as life, knowledge, friendship, and so on. Might there be goods that are good simpliciter that one ought to pursue, regardless of their bearing any internal relation to the human life form? Iris Murdoch argues along these lines that the starting point of ethical training must be aesthetic training.^[@murdoch1998sovereignty 90] One must cultivate the ability to see intrinsic value by first learning to see intrinsic beauty in art and nature, learning to appreciate it dispassionately. Just as the dispassionate pursuit of knowledge aims at knowledge of external realities (physical objects, animals, chemicals) and not just at knowledge of knowers, might not the pursuit of the good aim at external goods?
McDowell takes Murdoch’s thesis in a different direction. He argues that “the remoteness of the Form of the Good is a metaphorical version of the thesis that value is not in the world… The point of the metaphor is the colossal difficulty of attaining a capacity to cope clear-sightedly with the ethical reality which is part of our world.”^[@mcdowell1979virtue 347] For McDowell, this recognition of the difficulty of ethical training can benefit one “negatively, by inducing humility, and positively, by an inspiring effect akin to that of a religious conversion.” For McDowell, then, ethical (and aesthetic) training is not progress toward the discovery of objective value but toward the unfolding of one’s subjective or intersubjective values.
I am content to remain neutral with respect to these two options. Minimally, practical reasoning is the ability to judge the good of the human life form. This minimal ability is compatible with the paradoxical thought that what is good for humans is not merely the human good. What is good for humans might be the good simpliciter. It is needful, before examining this further issue, to defend the notion of basic, human goods. That is my aim here. Jennifer Frey summarizes:
…all practical reasoning is ultimately reasoning for the sake of attaining or maintaining these ends [i.e, basic human goods]. Consequently, all practical reasoning is ultimately for the sake of living the sort of life that pertains to man. Indeed for Aquinas, there could be no practical teleology without natural teleology, since there would be nothing to reason towards if the will were not by nature inclined towards the exemplification of human form.^[@frey 66]
To sum up the account thus far, all organisms incline toward the good of their life form, including those basic goods that enable the full actualization thereof. Various organisms express this inclination in various ways. For lower organisms, consciousness plays no part in this process; for higher organisms, consciousness does play a part. For humans, the essential difference is a sensitivity to the space of reasons, both evidential and practical. ‘Practical reasoning’ is the name for the whole complex process of perceiving certain salient facts as reasons to pursue or avoid some course of action, and comparing and ranking competing reasons in light of an overall conception of a good human life and acting accordingly. None of this is intended to deny that evaluative practical reasoning arises in a normal process of socialization. Rather, that our conception of how to live would arise that way is what we would predict for rational primates who speak and live in society.
I am now in a position to state and respond to three objections.
1. Procedural Reasoning: One challenge is the familiar notion that practical reasoning is a value-neutral procedure by which we line up means to our ends.^[For a discussion of this distinction, see: @hooker2004procedural.] On this view, moral reasoning is about the morally good and bad while practical reasoning is about something else entirely, such as the prudent or imprudent, the advisable or ill-advised. So how could an intellectual exercise be essential to moral virtue?
2. Reason, Practice, and Motivation: Another challenge comes from non-cognitivism (especially expressivism).5 The worry is that practical reasons by themselves can’t motivate us to act (without complementary psychological attitudes such as desires), while motivations to act cannot be rationally evaluated as true or false. Is practical reasoning really rational? And if so, is it really practical? It seems that it must be either one or the other.
3. Overriding Reasons: A third challenge is a familiar distinction between “moral reasons” on the one hand and “prudential reasons” on the other, where moral reasons are overriding reasons. On this distinction, one can be foolish by failing to act on some considerations, but one is immoral by failing to act on relevant overriding moral reasons. If practical reasoning is a process of identifying or inventing what is advisable or ill-advised (but not ultimately binding), then how does this process relate to an appropriate sensitivity to what is morally permissible or impermissible (which is ultimately binding)?
On Procedural Reasoning
According to the Procedural Reasoning objection, reasoning is not about ends but only about means. Practical reasoning is a procedural or instrumental process. The critic alleges that one may only criticize Smith as “irrational” when Smith fails to use the necessary means to her own ends, but one may not criticize Smith’s ends themselves as irrational. For example, if we define practical reasoning as the process by which one adjudicates the means to one’s own health, then any unhealthy action (e.g., eating delicious but less-than-healthy food) would be ipso facto irrational. Isn’t it problematic to build into the definition of rationality any specific, ready-made ends?
The first response to this challenge is that, even on the procedural view, practical reasoning must necessarily have a certain intelligible structure. The advocate of the procedural view, no less than the advocate of the substantive view, needs a sufficiently general starting point for procedural reasoning to even get off the ground. Frey’s candidate for that starting point is the maximally general conception that “good is to be done and evil avoided” or that “one must pursue one’s own good.” Her argument concluded that when practical reasoners act at all, they act by definition in pursuit of a particular object falling under a universal category. In order to construct any practical syllogisms as we do, one needs a sufficiently broad “major premise.”
A second response to Frey’s view is that it does not build in very specific ends. The built-in end is quite general: it is some conception of how to live in the way (or set of ways) that is good for practical, rational primates like us. This substantive good or set of goods is general enough to accommodate a variety of controversial details about what one ought to do or not do. In other words, the substantive view of practical reasoning allows for the possibility that, in a disagreement, both parties are basically rational, while one party may be more accurately identifying what is to be pursued or avoided.
Foot offers two additional considerations that support this Aristotelian account. When she wrote her famous “Morality as a System of Hypothetical Imperatives,” she argued that moral reasons are not overriding, categorical imperatives contrasted with every other kind of reason. She explains that, at the time, she had not discovered a way of showing “the rationality of acting, even against desire and self-interest, on the demand of morality.”^[@foot2001natural 63] What changed her mind was an argument from Warren Quinn to the effect that if practical reasoning is to be important at all it must be by definition the pursuit of some good. Quinn says:
Practical thought, like any other thought, requires a subject matter. And for human beings the subject matter that distinguishes thought as practical is, in the first instance, human ends and action insofar as they are good or bad in themselves… Practical thought deploys a master set of non-instrumental evaluative notions: that of a good or bad human act, a good or bad human life, a good or bad human agent, and a good or bad human action. Practical reason is, on this view, the faculty that applies these fundamental evaluative concepts.^[@quinn1993morality 223]
What Foot found so compelling is the change to “seeing goodness as setting a necessary condition of practical rationality and therefore as at least a part-determinant of [practical rationality] itself.” To one who objects, she points out that:
Many of us are willing to reject a ‘present desire’ theory of reasons for action because we think that someone who knowingly puts his future health at risk for a trivial pleasure is behaving foolishly, and therefore not well. Seeing his will as defective, we therefore say that he is doing what he has reason not to do. Being unable to fit the supposed ‘reason’ into some preconceived present-desire-based theory of reasons for action, we do not query whether it really is a foolish way to behave, but rather hang on to the evaluation and shape our theory of reasons accordingly. And it is exactly a generalization of this presumption about the direction of the argument on which I am now insisting. For what, we may ask, is so special about prudence that it alone among the virtues should be reasonably thought to relate to practical rationality in such a way?[@foot2001natural 63]
Quinn, Foot, and Frey are arguing that goodness is a “necessary condition of practical rationality.” Rational action is action in pursuit of some end, where “some end” is not merely an end (such as food, friendship, or knowledge) that is intrinsically desirable for practical rational primates like us. Identifying and pursuing such ends as desirable or undesirable is already a substantive evaluative judgment. Therefore, any rational action necessarily includes a substantive evaluative judgment.
If we accept this point, and I do not see how to avoid it, then we are already committed to a minimally substantive view of practical reason, rather than a merely procedural one. The alternative to aiming at the apparent good is not aiming at some value-neutral “end” or goal; the alternative to aiming at the apparent good is not acting at all.
While I summarized McDowell’s reply to the Humean critic above, I would like to return to the subject of motivation here. It will be useful to briefly situate my neo-Aristotelian account within the debate between motivational internalists and externalists, even if I cannot adequately engage the vast body of literature here.
In brief, the motivational internalist argues that any practical reasons “out there” must necessarily connect up with my motivational structure if they are to move me to action.^[@williams2007internal] The motivational externalist, by contrast, argues it is possible for there to be practical reasons “out there” such that I ought to be motivated by them, even if I am currently not. Indeed, the existence of binding practical reasons that I am ignoring or failing to act on is the prime explanation of immorality.
The danger of internalism is that it seems to allow that the amoralist who is not motivated to be moral is off the hook. By contrast, the externalist argues that the immoralist has reasons to φ even if she has no (current) motivation to φ.
On my view, motivational internalism gets this much right: one is motivated to pursue something that falls under a category that, within the existing motivational structure, one already judges to be desirable. However, the internalist too narrowly defines a “motivational structure.” If by “motivational structure” we mean my present set of broad psychological inclinations, then it is possible that we may not have the right motivational structure that would lead to moral action. But if by that we simply mean my overall practical disposition toward the worthwhile, desirable, and good, then it is quite uncontroversial to assert that one only goes in for φ-ing when φ-ing seems to be worthwhile, because to be a practical agent just means to be oriented to pursue the good and avoid the bad. Whatever may appear to me to fall under the description of ‘good’ I will, ipso facto, be oriented toward (whether I pursue it or merely approve of it and admire it). Whatever may appear to me to fall under the description ‘bad’ I will, ipso facto, oriented away from (whether I disapprove of it, or avoid it, or both).
What motivational externalism gets right is that there might be reasons to φ that I am not aware of and (hence) am not motivated by. For example, perhaps it is true that one ought to save for retirement, but I may fail to do so because I am unaware of that reason or am ignoring it in my attention to other reasons.
Seen in this light, it is obvious that on my neo-Aristotelian account practical reasons can and do motivate us. We can put the matter more strongly: according to Frey’s argument above, practical reasons are the primary meaning of the term ‘motive.’ Motivation is (I argue) a fundamentally rational state. It is true that sub-rational animals, plants, and insects are moved about by impulses such as hungers, thirsts, loves, fears, etc. And it is true that human animals are likewise moved about by such impulses. But for rational animals, there is an additional source of motion, namely practical reasons.
Hence, my contention is that our default view of practical reasoning creatures ought to be that practical reason is intrinsically capable of motivating. The whole process of discerning whether or not to φ is theoretical in much the same way that the process of discerning whether to believe that p, but it is also (by definition) practical. Practical reasoning is not something one does before resolving what to do, as one picks up an item in a store before purchasing it. Practical reasoning is the name we give to the process of resolving what to do, as checking out from the store is the process of purchasing. Just as the appraisal of overwhelming evidence for p is not utterly distinct from the affirmation that p, the deliberative conclusion that one ought to φ is not utterly distinct from the decision to φ. To borrow Gibbard’s unforgettable phrase, practical reasoning is “thinking how to live”.^[@gibbard2009thinking]
At the same time, there are goods we may not be pursuing (but ought to be) and evils we may not be avoiding (but ought to be). We acquire new motivations only when we successfully make new evaluative judgments about what is to be pursued and avoided. Our fundamental motivation is to pursue the good and avoid evil. We acquire new motivations when we come to identify and affirm new practical reasons. These practical reasons only motivate because they link up with the initial, existing motivation.
On Overriding Reasons
A final challenge that needs a response is this: moral reasons are sometimes treated as “overriding” or “verdictive” reasons that settle the question of what to do. Given the choice between, say, making a bit of easy money by fraud or making the same amount through honest but hard work, the prohibition against fraud is supposed to settle the matter. On my account, do prudential practical reasons weigh just as heavily as moral ones?
My answer is that the practical consideration that one ought never commit fraud is, in such a case, certainly overriding. However, sometimes prudential considerations are overriding, too. To take a different example, suppose Smith comes into a bit of money from an inheritance, and thus has a choice between spending it (innocently) on world travels or allocating it to a solid retirement plan. Even if Smith clearly needs more money in his retirement, the quandary ethicist would have no moral recommendation, because neither choice is obviously immoral in the sense that it violates one’s duties to others. The character ethicist would: the practically wise person takes the longer-term benefit of saving over the short-term benefit of traveling to be overriding.
A normal human life presents practical reasoners with many situations in which reasons pertaining to moral virtue (narrowly defined) play little or no part. One must be sensitive not only to such reasons but to the broad range of practical reasons. All practical reasons must be ranked and weighed before a final, verdictive reason emerges. Any reason to φ is a practical reason that can feature in an overall account of what to do. What Anscombe calls “the verdictive ought” is simply what Foot calls the thing to do “all things considered.”^[@foot2001natural 57] It often happens that one’s individual practical reasons conflict. McDowell is incorrect to persist in labeling the broader process of adjudicating such conflicts “virtue.” He ought to call it practical wisdom. The practically wise person is the one who coordinates all other virtues and executes them to good ends.
I should respond here to one final objection a reader might have. Has my neo-Aristotelian view of practical reason defined away the possibility of immorality? If everyone who acts is “aiming at the good,” doesn’t this exculpate an agent’s apparently immoral motives or ends? For example, someone might say, ‘It’s ridiculous to think that I always pursue the good because I sometimes do wrong.’ This objection misses the point. Of course practical reasoners sometimes do the wrong thing. The proper response is that we perceive the bad as overall worthwhile. If the immoral person acts wrongly, then he has misjudged the good. On the neo-Aristotelian view, immoral acts are rational mistakes.
But it remains true that if the immoral person acts at all then, according to the argument, he must by definition be pursuing some apparent good. To be practically rational necessarily means to pursue something as good, as desirable. Just as an epistemic agent might hold a false belief p without affirming the false as false, a practical agent might pursue a bad thing without pursuing it as bad. Rather, the immoral person fails in their practical reasoning to correctly rank and order specific goods. The imprudent person, for example, judges that it would be better to eat, drink, and be merry today rather than plan to avoid future ills. The cruel person judges that it would be better to cause suffering than to be kind.
Someone might say, “But sometimes I perceive the bad as bad and pursue it anyway.” My view is that we are sometimes able to include an end we know to be bad into an overall set of practical reasons, which we still judge is the thing to do, all things considered. One might judge, for example, that smoking cigarettes is bad and still start if one judges (with some conflict) that the potential pleasures overrule. (Having become addicted to nicotine, the smoker’s judgment that the potential pleasures are not worth it may not necessarily mean the smoker quits.)
I do not wish to suggest that identifying the thing to do is a smooth and easy project. It is no more or less difficult than the project of identifying what to believe is true.
The defendant of the procedural view is liable to point out that reasoning about ends is even messier than such theoretical reasoning. Indeed it is. But we must do it. People regularly argue, debate, and reason about ultimate ends. Suppose Smith says to Jones, “I’m concerned about you. You haven’t returned my calls. You lost your job, and you are not eating. What’s wrong?” It would be no consolation for Jones to respond, “Nothing’s wrong. I’m destitute, alone, and unhealthy, but that is what I am aiming for.” Smith would rightly judge that something had gone wrong for Jones to adopt such unhealthy and ridiculous aims.
Practical wisdom is the paramount virtue of practical rational animals. The upshot is that the foolish person – the habitually, incorrigibly foolish person responsible for his own folly – is, ipso facto, a bad practical rational primate. He is failing to do the thing to do.
What is good in this sense for human beings is specific to our species. The primary good of a kind for us is the human life form. The derivative goods for us are any and all things necessarily related to the human life form. In virtue of what we are, it is good for us to achieve humanity, to become fully human. We aim to become what we are. That is, we aim to become in actuality what we already are in potentiality. Some of these goods are basic human goods toward which we are naturally inclined: food, shelter, companionship, knowledge, etc. They are starting points without which human beings would not be motivated to do anything at all. \newline I should clarify that a thing’s status as a basic good is revisable. The normal process of practical reasoning about what to do can and sometimes does overrule the basic inclination toward a basic good in pursuit of some alternative good. The point is that this overruling judgment is not something over-and-above the practical pursuit toward the good but another expression of the same pursuit. For example, some people overrule their inclination toward the basic good of human companionship by becoming solitary monastics but they only do so in pursuit of other goods judged to be better. \newline Since practical reasoning is the process whereby we determine the “sort of life that pertains” to creatures like us, all particular practical reasonings about what to do in a given situation come to light as parts of this whole. This fits with the account of virtue defended above. There we saw that excellence in practical reasoning and rational practice aims at doing well with one’s whole life. In other words, every short-term choice fits into a context of long-term projects such as what career to pursue, whether or not to marry, what friendships to maintain, and so on. Furthermore, every long-term project fits into a broader context of one’s answer to the maximally general question “how should one live?”
This chapter has defended in detail a neo-Aristotelian conception of practical reasoning. It is an intrinsically normative and evaluative process that defines the life form of practical, rational primates. On my account, the structure of practical reasoning is akin to theoretical reasoning. Whereas theoretical reasoning is by definition a normative process in which the true is to be believed and the false to be disbelieved, practical reasoning is by definition a normative process in which good is to be pursued while evil is to be avoided. These “first principles” are known by all functioning human adults. And while particular rational inquiries aim at identifying good reasons to believe or disbelieve a claim, particular practical inquires are aimed at identifying basic goods intrinsic to human life. I argued that the procedural view of practical reasoning is itself committed to certain substantive normative judgments, including the judgment that one ought to do whatever will bring about one’s chosen ends. More to the point, I argued that the substantive view of practical reasoning is more plausible: we reason about apparent goods and bads and act accordingly. Nevertheless, my account leaves room for the commonsense insight that success in practical reasoning (like theoretical reasoning) is by no means guaranteed.
Success in identifying how to live and what to do requires a complex process of adjudicating between all the available goods known to one, sorting them, ranking them with care and wisdom, and forming them into a complete life plan. The virtuous person knows what to do. Hence, contra McDowell, practical wisdom is a kind of practical knowledge (a “disposition to act well”) while other virtues (as discussed in chapter 4) can include rational practices and habits formed over time that conduce to the realization of one’s life form. When practical reasoning is well-functioning, it constitutes part of the natural excellence of creatures like us. The vicious person is hindered by practical error — or perhaps ignorance — of what to do. Success or failure along these lines has a major influence on one’s other character traits. So practical wisdom is an essential part of living a fully human life.
How should one live? A virtuous person’s answer to this question is not just a proposition but a plan. More so, the virtuous person does not simply have a good plan but enacts a good life plan. The answer is not just a philosophy but a life.
Compare with @cuneo2014speech. ↩
“It might be thought that ordinary conceptions of rationality are Platonistic or intuitionistic. On the Platonistic picture, among the facts of the world are facts of what is rational and what is not. A person of normal mental powers can discern these facts. Judgments of rationality are thus straightforward apprehensions of fact, not through sense perception but through a mental faculty analogous to sense perception. When a person claims authority to pronounce on what is rational, he must base his claim on this power of apprehension.” See @gibbard1992wise 154. ↩
We all exhibit various dispositions to act in certain ways, to rank and organize our various motivations, to pursue certain things, or to make certain decisions rather than others. Such dispositions are clearly practical. They have the right kind of action-guiding force to explain why we act the way we do. On the other hand, there are dispositions. The term ‘disposition’ gets used in various ways: one can be disposed (say) to repay one’s debts (a moral commitment), or disposed to shout when angry (a temperament), or disposed to travel abroad every summer (an interest). But is a “disposition” a form of knowledge? ↩
Foot says: “It is my opinion that the Summa Theologica is one of the best sources we have for moral philosophy, and moreover that St. Thomas’s ethical writings are as useful to the atheist as to the Catholic or other Christian believer.” (Virtues and Vices, 2.) ↩
Non-cognitivism is motivated by metaphysical naturalism, which objects to normative realism about practical reasons. I shall address that objection in chapter 6. ↩