Squeezing the good into the right: the connection between virtue and reason

Integrando o bom ao certo: a conexão entre virtude e razão

ABSTRACT

Some of our reasons for acting are grounded precisely on the fact that we are not fully virtuous agents. This shows that the intuitive view that what we should do is what a virtuous person would do in the circumstances we find ourselves is false. Many take this to show that there is not a close connection between virtue and practical reasons. I hold that this is a mistake. I argue that a failure to act in light of a decisive reason always amounts to a flaw from the standpoint of virtue. This is the case even when it comes to reasons grounded in defects of character. The upshot is that our conception of virtue constraints the normative judgments we can accept. That is the case because every reason must be grounded in a value that is compatible with virtue. I conclude that even though we should not always act as a virtuous person, we should always act in response to values that a virtuous person could uphold.

Keywords:
virtue; normative reasons; value; dispositions; practical reasoning

RESUMO

Algumas de nossas razões para agir baseiam-se precisamente no fato de que não somos agentes completamente virtuosos. Isso revela que a visão intuitiva de que devemos fazer o que uma pessoa virtuosa faria nas circunstâncias em que nos encontramos é incorreta. Muitos interpretam isso como prova de que não existe uma conexão próxima entre virtude e razões práticas. Eu, porém, sustento que essa interpretação é equivocada. Defendo que falhar em agir conforme uma razão decisiva equivale sempre a uma falha do ponto de vista da virtude - mesmo quando a razão se baseia em defeitos de caráter. O resultado é que nossa concepção de virtude restringe os julgamentos normativos que podemos aceitar. Isso ocorre porque toda razão deve estar fundamentada em um valor compatível com a virtude. Concluo, portanto, que, embora nem sempre devamos agir exatamente como uma pessoa virtuosa agiria, precisamos sempre responder a valores que uma pessoa virtuosa poderia defender.

Palavras-chaves:
virtude; razões normativas; valor; disposições; raciocínio prático

1 Introduction

It may seem obvious that there is a close connection between virtue and practical reasons. If we hold that what one should do is what there is most reason to do and that one should act as a virtuous person would act, we may come to the conclusion that what we have most reason to do is that which a virtuous person would do if she were in our place.

Bernard Williams has shown this idea to be deeply flawed. When considering what we should do, one thing we should take into account is the ways in which we fall short of virtue. For instance, a virtuous agent would exhibit temperance. But if I know I lack temperance and may be unable to control my appetites in certain circumstances, I may have reason to avoid putting myself in those circumstance. And that is not something a virtuous agent would do or have reason to do (Williams, 1995, p.190). Furthermore, we cannot avoid the problem by including the lack of temperance in the specification of the circumstances in which I act, for then a virtuous agent would never find herself in those same circumstances. As Williams puts it:

It will be no good trying to accommodate this difficulty, of squeezing the good into the right through the tubes of imperfection, by putting all [of the agent’s] limitations in the account of the circumstances. If the circumstances are defined partly in terms of the agent’s ethical imperfection, then the phronimos cannot be in those circumstances. ( Williams 1995 , p.190)

Other examples of the same kind are easy to produce. We all have reasons to better ourselves, but only because we are flawed. A perfectly virtuous person would not work to better herself and would have no reason to do so simply because she has no defects of character on which to work. Or consider Watson’s (1975, p.210) description of an agent that just lost a friendly squash game to an irritating opponent. A virtuous person would display sportsmanship and politeness by calmly shaking the opponent’s hand. Watson’s squash player, however, is hotheaded and if he attempted to greet his opponent, he would be unable to control his anger and would end up beating him with the racquet. On account of his hotheadedness, therefore, he has a reason to walk away without greeting his opponent - something the virtuous agent would not do and has no reason to do.

What these cases show is that reasons can be grounded in defects of character. And if that is the case, the claim that a consideration provides us with a reason only if a virtuous person would decide to act in light of that consideration cannot be true. We cannot account for reasons grounded in defects of character in terms of the practical reasoning of a perfectly virtuous agent that finds herself in the same circumstances, because a perfectly virtuous agent would never be in those circumstances.

While Williams’ argument succeeds in showing that we cannot connect virtue and practical reasons in this straightforward manner, it does not show that there is no interesting connection between virtue and practical reasons.

In section 4.1, I argue that there is indeed a close relation between virtue and practical reasons. I hold that if we have decisive reason to perform a certain action and fail to do so, that entails we lack in character to some extent. It follows that a perfectly virtuous agent always acts according to her reasons. That is, if an agent has a decisive reason to perform an action, and that agent is perfectly virtuous, she will perform that action. This entails that our conception of virtue and our view about what we have reason to do must be aligned. The failure to act in light of a decisive reason must be a flaw from the standpoint of virtue. In this sense, an agent’s conception of virtue constrains the normative statements about reasons she can accept. We cannot ascribe a decisive reason to an agent unless we are prepared to hold that she lacks in character if she knowingly fails to respond to that reason.

This connection may not be enough to provide an analysis of the concept of reason in terms of virtue, but it is relevant nevertheless for it entails that one’s conception of virtue provides a test for reasons statements. That is the topic of section 4.2. If am I correct in holding that a failure to act in light of a reason must be a flaw from the standpoint of virtue, then we can rule out certain statements about reason by showing that failure to act in accordance with the putative reason would not be a flaw from the standpoint of virtue. For instance, we cannot accept the claim that someone has reason to be cruel to someone else unless we are prepared to accept that failing to act in that way would be a flaw from the standpoint of virtue. And if we hold a moral conception of virtue (according to which a perfectly virtuous agent is honest, considerate, just, altruistic, not cruel or selfish, and so on), that would be the case only in the extremely rare situations in which acting in a cruel manner would be the only way to uphold some other, more important moral value - so that acting in a cruel manner would not be immoral in the first place. In all other cases, failing to act in a cruel fashion would not be a flaw from the standpoint of virtue (quite the opposite, actually) and so our moral conception of virtue would rule out the claim that the agent has reason to be cruel to someone else.

The claim that we can test reason statements against our conception of virtue in this way is controversial. Some philosophers reject it. They hold that one can have reason to act in a way that is utterly incompatible with virtue, such that failing to act in that manner would not at all be a flaw from the standpoint of virtue (as they themselves conceive of it). That is a consequence, for instance, of both Bernard Williams’ internalism about reasons (1980) and Sharon Street’s humean constructivism (2010, 2012). If my arguments are correct, then they must be mistaken.

Williams’ point, however, remains: it is false that we should always act as a virtuous person, because we can have decisive reasons grounded in defects of character to act in ways a virtuous agent would never act. In what sense can normative judgments about reasons grounded in defects of character be aligned with our conception of virtue? In section 4.3, I argue that even reasons grounded in defects of character reflect values that are compatible with perfect virtue. This allow us to characterize the general connection between reasons and virtue: even though we should not always act as a virtuous person, we should always act in response to values that are compatible with virtue. We should not mimic the virtuous agent, but we should emulate her values.

My defense of this view is based on a number of connections between some of our most central normative concepts, namely, reasoning, reasons and value. So, before presenting my argument, I will articulate the relevant connections in sections 2 and 3.

2 Reasons and practical reasoning

Normative reasons are the reasons that there are for agents to act. These reasons are facts that count in favor of doing something.¹ If our reasons to act in a particular way are stronger than our reasons to act in any other possible way, then these reasons are decisive reasons and acting in that way is what we have most reason to do (Parfit, 2011, p.32). Often, we have sufficient reasons to act in two or more incompatible ways. In these cases, there is nothing we have a decisive reason to do. We have sufficient reason to do something when this reason is not outweighed by reasons to act in some other way. A sufficient reason to act is a good reason to act. Whenever we have a decisive reason to do something we also have a sufficient reason to do that and do not have sufficient reason to do anything else.

When we have decisive reason to do something, that is what we should do, at least in what Parfit calls the decisive-reason-implying sense of “should” (Parfit, 2011, p.33). The claims that we should do A and the claim that we have decisive reason or most reason to do A entail each other. If an agent has no decisive reason to act, but has sufficient reason to do A or B, and no sufficient reason to do something else, then she should do either A or B.

In what follows I will assume that practical reasoning is the kind of thinking in which an agent moves from some considerations to a decision to act in a particular way.² The considerations in light of which the agent decides to act are her motivating reasons. If these considerations count in favor of acting in that particular way, then they are also normative reasons for acting.

A piece of practical reasoning is bad, unsound reasoning when the agent’s motivating reasons do not give her sufficient normative reason for doing what she decided to do. A piece of practical reasoning is good, sound reasoning when the agent’s motivating reasons give her sufficient normative reason for acting as she decided to act.

With the notion of good practical reasoning in place, we can introduce the notion of good dispositions of practical thought.³ A disposition of practical thought is simply a disposition to engage in certain patterns of practical reasoning, that is, a disposition to make certain decisions in light of certain considerations when certain circumstances obtain. A good disposition of practical thought is a disposition to engage in good practical reasoning and, so, a disposition to move from what are actually sufficient normative reasons to act to a corresponding decision. A bad disposition of practical thought is a disposition to engage in bad practical reasoning, that is, a disposition to decide to act in light of considerations that do not provide a sufficient reason to act in that manner.

If an agent has a sufficient reason R to do A in circumstances C, then the piece of practical reasoning that moves from consideration R, perhaps combined with the consideration that circumstances C obtain,⁴ to the decision to do A is a good piece of practical reasoning. And the disposition to reason in that manner is a good disposition of practical thought. Given that the agent may have sufficient reason to act in other ways in the same circumstances, there may be other equally good or sound patterns of practical reasoning available in those circumstances and, consequently, other equally good dispositions of practical thought the agent could have exhibited in these circumstances.

If an agent has a decisive reason R to do A in circumstances C, then the piece of practical reasoning that moves from consideration R, perhaps combined with the consideration that circumstances C obtain, to the decision to do A is a good piece of practical reasoning. The disposition to reason in that manner is a good disposition of practical thought. Any disposition to reason in a different way in those circumstances is a bad disposition of practical thought, for it follows from the claim that the agent had decisive reason to act that she did not have a sufficient reason to act in any other way. So, if an agent has a decisive reason R to do A in circumstances C, she should exhibit the disposition to decide to do A in C in light of consideration R.

3 Reasons and value

An agent values something if, and only if, she is disposed to treat certain considerations as reasons to respond to that thing in a positive way (such as choosing, using, producing, preserving or admiring this thing).⁵ One treats a consideration as a reason when one decides to act in light of it, so that it is the motivating reason for which one acted. To value something, therefore, involves exhibiting certain dispositions of practical thought. How one responds to that which one values will be different in different cases. Sometimes to value something is to be disposed to decide to produce it or promote its existence. But not always. For instance, if one values friendship one will be disposed to be loyal to one’s friends, to care for them, and so on, but not to maximize the number of friends one has. A person that betrays one friend to make several new ones is not thereby shown to value friendship (Scanlon, 1998, p.89). To value the well-being of others is not necessarily to see well-being as a value to be maximized. Valuing the well-being of others may consist in seeing reason to guarantee a reasonable level of well-being to all, even if that means reducing the overall balance of well-being. And one who values the well-being of others may be completely indifferent to the small increases of well-being that may be accrued by satisfying the desires of those with expensive tastes. In the same vein, valuing something need not involve a disposition to respond to what one values in every opportunity one has, but only on suitable opportunities. For instance, an agent that values mastering a musical instrument is not one that is disposed to decide to practice whenever she has any free time, but one that is disposed to set up a more or less defined practice routine or schedule and to abide by that schedule in the absence of strong reasons to revise it.

One can value something one judges not be valuable. For instance, one can value the opinion of a person and be disposed to decide to act in ways that will please that person, while at the same time judging that one does not have a good reason to behave in that way. In that case, one is disposed to decide to act in light of a motivating reason one does not take to be a normative reason for acting in that way. So, the agent that values something she does not consider valuable judges the disposition of practical thought she exhibits to be a bad disposition of practical thought - one that manifests itself in unsound pieces of practical reasoning.

The claim that something is of value or is valuable commits one to the claim that the dispositions of practical thought displayed by an agent that values that thing are good dispositions of practical thought and that the corresponding pieces of practical reasoning are good. That is, it is to judge that there are good, normative reasons to respond in a positive way to the thing one deems valuable.

To say that something is valuable in this sense is not the same as saying that everyone should value it (pace Scanlon, 1998, p.95). The claim that something is valuable entails only that the disposition to decide to respond to it in some positive way is a good disposition of practical thought. That entails that one has sufficient reason to act in that manner, but not necessarily a decisive reason. If the agent has only sufficient reason, then she may have sufficient reason to act in other ways and, if that is the case, there are other good dispositions of practical thought she could exhibit. There are many valuable things one may not value without thereby failing to respond to a decisive reason. Achieving the physical excellence of an Olympic gymnast may be a valuable goal, but it is not the case that everyone should care about achieving that goal.

There are some values, however, that everyone should value. That is the case when the dispositions of practical thought exhibited by those who value it include dispositions to respond to decisive reasons. For instance, valuing human life is, in part, a matter of being disposed to respond to decisive reasons to preserve it and to not violate it in most circumstances. If human life is indeed valuable, then these are good dispositions of practical thought. Since the reasons at stake are decisive reasons, one should exhibit these dispositions. That is, one should value human life.

Values can conflict when the responses they call forth are incompatible. In that case, the normative reasons corresponding to the values will be in conflict. The reasons may be equally strong, in which case the agent has sufficient reason to act in response to either value. Or one of the conflicting normative reasons may outweigh the others, in which case the agent has a decisive reason to act in response to the corresponding value. We can say that the value in question trumps the other conflicting values or that it is more valuable.

If A is more valuable than B, then one should value A more than B. But, from the claim that an agent exhibits a good disposition of practical thought in valuing A more than B, it does not follow that A is more valuable than B. If an agent values A more than she values B, then, in the event of a conflict between A and B, she is disposed to act in response to A for the appropriate reasons and not to B (even though she is disposed to act in response to B for the appropriate reasons in a context in which there is no conflict). If that is a good disposition of practical thought, then either A is more valuable than B (in which case the agent has decisive reason to act in response to A and not B) or A is as valuable as B (in which case the agent has a sufficient reason to act in response either to A or B). So, to value something more than something else that is equally valuable is not necessarily a bad disposition of practical thought. The disposition that corresponds to valuing A more than B is a good disposition even if A and B are equally valuable. For if that is the case, the agent has sufficient reason to act in response to A and, therefore, the piece of reasoning that results in acting in response to A is good practical reasoning.

If values A and B conflict and A is more valuable than B but an agent is disposed to act in response to B and not to A, even though she also values A, then she values A less than she should. In this case, the agent values B more than A which is actually more valuable. The failure to value A to the extent that it is valuable is a bad disposition of practical thought.

4 The connection between virtue and reason

4.1 The normativity of virtue

In this section I argue that when an agent fails to respond to a reason while being aware of the fact that constitutes the reason, what explains that failure is the fact that the agent exhibits some defect of character or at least acted as someone who has that defect of character.

The basic premise of my argument is the claim that when an agent fails to act as she should, it should be possible to explain why she failed to respond to her reason and that explanation must point to an imperfection in the agent. The imperfection can consist in the agent ignoring some relevant fact that gives her reason to act. In this case she may be said to be ignorant. But if the agent failed to act as she should while being aware of the relevant facts, then she is not as she should be. The explanation of the agent’s failure to respond to her reason must show that there is something wrong with her - that she is, to some extent, flawed. It is that flaw that accounts for her failure to respond to her reasons. In other words, whenever an agent fails to respond to a decisive reason (while being aware of the relevant facts), she must expose herself to a personal criticism of some kind.⁶ We can formulate this premise as:

(Criticism) If an agent has decisive reason to act, is aware of the relevant facts, and fails to respond to that reason, then she exposes herself to some form of personal criticism that explains her failure to respond to that reason.

If she is aware that she has a decisive reason, but fails to respond anyway she is akratic. If she is aware of the facts that give her reason, but fails to acknowledge those reasons, she exposes herself to some other kind of personal criticism. But what kind?

The second premise in my argument is the claim that a perfectly virtuous agent is one that is irreproachable. The perfectly virtuous agent (whatever that turns out to be) is someone who does not exhibit the kind of imperfection that would expose her to a personal criticism that could explain a failure to respond to a normative reason. If someone does expose herself to that form of personal criticism, then there is some aspect of herself and her capacity for practical thought with regard to which she could improve. So, she is not perfectly virtuous.

Criticism entails that if an agent does not expose herself to any personal criticism, then she does not fail to respond to any decisive reason of hers. That is, as long as she is not ignorant, she always decides to act as she should act for the appropriate reasons. According to our second premise, a perfectly virtuous agent does not expose herself to any personal criticism. It follows that a non-ignorant perfectly virtuous agent always acts as she should act and does so for the right reasons. Having only good dispositions of practical thought is a necessary condition for perfect virtue.

So, if an agent should do something but fails to respond to the decisive reason at play (while being aware of the fact that gives her the reason), then she is not perfectly virtuous. That is, the agent lacks in character to some extent. That means that when a person fails to respond to a decisive reason (and does not ignore any relevant fact) she exposes herself to the kind of criticism that is directed to one’s character (assuming the claim that the agent is akratic also points to a lack in character). If one had a reason to be kind to someone else but failed to acknowledge that reason, then one is insensitive or perhaps cruel and that explains the failure to respond to the reason. If one had a decisive reason to help someone else but failed to acknowledge that reason, then one is callous or selfish. If one had a decisive reason to forgive, but failed to respond to it, one may be vindictive. And so on.

One could object that even truly virtuous persons may sometimes fail to respond to a decisive reason without thereby showing themselves not to be virtuous. They may simply act out of character. What we should say, then, is that whenever an agent fails to respond to a decisive reason, she thereby shows herself to lack in character or, at the very least, acts as if she lacked in character (which may very well be out of character for her). The main point remains: whenever one fails to acknowledge a decisive reason, one exposes oneself to a kind of criticism that is formulated in terms of defects of character.

Another possible objection is to hold that the notion of a perfectly virtuous agent is not coherent. One may hold, as Eylon (2014) does, that different virtues might prescribe different actions. Compassion and justice may pull us in different directions, for instance. If that is the case, then there may be situations in which it is impossible to manifest both virtues. If a perfectly virtuous agent is one that exhibits all the virtues, then that agent would be impossible as well. But that is the wrong way to characterize the perfectly virtuous agent. The perfectly virtuous agent is one that is irreproachable. Given Criticism, that entails that the perfectly virtuous agent responds perfectly well to her reasons. Now, when there is a conflict of virtues, there are two options. One possibility is that there is a correct answer about how one should act, in which case there is a decisive reason to which the perfectly virtuous agent will respond. If, say, in the event of a conflict between compassion and justice, one should act in a compassionate manner, that is what the perfectly virtuous agent will do. The perfectly virtuous agent not only has all the virtues but also exhibit them when and to the extent that they are appropriate. Another possibility is that in the event of a conflict of virtues there is not a single correct answer about what to do, so that there is sufficient reason to act in more than one way. If that is the case, what follows is that there is more than one way in which to be fully virtuous. If there is sufficient reason to be compassionate but also sufficient reason to rigorously enforce justice, then both the compassionate and the rigorous agent may be fully virtuous. Different agents, that exhibit different dispositions of practical thought may be equally irreproachable - there may be several ways of being perfectly virtuous.

This is an important point, worth emphasizing. Although I speak of the perfectly virtuous agent, I do no mean to imply that there is only one way in which to be an irreproachable agent. It is not the case that every perfectly virtuous agent always acts in exactly the same way and value the exact same things to the same extent. As I argue above, to value something more than something else that is equally valuable is not necessarily a bad disposition of practical thought. Perfectly virtuous agents may value different things to a different extent. One who values mastering the violin but does not value skiing can be just as virtuous as one who values skiing but does not care about violins. So, not every perfectly virtuous agent exhibits the same dispositions of practical thought. When there is sufficient reason to act in more than one way, each option is compatible with perfect virtue. And so, to each option corresponds a different perfectly virtuous agent. We can understand our ideal of virtue as a disjunctive set of descriptions of different and potentially incompatible ways of being an irreproachable agent. Leaving aside for now reasons that are grounded in defects of character, we have that if one has decisive reason to do A in circumstances C, every agent that conforms to the ideal of virtue does A in C. If one has sufficient reason to either A or B, then both the agent that does A and the agent that does B conform to the ideal of virtue and are perfectly virtuous, even though they have different disposition of practical thought and value different things.⁷

There would be a problem here (and this is Eylon’s main point) if we took the notion of the perfectly virtuous agent to be prior to the notion of normative reason. Then we would have to somehow determine how the perfectly virtuous agent would act in the event of a conflict without appealing to the notion of normative reason as I just did. I should emphasize then that while I am defending the claim that there is a necessary connection between judgments about normative reasons and judgments about virtue, I am not committed to the claim that one of the poles of this necessary connection has priority over the other. The relation here is one of reciprocity, not priority.

4.2 A negative test for normative judgments

Some reasons are grounded in the fact that the agent lacks in character. Watson’s angry squash player has a reason to walk away without shaking his opponent’s hand because he is an angry person. Someone who was not that angry and exhibited the virtue of sportsmanship would not have that reason. So, his reason is grounded on the fact that he lacks in character to some extent. This show that we cannot hold that the fact that an agent has a decisive reason to do A in circumstances C entails that that is what a perfectly virtuous agent would do in those circumstances. Nor can we hold that the fact that any perfectly virtuous agent would decide to do A in circumstances C for the reason R entails that that is what you should do.

Nevertheless, the fact that whenever one fails to act according to a decisive reason one is thereby shown to lack in character stands. If one has a decisive reason to act that is based on a defect of character, then it is good disposition of practical thought to decide to act in that manner in light of that fact that one lacks in character, perhaps combined with other pertinent considerations. In fact, if one has a decisive reason, then that is the only good disposition of practical thought in the circumstances one finds oneself. If the agent fails to exhibit that disposition, that is, if she fails to decide to act in that manner for the appropriate reasons, that reveals yet another way in which she lacks in character. If Watson’s squash player fails to respond to his reason to walk away, he is show to exhibit another defect of character in addition to his irascibility. What flaw he shows will depend on how he fails to respond to the reason. If he acknowledges the reason but moved by anger he fails to respond, he is akratic and lacks self-control. If he fails to acknowledge the reason, he may be a violent or perhaps an impetuous person, or perhaps he lacks self-awareness to a reproachable extent. These flaws explain why he failed to acknowledge the reason.

Acting accordingly to one’s decisive reasons, in a way that displays a good disposition of practical reasoning, always shows one to be closer to virtue than one would be if one failed to so act. We cannot, therefore, claim that an agent has a decisive reason to act in a particular way if we are not willing to hold that she lacks in character to some extent if she fails to act in accordance with that reason. In this way, normative statements about our reasons restrict the space of possible virtue. This is clearer in the case of reasons that are not predicated in any in defect of character, but it is also true of reasons that are predicated in defects of character. For lack of a better term, let us call the former perfect reasons and the latter imperfect reasons.⁸

Suppose that I hold that, in circumstances C, I have a perfect decisive reason R to do A, say, help a person in need. That commits me to the claim that if I were perfectly virtuous, I would do A in C in light of R because (i) R is a reason I would have even if I did not lack in character at all (for it is not predicated on any defect of character) and (ii) R being a decisive reason, I would lack in character if I failed to respond. So, assuming that the (disjunctive) ideal of virtue that applies to me applies to others as well, the claim that I have a perfect decisive reason to do A in C (which may, of course, include information about my preferences, substantive desires, personal relations, etc.) entails that any perfectly virtuous agent that found herself in the same circumstances would do A. If I were to hold that other virtuous agent could act differently in the same circumstances, I would commit myself to the claim that there are other good dispositions of practical thought that could be exhibited in those circumstances. And that would entail that R was at best a sufficient but not decisive reason to do A.

So, assuming that the ideal of virtue that applies to me is the same that applies to others, in holding that I have a perfect decisive reason act in a particular way, I commit myself to the claim that any perfectly virtuous agent would act in the same way. And in holding that I have perfect sufficient reasons to do either A or B, and no other sufficient reason, I commit myself to the claim that any perfectly virtuous agent would do either A or B. In this sense, normative statements about what we have perfect reason to do restrict the space of possible virtue. They commit us to the claim that perfectly virtuous agent would act in a particular way, thus restricting the realm of possible ways in which virtuous agents can reason practically and act. Another way of making the same point is saying that every statement about perfect decisive or sufficient reasons paints a partial image of virtue.

That means that our ideal of virtue and our normative judgments about perfect reasons must be aligned. Our normative judgments about perfect reasons must be in accordance with the possibilities our ideal of virtue acknowledges. Therefore, our ideal of virtue imposes a limit to the normative judgments we can accept. If we have an independent grasp on our ideal of virtue, we can test normative judgments against it. If a normative judgment restricts the space of possible virtue too much, excluding modes of behavior or practical reasoning we consider compatible with virtue, then (assuming we do not want to revise our ideal of virtue) we should not accept that judgment. The same is true if a normative judgment does not restrict the space of possible virtue enough, admitting in that space modes of behavior or practical reasoning that are actually incompatible with our ideal of virtue.

Taking our moral perspective for granted for the time being, assume that a perfectly virtuous agent is one that is altruistic, just, kind, considerate, honest, etc. Any acceptable judgments about perfect decisive or sufficient reasons will have to be compatible with this ideal of virtue. One cannot have a perfect decisive reason to be cruel in order to obtain a personal gain, for instance, for that would entail that all perfectly virtuous agents would act in that way. That is too restrictive, for it excludes from the space of possible reason compassionate and kind agents, and too permissive, for it includes in the space of possible reason cruel agents. One also cannot have a sufficient reason to be cruel for profit. That would entail that at least some possible perfectly virtuous agents eventually act in a cruel manner for profit - which is in disagreement with our ideal of virtue.

If we follow Scanlon (1998, p.53) in understanding maxims to be principles that specify which considerations provide adequate, decisive or sufficient reason to act in a particular way in certain circumstances (or, which is the same, which patterns of practical reasoning are good patterns), we can say our ideal of virtue provides us with a negative test for maxims - a test that allows us to declare some maxims unacceptable. Any acceptable maxim must be compatible with our ideal of virtue. Any acceptable maxim must capture a good pattern of practical reasoning, one that correspond to a good disposition of practical thought that a perfectly virtuous agent could exhibit.

4.3 Reasons grounded on defects of character

Normative judgments about imperfect reasons are also constrained by our ideal of virtue, but not in the same straightforward way. Consider an agent that is terribly corrupted, callous and cruel. If the argument of the previous section is correct, then (assuming our moral ideal of virtue) this agent cannot have a perfect reason to act in a cruel manner for personal gain. Could she have an imperfect reason to do so? The answer is no. Cases in which an agent has an imperfect reason come in two forms: (i) the agent has a reason to behave in a way in which a perfectly virtuous agent would not behave because she finds herself in a situation in which a perfectly virtuous agent would never be and (ii) given the agent’s defects of character, trying to act as a perfectly virtuous agent would backfire resulting in an action that would violate an important value. In both cases, acting in light of the imperfect reason is a manner of responding in the appropriate way to a value a perfectly virtuous agent could share.

If one has an imperfect reason R to do A, either decisive or sufficient, then deciding to do A in light of R is a good disposition of practical thought. If an agent exhibits a disposition of practical thought, it follows that she values something (or at least, is acting as someone who values that) and if the disposition is a good one, then that which she values is valuable. So, just as is the case with perfect reasons, in responding to an imperfect reason, the agent is responding to a value. Consider Watson’s squash player. If he were disposed to decide to walk away in light of the consideration that if he were to approach his opponent he would beat him with the racquet, then it would follow that he values, say, his opponent’s physical integrity (and values it more than he values abiding by the social conventions of the sport). If he does have a decisive reason to walk way, then that is a good disposition of practical thought. It follows that his opponent’s physical integrity is indeed valuable (and more valuable than the social conventions he violates by walking away). It is only because trying to act as a perfectly virtuous agent would result in a violation of that value that the squash player has an imperfect reason to act in a different way.

So, just like a disposition to respond to perfect reasons, a disposition to respond to an imperfect reason is a disposition to respond in the appropriate manner to some value. What makes it the case that an agent has an imperfect reason to act is not the fact that the landscape of values is different for a perfectly virtuous agent and an imperfect one, but rather the fact that the possibilities of action are different for the perfect and the imperfect agent. If a perfectly virtuous agent was in the same situation as the squash player (minus the anger issue), a course of action that respected both the value of the conventions of sportsmanship and the value of the opponent’s physical integrity would be open to her. The same course of action is not available to Watson’s squash player. And that is what explains why he has an imperfect reason to act some other way.

If imperfect reasons correspond to values in this manner, then they are also constrained by our idea of virtue. Something has value only if the dispositions of practical thought exhibited by someone who values it are good dispositions of practical thought. In other words, something has value only if and to the extent that valuing it is compatible with perfect virtue. The values that ground imperfect reasons must also be compatible with virtue. They must be values that at least some conceivable perfectly virtuous agents uphold. If the squash player has a decisive reason to walk away that is because the value of his opponent’s physical integrity outweighs the value of the conventions of sportsmanship. A perfectly virtuous agent would, therefore, value the opponent’s physical integrity over sportsmanship and acknowledge that the squash player should, in light of these values, walk away.⁹

Returning to our initial question whether one could have an imperfect decisive reason to act in a cruel manner in order to obtain some personal advantage, we can now see why the answer is no. If one had that reason, the corresponding disposition of practical thought would be good. Someone who exhibit it would value her own interests more than the well-being of others. It would follow that her own interest are indeed more valuable than the well-being of others. But these are not the values of a virtuous agent. No perfectly virtuous agent would exhibit that disposition of practical thought. So, one cannot have that reason.

5 Conclusion

I have argued that if every conceivable perfectly virtuous agent exhibits a disposition to decide to do A in circumstances C in light of consideration R, it follows that for any person P, R is a decisive reason for P to do A in C, unless P has a sufficient imperfect reason to act differently. And I have also argued that imperfect reasons are grounded on values that a perfectly virtuous agent could value. If this is correct, all our reasons have their source on values. Something has value only if valuing it is compatible with virtue. So, if a particular disposition of practical thought is a good disposition (as is the case when one has at least sufficient reason to act), then one who exhibits that disposition acts in response to values that are compatible with virtue. Virtue is normative for us in the sense that we should (in the decisive-reason-implying sense of “should”) always act in response to values that are compatible with virtue. This does not mean that we should always act as a perfectly virtuous person would, but that when we should act differently, the reason there is for us to act reflects values a virtuous person could have (even though in our current circumstances, responding to these values require us to act in ways a perfectly virtuous agent would not act). We should not mimic the virtuous agent, but we should emulate her values.

References

• ALVAREZ, M. 2016. Reasons for action, acting for reasons, and rationality. Synthese, 195(8): p. 3293-3310.
• DANCY, J. 2004. Ethics Without Principles Oxford: Oxford University Press.
• EYLON, Y. 2014. The virtuous person and normativity. In: VAN HOOFT, S. (ed). The Handbook of virtue ethics Bristol: Acumen.
• HIERONYMI, P. 2009. The will as reason. Nous - Supplement: Philosophical Perspectives, 23(1): p. 201-220.
• PARFIT, D. 2011. On What Matters: Volume 1. New York: Oxford University Press.
• SCANLON, T. M. 1998. What We Owe to Each Other Cambridge: Harvard University Press.
• SETIYA, K. 2007. Reasons Without Rationalism Princeton: Princeton University Press.
• SMITH, M. 1995. Internal Reasons. Philosophy and Phenomenological Research, 55(1): p. 109.
• STREET, S. 2010. What is Constructivism in Ethics and Metaethics? Philosophy Compass, 5(5): p. 363-384.
• STREET, S. 2012. Coming to Terms with Contingency: Humean Constructivism about Practical Reason. In: LENMAN, J., & SHEMMER, Y. (eds.). Constructivism in Practical Philosophy Oxford: Oxford University Press .
• WATSON, G. 1975. Free agency. Journal of Philosophy, (72): p.205-220.
• WAY, J. 2017. Reasons as Premises of Good Reasoning. Pacific Philosophical Quarterly, 98(2): p. 251-270.
• WILLIAMS, B. 1981. Internal and External Reasons. In: WILLIAMS, B. Moral Luc k. Cambridge: Cambridge University Press.
• WILLIAMS, B. 1995. Replies. In: Altham, J. and Harrison, R. (eds). World, Mind and Ethics: Essays on the Ethical Philosophy of Bernard Williams Cambridge: Cambridge University Press , p. 185-224.

• 1
  This view is widely accepted. So much so that Alvarez (2016) describes it as near consensus in philosophy of action.
• 2
  See Hieronymi (2009) for a defense of this view.
• 3
  See Setiya (2007) and Way (2017) for a discussion of dispositions of practical thought.
• 4
  Following Dancy (2004, p.39) we can distinguish between considerations that favor a decision and considerations that enable another consideration to favor a decision. If I have promised to do A that consideration counts in favor of doing A and is a normative reason for doing A. The consideration that my promise was not given under duress is not a further reason in favor of doing A, but an enabler - a consideration that enables the consideration that I promised to do A to count in favor of doing A. Most of the considerations pertaining to the characterization of circumstances C will count as enablers, and not reasons, in this sense. If that is the case, then not all the true premises of sound practical reasoning are normative reasons. But it is still the case that if a consideration provides sufficient reason to act in particular way, then it is a premise in a good piece of practical reasoning. If one has sufficient reason R to do A, then the piece of reasoning that moves from R and form the relevant enabling considerations, to a decision to do A, surely is a good piece of practical reasoning. I said that perhaps good practical reasoning incorporates enabling considerations as premises in order to leave open the possibility that in some occasions it does not. It seems that in some circumstances, moving directly from a sufficient reason R to the decision to A, without taking notice of the enabling conditions, may be good reasoning (as long as the enabling conditions do obtain). That seems to be the case in circumstances in which one has to act fast, and it is not possible to check whether all enabling conditions are satisfied.
• 5
  This account of valuing is put forward by a number of philosophers. See Scanlon (1998, p.95) and Street (2010).
• 6
  The kind of criticism at stake here is criticism pertaining to the agent’s capacity for practical reasoning - her capacity to make decisions and to respond to reasons. Criticism pertaining to other aspects of the agent’s personality (as say, the claim that the agent is not funny or has an annoying voice) are irrelevant.
• 7
  Therefore, even if we ignore reasons grounded on defects of character, from the claim that a perfectly virtuous agent would do A in C it does not follow that we all have decisive reason to do A in C, only that we have sufficient reason to do A if we find ourselves in C (which may include information about our preferences, substantive desires, personal relations, etc.). It only follows that we have decisive reason to do A in C if we assume the any perfectly virtuous agent would do A in C.
• 8
  This is not the distinction between perfect and imperfect duties. One can have a perfect reason to discharge a perfect duty - such as the reason no to kill. One can have a perfect reason to discharge an imperfect duty - such as a reason to perform a benevolent act. One can have imperfect reason to discharge an imperfect duty - as the reason to improve one’s own character. And one can have an imperfect reason to discharge a perfect duty - that is the case of Watson’s angry squash player.
• 9
  This view may seem similar to Smith’s adviser account of reasons (Smith, 1995). But that is not the case. According to Smith, what we have reason to do is what we would advise our actual self to do if we were fully informed and perfectly procedurally rational. What I hold is that one should always act in respond to values that a perfectly virtuous agent could value. The perfectly virtuous agent in question is not an idealized version of myself, but any agent that satisfies the ideal of virtue and is, therefore, irreproachable. Something is valuable not because our idealized selves would value it, but because valuing (and valuing to a certain extent) is compatible with the ideal of virtue.

Edited by

• Nome dos editores responsáveis pela avaliação:
  Inácio Helfer
  Luís Miguel Rechiki Meirelles

Publication Dates

• Publication in this collection
  13 Jan 2025
• Date of issue
  2024

History

• Received
  08 Sept 2023
• Accepted
  27 Aug 2024