H. P. Grice comments on Aristotle's two chapters on friendship in "Ethica Nichomachaea": reciprocal good will (eunoia) -- benevolence and self-love as emotions -- egoism vs. altruism

The topic of Books VIII and IX of the Ethics is friendship. 

Although it is difficult to avoid the term “friendship” as a translation of “philia”, and this is an accurate term for the kind of relationship he is most interested in, we should bear in mind that he is discussing a wider range of phenomena than this translation might lead us to expect, for the Grecians use “philia”, to name the relationship that holds among family members, and do not reserve it for voluntary relationships. 

Although Aristotle is interested in classifying the different forms that friendship takes, his main theme in Books VIII and IX is to show the close relationship between virtuous activity and friendship. 

He is vindicating his conception of eudaemonia as virtuous activity by showing how satisfying are the relationships that a virtuous person can normally expect to have.

Aristotle's taxonomy begins with the premise that there are three main reasons why one person might like someone else. 

The verb, “philein”, which is cognate to the noun “philia”, can sometimes be translated “like” or even “love”—though in other cases philia involves very little in the way of Griceian or Grecian emotion or feeling.

One might like or love someone because he is good, or because he is useful, or because he is pleasant. 

And so there are three bases for friendships, depending on which of these qualities binds friends together. 

When two individuals recognize that the other person is someone of good character, and they spend time with each other, engaged in activities that exercise their virtues, they form one kind of friendship. 

If they are equally virtuous, their friendship is perfect. 

If, however, there is a large gap in their moral development (as between a parent and a small child, or between a husband and a wife), although their relationship may be based on the other person's good character, it will be imperfect precisely because of their inequality.

The imperfect friendships that Aristotle focuses on, however, are not unequal relationships based on good character. 

Rather, they are relationships held together because each individual regards the other as the source of some advantage to himself or some pleasure he receives. 

When he calls these relationships “imperfect”, Aristotle is tacitly relying on widely accepted assumptions about what makes a relationship satisfying. 

These friendships are defective, and have a smaller claim to be called “friendships”, because the individuals involved have little TRUST in each other, quarrel frequently, and are ready to break off their association abruptly. 

Aristotle does not mean to suggest that unequal relations based on the mutual recognition of good character are defective in these same ways. 

Rather, when he says that unequal relationships based on character are imperfect, his point is that people are friends in the fullest sense when they gladly spend their days together in shared activities, and this close and constant interaction is less available to those who are not equal in their moral development.

When Aristotle begins his discussion of friendship, he introduces a notion that is central to his understanding of this phenomenon: a genuine friend is someone who loves or likes another person for the sake of that other person. 

Wanting what is good for the sake of another he calls “good will” (eunoia), and friendship is reciprocal good will, provided that each recognizes the presence of this attitude in the other. 

Does such good will exist in all three kinds of friendship, or is it confined to relationships based on virtue? 

At first, Aristotle leaves open the first of these two possibilities. 

He says that is necessary that friends bear good will to each other and wish good things for each other, without this escaping their notice, because of one of the reasons mentioned (1156a4–5)

The reasons mentioned are goodness, pleasure, and advantage.

And so it seems that Aristotle is leaving room for the idea that in all three kinds of friendships, even those based on advantage and pleasure (hedone, not pleasure) alone, the individuals wish each other well for the sake of the other.

But in fact, as Aristotle continues to develop his taxonomy, he does not choose to exploit this possibility. 

He speaks as though it is only in friendships based on character that one finds a desire to benefit the other person for the sake of the other person.

He says that those who wish good things to their friends for the sake of the latter are friends most of all, because they do so because of their friends themselves, and not coincidentally (1156b9–11).

When one benefits someone not because of the kind of person he is, but only because of the advantages to oneself, Aristotle says, one is not a friend towards the other person, but only towards the profit that comes one's way (1157a15–16).

In such statements as these, Aristotle comes rather close to saying that relationships based on profit (or utility) or pleasure (hedone, not a utility or a profit) should not be called friendships at all. 

But he decides to stay close to common parlance and to use the term “friend” loosely. 

Friendships based on character are the ones in which each person benefits the other for the sake of other; and these are friendships most of all. 

Because each party benefits the other, it is advantageous to form such friendships. 

And since each enjoys the TRUST and companionship of the other, there is considerable pleasure (hedone, not utility or profit) in these relationships as well. 

Because this 'perfect friendship' produces advantages and pleasures for each of the parties, there is some basis for going along with common usage and calling any relationship entered into for the sake of just one of these goods a friendship. 

Friendships based on advantage alone or pleasure alone deserve to be called friendships because in full-fledged friendships these two properties, advantage and pleasure (hedone, not a utility, profit, or advantage), are present. 

It is striking that in the Ethics Aristotle never thinks of saying that the uniting factor in all friendships is the desire each friend has for the good of the other.

Aristotle does not raise questions about what it is to desire good for the sake of another person. 

He treats this as an easily understood phenomenon, and has no doubts about its existence. 

But it is also clear that he takes this motive to be compatible with a love of one's own good and a desire for one's own happiness. 

Someone who has practical wisdom will recognize that he needs friends and other resources in order to exercise his virtues over a long period of time. 

When he makes friends, and benefits friends he has made, he will be aware of the fact that such a relationship is good for him. 

And yet to have a friend is to want to benefit someone for that other person's sake.

It is not a merely self-interested strategy, or procedure ('Strategy' should be reserved to its literary military use, when 'war' and competition is presupposed).

Aristotle sees no difficulty here, and rightly so. 

For there is no reason why acts of friendship should not be undertaken partly for the good of one's friend and partly for one's own good. 

Acting for the sake of another does not in itself demand self-sacrifice. 

It requires caring about someone other than oneself, but does not demand some loss of care for oneself. 

For when we know how to benefit a friend for his sake, we exercise the ethical virtues, and this is precisely what our eudaemonia consists in.

Aristotle makes it clear that the number of people with whom one can sustain the kind of relationship he calls a perfect friendship is quite small (IX.10). 

Even if one lived in a city populated entirely by perfectly virtuous citizens, the number with whom one could carry on a friendship of the perfect type would be at most a handful. 

For he thinks that this kind of friendship can exist only when one spends a great deal of time with the other person, participating in joint activities and engaging in mutually beneficial behavior; and one cannot cooperate on these close terms with every member of the political community. 

One may well ask why this kind of close friendship is necessary for eudaemonia, and here comes the Grecian and Griceian aporia. 

If one lived in a community filled with good people, and co-operated on an occasional basis with each of them, in a spirit of good will and admiration, would that not provide sufficient scope for virtuous activity and a well-lived life? 

Admittedly, close friends are often in a better position to benefit each other than are fellow citizens, who generally have little knowledge of one's individual circumstances. 

But this only shows that it is advantageous to be on the receiving end of a friend's help. 

The more important question for Aristotle is why one needs to be on the giving end of this relationship. 

And obviously the answer cannot be that one needs to give in order to receive; that would turn active love for one's friend into a mere means to the benefits (profit, utility, advantage) received.

Aristotle attempts to answer this question in IX.11, but his treatment is disappointing. 

His fullest argument depends crucially on the notion that a friend is “another self” (alter ego, tu), someone, in other words, with whom one has a relationship very similar to the relationship one has with oneself. 

A virtuous person loves the recognition of himself as virtuous; to have a close friend is to possess yet another person, besides oneself, whose virtue one can recognize at extremely close quarters; and so, it must be desirable to have someone very much like oneself whose virtuous activity one can perceive. 

The argument is unconvincing because it does not explain why the perception of virtuous activity in fellow citizens would not be an adequate substitute for the perception of virtue in one's friends.

Aristotle would be on stronger grounds if he could show that in the absence of close friends one would be severely restricted in the kinds of virtuous activities one could undertake. 

But he cannot present such an argument, because he does not believe it. 

He says that it is “finer and more godlike” to bring about the well being of a whole city than to sustain the happiness of just one person (1094b7–10). 

He refuses to regard private life — the realm of the household and the small circle of one's friends—as the best or most favorable location for the exercise of virtue. 

He is convinced that the loss of this private sphere would greatly detract from a well-lived life, but he is hard put to explain why. 

He might have done better to focus on the benefits of being the object of a close friend's solicitude. 

Just as property is ill cared for when it is owned by all, and just as a child would be poorly nurtured were he to receive no special parental care —points Aristotle makes in Politics II.2–5 — so in the absence of friendship we would lose a benefit that could not be replaced by the care of the larger community. 

But Aristotle is not looking for a defense of this sort, because he conceives of friendship as lying primarily in activity rather than receptivity. 

It is difficult, within his framework, to show that virtuous activity towards a friend is a uniquely important good.

Since Aristotle thinks that the pursuit of one's own happiness, properly understood, requires ethically virtuous activity and will therefore be of great value not only to one's friends but to the larger political community as well, he argues that self-love is an entirely proper emotion—provided it is expressed in the love of virtue (IX.8). 

Self-love is rightly condemned when it consists in the pursuit of as large a share of external goods—particularly wealth and power—as one can acquire, because such self-love inevitably brings one into conflict with others and undermines the stability of the political community. 

It may be tempting to cast Aristotle's defense of self-love into modern terms by calling him an egoist, and “egoism” is a broad enough term so that, properly defined, it can be made to fit Aristotle's ethical outlook. 

If egoism is the thesis that one will always act rightly if one consults one's self-interest, properly understood, nothing would be amiss in identifying him as an egoist, rather than an altruist.

But egoism is sometimes understood in a stronger sense. 

Just as consequentialism is the thesis that one should maximize the general good, whatever the good turns out to be, so egoism can be defined as the parallel thesis that one should maximize one's own good, whatever the good turns out to be. 

Egoism, in other words, can be treated as a purely formal thesis: it holds that whether the good is pleasure, or virtue, or the satisfaction of desires, one should not attempt to maximize the total amount of good in the world, but only one's own. 

When egoism takes this abstract form, it is an expression of the idea that the claims of others are never worth attending to, unless in some way or other their good can be shown to serve one's own. 

The only underived reason for action is self-interest.

That an act 'helps' another (benevolentia, beneficentia, cooperation, helpfulness) does not by itself provide a reason for performing it, unless some connection (or overlap) can be made between the good of that other and one's own.

There is no reason to attribute this extreme form of egoism to Aristotle. 

On the contrary, his defense of self-love makes it clear that he is not willing to defend the bare idea that one ought to love oneself alone or above others.

Aristotle defends self-love only when this emotion is tied to the correct theory of where one's good lies, for it is only in this way that he can show that self-love need not be a destructive passion. 

He takes it for granted that self-love is properly condemned whenever it can be shown to be harmful to the community. 

It is praiseworthy only if it can be shown that a self-lover will be an admirable citizen. 

In making this assumption, Aristotle reveals that he thinks that the claims of other members of the community to proper treatment are intrinsically valid. 

This is precisely what a strong form of egoism cannot accept.

We should also keep in mind Aristotle's statement in the Politics that the political community is prior to the individual citizen — just as the whole body is prior to any of its parts (1253a18–29). 

Aristotle makes use of this claim when he proposes that in the ideal community each child should receive the same education, and that the responsibility for providing such an education should be taken OUT of the hands of private individuals and made a matter of common concern (1337a21–7). 

No citizen, he says, belongs to himself.

All belong to the city (1337a28–9). 

What he means is that when it comes to such matters as education, which affect the good of all, each individual should be guided by the collective decisions of the whole community. 

An individual citizen does not belong to himself, in the sense that it is not up to him alone to determine how he should act.

He should subordinate his individual decision-making powers to those of the whole. 

The strong form of egoism we have been discussing cannot accept Aristotle's doctrine of the priority of the city to the individual. 

It tells the individual that the good of others has, in itself, no valid claim on him, but that he should serve other members of the community only to the extent that he can connect their interests to his own. 

Such a doctrine leaves no room for the thought that the individual citizen does not belong to himself but to the whole.