The paradox of friendship

This work is licensed under the Creative Commons | © Julian Pitt-Rivers. ISSN 2049-1115 (Online). DOI: http://dx.doi.org/10.14318/hau6.3.032


The paradox of friendship


Translated by Matthew Carey

This paper argues that the opposition between kin and friend is analytically misleading. Analogously, the notion of “fictive kinship” is unsuitable to describe forms of relatedness such as blood brotherhood, god parenthood, or “friendship” itself. Each of these social forms generates contradictions that ideas of kinship cannot fully resolve, and these problems would seem to be part of friendship's subject matter. The paradox of friendship, with its offers of help and implicit demands for a reciprocal counter-gesture, is more usefully compared to the paradox of the Maussian gift, with its studied spontaneity and dire consequences of refusal.

Keywords: kinship, friendship, fictive kinship, free gift, reciprocity

There were, until fairly recently, very few anthropological studies of friendship. This might seem surprising, both because of the intrinsic interest the topic holds for all of us and also because of its rich theoretical potential. After all, friendship truly is the “atom of social organization”—viz., the basic element beyond which it is impossible to reduce actions of welcoming or rejecting others. Friendship is the constitutive basis of all social ties for the simple reason that solidarity and cooperation cannot do without a degree of indulgence, and this is an aspect of friendly relations. It has even been suggested that in modern or peasant societies (and to a lesser extent in all societies) relations of kinship are in fact simply options that are activated when a relationship of friendship is superimposed on a preexisting kin tie. There is no shortage of examples of the word “friend” being used to describe a [444]kinsman, as if one could only feel friendship for cognatic relatives. Ireland is one classic example of this (cf. Arensberg 1934). In his study of a Mexican peasant village, Lilo Stern (1962), a student of Meyer Fortes, speaks of a “panel” of kinsmen from which one chooses one’s friends, thereby activating the kin tie. One is born a potential kinsman, but only becomes a de facto kinsman by virtue of a mutual act of choice (volonté); and if one chooses not to become friends, then the supposed kin solidarity is more likely to lead to family disputes than to cooperation. The existence of a legal kin tie can just as easily be a source of hostility as of friendship.

On the other hand, Fortes (1969) spoke of “kinship amity,” which is the friendship that is the essential principle of kinship and which inspires a solidarity both of interest and of the heart. This amity excludes any form of accounting between participants: yours and mine no longer exist; there is only ours. This world of kin solidarity is contrasted with the world outside, which is governed by a legal regime that demands accounts. When, however, one considers the number of family quarrels that end up in the courts, this idea of kinship seems somewhat idealized. But if the essence of kinship is indeed friendship, how are we to explain the frequent contrast between the two terms, not only in anthropological theory but also among natives themselves, who often see them as opposites?1Above all, how are we to explain that anthropologists have paid so little attention to the institution of friendship, while lavishing it on kinship?

One answer is that they only wanted—and only knew how—to deal with institutions that were clearly defined by the people they worked with, and so they restricted themselves to forms of friendship associated with a particular ritual: blood brotherhood, compadrazgo, et cetera.2 Such ties were long considered forms of “fictive” kinship—perhaps to justify anthropological interest in an otherwise neglected institution. However, as soon as one looks in detail at the forms of behavior and rights associated with these forms of ritual friendship, it becomes clear that they involve no sort of fiction and are essentially opposed to kin ties. Adoption, in contrast, can quite properly be called “fictive kinship,” because it involves assimilating somebody born outside of a family to one of its members, both emotionally and legally.

The error that consisted in describing everything that wasn’t natural kinship as fictive is doubtless a product of the fact that such relations almost invariably borrow their terminology from kinship, and natives themselves frequently make the link: compadres and blood brothers are, they say, “like brothers.” But they often add (and this is what betrays them) that they are, in fact, “closer than brothers.” Such statements are easily comprehensible if one considers the widespread motif of warring brothers. If these ties are closer, it is precisely because they are not brothers, share no family inheritance or duties, cannot replace one another, are not caught up in binding relations with one another’s respective kin, and are simply a matter of choice (volonté) and mutual feeling. In short, they are relations conceived outside of the prevailing social structure that, for this very reason, better embody the ideal of fraternity than relations between true brothers, who cannot choose one another.[445]

Besides, there is often no word in the vernacular for “friend,” as Donald Cole notes of the Bedouin (see, for example, Cole 1982), who make metaphorical use of a kin term to designate a relationship of friendship. It is easy to conclude that if there’s no word for a thing, it must not exist. Friendship, it has been claimed, is an invention of civilized peoples like us; primitives lack the requisite refinement of sentiment to feel emotional attachments outside of kinship. This in turn led to the idea that kinship and friendship are mutually exclusive forms of relation: one tastes the pleasures of friendship only insofar as one has slipped the chains of kinship. In fact, there is no need to think of friendship as an institution, or give it a generic name,3 when it is the very stuff of personal relations, and such bunkum is nigh on unforgivable when one recalls Raymond Firth’s (1960) description of Tikopian sworn friends, soa, who express their amicable affection with rare refinement.

To recap, in ethnographic descriptions of friend- and kinship, be the terms opposed or assimilated to one another, there is an invariable lack of an overview of the question of what friendship is per se.

Eric Wolf set out to remedy this some twenty years ago in an essay entitled “Kinship, friendship and patron-client relations in complex societies” (1966). Drawing on Ruben Reina’s (1959) excellent ethnography of relations of friendship in a small Guatemalan village, Wolf establishes a typology of friendship that distinguishes between two main forms, which he calls “expressive or emotive” and “instrumental.” And indeed, the contrast is clear in his comparison of emotive ties of friendship between adolescent Indians, which (without the least element of homosexuality) are affectionate to the point of exclusivity (one cannot have more than one friend), and friendly relations between Ladino (i.e., Hispanic, non-Indian) family men, petty traders, and entrepreneurs, which consist of tacit alliances of mutual help and are therefore all the more effective if they are part of a wider network of friendship: the more friends one has, the greater one’s worth as a friend. It is of just such friendships that the patron client networks of “caciquismo” were made—networks that dominated the Hispanic world in the nineteenth century and are far from vanished today. We find these two types of friendship more or less everywhere, and Aristotle already identified them in the Nicomachean ethics (2009). Rather than explain the difference as Wolf would have us do, however, in terms of a distinction between more or less open communities, more or less dominated by solidary forms of kinship, I think we should consider the age and social situation of the friends Reina describes, because we find both types in a wide range of societies, including our own.

The adolescent Indians are entering adult life with an as-yet unsullied amicable sensibility; they are free to form personal ties as they wish since they cannot engage in any entrepreneurial activity and have no family worries, being single and employed as agricultural laborers. Ladino family men, in contrast, have plenty of scope for entrepreneurial action, significant social and material ambitions, and numerous family worries. They nonetheless seem to set great store by the affective aspect of their friendships, apparently not cheapened in their eyes by its instrumental aspect. I think these examples are simply extreme cases of a general phenomenon that can be explained by the difference of age—Reina informs us that this intense emotive [446]friendship does not survive marriage—and of social situation, rather than by the type of society in question. When one is young, one thinks differently of friendship than later in life. The French bourgeoisie of yesteryear captured this idea very well by reserving the informal “tu” for childhood friends and fellow students, as if they no longer had the same emotional availability after the age of twenty-five. But this was, in fact, slightly misleading, as the use of “tu” was rather an expression of simple camaraderie than of brotherly love. It is nonetheless true that we are far more influenced by our childhood friendships, and remember them more fondly, than we do those contracted later in life. Some studies of friendship, notably those conducted by social psychologists, have examined the changing nature of friendship from the perspective of people’s age (Du Bois 1974: 23–25), but the problem of the comparative dimension of ideology and the practice of friendship in different cultures, as well as the different behavioral codes they impose, however obvious they may be, still lack a clear theoretical framework. The structures of friendship have yet to be defined, just as we are still to determine whether or not it is correct to treat it “as a residual category of social structure rather than as a major social form” (Leyton 1974: 93).

I would now like to turn my attention to friendship in contemporary urban society. In order to do so, I think it important first to identify what distinguishes it from friendship among peasants and primitives on the one hand, and from that of chivalric friendship on the other. Above all, contemporary, urban varieties of friendship need to be distinguished from ritualized friendship, referred to as compadrazgo in the anthropological literature, on which topic there is an abundant literature. For reasons that remain to be determined (thanks perhaps to the ritual that seals the compact), all these people seem better able than us to organize their friendships clearly and in such a way as to avoid the anxiety and disappointment that seems to surround contemporary friendship, at least in Anglophone countries. An American professor of my acquaintance, who taught a course on friendship at Harvard some thirty years ago, told me how his students would speak quite openly of their amorous relations but were immediately ill at ease when it came to discussing friendship. This is unsurprising, as modern romantic individualism only associates friendship with emotional states and neglects the duty of reciprocity present in any friendly relation, especially when it is emotive, as one more easily forgives its absence when it is the result of a material rather than an emotional shortfall. We are afraid of appearing calculating but do not wish to recognize the legitimate interests of our friends. The upshot is that we have lost our compass, and disappointments multiply without our being able to identify the cause.4 We wish to be everyman’s [447]friend without putting ourselves out for any of them. To this we can add that the ideology of universal friendship, which is, I might add, quite different from the religiously inspired notion of the Quakers (who address everybody as “friend” without distinction), seems to devalue the institution as a whole: a vague togetherness, a discount bonhomie, replaces the gemeinschaft of the antique community, which has dissipated in the face of urban life and increased spatial mobility, which make it harder to maintain effective relationships between old friends. We reject the formalism of traditional manners, which prevents the generation of instant intimacy; we introduce ourselves by our Christian names before offering our surnames, and we leave without bidding farewell. We do not wish to recognize the moral economy of friendship, so ably described by Aristotle, or to recognize that everyman’s friend is also no man’s friend.

I am not trying to downplay the importance of emotions in the study of friendship, or to deny its emotive aspect by explaining it away in practical or materialist terms, as some of my colleagues are wont to do. No more, however, do I think that we should adopt the perspective of moralists of the European literary tradition, such as David Hume, who distinguished between “the self-interested commerce of men” and “the more generous and noble intercourse of friendship and good offices” inspired by love. The distinction turns on the disinterested nature of the friendly gesture, but this in no way prevents the repayment of (equally disinterested) services rendered. The idea of contract is present in “self-interested commerce,” but not in nobler and more generous friendship.5 The critical element here is the notion of contractual obligation, because it removes the idea of any form of responsibility outside the contractual tie; it sets a limit to the obligation by specifying it, whereas compadrazgo or sentimental friendship more generally generate a diffuse but total sense of responsibility. When nothing is specified, when there is no longer even a tacit contract according to local mores, then the nature of friendship and the limits of each party’s obligations remain, as with tips in restaurants, “at the client’s discretion.” In the conceptual vacuum between different estimations of obligation, anything goes, from the most emotive to the most instrumental concept of friendship, which no longer depends on anything but individual personalities and moral sensibilities; and in this vacuum arise the unspecifiable claims and grievances that cause friendships to founder. Rather than exploring successful friendships in different societies, we ought to examine friendships gone wrong to assess the effect of different ideas of friendship.

European civilization’s problems of conscience cannot provide the basis for a comparative study of the topic that would allow for the elaboration of a universal [448]typology of friendship, and were one to attempt such an approach, it would fall prey to contradictions that are proper to the institution itself, which rests on a paradox. This paradox appears at the point of transition between the emergence of a feeling of sympathy for another and its expression in acts, between the domain of self and that of contact with the other, between the psychology and the sociology of friendship. The paradox is as follows: a friendship that is never given expression is stillborn, but how to express it except via a gesture that demonstrates a desire to please one’s potential friend: that one seeks his company, finds him admirable, is ready to trouble oneself for him, to despoil oneself in his favor by means of a gift, to sacrifice oneself? The true friend is not content to express this in speech but also in acts, because actions speak louder than words. However, the gesture of friendship demands a reciprocal countergesture (more or less immediately depending on the local mores), and if this countergesture does not come, it means that the friendship is refused and the initiator humiliated such that he will surely become an enemy. If the countergesture is offered within a reasonable timeframe, and matches the initiator’s expectations, then a tacit agreement is reached. But if the gesture was made, in the words of Hume, “with the prospect of advantage”—i.e., with the aim of provoking the countergesture—or if the countergesture was carried out inadequately or simply to avoid problems, then the friendship appears false because it was predicated on interested calculation rather than noble and generous love.

Friendship, then, must be expressed in words and especially in gestures. It must be reciprocated in like fashion, but it cannot be given expression with the aim of provoking the countergesture, nor reciprocated out of mere convenience or with an eye to profit. If, however, the countergesture does not come, one can justifiably take offence. Nevertheless, the act of taking offence will likely inspire suspicion as to the disinterestedness of the initial sentiments: “if you acted out of love and ‘without prospect of advantage,’ why complain about not having benefited from the action?” The trap is sprung! The paradox can be summed up thusly: to defend the purity of one’s sentiments one must act in (blind or hypocritical) ignorance of the consequences of one’s actions! As no member of a society that recognizes this emotional logic will ever admit his friendship to be interested, the possibility of distinguishing between emotive and instrumental friendship depends on the analyst’s assessment of the unavowable inner states of his victims (sic).

Whatever admiration the anthropologist may feel for David Hume, it is not to him but to Marcel Mauss that he must turn in this matter.

Mauss does not directly address friendship, but in his grand essay on The Gift (1990) he nonetheless touches upon its essence, as the gift is the simplest model of all the different forms of favor that friends may exchange. He pays great attention to the symbolic meaning of the gift and the system to which it belongs. A gift is not simply an object of more or less recognized value that one makes over but is always a gift of oneself, as part of oneself accompanies that which is offered; this is the very meaning of the act of offering, which is not the same as giving. This is why a peasant always offers part of his production rather than purchased goods. It is also why we mind how a gift is treated: one can be insulted via one’s gift. If one disposes of an object in a commercial transaction, one accepts payment without expecting thanks, but the gift opens up a line of moral credit, which is part of a system of exchange that is much more than merely economic. It is a moral system rather like the [449]exchange of women carried out in Africa or elsewhere through rules of marriage and that is often associated with a payment of livestock or even money. Anthropologists had great difficulty preventing missionaries and colonial administrators from referring to these nuptial transfers as “bride price,” as if they were commercial transactions. There are, of course, economic effects, but these systems of exchange are moral and social in their inspiration, and economic only in their effects. The gift, however, always has an eye toward the counter-prestation, especially when made to somebody possessed of the power to perform a favor or hand down a judgment. This is the moral basis of the problem of corruption that seems to dominate the contemporary world and that anthropologists have only begun to investigate.

The essential aspect of the gift is its free nature, as Mr. de la Palisse would say,6 and this is true of all gestures of friendship; friendship, like the gift that heralds it, is an exchange of grace. Just like religious grace, which is only accorded on the condition that the ritual participant truly wishes it (volonté), the secular grace that flows through rituals of friendship can only be realized if the donator wills it (volonté). And just as with religious sacrifice, the despoilment of oneself effects a conversion of having into being; one makes over the object one offers in order to become a friend. However, this free expression of friendship requires an equally free reciprocal gesture; the reciprocity of friendship is not merely material, as in a contract, but is also a reciprocal gift of oneself. Boris Oguibenine put it very well when he said, “any affective tie implies reciprocity” (1998: 189). One might add that this is invariably true, even when the reciprocity is not direct, as with parents and children, because the debt owed to the previous generation is repaid to the next: it is a form of reciprocity open to the future, but like all reciprocities of the heart, one that is outside accountability.

Mauss’ essay, though, speaks of the obligation “to give, receive, and reciprocate gifts” (1990). He does not concern himself with the problem of motivations, and rightly so, because in most human societies, amicable reciprocity does not pose any problems; it is an obvious social duty rather than a question of conscience, and people are not embarrassed to admit their interest in the counter-gift. Sometimes, this is said quite explicitly. Besides, people are quite aware of the possibility that a new friend may reveal himself to be false and are thus glad of the chance to cement the relationship by means of a ritual that ensures fidelity. One takes time to study carefully the character of a future compadre before asking him to be godfather to one’s child, because once the ritual is accomplished there is no way of annulling the rite. Thanks to the rite, the wrath of God will smite a faithless compadre. The blood of the brother-to-be ingested during the blood brother ritual will punish any betrayal with death. Religion or magic come to the rescue of the honor men lack. In such conditions, there are fewer reasons to appear disinterested, as the opposition between friendship of the heart and friendship of the head does not exist; the reciprocity of contract and that of sentiments sit well together; interest is not opposed to grace and amicable transactions are not separated into calculating versus diffuse because the world is not divided into distinct aspects (economic, legal, religious, moral, sentimental, etc.) each of which has its own distinct vocabulary and intellectual discipline.[450]

Thus we see that the categories adopted by Wolf to build his typology, and prefigured in the work of Hume, reflect a crisis of friendship grown acute in twentieth-century urban society, where heightened spatial and social mobility, along with significant uncertainty of community or class identity and loosened kinship ties, have thrown the individual upon his own resources—helped only by his friends.

Mauss insists on the ambiguity of the gift, as the donor’s intentions are open to more than one interpretation. One can humiliate the recipient with an excessive gift, because if the expected counter-prestation exceeds his means then he is humiliated. This is the very principle of the potlatch. While the gift honors the recipient, alms crush him, as one may give out of pity, and pity is scorn’s sympathetic sister. To avoid these possible implications, one must choose a gift with care. Perishable gifts are best because they cannot carry the implication that one’s friend wants for anything. One must be prepared to help but not hurry to do so too early or too obviously. One can always offer a woman flowers, but if one gives diamonds, she can reasonably wonder about the nature of the counter-prestation to which she subscribes by accepting them. She may well refuse them, since favors, like grace, cannot be bought. And yet, they can be rendered.

The refusal of a gift, like the refusal of a challenge (the rules surrounding gifts are as ambivalent as those surrounding honor), is a rejection of exchange, which demonstrates a rejection of association tantamount to scorn. When it is a question of honor, the same gesture can have two contradictory meanings, and in the same way, the gesture of friendship can express love just as it can express rivalry or hostility. The potlatch of the Canadian Northwest furnished Mauss with the perfect example of festivities offered out of hostility.

The ambivalence of the gift reflects the ambivalence of friendship, of which it is but a manifestation. The potlatch is a hostile variety of friendship, but the heart of the paradox, reciprocity, is still present, albeit negatively. Rather than being an expression of fellow feeling, the invitation to a potlatch is a challenge, as the festivities are a demonstration of the host’s economic might by a destruction of property in different forms. The guest of honor, for whom the potlatch is put on, is obliged to respond within a year and in like fashion, but with even greater destruction of property, or be forever dishonored.

The potential enemy hidden under the surface of friendship, like the hatred concealed in love’s center, is in this case exposed for all to see, and this inverts everything: there is, of course, exchange, but the things exchanged lose their habitual meaning. The gesture that is supposed to honor the other is tacitly transformed into a locking of horns; the expression of friendship is a form of aggression; the vaunted copper is not offered to the guest but broken and cast into the sea; oil is poured onto the flames and so wasted; the blankets supposed to keep people warm over winter are committed to fire to force the guest, knees already red from the flames, to move his chair back another few inches; the gift becomes a challenge. This is friendship reversed, transformed into a norm by the mores of a warrior people who had lost, under European aegis, the right to open combat.

This is surely an extreme example, but not a unique one. Mortal rivalry disguised as friendly festivities is not restricted to Canadian Indians. We might mention wealthy ladies of the house, or the rivalry around a card table, or the intimacy of a golf course. The exploitation of compadrazgo in Hispanic and Latin politics, [451]which uses the sacred ties of the baptismal fount to bind clients, is no secret. And to give a final example drawn from a very simple society, we can cite the portrait of blood brother–ship in a Kaguru myth from Central Africa admirably recounted by Thomas O. Beidelman (1963), which calls to mind La Fontaine’s fable of the fox and the stork. In this case, it depicts an eagle and a monitor-lizard, sworn blood brothers who can therefore refuse one another nothing. The monitor-lizard asks his friend for his eaglets’ down because he is cold, along with countless other inopportune demands, until the eagle finally seeks revenge on the monitor-lizard by asking for his skin to protect him from the sun.

The classification of different expressions of abstract notions such as honor or friendship, or any other idea that is formulated to deal with problems of behavior in our society and is therefore part of what we might call “its popular moral philosophy,” leads nowhere, for it proposes to treat as an objectively definable relationship (of, for instance, kinship) a relation that is only composed of subjective sentiments and evaluations of behavior whose objective analysis reveals inherent contradictions. What is a friend nowadays? Somebody who behaves toward me in what I deem to be a friendly manner. If his behavior alters, I may change my mind and say to him, “I was mistaken to think you were a friend.”

Rituals for establishing friendship aim to objectify the relation by transforming the mutual sympathy that first inspires friendly behavior into a statutory tie that can be defined and classified. In the absence of such a tie, however, the only way to study this “atom of social organization” that is friendship is as a series of atoms— i.e., as particular cases. From an analytical perspective, there is, to return to the title of this workshop, no such thing as friendship, only friendships.


Arensberg, Conrad M.1934. “A study of rural life in Ireland as determined by the functions and morphology of the family.” PhD thesis, Harvard University.

Aristotle. 2009. The Nicomachean ethics. Oxford’s World Classics. Oxford: Oxford University Press.

Beidelman, Thomas O. 1963. “The blood covenant and the concept of blood in ukaguru.” Africa 33: 321.

Cole, Donald.1982. “Tribal and non-tribal structures among the Bedouin of Saudi Arabia.” Al-Abhath 30:77–94.

Du Bois, Cora. 1974.“The gratuitous act: An introduction to the comparative study of friendship patterns.” In The compact: Selected dimensions of friendship, edited by Elliott Leyton, 15–32. Toronto: University of Toronto Press.[452]

Firth, Raymond William. 1960.“A Polynesian aristocrat.” In In the Company of Man, edited by Joseph Bartholomew Casagrande, 1–40. New York: Harper.

Fortes, Meyer. 1969. Kinship and the social order: The legacy of Lewis Henry Morgan. London: Routledge.

Hume, David. 1893. A treaty of morals. Boston: Ginn.

Leyton, Elliott, ed. 1974. The compact: Selected dimensions of friendship. Toronto: University of Toronto Press.

Mauss, Marcel. 1990. The gift: The form and reason for exchange in archaic societies. Translated by W. D. Halls. London: Routledge.

Oguibenine, Boris. 1998. “The cooking of poetic words.” In Essays on Vedic and Indo-European culture, 167–98. New Delhi: Motilal Banarsidass Publishers.

Pitt-Rivers, Julian. 1968. “Pseudo-kinship.” In International encyclopedia of the social sciences, Vol. 8, edited by D. L. Sills. New York: Macmillan.

Reina, Ruben. 1959. “Two patterns of friendship in a Guatemalan community.” American Anthropologist 61: 44–50.

Stern, Lilo. 1962. “An ethnography of Chiapilla, Chiapas.” Microfilm collection of manuscripts on American Indian cultural anthropology. Unknown Binding.

Wolf, Eric R. 1966. “Kinship, friendship and patron-client relations in complex societies.” In The social anthropology of complex societies, edited by Michael Banton, 1–22. New York: Frederik A. Praeger.

Le paradoxe de l’amitie

Résumé : Cet essai suggère que l’opposition entre parenté et amitié est analytiquement trompeuse. De même, la notion de «parenté fictive» ne convient pas à la description des formes de proximité telles que les amitiés de sang, le parrainage ou l’amitié en elle-même. Chacune de ces formes sociales crée des contradictions que l’idée de parenté ne peut entièrement résoudre, et ces problèmes font partie intégrante de ce que constitue l’amitié. Le paradoxe de l’amitié, l’aide qu’elle offre et la demande implicite pour un contre-geste réciproque, est plus utilement comparé au paradoxe du don maussien, avec sa spontanéité étudiée et ses conséquences dramatiques en cas de refus.

Julian PITT-RIVERS (1919–2001) was a British social anthropologist. He studied at Oxford under E. E. Evans-Prichard and taught at the London School of Economics and the École Pratique des Hautes Études, among others. His many publications include The people of the Sierra (University of Chicago Press, 1954) and The fate of Shechem or the politics of sex: Essays in the anthropology of the Mediterranean. A collection of his work, From hospitality to grace: A Julian Pitt-Rivers omnibus, edited by Giovanni da Col and Andrew Shryock, is forthcoming from Hau.


Editorial Note: This is the translation of “Le paradoxe de l’amitie”, an unpublished talk by Julian Pitt-Rivers, originally presented at the colloquium “L’Amicizia e le Amicizie” (Palermo, 1983) and forthcoming in From hospitality to grace: A Julian Pitt-Rivers omnibus, edited by Giovanni da Col and Andrew Shryock (HAU Books, 2017). The Editor wish to thank Françoise Pitt-Rivers for granting permission to publish, Matthew Carey for the accurate translation, Andrew Shryock for the crafty editing, and T. David Brent, Priya Nelson, Stéphane Gros for their help in retrieving the manuscript.

1. For the opposition between kin/friends, see Fortes (1969: passim).

2. See J. Pitt-Rivers, “Pseudo-Kinship” in Encyclopedia of the Social Sciences, Vol. 8.

3. Primitive societies rarely have a generic word for kinship in their languages either; they only have terms for specific kinship relations, such as uncle, cousin, etc.

4. This anxiety in the face of the uncertainties of friendship in contemporary England is well illustrated by a London weekly that, some twenty years ago, used to organize a weekly literary competition. In one instance, this required of entrants that they alter a single letter of a quotation and so endow it with an entirely new meaning; the winner put forward a stanza from Coleridge’s The Rime of the Ancient Mariner:

“As one who on a lonely road
Marches in fear and dread
And having once looked back goes on
Because he knows some fearful friend (originally FIEND)
Doth close behind him tread.”

5. “But though this self-interested commerce of man begins to take place, and to predominate in society, it does not entirely abolish the more generous and noble intercourse of friendship and good offices. I may still do services to such persons as I love, and am more particularly acquainted with, without any prospect of advantage; and they may make me a return in the same manner, without any view but that of recompensing my past services. In order, therefore, to distinguish those two different sorts of commerce, the interested and the disinterested, there is a certain form of words invented for the former, by which we bind ourselves to the performance of any action. This form of words constitutes what we call a promise” (Hume 1893: 168).

6. Mr. de la Palisse is the approximate equivalent of “Captain Obvious.”