Friendship
Friendship
Friendship is a voluntary, close, and enduring social relationship. The behavior of friends varies greatly among societies and situations and according to personality variables. Values about friendship vary less and can be summarized as involving closeness, solidarity, absence of ulterior ends, reciprocity, impulsiveness in mutual choice, and, perhaps, independence of social distinctions such as age, sex, and class. Friendship is intimate but less so than love and some family ties. Supplementing sexual and familialties, friendship is a residual cultural category subsuming close and expectedly enduring ties. Since friendship involves voluntary commitment, intimacy, and spontaneity, its consequences for the individual and for society, through individual growth and security, are presumably crucial. Possibly for this reason, to be without friends often involves shame.
Most other important social relationships exclude friendship. Even highly compatible and close brothers are brothers rather than friends, and friendship tends to be incompatible with such relationships as those of mother and child, lovers, and employer and employee. This incompatibility is probably due to the fact that the obligations and rights of friends typically are subject to overruling by other ties. The impulsiveness of choice and the reciprocity or symmetry of the friendship relation also rule out various choices. If friends are impulsively chosen, few brothers will be friends. Reciprocity and symmetry imply rough equality in mutual rights and obligations and in qualities and performances, requiring fairly equal status in significant respects between friends. In general, friendship can be logically and culturally expected to occur only when there is a low probability of higher or strongly sanctioned obligations intervening directly between friends.
Like distant kinship terms, “friend” is a relational designation: “friend” and “friendship” refer to a relationship between two or more persons rather than to the characteristics of one or more persons. This is in contrast to close kinship terms, which are simultaneously relational and categorical, and to occupational designations, which are on the whole categorical. Accordingly, while friendship is significant in personal terms, it is no less an interpersonal structure. It follows that variants of friendship structure may well be a characteristic of collective units.
In theoretical terms, friendship is definitely a relational phenomenon: it is impossible to assign meaning to statements such as “He is a friend” without implying to whom “he” is related in this way. However, empirically the matter is less clear, partly because the friendship role is vague. Also, the friendship relationship as a social type is made ambiguous by personal descriptions such as “He is friendly,” “He is a great friend,” or “He is every body’s friend.” The suggestion has been made that a highly differentiated society with a high degree of mobility and an emphasis on specific performance cannot also support enduring and important intimate relationships beyond those of the nuclear family. This may be the structural source of “pseudo Gemeinschaft”—that is, appeals, usually commercial, that use a presumption of closeness to transfer modes of behavior from a friendship setting to strangers, with the consequent growth of values of superficial friendship, friendliness, and popularity.
Friendship is a distinct institution. In Western societies (loosely the basis for the above considerations) it is a vague institution, whereas in other societies friendship is often more salient and differentiated. In either case, friendship is a low-order (as well as a crosscutting) institution: it is found everywhere but is not a distinct, comprehensive segment of society. It is comparable to money, language, and love rather than to religion, the family, or the economy. Partly for this reason, friendship is not at present a specialized field of inquiry in sociology. While few studies focus on friendship, many find it, since closeness to others is a pervasive potentiality in man.
Friendship as an institution
To say that friendship is an institution is to ask the cross-cultural question: What links are there between variations in the structure of friendship and in the structure of society? Two studies have initiated the general cross-cultural study of friendship (Eisenstadt 1956; Cohen 1961). At the least, these studies provide a spectrum of variation in friendship institutions; at their best, they provide hypotheses or findings about the place of friendship in encompassing social structures.
Variations between societies
S. N. Eisenstadt’s subject is “ritualized personal relations,” including blood brotherhood, blood friendship, “best” friends, compadre and godparent relations, and cases of contractual servitude. All these relationships are particularistic, personal, voluntary, and fully in stitutionalized (usually in ritual terms). They are both diffusely affective and instrumental, always in the economic sphere, often more broadly (politics, etc.). Eisenstadt hypothesizes that these kinshiplike but voluntary relations are to be found in predominantly particularistic (that is, kinshipdominated or caste-dominated) societies because they alleviate strains in and between the groups that constitute such societies. In effect, Eisenstadt proposes that ritualized personal relations, similar to friendship in Western terms in that they are voluntary and personal or intimate, are mechaisms of social integration. They are parallel to such institutions as kinship extension, extralineage kinship obligations, various types of associations, hospitality toward strangers, and joking relation ships in providing ties cutting across groups and categories. Ritualized personal relations are also mechanisms of social control. [SeeKinship, article on Pseudo-Kinship.]
Yehudi A. Cohen presents a typology of friendship institutions that loosely expresses a dimension of degree of commitment between friends. Inalien able friendship is a variant characterized by ritual or ceremonial entry and, ideally, permanence—in the main corresponding to Eisenstadt’s ritualized personal relations. Three other somewhat separate categories are close, casual, and expedient friendships. In a sample of about sixty societies, twenty have the inalienable type of friendship as the dominant form, thirty have close friendship, three have casual friendship, and four have expedient friendship (there are no data for ten societies, and some are counted twice because of structural changes). Activities in the friendship relation commonly include material exchange (and/or economic as sistance) and sociopolitical and emotional support; in some cases they include the more specialized activities of go-between in love affairs and marriage arrangements, homosexuality, sponsorship in rites of passage, mourning obligations, and exchange of children. Formalized friendship is far more frequent among men than women and is rare across sex lines. Inalienable friendship carries with it incest taboos in half the cases. It is almost invariably joined with just one partner, who in some cases has to be chosen inside the local solidary group and in other cases outside of it; in a few societies either choice is open.
Cohen predicts covariation between type of friendship and the nature of the significant soli dary grouping to which the individual is attached. He expects inalienable friendship to coincide with the maximally solidary community (generally, localized descent groups where nuclear families and households are socially, physically, and emotionally close as a societal nucleus sharply distinguished from other groupings). Close friendship is expected to be associated with the solidary-fissile community (where solidarity is split between kinship group and community). The nonnucleated society (where isolated, solidary nuclear families are tied loosely together) is expected to be associated with casual friendship. Finally, an expedient type of friendship institution is expected to occur in individuated social structures (where there is emphasis on individual amassing of wealth and relatively little solidarity even in the nuclear family).
Roughly speaking, Cohen’s hypotheses are suported by his data. There is an association between the degree of solidarity in the local community and the degree of commitment in the friendship institution (although the number of cases in extreme types of friendship and solidarity is small). Of 13 societies with maximally solidary communities, 11 have predominantly the inalien able type of friendship; of 35 societies with solidaryfissile communities, 27 have close friendship.
Cohen’s interpretation of this finding is in terms of holistic compatibility in culture and personality. A dimension of personal generosity versus with holding seems to be the basic variable that unites community structure with friendship structure. There is a correlation between childhood gratification versus deprivation, on the one hand, and adult food sharing versus individual amassment of food or money on the other. For the present, it remains an open question as to how this holistic pressure can be translated into detailed questions about the operation of social arrangements.
Cohen’s study raises formal questions about his variables: To what extent are his four-point variables adequate and reliable in use? Is inalienable friendship a more significant category than formalized friendship? Further, under what conditions, and with what effects, are intragroup and intergroup friendships mandatory or preferred? The most crucial and stimulating question would seem to be this: Is institutionalized friendship (i.e., ideal and/or dominant forms) a more relevant variable than distribution of actual forms of friendship? This question leads to interesting problems in Western societies. What, in various settings, is the distribution of close, casual, and expedient friendships? Is degree of differentiation of societies correlated with dominance of expedient friendships? Is a possible dominance of expedient friendships to be contrasted with some of our values concerning friendship rather than with our past—or has there been a real change in the friendship institution over, say, the last fifty years, concomitant with industrialization and increasing differentiation?
Variation within one society
Additional dem onstrations of the dependence of norms regulating friendship on an encompassing social structure emerge if we compare separate settings within one society rather than comparing different societies. In a study of the merchant marine in a society where it is a significant element economically and culturally, Aubert and Arner (1959) found that the culture and the social structure of the Norwe gian merchant marine include a near taboo on personal friendships. This trait is probably a conse quence of (as well as a contribution to) other structural elements: the ship is a “total institution” (Goffman 1958); top positions can only be reached from the bottom; there is an extraordinarily high rate of turnover; work roles dominate (to the extent that terms of address and reference are largely job titles, the alternative being home region); there is a cultural and realistic emphasis on crises requiring discipline; and, finally, there is a peculiar combination of equality and inequality (in a number of respects the crew are sailors “in the same boat,” yet each man and his position is unique through pay and shift arrangements). In outline, the occurrence and forms of friendship among the crew are reinforcing consequences of the nature of the ship as a place of work, in particular through its arrangements for interaction, physical closeness, and recruitment. In a significant con trast, friendship is a standard occurrence among the crew in the Hull distant-water fishing fleet (Tunstall 1962); the distant-water trawler lacks the character of the total institution that is inher ent in the Norwegian merchant vessel.
Friendship as interaction
Perhaps the most penetrating study of friendship yet to appear is W. F. Whyte’s Street Corner So ciety (1943), a description and analysis of life in an immigrant slum in the late 1930s. Whyte’s topic is the interaction between young men, the significance of this interaction for individuals, and its relationship to career, welfare work, and poli tics. In the first place, Whyte gives a vivid picture of voluntary association among “Cornerville” young adults. This association is marked by a strong in formal structure. Loosely integrated gangs, con sisting of small cliques, have a clearly hierarchic structure in terms of influence and prestige. Par ticipation and acceptance in these groups are cru cial for the balance of individual personalities. Changing and stable group structure is symboli cally expressed in interaction. The typical group ties the individual to his community in many ways; it takes membership in atypical groups to foster career ambitions beyond Cornerville. In the first place, then, Whyte describes the social realities, of which friendship is the predominant one, for young men in Cornerville. Second, he analyzes the interrelations of friendship and the Cornerville culture with racketeering and politics, which are intimate, as well as with traditional welfare work, which cuts itself off from the mainstream of Cor nerville life.
The kind of friendship structure that Whyte describes is, in all likelihood, unusual. Friendship groups are rarely as highly structured in a hier archy as in Whyte’s community; friends are rarely as dependent on leadership; they are rarely as significant for each other and so often and so regularly in each other’s presence; friends are rarely so sharply segregated from nonfriends; and friends of one person are rarely to such an extent also friends of each other. Whyte described a situ ation that was unusual, arising as it did from the historical accident of a major depression in a lower-class environment of second-generation im migrants.
Several themes above have been investigated in later, more specific studies of friendship. Elizabeth Bott (1957) has investigated friendship networks, stressing the difference between connected networks, where one’s friends are also friends of each other, and open ones, where friends do not make up an interconnected group. Her exploratory study of families and their friends indicates that if a married couple is involved in a close-knit set of friends, the couple tends to have a rigid separation of roles in the household. On the other hand, if the network of friends is loosely connected, separation of roles among husband and wife is at a minimum.
Adolescent friendship
In his study of friendship among young men, Whyte related variants of friendship to career ambitions in distinguishing between “corner boys” and “college boys.” Friendship among adolescents may be more significant than at earlier and later ages, both in general and specifically for career choice. What is the nature and significance of friendship among adolescents? In a discussion of David Riesman’s hypothesized “other-directed” personality type, with its assumed peer dependence among adolescents, Parsons and White (1961) see adolescent friendship as a mechanism for loosening children’s dependence on parents and as a channel for the testing of career choice. Relying on findings from studies of friendship choices in high schools, they conclude that the adolescent’s dependence on peers is far from a seeking of free-floating approval; relations among peers partially consist of commitments to normative standards. Parsons and White emphasize that there are two kinds of normative culture among high school students: one is rather hedonistic, characterized by much value on popularity and reluctance to accept the achievement orientation in the adult world, and the other, which is somewhat less fre quent, includes a strong commitment to mastery of this achievement orientation. The availability of friendship cliques (of either type of culture) is both a sorting mechanism and a testing ground for longterm educational and career commitments.
The finding of two distinct normative cultures among adolescents, one centering on hedonism and the other, nearly as important, on scholastic achievement, is unusual. Coleman (1961) found widely varying social climates in the ten high schools that he studied. In all of them, athletic achievement was the major basis for recruitment to the informal elite of the school; scholastic achievement was always a decidedly minor basis; the strongest position of all was held by the allrounder. Gordon (1957), in a case study of a Midwestern high school, found a salient structure of dominant cliques, within which close friendship might occur. Cliques centering on scholastic achievement were quite unimportant, although students showed increasing fulfillment of school expectations for scholastic achievement with each additional year in school. The students’ school and social life seemed to be dominated by a fierce competition for prestige that was achieved on the basis of conformity to highly developed normative patterns. Adolescent life in school was dominated by efforts to achieve a differentiated social status —that is, essentially a search for identity. Membership in a group gives the protection of being visibly and actively “somebody,” but friendship appears to be largely an instrumental and regulative structure rather than a supportive and permissive one. However, the detailed regulation of behavior implied in the adolescent culture is partly (as in dress) a symbol of autonomy relative to adult society and partly (as in puritan morals) an acceptance of explicit adult values.
In spite of variation in content of adolescent culture and friendship groupings, Parsons and White’s main points are valid: adolescent culture, with cliques forming around variant values, affords standards of right and wrong, affects career choice, and enforces a degree of independence from parents. Nevertheless, the central place of popularity in this adolescent culture leaves Riesman’s claim of a growing “other-directedness” an open issue.
Friendship in work groups
Adolescent friendship approaches being a way of life, although a transitional one, related to love and a future family as well as to work and a future career. Friendship is also important in careers. The influence of friendship on recruitment to and work in a career has been shown in community studies (Warner & Lunt 1941, pp. 188-199; Seeley et al. 1956, p. 135). The importance of friendship in one line of work was shown dramatically by Coleman, Katz, and Menzel (1957): doctors’ adoption of new therapeutic drugs depends to a considerable extent on membership in informal friendship cliques.
The theme of friendship in work has been a significant one in social science at least since the Hawthorne studies in the 1930s. In one setting it was found that a differentiated set of friendship relations among male workers supported a broad normative orientation, including a norm limiting productivity (Roethlisberger & Dickson 1939, part 4). In another setting, friendly relations among female workers were suggested as an explanatory variable behind a continuously climbing curve of productivity under varying and controlled external work conditions (ibid., part 1). The concept of an informal social structure in organizations has been with us since these studies. With respect to friendship or friendly human relations, these two findings have been duplicated over and over again, in industrial and bureaucratic settings as well as in research on problem solving in laboratory groups: friendly relations may lead to either increased or decreased productivity. The normative basis of friendship, particularly the definitions of relations to higher authorities, is presumably one decisive differentiating variable.
Selection of friends
The initial view of friendship given in this article is a romantic one; it is empirical in that it formulates surface values concerning friendship in our culture. However, while it may be true that friends are culturally expected to be chosen im pulsively, it is certain that choice follows socially structured paths. Friends tend to share social position. The tendency of friends to be alike is well illustrated by a finding in a study of a political election that compared voters’ choices with those of their best friends. A sample of voters was divided into four subgroups according to whether three, two, one, or none of their best friends intended to vote for the Republican party; among those whose friends were all intending to vote Republican, 61 per cent expressed strong intentions of voting Republican, and the percentage dropped to 37, to 23, and to 2 for the other three subgroups, respectively (Berelson et al. 1954, p. 99). Two explanatory principles are needed: friends select each other on the basis of similarity, and they in fluence each other to become similar. The system atic study of similarity and dissimilarity between friends was opened up by Lazarsfeld and Merton (1954) when they asked how selectivity comes about and how it varies for different kinds of attributes and within different kinds of social structures. While they clarify many issues in their article (for example, by distinguishing between status-homophily and value-homophily), their contribution has remained programmatic.
Similarity in attitude is one basis on which friendships are formed. Longitudinal studies (e.g., Newcomb 1961) indicate that the explanatory principle of selectivity in terms of similarity outweighs the principle of similarity resulting from friendship. The balance of the evidence for the view that friends select each other on the basis of complementarity of needs rather than on the basis of similarities is negative (Secord & Backman 1964). Findings based on sociometric studies of friendship concerning other bases of choice have been summed up as follows:
. . . a person is likely to choose the following individuals: (1) those with whom he has a greater opportunity to interact, (2) those who have characteristics most describable in terms of the norms and values of the group, (3) those who are most similar to him in attitudes, values, and social-background characteristics, and (4) those whom he perceives as choosing him or assigning favorable characteristics to him, (5) those who see him as he sees himself, and (6) those whose company leads to gratification of his needs. (Secord & Backman 1964, p. 247)
Two theoretical formulations with implications for the study of friendship have appeared in recent years (Thibaut & Kelley 1959; Homans 1961). Both are exchange theories of interaction, where basic terms are costs, rewards, outcomes, and comparison levels. This framework allows the analysis of steps in attraction processes (Secord & Backman 1964, chapter 7): friendship is the outcome of sampling and estimation, bargaining, commitment, and, finally, institutionalization.
Problems for research
Both sociometric findings on friendship and Secord and Backman’s exchange analysis of friendship focus on the initiation of friendship structures; that is, their central theoretical topic is attraction rather than established friendship relations. Research on friendship needs more concentration on the substantive contents of friendship itself. How, in fact, are rights and obligations in friendship experienced in various social environments? What are the institutional and actual encouragements and limits to friendship in various contexts—politics, business, everyday life? What are the major rewards and strains in friendships? Is the ambiguity of friendship a circumstance that serves to initiate as well as to terminate other kinds of relationships? Under what conditions will friendships end? Is friendship, more often than other types of relationships, a subjectively sustained reality in the face of decreased overt inter action? For partial answers to such questions, and for the generation of other specific and significant questions about friendship, analysis of detailed reports on the behavior and orientations of friends, acquaintances, strangers, and enemies in everyday life is required. It would seem that such studies, which demand more descriptive patience than we now see in social science, can give a rich yield, since friendship, as a kind of cement in personality and social fabrics, is probably more strategically related to other social relationships than research has indicated so far.
Odd RamsØy
[Other relevant material may be found inLeadershipandSociometry.]
BIBLIOGRAPHY
Aubert, Vilhelm;and Arner, Oddvar1959 On the Social Structure of the Ship. Ada sociologica 3:200–219.
Berelson, Bernard; Lazarsfeld, Paul F.; and Mcphee, William N. 1954 Voting: A Study of Opinion For mation in a Presidential Campaign. Univ. of Chicago Press.
Bott, Elizabeth1957 Family and Social Network: Roles, Norms, and External Relationships in Ordinary Urban Families. London: Tavistock.
Cohen, Yehudi A. 1961 Social Structure and Personality. New York: Holt.
Coleman, James S. 1961 The Adolescent Society: The Social Life of the Teenager and Its Impact on Education. New York: Free Press.
Coleman, James S.; Katz, Elihu;and Menzel, Herbert 1957 The Diffusion of an Innovation Among Physicians. Sociometry 20:253–270.
Eisenstadt, Shmuel N. 1956 Ritualized Personal Relations. Man 56:90–95.
Goffman, Erving1958 The Characteristics of Total Institutions. Pages 43-84 in Symposium on Preventive and Social Psychiatry. A symposium held at the Walter Reed Army Institute of Research in 1957. Washington: Government Printing Office.
Gordon, Cwayne1957 The Social System of the High School: A Study in the Sociology of Adolescence. Glen-coe, 111.: Free Press.
Homans, George C. 1961 Social Behavior: Its Elementary Forms. New York: Harcourt.
Lazarsfeld, Paulf.; and Merton, Robert K. 1954 Friendship as a Social Process: A Substantive and Methodological Analysis. Pages 18-66 in Morroe Berger, Theodore Abel, and Charles H. Page (editors), Freedom and Control in Modern Society. Princeton, N.J.: Van Nostrand.
Naegele, Kaspard. 1958 Friendship and Acquaintances: An Exploration of Some Social Distinctions. Harvard Educational Review 28:232–252.
Newcomb, Theodorem. 1961 The Acquaintance Process. New York: Holt.
Parsons, Talcott;and White, Winston 1961 The Link Between Character and Society. Pages 89-135 in Seymour Lipset and Leo Lowenthal (editors), Culture and Social Character: The Work of David Riesman Reviewed. New York: Free Press.
Roethlisberger, Fritzj.; and Dckson, Williamj.(1939) 1961 Management and the Worker: An Ac count of a Research Program Conducted by the Western Electric Company, Hawthorne Works, Chicago. Cambridge, Mass.: Harvard Univ. Press.? A paperback edition was published in 1964 by Wiley.
Secord, Paulfw.; and Backman, Carl. 1964 Social Psychology. New York: McGraw-Hill.
Seeley, Johnr. et al. 1956 Crestwood Heights: A Study of the Culture of Suburban Life. New York: Basic Books.
Thibaut, Johnw; and Kelley, Harold.H. 1959 The Social Psychology of Groups. New York: Wiley.
Tunstall, Jeremy 1962 The Fishermen. London: Mac-Gibbon & Kee.
Warner, W. Lloyd;and Lunt, Paul S. 1941 The Social Life of a Modern Community. New Haven: Yale Univ. Press.
Whyte, William F. (1943) 1961 Street Corner Society: The Social Structure of an Italian Slum. 2d ed., enl. Univ. of Chicago Press.
