Irwin, Terence H. (2000). Ethics as an inexact science: Aristotle's ambitions for moral theory. In Brad Hooker & Margaret Olivia Little (eds.), Moral particularism. New York: Oxford University Press. pp. 100-29.
4. Ethics and Variation
When Aristotle claims that ethics provides usual generalizations, does he primarily have in mind mere frequencies or natural norms? Does he recognize ethical principles parallel to the normative regularities of teleological biology?
Let us consider the example of wealth. Aristotle recognizes that wealth is not always, all things considered, good for everyone in every situation. But it is usually good, since it is good for those who use it correctly in normally favourable circumstances. Aristotle could hardly believe that wealth is usu ally good, however, if 'usually' were taken to imply 'more often than not. For it is not at all clear that most people are good enough, or that most cir cumstances are favourable enough, to ensure that wealth is more often beneficial, all things considered, than harmful to people who have it. This passage suggests that when Aristotle speaks of the usual, he really has in mind the normal, which may not be statistically the most frequent. If we are to fit his observation about the harms resulting from wealth into his general views about wealth, we must appeal to those claims that rely on assumptions about normality.
A similar point holds for bravery. In favourable conditions, bravery helps us to preserve and enjoy external goods. Brave people who have rea sonable luck do better in preserving their lives and cities than cowardly people do, but the relevant conjunction of moderately favourable circum stances cannot be guaranteed. Similarly, part of the reason for valuing jus tice is the fact that in a just community we are respected for our justice, justice leads to greater harmony between different groups in society, and so on; but we cannot count on these favourable conditions. The connections between virtue and external goods ensure that virtue secures happiness in normal conditions; but the example that Aristotle gives here refutes the unrestricted claim that virtue ensures happiness.
If the claim that virtue usually secures happiness refers to frequency, it implies that favourable circumstances and outcomes are more often encountered than unfavourable ones. Aristotle has no reason to believe this, and it is not especially relevant to his theory. If favourable circum stances are in our power to achieve, the fact that they cannot be taken for granted in most cases does not matter. Though admittedly Aristotle's the ory cannot work if it requires too many things to happen contrary to the way they most often do, the specific claims we are concerned with do not need to be true more often than not. In present circumstances, virtue may not secure happiness more often than not. Still, we have good reason to choose virtue if in normal circumstances it achieves happiness, and normal circumstances are within our power to produce.
In all three examples (about justice, wealth, and bravery), the usual gen eralizations that Aristotle alludes to are most plausibly taken to describe the normal situation rather than the most frequent situation. To see the con nexion between these claims about ethics and Aristotle's claims about the usual in nature, we must remember the connection between the usual and the natural. Aristotle believes that external goods are both good without qualification and good by nature (EE 1248b26-37). His conviction that they are good by nature does not depend on the conviction that most actual people in most actual circumstances actually benefit from them. It depends on his general theory about what is good for a human being.' 7 In ethics, as in the study of animals, the usual regularities are important because they describe natural norms, not because they describe frequencies.
If, then, we examine these specific examples, we can see what kinds of usual regularities Aristotle primarily has in mind in his remarks about the inexactness of ethics. He is primarily thinking of norms rather than fre quencies. He does not explicitly distinguish these two types of usual regu larities in his different claims about the usual. But we must notice the distinction; for we should attend especially to the consequences of treating ethical principles as statements of natural norms.
5. Usual Principles and Ethical Science
Why does Aristotle think it matters that ethical generalizations are usual? He might give either of two answers: (1) He emphasizes this to warn us that we ought not to take ethical generalizations too seriously. They cannot claim the status of scientific principles, but should simply be regarded as summaries of experience. (2) He emphasizes this to warn us that we ought not to be deterred from taking ethical generalizations seriously; for they are still scientific principles, even though they have exceptions.
To decide between these answers, we ought to remember, first of all, that Aristotle cannot take all ethical generalizations to be merely usual.18 It is easy to compile a list of generalizations that do not seem to have exceptions. Happiness is everyone's ultimate good; everyone's happiness consists in activity of the soul in accordance with complete virtue in a complete life; it is always better to aim at the mean than to aim at either the excess or the deficiency; it is always better to be brave than to be cowardly; it is right to care more about fine action for its own sake than about honour.
These claims may seem too schematic to be practically useful; but Aristotle is also committed to less schematic generalizations with no excep tions. He believes, for instance, that one ought always to be willing to face great danger if some important cause is at stake, and one ought never to be willing to face it for some trivial reason. He believes it is always bad to cul tivate the fearless attitudes of people who do not stop to think about dan gers, or of people who care so little about living that they do not care about dying (1115b24-9, 1117b17-20, 1124b6-9). He believes that we ought never to make fun of people simply to raise a laugh, without any regard for what is fine or expedient (1126b36-1127a6, 1128a4-7). These principles have some practical content, since they clearly prohibit the attitudes of some readily recognizable types of people.
If Aristotle is committed to many practically significant unqualified gen eralizations, he cannot consistently take his claims about the inexactness of ethics to rule them out. His claim that ethics states usual principles cannot mean that ethics states only usual principles; it must mean that ethics includes not only unqualified principles, but also usual principles.
If Aristotle believes that usual principles are not scientific, he must say that ethics includes both unqualified generalizations that are suitable for science, and usual generalizations that are unsuitable for science. If, how ever, he believes that some usual generalizations are scientific, he claims that the scientific part of ethics includes both unqualified and usual gener alizations. When he emphasizes that the science of nature includes usual principles, he means to affirm, not to deny, the scientific status of the study of nature. Why should we not say the same about ethics? We would have reason to hesitate if we believed that the usual generalizations available in ethics do not embody normal and natural regularities; but we have seen that these are precisely the regularities that Aristotle has in mind in the spe cific examples he gives.
I have emphasized these points because it is easy to suppose that Aristotle's remarks about ethical generalizations imply a sharp division between the epistemic status of ethical principles and that of the principles of theoretical sciences. This supposition is quite mistaken. The fact that ethics relies on usual regularities does not distinguish ethics from natural science, as Aristotle conceives it; and so we have no reason to suppose that he takes ethical generalizations any less seriously than he takes physical generalizations.
We can strengthen this conclusion once we recall that Aristotle normally takes the principles of a discipline to be 'better known by nature' (gnôrimôteron phusei) than the initial beliefs that provide the starting points for inquiry; these initial beliefs are better known 'to us, but not bet ter known by nature. The relevant aspect of this contrast, for our present purposes, is Aristotle's claim that what is better known by nature is also more universal, prior, and more explanatory (APo (Posterior Analytics), 71b29-72a7). If Aristotle in the Ethics is looking for principles that are bet ter known by nature than the starting points, he is looking for principles that are more universal than, prior to, and explanatory of the starting points.
The Ethics makes it clear from the beginning that Aristotle is looking for principles. He remarks that the road towards the principles is different from the road from the principles, and he insists that when we are on the road towards the principles, we must start from things that are better known to us. Aristotle contrasts, as he often does, what is known to us with what is known 'without qualification' (1095a36—b4), and hence known by nature.19 He implies that we are on the road towards principles (1095a30—b4).
Later, he claims to have found a principle by giving an account of hap piness. He accompanies this claim with a warning not to demand the same degree of exactness in all disciplines (1098a20—b8). This warning is not intended to qualify the claim to have found a first principle. On the contrary, it is intended to forestall an inappropriate objection to that claim, by showing that an inexact statement of a first principle is appropriate for the discipline whose principles we are seeking.
The beginning of the Ethics, then, gives us good reason to believe that Aristotle takes some ethical generalizations, including some usual ones, to be principles, and therefore to be prior in the order of explanation, justifi cation, and knowledge to the ethical beliefs that they explain. This conclu sion counts against the view that Aristotle's belief in the usual character of some ethical generalizations reflects any commitment to particularism.
6. Why are Ethical Principles Usual?
Aristotle does not suggest that the presence of usual principles in ethics simply reflects our ignorance or the incompleteness of our theory. He sug gests they are ineliminable, just as he assumes that they are ineliminable in natural science. Why is this?
In natural science we have a choice between two explanations: (1) These usual regularities cannot be eliminated because the behaviour of matter is essentially indeterminate, so that exceptions to teleological generalizations cannot be exhaustively specified. (2) Aristotle's preference for usual regular ities indicates his belief in the importance of teleological regularities and the unimportance (for these particular purposes, though not necessarily for all purposes) of exceptions to them, even if we can specify all the exceptions.
The first explanation asserts that Aristotle affirms merely usual prin ciples only because he believes that it is impossible to specify all the excep tions to teleological generalizations. The second explanation, by contrast, neither affirms nor denies that he holds this belief. It says that his holding it is not necessary for his affirmation of usual principles.
The two parallel explanations in ethics are these: (1) We cannot, even in theory, find all the qualifications that would be needed to formulate the appropriately qualified principles. (2) It is unwise, for practical purposes, to try to build all the qualifications into our principles, even if it is possible to build them in.
In natural science, the second answer is preferable. When Aristotle draws attention to teleological regularities in nature, it might simply be distract ing to try to list all the qualifications needed to make the generalizations accurate. His aims are better achieved if he sets out the natural norms and teleological regularities, recognizing that they have exceptions, but not specifying the exceptions in detail.
To see whether the second answer is also preferable in ethical theory, we must consider whether the purposes of ethical theory make it reasonable to stick to usual principles. Aristotle insists that ethical theory—the disci pline practised in his ethical treatises—essentially has a practical aim. The appropriate age for studying ethics is to be decided by considering the age at which this study is practically useful, 'since the end is not knowledge, but action' (1095a5-6). This practical purpose of ethics explains why we ought to be satisfied with principles that are stated only roughly. To demand more would be as misguided as a carpenter's seeking geometrical precision about right angles (1098a29-31). Nothing about wood and right angles prevents carpenters from finding out how the right angles they try to produce in wood fall short of being true right angles; but it would be pointless to occupy themselves with these questions. Similarly, the practical purpose of ethics implies that we should not try to build in all the qualifications that we would need to add to find exactly correct principles.
Aristotle's examples of usual principles suggest why we ought not to try to spell out all the exceptions. Virtuous people take the right attitude to wealth and to other external goods. They recognize that wealth is good without qualification, but not good for everyone, and they learn that virtue results in happiness in appropriate conditions, though not in all conditions without exception. It is more important to grasp these points about virtue, external goods, and fortune than to learn the more complex generaliza tions that would incorporate all the relevant exceptions to the unqualified generalizations.
Aristotle argues, then, that we need not list all the exceptions and quali fications that would replace our usual rules with exact generalizations. Nor should we take the usual rules any less seriously simply because they are usual. These usual principles will be useless to people who cannot guide their desires by reason, but useful to those who can guide desires by reason (1095a4-11). If I learn that brave action is always better than cowardly action, but that it only usually results in happiness, then I will not believe it is ever in my overall interest to prefer the cowardly action. My confidence in preferring the brave action will not be shaken if I recognize that in this particular case the brave action will not result in happiness for me. The combination of the universal and the usual generalizations about bravery and brave action will strengthen my confidence in acting virtuously; the same will be true for every other virtue. Our recognition of the usual char acter of some rules helps us to take them seriously; for we will not be dis concerted to find that they have exceptions.
If this is the point of recognizing usual principles, it does not imply that any special doubt or difficulty arises in deciding whether or not to do the brave action in this or that particular situation, or that something more than the theoretical principles is needed if we are to find the right thing to do in particular situations. Aristotle does not suggest, for instance, that the merely usual status of the generalization that bravery usually results in hap piness ever gives us a good reason not to act bravely, or that we need any special exercise of perception to see that we must act bravely in this partic ular case where brave action will involve significant harm. These remarks on the inexactness of ethical principles do not support particularism.