Introduction

Few concepts have become so entrenched in sociologists’ minds as the notion of social role. We struggle in trying to imagine how sociology could do without the idea.Footnote 1 Yet until the late 1930s, social role played little if any part in the thinking of most sociologists (and likewise in the thinking of most anthropologists and social psychologists). Several questions accordingly arise. How did sociologists manage without the concept social role? How did they otherwise describe the social agent and his agency? When and in what circumstances was the term social role initially formulated and when did it enter the vocabulary of social science?

This article casts light on the history of the term role and its concepts in sociology (and related disciplines) through the first four decades of the twentieth century. It is shown “role” was not used in one uniform way, sociologists having diverse understandings of the idea, and through those decades “personality” was a more important term than “role” in the lexicon of social science. The article describes how the contents of sociology and psychology altered in the 1930s, being part of the context in which role became more clearly delineated while “personality” established itself in the habitat of specialised branches of psychology, leaving “role” a niche to occupy in sociology.

The suggestion that “role” appeared infrequently in sociology in the first four decades of the twentieth century may be questioned by some readers in light of English culture’s long inclusion of the image of Shakespeare’s comedy “As You Like It” (1599 or 1600): “All the world's a stage, And all the men and women merely players: They have their exits and their entrances; And one man in his time plays many parts, His acts being seven ages” (the ages of infant, schoolboy, lover, soldier, rotund old age, “slipper’d pantaloons”, senility). Shakespeare did not mention role as such, but he employed the language of theatre in referring to “parts”, “players”, “stage”, “exits”, etc. Shakespeare’s term “parts” includes some of the meaning of the thespian “role” along with suggestions of the parts being defined in and by society, and also these parts corresponding to biological tendencies of people to behave in certain ways at each stage of the human life cycle. It is widely accepted in sociology today that social role is an analogical extension from theatre, with agents performing roles in social institutions as actors perform them in plays. The analogy, however, is not essential (not all sociologists invoke it), and it cannot be pressed hard (thespian roles include explicit scripts but not rights and responsibilities and in these respects they are the opposite of social roles). Given the long history of Shakespeare’s idea of parts performed on the social stage it may be argued the understanding of role that is common in sociology these days has always been part of it, and therefore the topic of this article is misconceived. The following discussion based on a clear array of evidence should serve to rebut any such objection.

Context

Roger Smith (1997, 565–566), in what Jill Morawski (2012, 22) rightly describes as his “exquisitely compiled history of the human sciences”, points out that “from the 1890s to World War 1” sociology became an established academic discipline in the US and a significant occupation besides, with the discipline taught in approximately 400 American colleges. These were also years of the coming of age of psychology in the US. America’s first fulltime professor of sociology (1894) was Franklin Giddings at Columbia, The American Journal of Sociology was founded by Albion Small in 1895, and the American Sociological Society was formed in 1905 with Lester Ward its first president. At Chicago Small headed the first department of sociology in the US, seeing to its expansion into a powerhouse of American sociology.

In America, sociologists had different views of relations between their discipline and psychology and other social science disciplines. The same was true in Europe where, for example, Emile Durkheim stressed sociology’s content is qualitatively different from, and irreducible to, that of psychology, whereas Georg Simmel – influenced by Wundt’s understanding of Volkerpsychologie, and an influence on Small and G. H. Mead in America – blended psychology with sociology (Smith 1997, 564). Mead is emblematic of the complex relations then existing between disciplines: he was located in Chicago’s philosophy department, taught social psychology as well as philosophy, and exerted an influence not only in these disciplines but in sociology as well.

The formation of social psychology at the interface of sociology and psychology represents important context of the argument of this article. American social psychology evinced, and continues to evince albeit to a lesser degree, two orientations, one psychological and the other sociological, each disposing its supporters to favour certain theories and methods. American sociologists felt a special affinity toward social psychology, sharing its interest in agents’ “purposes, needs and activities …[and] the modes of interaction and group process which they develop” (Fine 1979, 108–109). George Lundberg (1931–1932) noted that the majority of members of the American Sociological Society in 1930 nominated social psychology as their principal area of interest (Good 2000, 391).

The first two textbooks of social psychology are often cited in evidence of a polarity in the discipline. The Englishman, William McDougall’s Introduction to Social Psychology (1908), grounded social behaviour in the individual agent, with social environments assigned “little explanatory role” and the agent’s behaviour emanating from “a more primary, instinctual basis” (Gergen 2012, 141). Edward Ross’ (1908, 1) textbook Social Psychology assumed a sociological stance, describing social psychology as studying “the psychic planes and currents that come into existence among men in consequence of their association”, investigating “uniformities in feeling, belief, or volition – and hence in action – which are due to …social causes”.

Edward Tolman (1923), Emory Bogardus (1924), Charles Ellwood (1925) and Otto Klineberg (1940) were among the scholars in agreement with McDougall that understanding social life required explanations given in terms of instinctive tendencies. Critics of social instinct theory – including Knight Dunlap (1919), John Dewey (1922) latterly, and Floyd Allport - bore a particular animus to what they believed is its doctrine of the fixity of human nature (Collier et al. 1991, 32, 33, 92).Footnote 2

Floyd Allport was committed to making social psychology scientific, following the example of his Harvard teachers, particularly the experimental approach that Hugo Munsterberg advocated, and the behaviourism of Edwin Holt (Parkovnick 2000, 430–431). Averse to notions of the “group mind”, Allport (1924: 4) insisted “Social psychology must not be placed in contradistinction to the psychology of the individual; it is a part of the psychology of the individual, whose behavior it studies … Within his organism are provided all the mechanisms by which social behavior is explained.” From the mid 1920s Allport used his publications and his position as associate editor of The Journal of Abnormal Psychology and Social Psychology to promote experimental research and the behaviourist perspective in social psychology. “The establishment of experimentation as the distinct psychological approach to social psychology was closely connected with the general turn toward behaviorism. Reflecting the behaviorist view of psychology as a natural science, Floyd Allport played a key role in proselytizing for an experimentally based social psychology” (Collier et al. 1991, 94–95).

Collier and his coauthors (1991, 110, 125, 143) indicate that through the 1920s the individualist perspective was to the fore in American social psychology whereas Junius Brown (1936) and others exemplified a countertrend of American social psychology of the 1930s that deemphasized individual initiative and autonomy and stressed social conditions of, and group responses to, issues of poverty, prejudice, aggression and inter-group conflict. The “socialising” trend was apparent in studies of the social environment’s impact on agents’ attitudes, behaviour, perception, and personality disorders. Social psychologists were confident their discipline could help in ameliorating social problems, reminiscent of the outlook of Ross and others of the Progressive Era (Nicholson 2003, 192, 194). One needs to guard against overgeneralizing this development. A Floyd Allport, for example, might on the face of things seem to have gone along with the socialising trend of the 1930s by investigating social processes of conformity and facilitation. But his confidence in ontic and explanatory individualism remained unshaken and it was in this light that Robert Lynd (1939, 22 n.8) complained about Allport remaining oblivious to “the special qualities of behavior in group situations”, and misrepresented “institutions as derived from the behavior of individuals by a simple additive process”.

Muzafer Sherif and Kurt Lewin each undertook experimental investigations of small artificial groups formed ad hoc, contributing to the fact that by “the early 1940s experimental social psychology had moved into the forefront within psychological social psychology, and it now dominated the study of both individuals and groups” (Collier et al. 1991, 141). Emphasizing experimental methods, the psychological orientation would eventually so dominate the discipline that John Greenwood (2004) could obituarize the “disappearance of the social in American social psychology”.

Another notable trend in social psychology was an increase in the number of studies made of personality. There had been study of personality in nineteenth century French psychology and American social psychologists became interested in the topic early in the twentieth century (Barenbaum and Winter 2008, 5, Smith 1997, 600, Lombardo and Foschi 2003, 125ff.). Freud produced the first comprehensive theory of the formation and nature of personality and he lectured on it in the United States in 1909 where the response to his ideas was mixed (Parkovnick 2000, 431–432).

Psychological interest in personality followed the pioneering efforts of Walter Dill Scott in devising tests for helping business to recruit staff with personalities suited to sales careers and, in World War I, for helping the US army to recruit civilians suited to military service. “The use of tests grew rapidly after 1920 in commerce and industry …as well as in education …[and the] interest in personality was the background to [testing as] a major growth area of psychology in the interwar years” (Smith 1997, 599). Kurt Danziger (1990, 163) observes “The construction of personality traits as objects of investigation led to the emergence of a brand new area of psychological research. In the American Psychological Association Yearbook for 1918, not a single psychologist listed ‘personality’ as a research interest, but by 1937, 7 per cent did so.”

Allport’s 1924 textbook cited Freud’s discussion as deserving of investigation, the textbook including two chapters on personality. Luther Bernard’s (1926) Introduction to Social Psychology took up the subject of personality, and Gardner and Lois Murphy’s (1931) textbook, the first in experimental social psychology, included personality formation in its coverage. In cultural anthropology Franz Boas, Ruth Benedict and Margaret Mead led the way in studying the cultural conditioning of personality, often assuming Freudian psychoanalytic theory and in some instances testing its implications (Nicholson 2003, 195). Kimball Young’s (1930, 1) social psychological textbook revolved the discipline around studying “personality as it develops in relation to social environment”. The specialist journal Character and Personality was launched in 1932, and Psychological Abstracts added personality to its list of categories in 1934, while Lewin’s A Dynamic Theory of Personality appeared in 1935. Gordon Allport’s (1937, 48), Personality book revealed forty nine different definitions of personality before proposing the term be defined as “the dynamic organization within the individual of those psychophysical systems that determine his unique adjustments to his environment”, specific individuals being his focus.

“Precursors”?

Its deserved classic status notwithstanding, Thomas and Biddle’s (1966) history of the idea social role stands in need of correction and augmentation. They (1966, 4) cite a number of “precursors” of (“contributors” to) what they describe as the “role perspective” of members of society acting under the influence of “social prescriptions and behavior of others.” “To our knowledge”, Thomas and Biddle (1966: 6) write, “it was not until the decade of the 1930’s” that American social scientists began using the term role “technically in writings on role problems”, with J. L. Moreno (1934), G. H. Mead (1934) and Ralph Linton (1936) cited as the chief exponents.

Moreno used “role” on several occasions and in various ways in his major work, the pioneering sociometric, Who Shall Survive? (1934). For example he (1934, 15, 60, 196, 226, 230, 232, 295) wrote of the “role of submission” (role pertaining to a state or condition a person may be in), of a child’s “role …in the group” (role as position), role as “played” (an agent’s part), “the role of leadership” (capability or capacity), “role in this game” (role as position and/or part), “reciprocating role” (role as contribution), “role in other tests” (as function), and so on. Moreno’s interest in roles arose out of his psychodramatic approach to therapy involving patients acting “out their conflicts by playing roles in a social context” (Wepman 2000). Moreno arranged for patients to act “on a stage with a therapist suggesting actions or scenes …to perform, and sometimes includ[ing] dramatic techniques” such as “role reversal” whereby the patient would assume “the role of a person with whom he or she is in conflict, or interacting with other patients serving as foils.” By re-enacting and reflecting on their experience, patients gained “insights into their situations and relationships and” were provided with “methods of coping with interpersonal problems” (Wepman 2000).

In his posthumously published book Mind, Self and Society (1934) G. H. Mead repeatedly spoke of “taking the rôle of the other”. In common with most readers of Mead (e.g. Burr 2002, 67, Emmet 1966, 156–157, Swingewood 2000, 168–169), Thomas and Biddle (1966, 6–7) construe “taking the rôle of the other” as Mead referring to the social parts people play. Gary Cook’s (1993, 78ff.) exegesis refutes that interpretation, highlighting Mead’s clear dissociation of “role” from agents playing parts in social life. The understanding of rôle as social part, Mead (1934, 161) specifically cautioned, “is actually more sophisticated than …[the idea of role] which is involved in our own experience. To this degree it does not correctly describe that which I have in mind.” “Taking the rôle of the other” in Mead’s (1934, 162–163) text means a social agent accepts his community’s basic attitudes, adopting its dispositions to respond to events in specific ways. People without these shared attitudes, Mead (1934, 163) considered, may comprise an assemblage or loosely structured “group” but they will lack the integration needed to form a “community”. Being able to take the role of the other, Mead means members of a community have a common outlook, enabling them to understand and appreciate one another’s actions. The attitudes held in common with other members of the community are the source of an agent’s “personality”, “character” or “self” (Mead 1934, 162–163). The Meadean agent is “self-conscious” when he “call[s] … out” (summons or brings to mind) the “responses which go to make up the members of the community” (Mead 1934, 163). “The structure …on which the self is built is this response which is common to all, for one has to be a member of a community to be a self” (Mead 1934, 162, also 254–255). (Cook’s understanding of Mead’s role concept is comparable with that of Faris in his 1937 discussion of Mead’s social psychology.)

For reason of expository convenience we will hold off considering the contribution of Ralph Linton as the third of Thomas and Biddle’s main “precursors” of the role perspective.

Role in Social Science in the Early Decades of the Twentieth Century

In the first four decades of the twentieth century few scholars used the term “role” and those who used it did so in diverse ways. The scholars of these decades who mentioned social roles typically assigned the term their own special meanings, but with two commonalties, the first being they typically subordinated “role” to “personality” and the second that they coupled it with “status”.

Besides misconceiving or at any rate exaggerating Moreno and Mead’s significance as “precursors” of the role perspective, Thomas and Biddle overlook significant writings of several scholars that are indeed entitled to be described as “precursors” of it. Thomas and Biddle mention Georg Simmel’s 1920 essay “On the philosophy of the actor” (“Zur Philosophie des Schauspielers”) but ignore his much more substantial and influential writing, “How is Society Possible?” of 1910, being among the Simmel writings Albion Small translated for publication in the American Journal of Sociology. Simmel in “How is Society Possible?” mentioned “role” only once but severally referred to “position” which in essence is his idea of social type as exemplified in his discussions of the stranger, the spendthrift, the miser, the metropolitan type, the poor, etc. Simmel had the concept social type although it seems he never used the term itself (either in English or German) (Silver 2007; Almog 1998). In Almog’s (1998) thoughtful analysis, Simmel envisaged each social type as a complex “of certain psychological traits - as a specific personality, temperament, or mentality formed by structural forces, …human situations, and ecological conditions”. Coser (1971, 182) writes of the social type of Simmel’s agent being formed “through his relations with others who assign him a particular position and expect him to behave in specific ways. His characteristics are seen as attributes of the social structure.” Almost certainly thanks to Simmel’s influence, “social type” is a notion that appears time and again in the literature to which the present article recurs.

Thomas and Biddle make no mention of philosopher-psychologist James Mark Baldwin’s book of 1911, The Individual and Society or Psychology and Sociology.Footnote 3 Envisaging society and the individual, sociology and psychology, as complementary pairs, Baldwin (1911, 87) explained how younger members of society become equipped to earn a living by serving an apprenticeship that is “very conventional and …stereotyped” after which they come to be “classified as carpenter, butcher, clerk or telegraph operator.” Younger people are “trained in the requirements of actual life, and made ready for the roles of citizen, parent, wage-earner, etc.”, having to learn “from parent and teacher, the lessons of self-control …by which the status of each in his class and place are [sic.] established and maintained” (Baldwin 1911, 125 emphasis added). Baldwin suggested an individual’s roles are more fundamental than, and help in fixing, his statuses. Roles for Baldwin are forms of conduct deemed appropriate in a society and statuses are positions in social life corresponding to socially sanctioned actions.

From the 1920s the Chicagoans Robert Park and Ernest Burgess made a number of statements in trying to come to grips with the idea, social role. Their anthological textbook of 1921 Introduction to the Science of Sociology, in which Simmel has a significant presence, affirmed “The individual's self-consciousness — his conception of his role in society, his ‘self,’ in short … is based on his status in the social group or groups of which he is a member” (Park and Burgess 1921 , 55 emphasis added; Levine et al. 1976, 824). As had Baldwin, Park and Burgess connected role with status but unlike him they considered status to underlie role. A person has status in a social group, and different statuses in the various social groups to which he belongs and, in tracing out the implication of Park and Burgess, we see his status (or statuses) indicates the person’s “role in society” (singular) understood as the person’s broad view of where and how he fits into society and contributes to it.

A short essay by Park (1926, 135), “Behind our Masks” (again, uncited by Thomas and Biddle), explained recent Japanese-American racial tensions in reference to “what the Chinese call ‘face’”. Park explained that etymologically “person” originally signified “a mask”, reflecting the fact that members of society are invariably “more or less consciously, playing a role” (Park 1926, 137 emphasis added). “We are parents and children, masters and servants, teachers and students, clients and professional men, Gentiles and Jews” and “it is in these rôles that we know ourselves.” Describing our faces as “living masks” that tend “to conform to the type we are seeking to impersonate”, Park (1926, 137) understood the “mask” as representing our self image and as “the rôle we are striving to live up to …We come into the world as individuals, achieve character, and become persons.” In Park’s understanding a person plays several roles, each one having its normative dimension which the player is expected to embody and express. Among his examples are occupations, family positions, and religious affiliations of Jew and Gentile. Status is a term appearing once only in this writing of Park (1926, 135) and it contributes nothing to his argument.

Burgess’ (1929, 122, see also Burgess 1923) essay “The Family and the Person” depicts “roles” of “husband and wife, parents and children” as constituting the family, describing them as “cultural patterns” with “a history” and as “subject to social change. At the same time”, Burgess thought, “these familial roles are idealized by the members of the family. The stern but just father, the loving and prudent mother, the dependable and honest son, the dutiful and virtuous daughter are ideals toward which conduct is directed and by which shortcomings are measured.” Family members “assume roles” and “the family itself sets up claims and obligations which tend to …transcend the rights and even the individuality of its members.” The member of Burgess’ family receives rights, and incurs obligations, in virtue of assuming one of its historically patterned roles. Whether the family is representative of other social objects in respect of its having a role structure Burgess did not specifically say.

Edward Sapir included “role” in the psychoanalytically informed course of lectures “The Psychology of Personality” he delivered at Columbia and Chicago universities, and at Yale following his move there in 1931, his intention being to eventually turn the script into a book that would present “a major theoretical statement of Boasian cultural anthropology” (Irvine 2002, 7). Sapir (2002, 143–144) suggested the plurality of an agent’s roles in referring to the “series of roles, or [modes of] participation in society” that he “carves out for himself or takes part in.” Some roles, as Sapir saw it, are created by agents and others already exist for agents to occupy.

In his “Discussion” of The Jack-Roller, Clifford Shaw’s (1930) life history (“own study”) of the genesis, and experience, of the delinquent, Stanley, Burgess (1930, 193–194) referred to the role that “a person assumes and to which he is assigned by society” as creating his “social type” and, in a writing of the following year, explicating Park’s point that “‘man is not born human’”, Burgess equated a person’s “role” – his achieved social position (social place) – with his “personality” (Burgess 1931, 193–194). Burgess on these occasions was envisaging a person as having one role only.

Kimball Young (1934, 77 emphasis added) invoked the dramaturgical analogy in An Introductory Sociology, explaining how the social agent plays “his various parts upon the stage of life”, passing “out at his last scene, leaving the drama in the hands of those who come after him.” Notwithstanding his having spoken of the agent’s “various parts”, Young (1934, 77 italics added) proceeded to suggest the agent has one part only (shades of Burgess in 1930 and 1931), “society and culture prepar[ing] and fix[ing] the individual’s part in this drama”. Perhaps confused and surely confusing, Young (1934, 88, 93 emphasis added) went on to note how “Each group lays out various roles for the members” of society (suggesting the individual in belonging to different groups has several roles), adding that each adult person has a “dominant social type or role”, being largely determined by “culture and social organization.” Role as “dominant social type” represents the “pattern or type of behaviour which the child and adult builds up in terms of what others expect or demand of him”, Young (1934, 88) citing a parent who repeatedly “tells Johnny he is ‘no good’ or is a ‘black sheep’ or, on the contrary, builds up in the boy an expectation of ‘good’ conduct.” Repetition of the message, Young (1934, 89) believed, encourages a child to conform his conduct with “these definitions. The role, therefore, is related to one's acceptance of the definition of the situation by others.” Aware that “role and status have often been used almost interchangeably in sociology”, Young (1934, 89 emphasis added) distinguished them as having different referents. “Status has to do with the position, the standing of the individual within the group accorded him by his fellows” whereas “The role is what you do or do not do. It is concerned with activity. Status is the resultant place [anywhere from low to high] on the prestige scale” (Young 1934, 89). This was clarification of a sort, although some readers might complain Young’s understanding of status was a composite and confusion of different things, one of them being a person’s social position qua occupation as an objective fact and the other being the quantum of prestige that other agents subjectively accord him and his social position. And to what type of role was Young referring when he observed “The role is what you do or do not do”? Was it the singular role simpliciter, or the roles provided the agent by the different groups to which he belongs, or was it the individual’s role in the sense of his “dominant social type”? He may have had one, some or all these types of role in mind.

Young (1934, 94) confirmed Burgess’ immediate influence, and Simmel’s likely mediate influence, on him in quoting from Burgess’ “Discussion” of Shaw’s (1930) Jack-Roller: “The term,social type’” refers to “attitudes, values, and philosophy of life derived from copies presented by society. The role which a person assumes and to which he is assigned by society creates the social type”. So a person of a particular social type – for example “‘a professional runaway,’ ‘a delinquent,’ ‘a criminal,’” – has that type with its associated attitudes, values and outlook, in consequence of the corresponding role having been issued him by society, which role he has taken on (Young 1934, 94 quoting Burgess 1930, 193–194). “His acceptance of the criminal code and the orientation of his ambitions to succeed in a criminal career have to do with attitudes and values and are elements that enter into the creation of a social type” (Young 1934, 94, quoting Burgess 1930, 193–194). Young’s idea of social type as a person’s salient social identity seems clear enough whereas his idea of a person’s “role” remains hazy and is in the background of his discussion.

Personality

Used significantly by relatively few scholars –Mead, Burgess, Park and Young being the most notable - role was uncommon and marginal in sociology before the 1940s. Most sociologists in the so-called “second period of American sociological theory” (Hinkle 1994, ix) were generally unaware of the term or else they could find no place for it in their analyses.

Even the scholars we have identified as precursor or pioneer users of “role” said little about it and seldom employed it. We noted earlier how the concept personality entered American social psychology in the 1920s. Mead and the other scholars just cited were representative of a good many other social psychologists and social scientists in favouring “personality” as their principal term for discussing people as members of society, or else including “role” among their preferred terms along, and on a par, with personality, person, individual, character and self. Among the senses of “personality” displayed in The Oxford English Dictionary, the best approximation to the meanings of social scientists in the first four decades of the twentieth century is “The quality or collection of qualities which makes a person a distinctive individual; the distinctive personal or individual character of a person.” On the other hand, some social scientists took a leaf out of William Thomas and Florian Znaniecki’s The Polish Peasant in Europe and America, preferring to investigate “personality types” rather than specific personalities of individuals.

Baldwin (1911) made extensive use of “personality” along with “individual” and “self”. “Personality”, “person” and “character” appear frequently in Park and Burgess’ (1921/1924) textbook (especially chapter 2, “Human Nature”, sections C and D), followed by the terms “self” and “individual”.Footnote 4 Park and Burgess (1921, 70, see also 114) explained “The word personality is derived from the Latin persona, a mask used by actors. The etymology of the term suggests that its meaning is to be found in the rôle of the individual in the social group”. Some readers may have taken this to mean that role explains personality, but Park and Burgess had a different understanding in mind, explaining that personality represents “the sum and organization of those traits which determine the rôle of the individual in the group”, citing traits of physique, temperament, character, and prestige. For Park and Burgess at this rate, role is the product of personality traits.Footnote 5

Writing in “The Study of the Delinquent as a Person” Burgess (1923, 662–663) declared “In sociology the distinction is now clear between the individual and the person. The study of the individual, of the reaction of the organism to its environment, falls in the fields of psychiatry and psychology. The study of the person, the product of social interaction with his fellows, lies in the domain of sociology. Park thus defines the person …[as] an individual who has status. We come into the world as individuals. We acquire status, and become persons. Status means position in society. The individual inevitably has some status, in every social group of which he is a member. In a given group the status of every member is determined by his relation to every other member of that group.” Status as a person’s position in a group is determined by the tissue of his social relations with others of the group.

Park (1926, 137 emphasis added) believed “our conception of our role” eventually becomes “an integral part of our personality” while Sapir (2002, 139–190) suggested personality and personality type summarise an agent’s roles, Sapir’s discussion including points of which we have been apprised from other sources, as for example the etymology of “personality” having its roots in the Latin “persona” (mask) and personality being linked historically with status, with each person functioning “in the part laid out for him by society” (2002, 142 last emphasis added and 144). Sapir (2002, 144) noted the personality idea expressed in Shakespeare, and he distinguished current uses of “personality” according to different disciplinary definitions, the prevalent definition of sociology equating personality to the “roles, or [modes of] participation in society, which a person carves out for himself, or takes part in.” Sapir believed this definition needed augmentation with the psychological fact of “the feeling of consciousness.”

Pitirim Sorokin’s Contemporary Sociological Theories (1928, 55–56, 85–92, 713–728) included no discussion of role in any sense that bears on our topic, but his consideration of “psycho-sociologistic” thinking and other tendencies of social thought included some reference to personality, including distinctions drawn by social scientists between personality types. He (1928, 474) noted also that Emile Durkheim in his The Elementary Forms of Religious Life had included personality among the general, socially induced ideas of the human mind.

Personality was the key concept in Burgess’ 1929 essay (see above), the essay appearing in a collection titled Personality and the Social Group, and the following year Burgess’ (1930, 193) “Discussion” of Shaw’s (1930) The Jack-Roller separated role (social type) from personality, including compounds such as “personality traits”, “personality type” and “personality pattern”, with “role” seldom mentioned. Based on Shaw’s life history of Stanley (the life history method being commonly employed by Chicago sociologists of the first two generations (Murray 1986, 242–243)), Burgess (1930, 190–192) drew a “clinical picture of …[Stanley’s] egocentric personality” as revolving around his burning “sense of injustice” with traits of self-pity, hypercriticism, narcissism, suspicion, rashness and resentment; Stanley maintaining “his ego” in the face of social adversity so as to better cope with the “disappointments of life”. Burgess delineated the essentials of four different personality types and instances of these types he saw as crystallizing in childhood and proving resistant to major change in adulthood. “Stanley, no more than anyone else, is neither to be praised nor blamed for his personality traits. They were formed for him before he gained conscious control of his destiny. The point to be grasped is that the formation of the personality pattern is a natural product of forces in the constitution of the individual and in his childhood situation. Once this conception of behaviour is clearly understood, we will learn to accept people as they are and work with, rather than against, the basic set of their personality” (Burgess 1930, 193). Young took a number of these thoughts across to his sociology textbook of 1934.

In The Concepts of Sociology, his exceptionally extensive work of survey and commentary of 1931, Earl Eubank (106) rendered “self”, “person” and “personality” as closely connected concepts, one of his sections dealing with “The Single Human Being as a Combination of Persons”, including a subsection explaining “The Concept of the Person: The Situation Self”. Eubank found that “self-examination” assures us “that we are many [selves], as many in fact as there are groups or situations calling upon us.” This echoed William James’ (2001/1950, 294) understanding of a man having “as many social selves, as there are individuals who recognize him and carry an image of him in their mind.” By way of illustration Eubank (1931, 107) commented that on “Entering the university campus[,] each of us becomes a different self as we encounter our university associates. We even change selves” between class rooms, drawing upon the “particular aspects and knowledge” of each class which “elicits its particular responses ... Each one of these situations has called upon certain elements of the self and from them created a combination of attitudes which are characteristic of it alone.” These quotations show Eubank was uncertain whether each audience elicits different “elements” of an agent’s self (in this case a lecturer’s) or a different self.

We find in Young’s sociology textbook of 1934 (84) the same tendency we have revealed in a number of the other social scientists who were given to affirming social roles, his discussion of role proceeding in the context of, and in subordination to, the concept personality. Young (1934, 84) differentiated between “social self” and “personality”, envisaging personality in terms “of the habits, attitudes, and ideas which are built up around both people and things”. He (1934, 94) accepted the distinction of “personality type” and “social type” as drawn by Burgess (1930, 193–194) in his “Discussion” of Shaw’s (1930) The Jack-Roller, quoting Burgess’ definition of “‘Personality pattern [personality type] … as the sum and integration of those traits which characterize the typical reactions of one person toward other persons’ and of ‘social type’ as the ‘attitudes, values, and philosophy of life derived from copies presented by society. The rôle which a person assumes and to which he is assigned by society creates the social type.’” It is a depiction implying a social agent has one role only and this role – socially assigned and personally taken up - forges the compound of psychology and culture that forms his “social type”.

This in essence is the idea Park and Burgess expressed in their 1921 textbook which we noted earlier in the article. Young (1934, 93–94) recalled “W. I. Thomas and Florian Znaniecki in their Polish Peasant in Europe and America (1918-1920) depicting three social types: Bohemian, Philistine and creative”, and he noted how Burgess had been encouraged by that typology to elaborate the distinction of “the personality type and the social type”. (“Personality type” appeared in a number of the contributions to Burgess’ edited work, Personality and the Social Group (1929).)

The term “personality” was more commonly used and more prominent than “role” in the sociology and social psychology of the first four decades of the twentieth century and its reference differed to that of role in that personality was used more to emphasise agents’ “mental qualities” than the pressures institutional surroundings exert on agents (Linton 1936, 464, see also Murray 1986, 243ff., Heine 2008, 55). Linton’s book of 1936 included “personality” as a suggestive and useful idea but before long the personality topic would become less visible in sociology.

Linton

A professor in the Department of Sociology and Anthropology at the University of Wisconsin, Linton (1936, vii) wrote The Study of Man as an introduction for undergraduate anthropology students to the body of anthropological knowledge.Footnote 6

Linton’s (1936 100–106, 116) major premise affirmed people have to be recruited as positions become available in society to fill, the specialised training required for each position requiring imitation as well as explicit instruction as to society’s conscious “ideal patterns” (“systems of ideas” or exemplary models), each such pattern indicating the rights and duties associated with a position. Each pattern prescribes in which activities the position holder is required to engage and in which reciprocal relations to enter, the agent in modern society having to learn several of these patterns with the rights and duties they define (Linton 1936, 107–108).Footnote 7

The “ideal patterns”, Linton (1936, 99, 102) explained, originated in past behaviour, exist in the minds of many individuals, and evolve through innovative conduct as when “certain women, as individuals, decided that they wanted to work in offices and did so in spite of the fact that they were violating the accepted patterns for ladylike behaviour.’” Since each of their personalities and circumstances are unique in some respects, social agents only approximately conform to, and never perfectly replicate, the prescribed patterns of activities (e.g. bachelor, student, politician) and social relationships (e.g. between husband-wife, siblings, doctor-patient, employer-employee).

Linton’s (1936, 114–115) direct discussion of role (Chapter 8, “Status and Role”) is confined to a couple of pages, status (status position) being his primary concept and receiving much more of his attention. He (1936, 113) envisaged statuses as “polar positions” in “patterns for reciprocal behaviour” and he understood each person as having several statuses. In exercising the rights and duties of a status position for which he has been trained, Linton’s (1936, 114) agent is “performing a role.” Statuses and roles are tied together, each status-role pair deriving from, and expressing, a pattern.

He (1936, 115–118, see also 119–132) distinguished statuses into ascribed (the prevalent kind in modernity, including statuses that “take care of the …day-to-day business of living”) and achieved. The distinction is respectively between statuses, their occupations and activities that in effect are assigned to people based on natural or social “reference points” (e.g. sex, age, family relations, class or caste) and, on the other hand, statuses for which people have to compete (e.g. President of the United States, conductor of a symphony orchestra). This distinction came to be widely accepted (e.g. Parsons 1951, 64, Emmet 1958, 189n.).Footnote 8

How original was Linton’s account of role? Not especially. How important? Including Linton in their list of major “precursors” of role theory, Thomas and Biddle suggested his discussion of role was similarly nascent, preparatory and inchoate to those of Moreno, Mead and others. Linton (1936, viii) may have encouraged such a view by describing his book as summing up knowledge that “has already [been] accomplished” in past scholarship. It would, however, be wrong to think that Linton’s book was devoid of innovative conceptual adjustments, formulations and sharpening, some of them perhaps having escaped his notice. Role can be cited as a case in point. There can be no doubting that Linton benefitted from the writings of Young, Burgess and other scholars discussed above, but at the same time he was no parrot of their ideas and explanations. Linton would be done a disservice were one to pass over the fact that in his hands, in The Study of Man, social role emerged as a honed concept, at once clearer, more distinct, more detailed and fulsome than it had been in the renderings of his predecessors.Footnote 9 It was Linton’s definition of social role that rising stars in the firmament of sociology would soon be citing, contributing to its eventual paradigm status. (Interestingly, Linton made no mention of theatre as an analogue of social life, and he followed other social scientists in retaining personality as an important category of analysis, and in yoking role with status.)

Personality more Prominent in Psychology, Role more Prominent in Sociology

We have noticed Park and Burgess’ concepts of role, and the effect Burgess’ understanding of role had on Kimball Young. We found each of these scholars assigning “personality” a superior place to “role” in their theories, and in so doing they typified a good many other sociologists of these decades whose works include ‘personality’ among their leading categories (e.g. Giddings 1901, Hughes 1928, 1937, Eubank 1931, 92–93, 103–106, Burgess 1939).

The tendency of sociologists to think in terms of personality was encouraged by the fact that in the minds of many of their practitioners, sociology and social psychology were inseparable, overlapping and complementary disciplines, Eubank (1931, 93, also Oishi et al. 2009, 334, 338, Hinkle 1994, 200) observing, “Sociology originated and has largely developed the Socio-Psychological interpretation of human society”. Sociologists and social psychologists often wrote on common subjects, sharing theoretical insights and vocabulary. E. A. Ross, author of one of the earliest social psychology textbooks (1908), was a sociologist. Thomas and Znaniecki (1958, 1831) described the second volume of their The Polish Peasant in Europe and America as applying “the methods of social psychology to an evolving human personality”. Charles Ellwood (1924, 5) defined social psychology with reference to “collective human behaviour”, deemed social psychology as comprising “a very considerable part of sociology” and titled his textbook of 1925, The Psychology of Human Society: An Introduction to Sociological Theory. Burgess’ essay of 1929 (vii) appeared in a collection edited by him - Personality and the social group – comprising essays written by “men in social psychology and sociology” with an interest in “studying personality …as a product of group life.” G. H. Mead’s interests included sociology and social psychology, and Edward Tolman (1938, 239) was convinced that study of psychology could not be undertaken unless it were suffused with sociological knowledge (Oishi et al. 2009, 335). Young’s PhD was in psychology and he taught sociology and social psychology at different times in his career as well as writing textbooks in both disciplines: Social Psychology (1930), Personality and Problems of Adjustment (1941) and An Introduction to Sociology (1934). Linton looked on Young as the scholar who “perhaps helped him most in developing his view of social organization and its relation to individual personality formation” (Sharp 1968 emphasis added). He (Linton) devoted a chapter of The Study of Man to discussing “Culture and Personality” and that together with his book of 1945, The Cultural Background of Personality, testified to his unflagging interest in the topic of personality (Manson 1986, 81–91). Talcott Parsons’ The Structure of Social Action of 1937 amply indexes “personality” (see also Parsons 1968, xi), and Robert Merton’s “Bureaucratic Structure and Personality” appeared in 1940, while Parsons’ collection of his early essays, Social Structure and Personality appeared in 1964a, b.

The publication of The Study of Man was propitiously timed for the advancement of the role concept, coinciding with two important interconnected developments in psychology. Contemporaneous with the appearance of Linton’s book, the concept personality was losing currency in sociology at around the time it was acquiring more of it in psychology. This acquisition was occurring most obviously in personality psychology, a specialism that was crystallizing in the second half of the 1930s with the publication of Gordon Allport’s (1937) Personality, the field’s “first authoritative textbook” (McAdams 1997, 124), and the publication of textbooks by Ross Stagner (1937) and Henry Murray (1938) (Smith 1997, 599–606). Danziger (1990, 239 n. 11) writes of how Allport’s “influential text” helped establish “the scope and definition of’ personality psychology as a “new field”. The creation of the specialist journal, Character and Personality (1932), and Psychological Abstracts’ inclusion of personality in its list of categories from 1934 were signs of personality being recognized as a specialised field of psychology. Allport (Allport and Allport 1921) had worked tirelessly to advance personality psychology since 1921 and his efforts were now being rewarded. Personality was being “relocated” from sociology to social psychology as Edward Reuter (1940, 295) made clear: “The various questions centering about the determination of …personality have come to be more or less segregated in the category ‘social psychology’”.

More social psychologists were modelling their work on natural science, emphasising controlled laboratory experimentation, and disaffected with sociology whose variables they found to be resistant to manipulation (Oishi et al. 2009, 337). Presenting itself as studying personality and other matters scientifically, social psychology was claiming a cachet above that of sociology (Parkovnick 2000, 433). Greenwood (2004, 4) documents in detail how from the 1930s the character of American social psychology underwent substantial alteration. Social psychologists gravitated away from the paradigm of Wilhelm Wundt, Oswald Kulpe, and Frederic Bartlett of studying agents’ psychological properties as grounded in and conditioned by their communities (Greenwood 2004, 2–4, 19). They adopted a reductionist, individualized social psychology for which the likes of Floyd Allport (1924, 4) – “there is no psychology of groups which is not essentially and entirely a psychology of individuals” - had been advocating since the 1920s, studying “cognition, emotion, and behaviour” asocially, in terms of states and properties of the individual (biology and individual learning). “From the 1930s onward, the social dimensions of psychological states and behaviour came to be increasingly neglected by American social psychologists” (Greenwood 2004, 5). Oishi (2009, 338) and his coauthors write of “the loss of sociological perspective in social psychology” from around this time, social psychologists redirecting their attention away from social conditions towards individuals’ interpretations, perceptions and behaviour.

It would be exaggerating to say sociology and psychology underwent a mutation and completely separated from each other (after all the Social Psychology Quarterly journal continues to be published under the auspices of the American Sociological Association) but there was a discernible shift of interest and emphasis among psychologists that altered social and personal psychology as specialisms, their relations with sociology, and sociology itself. “Personality” as a topic did not disappear from sociology after the 1930s (see for example Davis 1948, Gerth and Mills 1953, Parsons 1951, 1964, DiRenzo 1977), but fewer sociologists addressed it, and tended to deal with it less often.Footnote 10 Revealingly, “personality” appeared twice only in Erving Goffman’s The Presentation of Self in Everyday Life of 1959, while Gordon DiRenzo in 1977 (261) has written of the “pervasive …attitude that” personality is not a “legitimately sociological” concern, Robert Levine (2001, 803) of anthropologists and sociologists’ rejection of “culture and personality studies” since the early 1950s, and a contemporary textbook such as Anthony Giddens’ Sociology has only a couple of pages for “personality” in its impressively detailed index.

With personality receiving less attention in sociology and more in personality psychology and social psychology, a space opened up in sociology to accommodate the role concept. Whether nature abhors a vacuum can be left for physicists to decide, but on this occasion sociologists seem to have had an abhorrence of empty space in their conceptual field. Linton was presenting sociology with a concept that could replace that of personality, and a concept that arguably was in better tune with the tenor of sociology than was personality. Personality, for Allport (1937, 48 emphasis added), referred to mental qualities of actors and to “unique adjustments by individual agents” (the “unique”, “individual person” (McAdams 1997, 4, 5, 8)) whereas roles represent positions in social institutions that agents enact in conformity with social norms specifying rights and responsibilities. An aspect of Talcott Parsons’ work gives helpful illustration of the changes then occurring: role was absent from the index of The Structure of Social Action of 1937 and from essays of his appearing approximate to that book, such as “The Professions and Social Structure” (1939), whereas the index of Parsons’ The Social System of 1951 points the reader to well over fifty pages of the book on which the word “role” appears.

Linton’s discussion of role became paradigmatic. By 1948 Kingsley Davis (118, see also Hinkle 1994, 184) was praising Linton’s “excellent presentation of role and status” and noting its “wide influence”. Frederick Bates in 1956 (313) considered “social status …and social role” as among the most common “ideas in social science since” Linton’s formal introduction of them “into the lexicon of social science”. Ralph Turner (1956: 316) cited “Linton’s famous definition” of role. Merton opined in 1957 (110) that “However much they may differ in other respects, contemporary sociological theorists are largely at one in adopting the premise that social statuses and social roles comprise major building blocks of social structure. This has been the case, since the influential writings of Ralph Linton on the subject, a generation ago.” Since Linton’s formulation (and that of Mead), S. A. Nadel (1957, 22) averred, role has come to be “widely used by social scientists”.

Conclusion

Linton was by no means the first social scientist to suggest roles are embedded in social life, and he was not original in coupling role with status. But notwithstanding his discussion of role was confined to a relatively small piece of text in his largish book (The Study of Man) and that he gave greater coverage to the complementary idea, status, Linton built on the work of other scholars, analysing role more directly and incisively than they, assigning it a clearer and more definite meaning. Sociology was made particularly receptive to the role concept by changes in psychology occurring around the time Linton’s book was published. His account of role had sufficient suasive appeal to encourage sociologists in ever larger numbers to incorporate the idea in their inquiries. Eighty years on, role remains among the most durable and pervasive of all the terms of sociology.