Download Symbolic lnteractionism:Themes and Variations and more Lecture notes Social Theory in PDF only on Docsity!
1 SHELDON STRYKER -
Symbolic lnteractionism:Themes and Variations
Introduction Of the theoretical orientations underlying work in social psychology, it is symbolic interactionism that has had its major development among sociologists and that has had major appeal to sociologjsts. In part, this reftects particulars of the history of the orientation: its early elaboration took place at the University of Chicago during the time that institution played a dominant role in the production of sociologists. In part, however, the appeal of symbolic interactionism to sociologists reftects the fundamental compatibility of this social-psychological perspective with the structural concerns of sociology proper A theme of this chapter is fit, sometimes neglected and only now being thoroughly exploited, between more general sociological theory and symbolic interactionism as social psychological theory' This theme is present particularly in a later section of the chapter, where the link being forged between traditional symbolic interactionism and role theory is emphasized and where it is argued that the concept of role serves as the point of articulation-the bridge-between theories that have to do, respectively, with the social structure and with the social person It is present as well in the section of the chapter that treats a version of current symbolic interactionism that eschews role concepts as too static, non-processual, and insufficiently attuned to the constructed character of social life, and uses the concepts of negotiation and negotiated order to link person and social organization. As the foregoing suggests, there is considerable internal variation in the content of symbolic interactionism. While there is a core set of theoretical assumptions and concepts which most, if not all, working within this framework accept and use, there are other theoretical ideas relatively peculiar to one or another version. This is equally-perhaps more -true of methodological ideas; the methodological stances of symbolic
4 SHELDON STRYKER
interactionists range from a thoroughgoing rejection of the ordinary conventions of science as commonly understood to a complete acceptance of these. Such internal variation is another theme of this discussion. This chapter is concerned more with ongoing and future developments in symbolic interaction ism than with history. Some critical sense of the history of this perspective is essential, however, if current emphases are to be understood; and sufficient history must be presented to permit that understanding. What follows this introduction is an abbreviated and selective history of symbolic interactionism that begins with the Scottish moral philosophers and carries the story to the very recent past.2^ To the degree that symbolic interactionist theory in a technical sense exists,^3 it does so in the form of small-scale explanations of relatively limited scope. Although a few such explanations will be briefly treated, the primary concern of this chapter is with symbolic interactionism as a theo- retical orientation or as a conceptual framework. That is to say, this chapter is concerned with delineating an approach to the social-psychological world in general, a frame that suggests the terms of and the ways in which explanations of social-psychological events and processes can be formulated. The distinction between theory and theoretical orientation is fundamental; the latter can only be judged on the basis of logical coherence and fruitfulness in suggesting theories that withstand empirical test. Obviously, this assertion implies the belief that specification and test of theories are indispensable to continued adherence to and utilization of a framework or orientation. This chapter is written in the spirit of that belief. Despite that belief, however, relatively little reference to concrete research will be made. In part, this is because a framework is logically prior to the formation of testable theory and thus to the research that tests theory. In part, it is because a choice had to be made: available space does not permit review and intensive critical evaluation of research; and to simply list researches or present findings uncritically does little justice to the complexities of relating findings of research of varying degrees of sophistication and relevance to theoretical issues. In part, it is because -although there is more good research done from a symbolic interactionist frame than its critics allow-a strong research tradition premised on a symbolic interactionist orientation is still emerging. Thus, the choice made was to concentrate on the framework itself. In the same vein, many of the applications of the symbolic interactionist framework have been in the substantive areas of deviance, of the family, of work (including the professions and occupations), and of collective behavior. Although brief recognition will be given such applications, by and large the focus of the chapter is on the framework abstracted from the substantive areas. Some years ago, Mullins (1973) essentially wrote off symbolic interactionism as a viable perspective within sociology. His concern was with symbolic interaction ism as broader, sociological theory, and he failed to appreciate the degree to which symbolic interactionist ideas have been absorbed into the sociological mainstream. Nevertheless, there have been periods in which symbolic interactionism has waxed, others when it has waned. It waned considerably during the ascendance of a "sociol-
Symbolic lnteractionism: Themes and Variations 5
ogy as hard science" aspiration, doing so in an important degree because many of its most visible adherents were explicitly opposed to the development of rigorous methods espoused by the hard-science advocates. It waned during the period in which functionalism was both intellectually and politically dominant in American sociology.^4 It is now waxing, sparked by the invigorating effects of current efforts to go beyond earlier concern with "proper" interpretations or transliterations of the "masters" and to go beyond sterile debates in which doctrinaire and stultifying positions were taken; by a new concern with theory construction and with test of theory abetted by new and successful attacks on difficult measurement problems; and by new forms of social organization.^5
The Early Development of Symbolic Interactionism
The fixing of origins of any complex line of thought is arbitrary. Although the label, symbolic interactionism, is a relatively recent invention,^6 the line of thought the term represents can conveniently be traced to the Scottish Moral Philosophers.^7 These eighteenth-century thinkers, as Bryson (1945:1) notes, sought to provide an empirical basis for the study of man and society. While holding diverse views of what was fundamental about the human mind, they-Adam Smith, David Hume, Adam Fer- guson, Frances Hutcheson, and others-argued in common that the facts of human association had to be taken into account if a science of man was to be achieved. They turned their attention to communication, sympathy, imitation, habit, and custom in their attempts to develop principles of human behavior. Thus, Adam Smith (1759) writes (discussing the consequences of isolation from others): "Bring him into society, and he is immediately provided with a mirror which he wanted before. It is placed in the countenance and behavior of those he lives with. This is the only looking glass by which we can, in some measure, with the eyes of other people, scrutinize the propriety of our own conduct."^8 David Hume -as well as Smith-saw in "sympathy" the principle through which humans develop their sense of membership in and benefits to be derived from society, and through which they come to be controlled by others. Sympathy, as these Scotsmen conceived it, allows persons to put themselves in the place of others, to see the world as these others do; and sympathy makes possible the communication that initially forms and subsequently reshapes (as we seek the approval of others) who and what we are. As this suggests, society is viewed as a network of interpersonal communication, connecting persons organically. In this manner, the Scottish moral philosophers foreshadowed the symbolic interactionist view of the basic nature of society and of the source of self in society. In addition, they collectively made viewing man as a natural object legiti- mate, and emphasized the scientific importance of everyday experience. Approaching human behavior from the standpoint of society rather than biology, they appreciated mind as instrumental in human adaptation,
8 SHELDON STRYKER
Once developed, the self becomes critical to the understanding of behavior. Mead envisions two aspects of self: the "me," or the organized attitudes (expectations) of others incorporated into the self; and the "I," or the responses of the person to the organized attitudes of others. For Mead, the "I" represents the creative, spontaneous aspect of human behavior. Important here is the idea that creativity and spontaneity occur within the social process, not outside of it; behavior is the outcome of a dialectic in which the attitudes of others are responded to by the person are responded to by the attitudes of others, ad infinitum. Social control is a necessary condition for the emergence of self-control. Obviously, for Mead, the development of self is of central importance. In general terms, as noted, the self develops as does any other object. More particularly, it does so through an early "play" stage and a later "game" stage. In play, the child takes the role of particular others. But social life is complex, and for it to proceed one must respond to an intricate pattern of related behaviors by multiple others; thus, Mead's metaphor of the game. To play a game, one takes the role of the "generalized other," the attitudes of the "organized community." One does so, of course, through the symbolic capabilities characterizing humans. Selves arise out of interaction of persons in organized groups; the prior existence of organized groups is thus implied. But, according to Mead, as society shapes the self, so does the self shape society through the I-me dialectic. Society is through this process continuously being created and recreated; in contemporary language, social interaction is constructed. Social order and social change are aspects of the larger social process. With Mead's (1934) synthesis, symbolic interactionism entered a period of exegesis, debate with respect to "proper" interpretation, application of the perspective to a variety of issues in sociology and social psychology, the working out of methodological positions, and -- to some extent -- conceptual development and research designed to examine and test fundamentaI assumptions. A discussion of the work of two persons, Herbert Blumer and Manford H. Kuhn, will serve both to characterize the period and to bring us closer to the point of contemporary developments. Self-presented and commonly taken as a straightforward elaboration and specification of Mead, a view devastatingly challenged recently (McPhail and Rexroat 1979),^11 the writings of Herbert Blumerl2^ have heavily influenced (particularly, perhaps, in methodological terms) the thought of many sociologists working within the symbolic interactionist frame. Since much of Blumer's writing has the character of a polemic against a sociology he defines in contradistinction to symbolic interactionism, that influence has indeed been consequential both for current issues within sociology and for the character of current attempts to cast symbolic interaction ism in more social structural terms. Mead's analysis of the bases of symbolic interaction, suggests Blumer (1962), presumes that society is composed of individuals who have selves, whose action is constructed through a process of reflexively guiding that unfolding action by indicating the meaning of objects in a social context for prospective action. It also presumes that group action involves the fitting together of individual lines of action through role- taking. Society must be conceived in a manner consistent with these "easily verified"
Symbolic Interacttontsm:Themes and Variations 9
premises, he believes, and he conceives society as consisting of people's actions taking place in and with regard to a situation and constructed by interpreting the situation, identifying and assessing things that must be taken into account, and acting on the basis of the assessment. He insists that even in situations in which there exist common understandings or definitions developed through prior interaction, an interpretive process occurs in which the actions of participants are constructed. He contrasts this view of society with the view held by conventional sociologists, for whom society is a structure or organization. According to Blumer, symbolic interactionism sees social organization as entering action only to the extent that it shapes situations and provides symbols used in interpreting situations. While profound in stable and settled societies, the influence of social organization is less in modern society where criss-crossing lines of action mean situations for which there are no prior standardized actions. From his point of view, seeking to link social behavior to role requirements, expectations, rules, attitudes, and so forth, is inconsistent with recognizing that the human is a defining, interpreting, and indicating creature; Blumer (1969a:l-60) argues that to do so is to have no place for people with selves through which their worlds are handled and action constructed. From his point of view, the articulation of individual lines of action constitutes the social organization of action. Failure to recognize this blinds analysts to the fact that established and repetitive forms of action have to be continuously renewed through interpretation and designation. Analysis in terms of concepts such as culture, social order, norms, values, rules misses the basic point that it is group life that creates and maintains rules, not the other way around. Blumer's methodological principles are drawn from this vision of the person, organized action, and the environment as fluid, continuously constructed and reconstructed through definitional and interpretative processes; this vision represents "the nature of the empirical world" that Blumer instructs us to respect and to organize our methodological stance to reflect. It is in the name of this vision of the nature of the empirical world that Blumer abjures what he sees as the current, conventional methodology (1969:28-34): adhering to scientific protocol, engaging in replication of research, relying on the test of hypotheses and employing operational procedures. It is also in the name of this vision that he (1954; 1956) instructs sociology to abjure the use of "definitive concepts" and "variables." Definitive concepts refer to what a class of objects have in common through a clear definition in terms of attributes or fixed bench marks. They contrast with "sensitizing concepts," which, according to Blumer, "merely suggest directions along which to look" rather than "prescriptions of what to see." Blumer suggests that analysis in terms of variables, among other deficiencies, leads one to ignore the processes of interpretation and definition by assuming that an independent variable automatically affects a dependent variable. This assumption fails to recognize that anything that is defined (and everything that is of consequence in social life is, from this point of view) can be redefined, which implies that relationships among "variables" have no intrinsic fixity and that interpretations cannot be given the qualitative constancy required of a vari-
10 SHELDON STRYKER
able. It is this same quality of interpretations that rules out the use of definitive concepts.
In place of the experiments, surveys, refined measurement instruments, census data, computer simulation, and "crucial empirical data to test hypotheses" he sees as dominating the methods of conventional sociology, Blumer calls for direct examination of the empirical world of everyday experience. But one must do more than look. Blumer recommends two modes of inquiry to us: exploration and inspection. By definition a flexible procedure not tied to particular techniques, exploration is guided by the maxim to get a clearer picture of what may be going on in an area of social life by any ethical procedure: observation, informal interviewing, listening to conversations, getting life-histories, using letters and diaries, arranging for group discussions, consulting public records, using a resource group of informed persons. In exploration, one constantly tests and revises images, beliefs, and conceptions of the social world being studied. Ultimately, one constructs a comprehensive and intimate account of what takes place in that empirical social world. Having done so, one then turns to inspection in order to meet the requirements of scientific analysis for clear, discriminating analytic elements and the isolation of relationships among these elements. This involves casting a problem in theoretical form, sharpening connotative referents of concepts, unearthing generic relations and formulating theoretical propositions. Again, the contrast Blumer draws is with his picture of conventional social research, which presumably starts with a theory or model framed in terms of relationships between concepts, uses the theory to select a problem, converts the problem into independent and dependent variables, uses precise techniques to obtain data, discovers relationships between variables, and uses the theory or model to explain these relations. Inspection is the antithesis of such methods; it is flexible, imaginative, creative, unroutinized. It involves looking at empirical instances of given analytic elements in a variety of different ways, viewing them from different angles and from the standpoint of many different questions. One is indeed to test empirically the basic premises of symbolic interactionism, according to Blumer, but one cannot do so by the "alien criteria of an irrelevant methodology" (that is, by conventional methods). Rather, these premises "can be readily tested and validated merely by observing what goes on in social life under one's nose" (1969:49-50). Given that the premises of symbolic interactionism are validated, suggests Blumer, certain methodological implications follow: see objects as people see them (since they act on the basis of the meaning the objects have for them) in order to understand their behavior; social interaction must not be compressed into pre-existing forms, rather the forms it takes must be empirically discovered; social action must be analyzed by observing the process of construction, noting how the situation is seen by the actor, what the actor takes into account and how this is interpreted, trying to follow the interpretation that leads to a selection of particular acts. From Blumer's point of view, the study of complex organization or complexly organized social life poses no methodological problem different from those posed when studying individual action.
Symbolic Interactionism: Themes and Variations 11
Many of the tensions, debates, and variations within symbolic interactionism have been described by delineating two "schools," one labeled the "Chicago School" and identified with Blumer and his students, the other labeled the "Iowa School" and identified with Manford H. Kuhn and his students.^13 The contrast between the two has been summarized (Meltzer, Petras, and Reynolds 1975:123, note 4) by saying that the " 'Chicago school' emphasizes process not structure, sympathetic introspection not attitude scales, indeterminacy and emergence not determinacy." As this not entirely apt description does suggest, Kuhn's symbolic interactionism is much more oriented to the development of precisely stated theory and rigorous empirical test of that theory than is Blumer's, and in that sense is considerably closer to the viewpoint of this chapter. To emphasize his interest in developing generalizations tested by empirical research, Kuhn (1964) chose the label self-theory for his version of symbolic interactionism. He regarded his theoretical and conceptual ideas as "orthodox" in their derivation from Cooley, Dewey, and Mead; but he sought to assimilate diverse materials to his perspective. In particular, he assimilated role theory and reference group theory, finding it difficult to distinguish these from symbolic interactionism. He adopted role theory's conception of social structure as consisting of networks of positions and associated roles (expectations). Agreeing that social structure is created, maintained, and changed through symbolic interaction, he also viewed social structure, once created, as constraining interaction. But his identification of role theory and symbolic interactionism is not total; his own emphasis is on role-taking, and on the self as mediating the relation between social structure and behavior. Noting the absence of determinacy in the relation of role expectations and performances, he is more inclined to see determinacy in the relation of self to behavior. Kuhn treats the self as an object. Observing that for Mead objects are plans of action, or "attitudes," he proposes that the "self" be conceptualized in these terms. The significance of self lies in the need to know subjective definitions of identity in order to predict how people organize and direct their behavior, a requirement growing out of the looseness in secular society of ties between social systems and the individual occupants of statuses in these systems. Attitudes toward self as object are the best indexes of plans of action in general; thus, the most significant object to be defined in a situation is the self. The core self, a stable set of meaning attached to self as object, takes on central importance in Kuhn's theorizing. It is the core self that provides structure and relative stability to personality and provides continuity and predictability to behavior. However, Kuhn visualizes the self as having a large number of component parts and related aspects, including subjective identification in status terms, attributes and traits, roles, role preferences, role avoidances, role expectations, areas of self-threat and vulnerability, self-enhancing evaluations, patterns of reference selections. Given the complexity of self, Kuhn clearly anticipates no simple relation of self to behavior. The core self constrains behavior; stability in self results in stability in interaction. Such stability, however, is relative. Kuhn sees opportunity for creativity through the role-taking process, and through the self-con-
14 SHELDON STRYKER
of a symbolic interactionist framework is enormous; if systematic and rigorous inquiry is required, the eligibles are relatively few. Further, if systematic and rigorous research that can be interpreted within a symbolic interactionist frame, as differentiated from that which self-consciously and explicitly stems from that frame, is included, the candidates for mention expand greatly. The attempt in the following is only to illustrate various research issues and genres, to make the point that, while the ratio of research to conceptual and theoretical discourse might well have been higher, there has been a continuous stream of research motivated by symbolic interactionism. One part of that stream seeks to examine the basic premise of Mead's thought, by asking whether the responses of others do indeed shape the self; much of this literature utilizes a surveyor an experimental format.^20 The complementary question, whether and how self-concepts do in fact affect further behavior, is studied in a variety of settings and uses a wide variety of research procedures, from informal interviewing to observational to schoolroom questionnaires. So, for example, Alfred R. Lindesmith (1947) uses informal interviewing and an analytic induction procedure to develop his theory of opiate addiction, in which self-concept as addict plays an important role. Donald R. Cressey (1953) uses similar methods and ideas to examine the processes by which persons become embezzlers, as does Howard S. Becker (1953) with respect to marijuana use. Walter Reckless and his associates (1956) focus on the ways in which self concepts as "good" and "bad" boys contribute to the making of delinquents, generating data through questionnaires, as do Michael Schwartz and Sheldon Stryker (1970). Becker (1951) ties the reactions of musicians to their audiences to their conceptions of themselves as musicians, basing his argument on participant observation and informal interviewing. And Schwartz, Fearn, and Stryker (1966) investigate the way in which stable self-conceptions as disturbed contribute to confirming children in emotionally disturbed roles, with data provided by structured instruments. Since the shaping of the self, and self-concept change, is the heart of the problem of socialization from a symbolic interactionist standpoint, studies already cited are germane to that research topic. So, indeed, are perhaps most researches developing from the frame, particularly if socialization is recognized as a lifelong process. (It is an interesting fact that, until comparatively recently, symbolic interactionists did little with childhood socialization other than cite Cooley's early observations of the evolution of self in his own children and repeat Mead's dicta.) ll1ustrative researches with direct interest in socialization per se are those of Becker et al. (1961) dealing with medical school students, and Olesen and Whitaker (1968) dealing with nurses, works primarily based on observational and informal interviewing; Brim's (1958) statistical analysis of data relating family structure and the learning of sex roles by children; Norman K. Denzin's (1972, 1975) observational studies of the emergence of self in early childhood; and Thomas and Weigert's (1971) cross-national analysis of adolescent conformity to the expectations of significant others using data from questionnaires administered in classroom settings. Central to the symbolic interactionist framework, along with the concept of self, is the concept of role-taking. While relatively few researches
Symbolic Interactionism: Themes and Variations 15
directly focus on role-taking behavior, some do, among them O'Toole and Dubin's (1968) behavioral demonstration of the reality of the phenomenon in a systematic observational study of mothers feeding their infant children and in an experimental study of body sway; Cottrell's (1971) experimental study of muscular tension in response to observing others' muscular tension; Thomas, Franks, and Calconico's (1972) questionnaire study of the relation between role-taking and power in families; and Stryker's (1956, 1957 work, using structured interviews, on the sources and consequences of accuracy in role-taking. Finally, this review of research stimulated by a symbolic interactionist framework during the period between the work of the founding fathers and the relatively recent past would be seriously misleading without reference to what have been taken to be characteristic examples of the research style associated with symbolic interactionism (Lofland, 1970): qualitative case studies of interaction in diverse social contexts. Examples are Glaser and Strauss's (1964, 1968, 1971) study of the interactions of hospital personnel, family, and dying patients in a hospital setting; Julius Roth's (1963) study of the passage of patients through hospitals, Egon Bittner's (1967) examination of the ways in which police deal with the mentally ill, and-perhaps epitomizing the genre for many^21 -Goffman's (1963b, 1967, 1971) many reports on strategies of interaction.
Current Developments in Symbolic Interactionism
Having traced the history of the symbolic interactionist framework from the Scottish Moral Philosophers to the relatively recent past, it remains to discuss more or less current developments. As a prelude, it will be useful to summarize that framework and to present various criticisms of the framework as it has developed. As previously remarked, there is no single symbolic interactionism whose tenets command universal acceptance; thus, it ought not surprise that no one summary statement will be acceptable to all. Stone and Farberman (1970:1) delimit the field of social psychology from the standpoint of symbolic interaction in terms of six questions: What is meaning? How does the personal life take on meaning? How does the meaning persist? How is the meaning transformed? How is the meaning lost? How is meaning regained? Manis and Meltzer (1978:5) suggest that the fundamental elements of symbolic interactionism include the meaning component in human conduct, the social sources of humanness, society as process, the voluntaristic component in human conduct, a dialectical conception of mind, the constructive and emergent nature of human conduct, and the necessity of sympathetic introspection. Meltzer, Petras, and Reynolds (1975:54) assert that all varieties of symbolic interactionists take as basic premises that humans act toward things on the basis of the meaning those things have for them, that meanings emerge from social interaction, and that meanings are modified and dealt with through an interpretative process used by persons when responding to things en-
16 SHELDON STRYKER
countered.22^ Jonathan Turner (1978) characterizes the core of symbolic interactionism as consisting in the assertions that humans create, use, and communicate with symbols; they interact through role taking, which involves the reading of symbols used by others; they are unique as a species through having mind and self, which arise out of interaction, and which allow for the interactions that form the basis of society.
The writer (1959) has characterized the common elements
uniting symbolic-interactionist thinking in terms of a set of
assumptions and a predilection. The assumptions are that human
beings must be studied on their own level, and that reductionist
efforts to infer principles of human behavior from the study of
nonhuman forms are misguided; that the most fruitful approach to
the study of human social behavior is through an analysis of
society; that the human infant enters life neither social nor
antisocial, with the potentialities for human development; and that
the human being is an actbr as well as a reactor. The predilection
is to stay close to the world of everyday behavior, both in the
development of the framework and in its application.
George J. McCall (1977) offers the following principles as underlying symbolic interactionism, and the set can be taken as his summarization of the framework:
I. Man is a planning animal, constructing plans out of bits and pieces supplied by culture.
- Things take on meaning in relation to plans; the meaning of a thing is its implications for plans of action being constructed, so a thing may have different meanings relative to different plans.
- We plan toward things in terms of their meanings; a plan of action is executed contingent on the meaning for that plan of things encountered.
- Consequently, every thing encountered must be identified and its meaning discovered. .
- For social plans of action, meaning must be consensual; if meanings are not clear, they are hammered out through the rhetoric of interaction resulting in the creation of social objects.
- The basic thing to be identified in any situation is the person himself; identities of actors in a situation must be consensually established.
- Identity, meaning, and social acts are the stuff of drama; as drama involves parts to be played, roles implicit in the parts must be conceived and performed in ways expressive of the role. The construction of social conduct involves roles and characters, props and supporting casts, scenes and audiences.
- Thus, identification of persons is most often in terms of roles and characters. We identify by placing things in systematically related categories of role systems, status systems, systems of social types, or contrastive sets of social categories.
Another generalized statement of one version of symbolic interactionism has been offered by the writer (1980:53-55). It has the advantage of being less terse and so perhaps more understandable as a summary description of the framework; and it has an advantage in that it incorporates important aspects of recent developments in the framework. Accepting the fundamental reciprocity of society and person, the statement arbitrarily begins with the impact of society on person. I. Behavior is dependent upon a named or classified world. The names or class terms attached to aspects of the environment, both physical and social, carry
Symbolic lnteractionism: Themes and Variations 17
meaning in the form of shared behavioral expectations that grow out of social interaction. From interaction with others, one learns how to classify objects one comes into contact with and in that process also learns how one is expected to behave with reference to those objects.
- Among the class terms learned in interaction are the symbols that are used to designate 'positions,' which are the relatively stable, morphological components of social structure. It is these positions which carry the shared behavioral expectations that are conventionally labelled 'roles.'
- Persons who act in the context of organized patterns of behavior, i.e., in the context of social structure, name one another in the sense of recognizing one another as occupants of positions. When they name one another they invoke expectations with regard to each others' behavior.
- Persons acting in the context of organized behavior apply names to themselves as weU. These reflexively applied positional designations, which become part of the 'self,' create internalized expectations with regard to their own behavior.
- When entering interactive situations, persons define the situation by applying names to it, to the other participants in the interaction, to themselves, and to particular features of the situation, and use the resulting definitions to organize their own behavior in the situation.
- Social behavior is not, however, given by these definitions, though early definitions may constrain the possibilities for alternative definitions to emerge from interaction. Behavior is the product of a role-making process, initiated by expectations invoked in the process of defining situations but developing through a tentative, sometimes extremely subtle, probing interchange among actors that can reshape the form and content of the interaction.
- The degree to which roles are 'made' rathep than simply 'played,' as well as the constituent elements entering the constructions of roles will depend on the larger social structures in which interactive situations are embedded. Some structures are 'open,' others relatively 'closed' with respect to novelty in roles and in role enactments or performances. All structures impose some limits on the kinds of definitions which may be called into play and thus the possibilities for interaction.
- To the degree that roles are made rather than only played as given, changes in the character of definitions, the names and the class terms utilized in those definitions, in the possibilities for interaction can occur; and such changes can in turn lead to changes in the larger social structures within which interactions take place.
This version of the symbolic interactionist framework obviously gives greater weight to social structure than do some alternative versions, and does so in a way that permits the elaboration of structural concepts to reflect the complexities of the social world in which humans exist. It also, perhaps not quite so apparently, leaves more room for the routine, habitual, and customary in human behavior than is generally true of contemporary symbolic interactionists despite the importance of such phenomena for their forebears. It opens the door to serious theorizing about the reciprocity of self and society, a basic theme of all symbolic interactionism, but one which-because of the way in which it traditionally has been formulated-has not moved much beyond the level of truism to specification of linkages. In part, recent developments in symbolic interactionism have occurred in reaction to critical appraisals of the framework.^24 Given the internal variations, a portion of these appraisals represent critiques of proponents of one version by proponents of another. Given that internal variation, as well, critiques addressed to one version mayor may not be applicable to another,^25 a fact not altogether appreciated by persons evaluating the
20 SHELDON STRYKER
point) leaves relatively undeveloped the conceptualization of the structural context within which negotiations take place,^36 and because the frame seems more oriented to developing understandings of specific negotiations than explanations couched in theoretical terms.^37 A more likely means by which this articulation can proceed is through role theory; and Ralph H. Turner has, over recent years, been concerned precisely with that articulation. Turner (n.d. c ) is working toward the synthesis of symbolic interactionist and role theoretic elements in "something akin to axiomatic theory." His starting point is with criticisms of role theory: the theory offers an overly structured view of human behavior as the enactment of normative scripts; it neglects normal processes of social interaction in its focus on role strain and role conflict; it fails to make adequate use of the concept of role-taking.^39 He seeks to provide role theory with a proper appreciation for the role- taking concept, and so to correct role-theory's overly structured and conformist quality, seeing this core concept as key to the development of theory that can handle both stable, structured forms of social organization and less structured, fluid forms as well. Human beings act as if others they meet are playing identifiable roles, role-taking to identify these roles. But cultural cues to roles are often vague and contradictory, and so provide only a general outline within which lines of action can be constructed. Under this circumstance, actors make their roles and communicate what roles they are playing in order to permit and to facilitate interaction (Turner n.d. a; n.d. b ). Actors will behave as though they and others with whom they interact are in particular roles as long as the assumption works by providing a stable and effective framework for interaction. They test the assumption by continuously assessing one another's behavior, checking whether that behavior verifies or validates the occupancy of a position by corresponding to expectations and by demonstrating consistency. In addition to this emphasis on the concept of role in modified form, Turner retains the symbolic interactionist's emphasis on self. Self-responses emerge from interaction with others, and we present ourselves to others via our self-conceptions. Seeking to infer the roles of others, we seek to inform others through gestures of the role being played, and whether the roles being played are consistent with and invested with self -in brief, we seek to inform others of the degree to which self and role "merge" (Turner 1978). Unsatisfied with role theory's disparate and unrelated propositions, Turner offers a strategy for theory building: begin with sensitizing concepts, narrow propositions and hypotheses drawn from the research literature, and move to precise definitions and to general, formal theoretical propositions linking empirical regularities and expressing major tendencies of those regularities. Then look for determinants of variation in the regularities, and group-related regularities. Finally, seek common principles to explain why the groupings of regularities should occur. He offers two general explanatory propositions: roles are used to achieve ends efficiently; the playing of roles is a means of achieving personal reward in the form of validation of self, self-esteem, and reinforcement from others.
Symbolic Interactionism: Themes and Variations 21
This strategy is illustrated in his (1978) discussion of role-person merger. Conceptualizing persons as consisting of a hierarchical ordering of all roles in their repertoires, and noting that doing so relates the person meaningfully to social structure, he suggests three criteria of role-person merger: playing a role in situations in which the role does not apply; resisting abandonment of a role despite advantageous alternatives; and acquisition of attitudes and behaviors appropriate to a role. Delineating two types of determinants of mergers, interactive and individual, he then asks: what functions are served for those interacting by viewing one another as persons (that is, as playing roles)? And he concludes that the concept of person is related to the requirements of social control, since effective and lasting social control requires a more stable object than an actor who simply plays a particular role. This interactive function of role-person merger leads to three interactive principles: in the absence of contradictory cues, people tend to accept others as they appear (the appearance principle); the disposition to see people in terms of their role behavior will vary with the potential effect of the role on interaction (the effect principle); and people will accept the least complicated view of the person that facilitates interaction ( the consistency principle). Propositions are then derived from these principles: for example, the more inflexible the allocation of actors to roles, the greater the tendency to conceive the person as revealed by the role; the greater the potential power vested in a tole, the greater the tendency to conceive the person as revealed by the role. A similar analysis of individual functions of role-person merger (to facilitate understanding, predicting, and controlling others by becoming more understandable and predictable to them, to economize effort when playing many roles, to facilitate control and autonomy, to make possible the playing of roles providing gratification, to allow the individual to obtain rewards commensurate with investment) leads again to guiding principles consistent with these functions: people tend to merge their persons into roles by which significant others identify them (the consensual frame of reference principle); selective merger will occur to maximize autonomy and self-evaluation (the autonomy and favorable evaluation principle); and individuals will merge into person those roles in which greatest investment is made or for which return on investment is still to come. Again, propositions are inferred from principles; for exampIe, the more intensely and consistently significant others identify individuals on the basis of a certain role, the more likely will those individuals merge that role and their persons; individuals tend to merge positively evaluated roles with their persons. Role-person merger speaks to the link between self and social structure. The implications of that link are developed by Turner (1976) in an essay on the "real self': the subjectively held sense that people have of who and what they really are. The link between real selves and social structure should be significant in the functioning of and change in societies, he suggests. To the degree that self has an "institutional" focus, people will see their real selves in feelings, attitudes, and actions that are anchored in institutions, they will recognize their real selves in action when accepting group obligations. To the degree that self has an "impulse" emphasis,
22 SHELDON STRYKER
people see their real selves in untamed impulse with conformity to institutional norms occurring at the expense of their true selves. Turner speculates that the past several decades have witnessed a major shift in the locus of self from institution to impulse in American society. If that is true, it has serious implications for the nature of social control and of societal order. Conventional sociological theories of control and order presume actors who locate their real selves in institutions. Turner hypothesizes that the locus of self correlates with a disposition to see either values or norms. Selves tied to institutions lead people to perceive values, and this facilitates control systems of the sort the sociological literature describes. Selves tied to impulse lead people to perceive norms and this does not fit with conventional control systems. The importance of the foregoing lies, not in whether the substantive theory either stated or implied is correct, but in the ways in which social structure is introduced into social psychological theorizing. Turner uses the concepts of position and normative expectations, albeit cautiously, in order to avoid an overly structured stance. He views larger social struc- ture as both constraint on self and social interaction and as product of self and social interaction. The larger structure organizes relationships, bringing some social circles^40 together and keeping others apart. The articulation of real selves with social structure is a major link in the functioning and change of societies. A complementary vision of the relation of person and social structure is contained in an essay by Eugene A. Weinstein and Judith M. Tanur (1976). Visualizing the strength of the symbolic interactionism in its sensitivity to the emergent properties of interaction, these authors see this strength as the source of the excesses of one wing of symbolic interac- tionism. Among the excesses remarked^41 is neglect of the connectedness of the interactive episodes in which social structure finds its concrete expression. Weinstein and Tanur suggest that it is the aggregated outcomes of many prior episodes of interaction in the form of informal understandings, shared meanings, codified rules, and material resources that serve as frameworks for interactions and so link episodes of interaction; it is these aggregated outcomes that give meaning to social structure. Agreeing that the concept of role as it has been used in structural-functional analyses carries too great a theoretical burden, and that the degree to which a social encounter exhibits "role-ness" is variable and problematic, they nevertheless assert that norms and roles are part of the meanings accessible to participants in interaction and usable as resources in that interaction. In brief, the extent, conditions, and means by which social structure is introduced into interaction are to be subject to investigation, but social structure is not to be ignored. A discussion by the writer (1980) reinforces these themes. Whatever their creative potential, most interactions tend to be with the same or slowly changing casts of others doing the same things on a repetitive basis. Structural concepts like group, organization, community refer to patterns of social life tying particular subsets of persons together and separating others. Structural concepts like class structure, power structure, age structure, and so forth refer to the more abstract social boundaries that operate in similar fashion. The important implication of the
Symbolic Interactionism: Themes and Variations 23
generic concept of social structure is that societies are differentiated entities whose differentiation has the cons~quence that only certain persons interact with one another in certain ways with certain resources in certain settings. Persons do not relate randomly, and the opportunities for and circumstances of social relationships are not randomly distributed. The person is shaped by interaction, but social structure shapes the , interaction. Conversely, when persons creatively alter patterns of interaction, ultimately social structure can change. These obverse assertions define the tasks of a sociologically oriented social psychology and (in part) of sociology, and require the bridging of social person and social structure. It is to meet this latter requirement that aspects of role theoryare being drawn into symbolic interactionism. A root idea of symbolic interactionism from its very beginnings has been the reciprocity of self and society. One way of expressing the criticism of the framework to which the attempt to draw in social structure responds is to say that it failed to respect the complexities of "society" in its conceptualization; and one way to describe sociology's treatment of "society" over recent years is to say that it has incorporated these complexities by imaging society as a multifaceted mosaic of interdependent but highly differentiated parts-groups, institutions, strata-whose relationships run from cooperation through" conflict. If society is highly differentiated, and if self reflects society, self, too, must be highly differentiated. It is this insight that underlies another major development of symbolic interactionism, the emergence of identity theory.^42 Identity theory capitalizes on William James's contention that people have as many selves as there are others who react to them. While Mead shared this sense of the human being with multiple selves, his philosophic premises and hopes led to an emphasis on self as a global, undifferentiated unity. But such an approach to self does not square with the basic symbolic interactionist dictum that self reflects society when the society at issue is complex, as it certainly is. A complex, differentiated society requires a parallel view of self on theoretical grounds. Empirically, there are issues whose resolution calls for a conception of self as complex and differentiated, yet organized.^43 In particular, there are issues of both behavioral consistency and inconsistency across situations, of explaining the choices that are made when persons are faced with conflicting role expectations, of dealing with the greater or lesser resistance to change exhibited by persons in the face of changing social circumstances. To meet both theoretical and empirical needs, the concepts of identity,^44 identity salience, and commitment are introduced. Identities are "parts" of the self, internalized positional designations that exist insofar as the person participates in structured role relationships, the consequence of being placed as a social object and appropriating the terms of placement for oneself. Persons may have many identities, limited only by the structured relationships in which they are implicated. "Identity salience" is one theoretically important way in which discrete identities making up the self can be organized. That is, identities are conceptualized as being organized into a hierarchy of salience
26 SHELDON STRYKER
find ways to deal with the dynamics of the relations between the two, and among them and performances. Finally, Burke refines the idea that identities motivate through defining behavior and through the action implications of their meanings. If identities as meanings located in semantic space have action implications, identities close to one another in that space ought to have similar action implications. Further, acts have meanings, and those in the same semantic locations as identities ought to carry implications for those identities. Implied is a measurement procedure that measures both identities and actions in common terms, specifically by locating them in the same semantic space.
Conclusion
It ought to be clear from the preceding pages that symbolic interactionism is alive and at least reasonably well, and that it is pursuing a course in its development that serves to integrate within its general stance a reasonable conceptualization of social structure. In doing so, it is fulfilling both its early promise and the promise of a sociologically oriented social psychology. It is certainly as true today as it was during the height of the Blumer-Kuhn "debate" that there is no symbolic interactionist orthodoxy, no single vision of what the framework "means." In particular, perhaps, the divisions are methodological in the broadest sense of that term. Let us hope, however, that there is more tolerance for alternative styles of work and greater appreciation of the virtues (as well as the limitations) of the various styles. If that is true, it is indeed a hopeful sign, for it implies that less time will be spent in sterile argument addressed to the unwashed both inside symbolic interactionism and outside; that more effort will be expended in the research enterprise on which the framework ultimately rises or falls; that we will exhibit greater willingness to let that research tell us what is and is not useful in the framework; and that there is growing understanding that others-even others writing from the perspectives of alternative frameworks-may have something of value to say. And, if these implications hold, we can look forward to the continued influence and continuing development of the symbolic interactionist framework.
NOTES
- This theme is also the motif of my more extended treatment (Stryker, 1980) of particular versions of symbolic interactionsim, which treatment is used throughout this chapter. See also Handel (1979).
Symbolic Interactionism: Themes and Variations 27
- For more thorough treatments of the historical development of symbolic interactionism, see Stryker (1980) and Meltzer, Petras, and Reynolds (1975).
- A theory, in a technical sense, is a set of propositions about some part of the empirical world specifying how this part presumably works, emerging from a set of assumptions or postulates and from a set of concepts used to describe the part of the world the theory purports to explain, and open to checking against empirical observations of that world.
- I am indebted to Ralph H. Turner (personal communication) for these insights into the relative decline in the influence of symbolic interactionism as a perspective in sociology. Turner sees part of the more recent increased interest in symbolic interactionism as reflecting its role as an alternative to polarization between functionalism and conflict theories.
- The reference here is to a relatively new Society for the Study of Symbolic Interaction, and its publication Symbolic Interaction, as well as to a revived Section on Social Psychology in the American Sociological Association which over the past few years has served to showcase panels on symbolic interactionist themes and topics.
- Herbert Blumer is responsible for the term, initially using it in a chapter in Emerson P. Schmitt (1937). For some, Blumer's theoretical writings are taken to define symbolic interactionism. For others, the label represents a tradition of thought to which Mead, Cooley, and Thomas are the preeminent early "sociological" contributors, in which Blumer takes his place as an important but not necessarily decisive figure, and which contains a number of contemporary versions methodologically and theoretically distant from Blumer. It is obvious that this chapter is the product of one who holds the latter view.
- For a detailed argument developing the import of the Scottish Moral Philosophers to American sociology and socia) psychology, see Bryson (1945).
- This passage, emphasizing the dependence of conceptions of self on others, anticipates Cooley's discussion of the looking-glass self.
- See also Dewey (1896, 1920).
- To use the title of Mead's (1934) influential work, which is actually a compilation of his lecture notes published posthumously. II. See also the challenge to the common assumption that Blumer's work epitomizes the tradition of Mead implicit in Stryker (1980). Blumer (1980) has written a rejoinder to McPhail and Rexroat, and has in turn been answered (McPhail and Rexroat, 1980 ).
- Blumer's views as presented here are taken from essays written relatively late in his career. While these are not always consistent with essays written earlier (or, indeed, internally consistent), their characteristic emphases seem to me to represent accurately the nature of Blumer's influence. Many of his earlier essays reappear in Blumer (1969).
- Neither label referring to the locales-for much of their careers-of those two symbols of contrasting views fits precisely in the intellectual pedigrees of persons tending to the respective positions the labels are intended to describe. Further, there is a large set of persons whose work is in a symbolic interactionist vein who cannot easily be fitted into either camp. Finally, as will be suggested later, one way of characterizing present developments in symbolic interactionism is to suggest that the divisions implied by the "Chicago" and "Iowa" labels are being bypassed. Too much intellectual energy has been expended in argument that seems to view one or another version of symbolic interactionism more in social movement terms than as a social psychological framework per se.
- Kuhn and his students were responsible for the development of the Twenty Statements Test, an attempt to achieve a standardized test identifying and measuring self- attitudes. While some equate that specific test and Kuhn's methodology -see Tucker (1966)- and discredit the latter by noting problems of the former, Kuhn's methodological stance does not depend on the success of any particular measurement device. .15. Why the relative neglect of others, some of whom are cited in this chapter, is an interesting question. Perhaps, as with Park, the messages were ones that Blumer tended to deny in his polemical writings; thus, insofar as Blumer is identified with symbolic interactionism, it became "inappropriate" to incorporate
28 SHELDON STRYKER
these others. Perhaps it was because Blumer was something of a charismatic figure, at least some of whose students tended to be "disciples."
- I do not mean to imply, in asserting that Waller adopted Mead's framework, that Mead was the only influence at work on Waller. An identical caveat holds for every other citation in this section.
- For an intersting and telling response to critics of the "societal reaction" theory of deviance, in which he calls for a much more subtle, structural analysis of society than labeling theorists typically provide, see Lemert (1974).
- And, indeed, the distinction between this literature, the applications literature, and the conceptual developments literature is itself highly arbitrary.
- There is a problem here that grows out of style of research and researchreporting. Goffman's research may well be systematic and rigorous in the extreme; yet the style of reporting is such that his work looks more like an art form 1 than science. A fair proportion of the symbolic interactionist literature has thischaracteristic.
- See, for example, Miyamoto and Dornbusch (1956); Couch (1958); Reeder, Donohue, and Biblarz (1960); Videbeck (1960); Quarantelli and Cooper (1966); Sherwood (1965).
- Perhaps not entirely fairly. Thus, Glaser and Strauss (1967) draw a strong distinction between the apparently undisciplined approach underlying Goffman's work and their own careful use of what they call the constant comparative method of grounded theory. Glaser and Strauss (1967) has become something of a bible -- and the term, grounded theory, something of a shibboleth and battle cry -- for those whose predilections are toward qualitative research; the book was written in the attempt to undergird qualitative work with the discipline and rigor its detractors have typically denied it.
- These premises are taken from Blumer's (1969:2-6) characterization of symbolic interactionism.
- Symbolic interactionists have traditionally asserted the uniqueness of the human species, premising their argument on the presumably unique capacities of humans and the emergence of self. That these presumably unique aspects of humans are indeed unique is severely challenged by recent work with chimpanzees. Apart from undergirding an extreme form of denial of the relevance of work on nonhuman animals for understanding human behavior, however, it is difficult to see what is gained from the assertion of human uniqueness.
- Space limitations preclude a full review of these critical appraisals. For more complete reviews, together with references to the critical literature, see Stryker (1980) and Meltzer, Petras, and Reynolds (1975).
- It must be more than apparent by now that criticisms directed at Blumer , will not, in general, hold for other versions, and vice versa.
- These five are reviewed in detail in Meltzer, Petras, and Reynolds (1975).
- See Stryker (1980: chapter 5) for such a defense.
- The neglect is relative, not total. See for example the discussion of interactionists' treatment of emotion in Meltzer, Petras, and Reynolds (1975:92), Kirkpatrick's (1963) and Turner's (1970) formulations of an interactionist theory of love, and Shibutani's (1961) discussion of emotion as response to blocked lines of action. See also the
suggestion in Stryker (1968) that the self be treated as having conative and cathectic modalities as well as the cognitive modality typically emphasized. There is some indication that more serious work on the emotions is an upcoming item on symbolic interactionism's agenda.
- This is not true of Huber's (1973) claim of bias in symbolic interactionism based on the argument that an unwillingness to be explicit in its theory leaves symbolic interactionism in the position of tacitly accepting the existing (power) structure of society. Not all symbolic interactionists are unwilling to be explicit in their theorizing.
- Blumer would fit the first category; the writer, the second.
- I think Maines overstates the case by ignoring the questions of the depth and adequacy of the treatment of these matters by Blumer and others.
- Persons seeking to defend symbolic interactioism against the claim that it precludes a concern with macro-social structure cite the work of Hall (1972), Farberman (1975), and Denzin (197V as counter-instances. These studies do indeed incorporate both traditional concepts of symbolic interactionism and concepts referring to macro-structure. Whether, however, they resolve theoretically the issues involved in linking the two is moot.
- See, for example, Bucher and Strauss (1961); Bucher (1962); Strauss, Schatzman, Ehrlich, Bucher, and Sabshin (1963); Glaser (1968); Glaser and Strauss (1965); Glaser and Strauss (1971); Strauss (1971).
- This is the explicit starting point in Strauss, et al. (1963), which refers to Mead (1936:360-61).
- While Strauss clearly thinks these aspects of negotiation contexts are reasonably general, the basic principle of the grounded theory approach is to make such matters open and dependent on emergence from the specific situation(s) being researched.
- My reading of the negotiated order literature leads me to conclude that conventional sociological concepts, such as position, role, role relationship, status, norm, and so forth, are implicitly incorporated into what is said about the social person, interaction, and social structure, but are not explicitly recognized in ways that would be useful in the task of articulating symbolic interactionism and more general sociological theory.
- I arrive at this judgment in spite of the explicit claims to the contrary to be found in Strauss (1978), through my sense of the basic focus of attention in relevant works.
- This articulation is also the point of my own efforts, most recently in Stryker (1980). See also Stryker (1964, 1973) and Handel (1979).
- Turner has another criticism of role theory, that its propositions have little connection with one another, which is less relev~nt in the present context.
- Turner uses the term "social circles" to mean what some others mean by "groups," and "social networks," or "role set." His usage stems from Znaniecki (1965) and focuses on those persons involved with the person in the carrying out of her or his role.
- The other is the injunction against quantitative research as a matter of faith rather than style, and the consequences of that injunction.
- The outlines of identity theory appear in a paper I presented in 1966 at a meeting of the American Sociological Association which was subsequently published (Stryker, 1968). The same fundamental ideas are elaborated by McCall and Simmons (1966), although with some shifts in terminology and emphasis. The "invention" was entirely independent. I choose to present the theory in my own terms here.
- The contemporary call for an elaborated conception of self derives only in part from the requirements of the symbolic interactionist framework. For an excellent and compatible vision of the complexities of the self-concept that stems fundamentaJly from the attempt to deal with empirical puzzles, see Rosenberg (1979).
- McCa11 and Simmons (1978) as well as Burke and Tully (1977) use the term role- identity in an equivalent way, thus stressing the intimate linkage of self and role.
- This usage derives from Kornhauser (1962).
- That potential is now being examined in a research program under way at Indiana University under the writer's direction; preliminary results are highly encouraging.
- These are taken from Stryker (1980); this statement is based on Stryker (1968).
