argument and demonstrate

Craig R. Scott Steven R. Corman George Cheney

Communication Theory

Eight: Three

August 1998

Pages: 298-336

Development of a Structurational Model of Identification in the Organization

In response to the growing interest in issues related to attachment in organi- zations, this paper develops a theory of identification in the workplace based on three key aspects of structuration theory. In this model, the identification process is treated as a duality involving identities that create and are created by identifications, which are themselves observed in social interactions with others. The structural component of this model is composed of several pos- sible identities conceptualized primarily in terms of regions varying in size or position and tenure, possessing front and back regions, and displaying both unique and overlapping regions with one another. The existence of multiple identities implicates multiple corresponding “targets” of attachment and ex- pressions of connection. Activity and activity foci are included in the model to help define the situation and thus account for identification with one tar- get as opposed to or along with others at various times. In this situation-sen- sitive view, activity is presumed to “link” to a certain identity region (or set of overlapping regions) more so than to others. The essay closes by discussing some of the advantages o f a structurational view of identification.

Research on issues of organizational attachment, in organizational com- munication and in organizational studies generally, “has mushroomed during the past decade” (Bullis & DiSanza, 1995, p. 1). This interest in the issue stems from the fact that reasonable levels of attachment (in- cluding identification, commitment, and loyalty) have been connected to many positive outcomes for the individual and the organization (for reviews, see Ashforth & Mael, 1989; Cheney, 1983b; Cohen, 1993; Meyer & Allen, 1997; Tett & Meyer, 1993). This concern about the individual- organization relationship and about our attachments to various facets of organizations has motivated productive research programs in and across several fields, including work on organizational identity (e.g., Albert & Whetten, 1985; Carroll, 1995; Cheney, 1991; Cheney &

Copyright 0 1998 International Communication Association 298

Identification Theory

Christensen, in press; Christensen & Cheney, 1994; Czarniawska-Joerges, 1994; Dutton & Dukerich, 1991; Dutton, Dukerich, & Harquail, 1994); identification and unobtrusive organizational control (e.g., Barker, 1995; Barker & Tompkins, 1994; Bullis, 1991; Bullis & DiSanza, 1995; Bullis & Tompkins, 1989; Cheney, 1983a, 1983b, 1991; Cheney & Tompkins, 1987; Tompkins & Cheney, 1983, 1985); dual commitment to union and company (e.g., Barling, Wade, & Fullager, 1990; Beauvais, Scholl, & Cooper, 1991; Bemmels, 1995; Fullager & Barling, 1991; Reed, Young, & McHugh, 1994); and multiple commitments generally (e.g., Abrahamson & Anderson, 1984; Becker, 1992; Becker, Randall, & Riegel, 1995; Gregersen, 1992, 1993; Hunt & Morgan, 1994; Reichers, 1985; Yoon, Baker, & KO, 1994). The timeliness of this issue is further fueled by the declining level of loyalty organizational members feel toward their employers (Kleinfeld, 1996; Moskal, 1993) and popular press procla- mations regarding the “death of corporate loyalty” (Death, 1993). Such concerns seem likely to persist given the current organizational climate of downsizing and other forms of restructuring.

For us, the attachment process is largely symbolic and is shaped by both individuals and the social contexts of which they are part. We be- lieve this issue of attachment concerns the linkage between an individual and some “target”’ or social “resource” based on perceived social mem- berships and the manifest behaviors that produce and are produced through those perceived memberships (see Patchen, 1970). Similarly, identification has been defined as both “the perception of oneness with or belongingness to [a collective], where the individual defines him or herself in terms of the [collective] in which he or she is a member” (Mae1 & Ashforth, 1992, p. 104) and as occurring when an organizational member “desires to choose the alternative that best promotes the per- ceived interests of that organization” (Tompkins & Cheney, 1985, p. 194). Thus, we focus here on identification, rather than other forms of attachment, as most useful for illustrating the attachment process. We do this for several reasons: (a) loyalty is but one dimension of identifica- tion (Cheney, 1982; Patchen, 1970), (b) attitudinal commitment2 as de- veloped primarily in the organizational behavior literature shares much overlap with identification (see Motazz, 1989; Sass & Canary, 1991), and (c) both identity and communication, which we will discuss in much greater detail, have been more clearly linked to “identification” than to other forms of attachment (Burke, 1969; Cheney & Tompkins, 1987).

Because of the sometimes divergent work in this area and the growing importance of attachment to various aspects of organizational life, we see the need for a conceptual or theoretical framework that helps ex- plain organizational attachment (even granting various parameters around generalizability). We argue that organizational identification re- search, and attachment research more generally, could benefit from an

299

Communication Theory

integrative model that incorporates three primary components taken from existing research on organizational attachment, as well as from broader advances in organizational communication studies. To that end, we ad- vocate the use of a structurational model (see Giddens, 1984) of identi- fication that highlights three of its important aspects. First, we incorpo- rate Giddens’s duality of structure as a means of both linking identities with identifications and addressing tensions between action- based and structure-oriented views of identification (and identity) in the literature. Second, we elaborate on the regionalization of multiple and intercon- nected (though often distinct or even conflicting) identities that make possible and depend upon identification with various targets within the organizational context (e.g., organization, occupation, work team, etc.). Third, we introduce situated activities as a means to explain when cer- tain identities may be more o r less salient for the organizational mem- ber, and thus explain the strength and nature of one’s identification with varied targets in different situations. In sum, we utilize certain aspects of structuration theory as a guiding framework to present a situational view of multiple identifications (and identities) in organizations.

The three main sections of this paper address the issues of duality of structure, regionalization of multiple identities, and situated activity, re- spectively, as they apply to the attachment process. We conclude by dis- cussing implications of this model and directions for future identifica- tion research. However, before we more specifically define some of our key terms and embark on our own theoretical narrative, it is important to clarify our goals and to delineate clearly the bounds of our eff0rt.j To keep our essay both manageable and focused, we will exclude from con- sideration here broad institutional identifications such as those associ- ated with class, religion, nation, ethnicity, etc. (see, e.g., Alonso, 1995; Devereaux, 1975; De Vos, 1983; Gray, 1989; Pye, 1962; Stoecker, 1995; Stohl & Perrucci, 1997; Willis, 1977; Winter, 1968). At the same time, however, we do not mean to suggest that these identifications (and asso- ciated identities) are unimportant; rather, in this essay, we simply wish to emphasize those identifications linked to such distinctively organiza- tional phenomena as position, work group, organization, profession, e t ~ . ~ Also, we will, for the most part, not consider psycho-sociological and developmental perspectives o n identity and identification as being within the scope of this essay (even though those form much of the basis for both the organizational attachment literature and our individual at- tachments to others). Thus, we seek to offer a middle-range theoretical and heuristic framework for understanding better how organizationally related identities and identifications serve to structure one’s experience, how they become meaningful in action, how they are evoked situationally, and how they relate to one another and to some sense of an overall identity (with a “working” level of coherence). Following Clegg’s (1994)

300

Identification Theory

suggestion, we are trying to offer an analysis of organizational life that allows structure and process to be conceptualized simultaneously and in conjunction with one another. This is precisely why we turn to Giddens’s writings on structuration (e.g., 1984, 1991a) for assistance in develop- ing our ideas about identities and identifications within the context of situated organizational activities, while still acknowledging the merits of several other theories.

Identification and Structuration Theory Given our interest in the attachment process, we find structuration theory (Giddens, 1984) a generally useful perspective. Although structuration theory has been used to understand better a variety of organizational constructs, including climate (see Bastien, McPhee, & Bolton, 1995; Poole, 1985; Poole & McPhee, 1983), culture (Riley, 1983), networks (Corman & Scott, 1994), and small group development (Poole, Seibold, & McPhee, 1985), that alone is hardly enough t o warrant its use here. Structuration theory is appropriated in this essay as a framework that provides us with three concepts we wish to apply to the overall identifi- cation process: duality of structure, regionalization of structures, and situated activity.

Giddens’s development of structuration theory focuses on the duality between structure and system, which involves the process by which hu- man action both produces and is mediated by structure. That is to say, the product of our actions today may well be a structural resource for interaction that we rely upon in the future. Structures are viewed as generative rules and resources available to actors that are capable of bringing about certain states of affairs (Giddens, 1981). Giddens (1984) suggests these structures are usefully understood as formulas or general procedures that tend to span time and space. McPhee (1985) offers a valuable operationalization of this view when he describes organiza- tional structures of various types as “substitutes” for larger communica- tion processes. As another example, language use depends in part o n relatively enduring formulas and procedures for the meaningful arrange- ment of symbols (grammar, syntax, and lexicon). Language use itself, however, can sometimes modify those structural resources, thereby pro- ducing new bases for future interactions. We shall begin linking identifi- cation and structuration theory by focusing o n this duality as it describes the attachment process.

A second portion of structuration theory we draw upon is Giddens’s (1984) notion of regionalization, which refers to the “temporal, spatial or time-space differentiation of regions either within or between locales [settings for interaction]” (p. 3 7 6 ) . As we view it, the various rules and resources available to an agent are regionalized, o r grouped, into certain

301

Communication Theory

identities. We will argue that the multiple identities of organizational members, which are “structures” in this theory, are usefully understood through regionalization. Thus, the second portion of the essay links the attachment process to the idea of regionalization as found in structuration theory.

The other portion of the duality in this theory, labeled “system,” rep- resents the reproduced relations between actors organized as regular social practices. System-level phenomena regularly take the form of so- cial interaction situated in time and space. The situation may be defined, at least partly, in terms of the activities in which organizational actors are engaged. In structuration theory, structure is instantiated in social interactions (systems) that are situated contextually. Thus, our third and final connection between structuration theory and identification will ar- gue for a situationally changing view of the attachment process.

Even though we are drawing o n Giddens’s structurational perspective in order to frame our arguments, this should not be taken to mean we are following exactly the theoretical agenda offered in Giddens’s (1991a) writings on identity. Giddens concentrates on the development and mo- tivation of individual agents o n the one hand, and the effects of large- scale institutional processes on the ~ t h e r . ~ In this essay, we take a some- what different approach, allowing both microindividual and macrosocial phenomena to recede into the background at least temporarily. We bring to the fore the question of how identity functions as a resource for inter- acting with others in social settings like groups and organizations. In a nutshell, we argue that agents draw on a plurality of structural identity resources in making certain identifications. These produce or reproduce the structural resources and differentiate them into various identity re- gions. It is in terms of this “double structuring” (Shotter, 1983), along with emphasizing the mediation of identity by other structures of com- munication and activity (such as work, which we shall discuss in detail later), that we claim our allegiance to structuration theory. Certainly, Giddens’s work may be interpreted as a worldview (Kilminster, 1991) applicable to a wide range of issues; indeed, Giddens himself says the concepts in his theory “should for research purposes be regarded as sen- sitizing devices . . . useful for thinking about research problems and the interpretation of research results” ( 1984, pp. 326-327). It is in this heu- ristic spirit that we utilize the theory.

The Duality of Identification Structuration theory provides the framework to describe identification as both a process of attachment and as a product of that process. The duality of structure allows us to discuss the attachment process that involves both identification as system and identity as structure. Indeed,

302

Identification Theory

we shall endeavor to use the term “identification” when we mean inter- action or other behaviors illustrating one’s attachment; we shall speak of “identity” as a set of rules and resources that function as an anchor for who we are; and we will speak of the “identification process” or the “attachment process” when we wish to address the relationship between identity and identification during attachment. In an attempt to clarify how we view identity and identification, we begin by describing them separately for analytical purposes; however, we remain more interested in their duality than in sharp distinctions between them. Finally, we il- lustrate how a structurational model addresses the action-structure de- bate in the identity and identification literatures. Identity Identity has become a tremendous preoccupation of organizations as well as individuals in the modern world. Identity is elusive in today’s society precisely because it cannot be established once and for all, for any individual or group, yet it is constantly and fervently being sought (Cheney & Christensen, in press). Identity, especially in the lay sense, involves central, distinctive, and enduring characteristics (Czarniawska- Joerges, 1994). Thus, any identity includes core beliefs or assumptions, values, attitudes, preferences, decisional premises, gestures, habits, rules, and so on. Identities, ideally speaking, provide us with relatively stable characteristics that make up the self. Each identity references certain norms and other ideas about who we are, how we are to act, and what is important to us. An identity makes further sense in terms of defining who we are and who we are not on the basis of social memberships; for example, one’s identity as a teacher carries with it certain assumptions, values, and beliefs about what it means to belong to this social category labeled “teacher.”

Although the identity of a person (or an organization) suggests its enduring qualities or perhaps even its “essence,” the emphasis on unique- ness or special qualities marks a distinctively Western, individualistic cast that identity has in much public as well as private discourse. We prefer instead to stress the function of identity as an “anchor” of the individual or collective self. In this way, identity is something we all need and make reference to, even when we acknowledge changes in the identity of oneself or another. To adapt Giddens to our view, identity is usefully seen as structure; that is, each identity constitutes a set of rules and resources that may be drawn upon by an organizational member. These identities not only help define who we are, but also provide us with the necessary resources we need to interact with others. Identity represents a type of knowledge about a part of our self that helps to produce and to reproduce behaviors in specific social situations. Thus, important parts of any identity are usefully thought of as “memory traces” (Giddens, 1984, p. 25) in the individual. However, such a view is incom-

303

Communication Theory

plete in that it suggests a rather cognitive and reified view of the identity construct. Thus, we add several clarifications. First, identity is constantly being produced, reproduced, and altered via external presentations of our identity. More specifically, identity is shaped by and revealed through discourse, as discussed in much greater detail below (see H a r r t & Gillett, 1994, for a detailed discussion o f the discursive turn). Second, many aspects of identity may be institutionalized (e.g., roles) and thus located in places other than memory or cognition (e.g., established common rules, rituals, handbooks, plaques, etc.). Third, an identity may not be “recognized” for or by a person until activated in a certain situation (as described in the final section of the essay). Fourth, one can also treat identity as a feature of organizations (although the most relevant identi- ties for an organization would differ from those of the individual and might include industry, community, and nation, among others), which again suggests a less cognitive view of identity.

This structurational view suggests a somewhat different view of iden- tity than is assumed in much of the relevant organizational literature. Giddens (1984, p. 282) recognizes that social identities “are ‘markers’ in the virtual time-space of structure. They are associated with norma- tive rights, obligations and sanctions.” Although some definitions of identification (especially those based on social identity theory) stress that identification “is the perception of belongingness to a group classifica- tion” (Mael & Ashforth, 1992, p. 104), it is perhaps more accurate to say it is really a social identity that indicates our perceived membership in a social collective or some other object of attachment. We develop an organizational identity when we perceive having joined an organiza- tion, various workgroup identities as we become members of project teams or departments, and so on. Although we d o not wish to get focus o n the fuzzy distinctions between perceived and actual membership, we believe perceived membership is all that is necessary for the initial for- mation of an identity associated with that social category. For example, to the extent I perceive that I am a “professional,” I have that identity. Of course, one’s own view of his or her identity may be challenged at any point by another person o r group. Thus, there is always the poten- tial for negotiation of an identity in interaction (actual or imagined). It is to this realm of social interaction that we turn next. Identification Within this theoretical context, identification is the process of emerging identity. Identification, especially as expressed in symbolic terms, repre- sents the forging, maintenance, and alteration of linkages between per- sons and groups. Often made manifest in social interaction, identifica- tion in a structurational sense represents the type of behavior produced by and producing identity. Thus, identifications are situated in contexts of interaction in the presence of other social actors (or in reflection of or

304

Identification Theory

in anticipation of such). Identifications are more readily subject to change than are more enduring identities (as structures). Identifications, as a system-level construct, occur in socially recognized behaviors, usually involving others, that signify some degree of attachment to various so- cial collectives or other targets. Such behaviors indicating identification might include decision-making behaviors, actions done with consider- ation of various targets’ interests, continued membership in a collective, and citizenship behaviors toward one or more targets.

For us, perhaps the most important indicators and expressions of iden- tification are found in language. Thus, communicative manifestations of identification are emphasized as they occur in social interaction with others a t the system level. Although some applications of social identity theory tend to suggest that identification may occur in the absence of interaction (Mael & Ashforth, 1995), we see actual, hypothesized, or even retrospectively examined interaction as being essential to the devel- opment of identification. Our “accounts of personhood” (Shotter, 1984) necessarily involve other actors (individuals and groups) and our rela- tions with them. It is through communication with others that we ex- press our belongingness to various collectives, that we assess the reputa- tion and image (Dutton, Dukerich, & Harquail, 1994) of that collective, and that the social costs and rewards of maintaining various identities are revealed. The story we tell of ourselves in interaction ( o r posit with respect to interaction) with others is the essence of identification. Never- theless, we are often engaged in such “conversations” even when inter- action is not taking place (see, e.g., Tompkins & Cheney, 1983). Others are “present” in symbolic form, as representations (e.g., in the form of Mead’s, 1934, “generalized other” or in the form of targets of potential accounts for behavior).

In general, close analysis of the communicative expression of identifi- cation has not been widely considered in the identification literature, despite Cheney and Tompkins’ (1987) contention that “the process of identification is conducted primarily with language, and the product of identification is expressed primarily with language” (p. 1 1 ) . Two excep- tions are Cheney’s (1991) analysis of the development of the U.S. Catho- lic bishops’ 1983 pastoral letter o n war and peace, which explores mul- tiple levels and arenas for the expression of “we”; and Bullis and DiSanza’s (1995) research o n U.S. Forest Service documents, which analyzes iden- tification strategies in internally and externally directed documents of the agency. Hecht (1993) argues that the new stance on identity views identification as inherently communicative because “messages are sym- bolic linkages between and among people that, at least in part, are en- actments of identity” (p. 78). The commitment literature, especially that focusing on behaviors, has noted that “talking up” the organization is indicative of commitment (see especially Kaufman, 1960). Even here,

305

Communication Theory

though, the focus tends to be on traditional outcome behaviors (Larkey & Morrill, 1995) such as job turnover and absenteeism. Larkey and Morrill, who examined three case studies of organizational commitment, concluded that “narratives of identity may provide clues to the ways in which people manage the shifting and multiple identifications with their organization and its many faces” (p. 2 0 9 ) . Although we firmly believe that examination of communicative expressions of identification is im- portant, such empirical or interpretive work is beyond the scope of this theoretical essay.

In sum, identification represents the dynamic social process by which identities are constructed, through which they guide us, and by which they order our world. This dynamic process involves manifest behaviors in explicitly or implicitly social settings that illustrate one’s linkage to some “target” (usually, a social collective). Thus, we have attempted to illustrate the importance of “others” in social expression of identifica- tion. Indeed, identification behaviors (especially communicative expres- sions) are most meaningful in a social context. More importantly, structuration theory draws attention to the relationship between struc- ture and system. Thus, it is t o the relationship between identity and identification that we turn next. Identity-Identification Duality Although it is necessary to make the analytical distinctions between iden- tity and identification, these concepts are dialectically interrelated, espe- cially as seen from a structurational standpoint. Thus, we are most in- terested in the broader process relating the two. Structures and systems imply one another, so the real emphasis is o n the duality between them. This duality of identity and identification accounts for the perceived linkage between (re)sources of identity and our (re)presentations of iden- tification. Thus, in a structurational sense, this duality involves the ap- propriation of identities in the expression of identifications, which in turn serve to reproduce, regionalize, and unify identities (as discussed in the following section).

It is perhaps relatively easy to imagine how one might appropriate identity structures in the creation of identification. For example, one is able to tell one’s “professor” narrative by drawing o n the values, beliefs, premises, and so on, that are his or her occupational identity. Further- more, several scholars have noted that what we refer to as identification may in turn reinforce identity (Dutton, Dukerich, & Harquail, 1994; Mae1 & Ashforth, 1992). For example, statements about work-team membership may further establish one’s belonging t o that team, which in t u r n prompts further expressions of team identification, etc. Czarniawska-Joerges (1994) and Christensen and Cheney (1994) point to the importance of narrative in the ongoing construction of identity: Individuals and collectivities regularly engage in reflections on the “self,”

306

Identification Theory

ACTIVITY IDENTIFICATION

Figure 1. Sructurational Model of Identification During Siuated

1 Activities

IDENTITY

with their past commentaries being incorporated (in one way or an- other) into the larger story that, in some sense, becomes central to iden- tity. One’s social actions illustrate this duality as well. As one of the reviewers of this manuscript explained, not only are identity and identi- fication products of one another, but they make sense,of one another. I am a father because I act in fatherly ways, and my fatherly ways make sense because I am a father.

Although the duality of structure allows for this reinforcing influ- ence, structuration theory also permits ironic appropriations of struc- ture and unexpected consequences of doing so. An organizational mem- ber may in fact draw on a work-team identity to express a lack of iden- tification with that group (as when he or she is unhappy with how the team is doing). Furthermore, those expressions of &identification may have the unexpected effect of greatly increasing the salience of an orga- nizational identity that is perhaps only loosely coupled with the work- team identity. The identity regions model we present in the next section helps account for such ironic appropriations and unexpected conse- quences, but for now we simply wish to say that identities and identifi- cations need not be reproduced the same way over time (though fre- quently they are), as there may be unintended consequences of appro- priating certain identity structures.

The right half of Figure 1 presents the identity-identification duality, which we refer to as the attachment modality,6 or process. To repeat, this duality treats the available identities an organizational member may have access to as a set of rules and resources (structure) that can be drawn upon in the production of identification (system-level construct)

307

Communication Theory

with some corresponding target. For instance, “talking up” one’s work- team involves drawing o n one’s work-group identity. This usage of the terms is similar to Cheney and Tompkins’s (1987) discussion of identifi- cation as the appropriation of identity. Structuration theory provides a useful framing device for linking these two related constructs in a way that does not give priority or bias t o structure. In a sense, we can enter the discussion of identity o r identification a t any point by applying con- cepts from structuration theory. Identities, as relatively or comparatively fixed markers in time and space, are social structures, though they may well be connected t o nonsocial aspects of “personhood,” such as inter- nal (physiological or partially physiological) drives toward maintaining a delineation between self and environment (Christensen & Cheney, 1994). As social structures, however, identities are not immutable-even for tradition-bound organizations such as the Roman Catholic Church. Structure, in Giddens’s terms, is both a product of interaction and a resource for interaction. In this way, one’s identity as a professional may be both a result of socialization into one’s field and a source of guidance or even inspiration t o behave in a certain way in the future. Identities represent ways of setting aside certain issues for the individual or group: By treating certain features as fixed or as decided upon, we are liberated from revisiting myriad options for defining the self and for acting in accordance with our self-images. By deciding o n an identity-even for a momentary situation or a well-defined role or a specific activity-we narrow the “range of vision” in further decision making (Simon, 1976), reduce the need for either self-deliberation or open-ended conversation with others, and thereby “locate” oneself in time and space. However, the need or struggle to d o this is constant and, in fact, becomes all the more urgent for the individual or organization in today’s communica- tive environment, given the plethora of options for self-definition and the inescapable tension between uniqueness and being part of the cul- tural crowd.

As a simple, touchstone example of this. duality that can be elabo- rated on as we progress, we offer the following. John is an associate professor of communication a t a large university. He has been employed there for a number of years and received his tenure there several years ago. For John, part of being a n associate professor means being knowl- edgeable and rigorous, and not being afraid to take on challenging or sensitive topics in class. H e tends to think of himself in these terms, too, and acts accordingly when planning and presenting class material. His attachment t o his job and what that means is revealed in much of what he says t o others. As we all know, students learn about their professors during their classroom experiences, and they share this information with other students. As a result of this process, John has a widely recognized reputation among students and colleagues as a rigorous teacher who is

308

Identification Theory

willing to take risks, thus reinforcing his own view of himself. His iden- tity as an associate professor both shapes and is shaped by the social interactions that reveal his identification with that target. Action-Structure Issue As suggested above, one important outcome of a structurational view is that it allows for the expression of identities in social interaction where they may be reproduced in familiar o r in new ways. However, structurational views have also been offered as one way to overcome tensions between action and structure biases (Bryant & Jary, 1991; Van de Ven & Poole, 1988). This issue is crucial because it represents an- other way of framing the related debates between change-stability, voluntarism-determinism, and psychologism-holism (Haines, 198 8 ) . We believe one of the strengths of our proposed structurational model is that it recognizes the dual influences of both individual agency and in- stitutional structures without assuming superiority of one over the other as a defining mode of explanation. Briefly, we attempt here to illustrate this tension in the literature and then discuss how our model addresses it.

As for the identity literature, Czarniawska-Joerges ( 1994) observes that the conventional school of thought sees identity as found in indi- viduals as part of their “genotype” o r “soul” (p. 195); conversely, the social environment school claims that society wholly determines iden- tity. Structuration theory allows for both views without necessarily hav- ing to favor either. “Identities . . . are the result of structuration pro- cesses” (Czarniawska-Joerges, 1994, p. 216) in that they both shape and are shaped by social interactions. Christensen and Cheney (1994) also try to advocate a balanced view of identity by claiming it “is not just an external or social construct but is also an internal drive” (p. 227), acknowledging among other things the role of individual and so- cial “bodies” and their boundaries. Such a balance is a t least closely approximated in structuration theory.

The action or structure issue is also apparent in the organizational attachment literature, though it has not been so explicitly discussed there. Much of that literature seems to take an implicitly action-oriented ap- proach that assumes it is individuals who freely choose with which “target(s)” they will identify. Nearly all of the definitions of identifica- tion refer to an activity where the individual “identifies,” “accepts,” “desires,” “has an impression,” “defines his o r her membership,” and so forth. In this sense, much of the literature on identification and com- mitment in the organization adopts a n individualistic bias that can pre- vent us from seeing how these very same individuals are operating within “fields of discourse” that culturally and historically set parameters for identity construction (Foucault, 1984).

However, some views of identification offer a more structurally ori- ented explanation, which suggests that the organization or other social

309

Communication Theory

structures are largely responsible for shaping individual action.’ Larkey and Morrill (1995) argue that in the commitment literature “the indi- vidual is conceived as a relatively passive agent who ‘gets committed’ as a result of immediate situational factors either directly or indirectly ma- nipulable by the organization’’ (p. 195). Mae1 and Ashforth 01995) re- port support for their claim that “certain characteristics and life experi- ences [serve to] predispose one to identify with a given organization” (p. 313), a conclusion we see as supporting a structural view of identifica- tion. A more frequent structural explanation is that one’s organizational position, or role, may strongly shape identification (McCaul, Hinsz, & McCaul, 1995; Robertson &Tang, 1995). Tompkins and Cheney (1985) note that identification in one role may not be relevant in another role, which leads to their operational redefinition of identification (cf., Simon, 1976): “A person identifies with a unit when, in making a decision, the person in one or more of hidher organizational roles perceives that unit’s values o r interests as relevant in evaluating the alternatives of choice” (p. 193).

We see neither of these divergent positions as capturing the whole picture, and structuration theory can help to account for limitations of each (Knorr-Cetina, 198 1 ). A structurational theory of identification would see both individuals as actively identifying themselves with vari- ous targets and view organizational or other structures as influencing the identifications of members. The very structural forces that shape our identifications were (and continue to be) shaped by individual actions. Thus, a structurational view of identification presents a conceptual frame- work for seeing this process as a duality, that is, the attachment process involves both individual action and institutional structures. In this sense, identities (structures) are both constraining and enabling when it comes to human agents actively identifying.

Indeed, many acts of identification on the part of individuals or groups are active and even strategic. For example, a person consciously chooses to continue membership on a project team or voluntarily talks about what it means to be a member of his or her profession. However, iden- tity and identification can also be less active in the sense of being signifi- cantly outside the realm of conscious strategy on the part of an indi- vidual or collectivity. The person who identifies with technology to the extent of mistaking its powers for his or her own (Burke, 1973) is, in a sense, appropriating the material and symbolic aspects of a structural resource for identity. The individual whose new managerial position comes with the prescription of greater organizational loyalty would also suggest a more structural (and less active) influence on identification. Again, we believe a structurational view helps to highlight the dual in- fluences of action and structure in the attachment process. Thus, our associate professor John identifies with his occupation in part out of

310

Identification Theory

individual choice and in part due to his structural position and title in the organization-and that individual choice both constitutes and is constituted by the structural position and its attendant expectations.

Regionalization of Multiple Identities In addition to the duality of structure, Giddens’s writings on “regionalization” provide a useful means of helping .to conceptualize identities (i.e., structures), their relations with one another, and their relations with corresponding identifications. Although we focus here on the identities of any given organizational member in an attempt to make our arguments as clear as possible, we do not support a unidirectional view of the attachment process nor wish to give identity any sort of ultimate primacy. The duality of identity-identification clearly indicates that identities construct and are constructed by identifications, thus con- necting individual members to larger social collectives. We begin this section by briefly reviewing the literature on multiple identities or iden- tifications. From there, we discuss Giddens’s notion of regionalization and use it to describe four key properties of identities that both shape and are shaped by identifications. Multiple Identities There are surely a large number of potential identities for any one per- son in modern, Western, industrialized societies, as recognized very early by Durkheim (1933). As Weigert, Teitge, and Teitge (1986) explain, the availability of identities shows a striking expansion in the 20th century and, “through all the change and complexity, modern individuals struggle to control their lives by organizing the mix of personal identities into a meaningful arrangement of biographical importance and situational flex- ibility within an increasingly rational and abstract social context” (p. 57). As they explain, what is sometimes termed an “identity crisis” for the individual, in the sense of the search for or dearth of meaning, can also be framed as a choice from among too many resources for or tar- gets of identity.

Indeed, from the standpoint of social identity theory, which several scholars (Ashforth & Humphrey, 1993; Ashorth & Mael, 1989; Dutton et al., 1994; Mael & Ashorth, 1992,1995; Scott, 1997) have used in the context of organizational identification, a person’s self actually consists of a personal identity and multiple social identities, each of which is linked to different social groups (Dutton et al., 1994). This perspective can be seen as an elaboration of James’s (1950) classic formulation of identity as both a highly social and a truly individual matter: In his view, one’s identity consists of references to groups whose opinions matter to the individual. Thus, it is useful to conceptualize individuals as having organizational, gender, class, occupational, ethnic, work-team, national,

311

Communication Theory

and several other identities. This is consistent with Burke’s (1937) views of various “corporate” identities, Larkey and Morrill’s (1995) view of commitment as involving the interplay of multiple identities of the self, and even Kahn’s (1990) premise that people have dimensions of them- selves that get used to various degrees in engaging o r disengaging at work. These different identities then correspond to different identifica- tion targets, or resources of identity. For example, we draw upoa ’ i s well as [relcreate) the resource of an organizational identity in express- ing organizational identification. Social identity theory views the vari- ous identifications as extensions of the identities that compose the indi- vidual (Mae1 & Ashforth, 1995). Thus, what may be conceived on one hand as a target toward which an individual aims his or her attachments should simultaneously be viewed as a resource that is appropriated by the individual for construction of the self and for self-expression (see, e.g., Cheney & Tompkins, 1987). Indeed, this is apparent during antici- patory organizational socialization, where a new recruit both shapes behavior toward the organization and uses the organization as a reason for various actions and attitudes.

Extant research suggests a multitude of possible identification targets relevant to organizational members. In addition to organizational iden- tification, which Rotondi, Jr. ( 1975) defines as “identification with the employing organization as a whole” (p. 97), previous identification ef- forts have labeled occupations o r professions, individuals or persons, work positions, task groups, reference groups, subgroups, task, depart- ment, hierarchy level, work, and invisible colleges as possible targets (see Cheney, 1991; Hall, Schneider, & Nygren, 1970; Kaufman, 1960; March & Simon, 1958; Rotondi, Jr., 1975; Simon, 1976; Tompkins & Cheney, 1983). In the 1960s and 1970s, extensive research explored the tensions between organizational identification and professional identifi- cation (e.g., Hall & Schneider, 1972; Hall et al., 1970; Patchen, 1970; Rotondi, Jr., 1975). Several more recent identification studies in organi- zational communication also illustrate a pluralistic approach in their empirical examinations of two or more identification targets, especially in terms of multiple organizational levels (Barker & Tompkins, 1994; Cheney, 1991; May, 1995; Ruud, 1996; Scott, 1997). Additionally, the commitment research reveals targets including coworkers, superiors, jobs, work, subordinates, occupations, careers, professions, unions, and other groups o r individuals in the workplace (see Becker & Billings, 1993; Koslowsky, 1990; Meyer & Allen, 1997; Morrow & McElroy, 1993). The work on dual commitments (see, e.g., Beauvais, Scholl, & Cooper, 1991; Bemmels, 1995) and multiple commitments (see, e.g., Gregersen, 1993; Hunt & Morgan, 1994; Reichers, 1985) emphasizes a pluralistic view of attachment in that literature.

Combining social identity theory, which emphasizes a personal iden-

312

Identification Theory

tity and various social categories in which we may hold membership- either formally or quite loosely-with the existing research on identifi- cation and commitment targets suggests, to us, four especially relevant identities in organizational life: individual (personal interests that put the individual’s well-being above more social considerations), work group (e.g., team or department, where the interests of an immediate and inter- acting group are strongly considered), organizational (where the inter- ests of the employing or primary organization are most salient), and occupational or professional (e.g., where consideration is made about the effects of one’s actions on their industry, professional associations, unions, or job types). As discussed previously, these by no means repre- sent all possible identifications. Also, as we shall explain, even though each of these identities may overlap substantially for organizational members, each identity is analytically distinct and practically important. Furthermore, we recognize that each of these targets can be somewhat vague, especially as the collective becomes larger; therefore, “organiza- tion” can mean substantially different things in multidivision, global organizations, and occupation o r profession may at times seem equally ambiguous. Nevertheless, these four d o represent a useful starting point for organizational analysis given their salience in the contemporary or- ganizational world, and we turn next to a discussion of how those iden- tities are “regionalized” and how they relate to one another. Regionalization As explained earlier, Giddens’s (1984) notion of regionalization refers to the “temporal, spatial or time-space differentiation of regions either within or between locales [settings for interaction]” (p. 3 7 6 ) . He notes that regionalization is important because it counterbalances assumptions about homogenous, unified societies. Although we d o not here address the matter of broader forms of social differentiation, we d o believe regionalization is a useful way of describing identities and how they relate t o identification. Thus, like society, or even organizational cul- tures, identities need not be homogeneous o r unified. Again, as we view it, the various rules and resources available to.an agent get regionalized, or grouped, into certain identities. The self is usefully thought of as be- ing regionalized into these identities. We d o not wish to suggest these regions exist most of the time in a literal physical sense, although they are in fact sometimes conceived that way by individuals. Rather, the power of the regions metaphor is that it allows us to visualize the ways in which these identities are interrelated, synchronically speaking. Thus, we shall focus here primarily on the spatial dimension of regionalization, saving a somewhat more detailed discussion of temporal changes to the final section of our essay.

Giddens’s (1984) writings o n structuration offer four modes of regionalization. Form represents the form of the boundaries that sepa-

313

Communication Theory

rate each region. Character, as Giddens explains, has to d o with the front and back regions and the importance of each. Span or scale repre- sent the size of the region, with larger spans often associated with insti- tutionalization. Duration refers to the endurance of a region over time. For our purposes, these modes correspond to four characteristics of iden- tity regions: unique and overlapping regions, front and back regions, identity size and position, and identity tenure. The bottom right portion of Figure 1 presents an example of our four especially relevant identity regions for an organizational member. Of course, identities are not nec- essarily arranged for the individual in a neatly ordered hierarchy. Never- theless, we believe this sort of spatial model can focus attention on these four characteristics of these connected identities. Overlapping and Unique Regions. With the recognition of multiple iden- tities, it becomes necessary to address the degree of compatibility and tension between and among them. In the identification literature, em- pirical evidence suggests some moderate, but positive, correlations be- tween and among various organizational targets of identification (see, e.g., Barker & Tompkins, 1994; Scott, 1997). Yet, others have claimed that resources of identity are regularly in competition with one another, from an administrative or managerial standpoint (Simon, 1976; Tompkins & Cheney, 1985; Van Maanen, 1976) as well as from a nonmanagerial one (see Bullis & DiSanza, 1995).8 ’The organizational commitment lit- erature also suggests both conflicting and compatible attachment tar- gets (Meyer & Allen, 1997). As Vandenberg and Scarpello (1994) ex- plain, the traditional view sees attachment t o one target as inherently incompatible with commitment t o another; however, their more con- temporary view suggests that various forms of commitment are posi- tively, not negatively, associated (a view supported in Wallace’s 1993 meta-analysis of the issue). Moreover, it is not at all clear that a “zero- sum” model of individual or group identity is the best one. Thus, we believe it would be a mistake to force linear relationships upon the com- plex interrelations between and among identities. We, of course, recog- nize that there are likely upper limits on how strongly one can be identi- fied with multiple different targets given that each takes time and en- ergy; however, we simply wish to emphasize that increased identifica- tion with a target need not be accompanied by a corresponding decrease in identification with some alternate target.

Our take on this compatibility and competition issue is that multiple identities (and identifications) are likely both. In sum, we find wisdom in Burke’s (1973) observation that organizational members may develop a variety of corporate identities, which are partially compatible and par- tially conflicting (see also, Cheney, l983a). Moreover, if we were to de- pict any one individual’s multiple identities (geo)graphically, we would likely find somewhat overlapping areas of attachment. For example, in

314

Identification Theory

an employee’s view work group and organization may be identical, par- tially overlapping, or entirely separate from one another. Creative lan- guage use, of course, allows for almost innumerable possibilities in terms of connections between and among identities, both synchronically and over time (Cheney, 1991).

Regions of overlap should be common since an individual would nearly always experience some overlap of his or her various identities. For example, it is difficult to imagine an individual being associated in any way with a profession and an organization if the two are completely incompatible. In fact, for a particular person, all identities may overlap extensively. An engineer, for example, may feel that his or her work is personally rewarding, central to the project group’s goals, consistent with the organization’s mission, and very much aligned with what the engi- neering profession is all about. Such instances demonstrate the overlap of all four relevant identities here. In such cases of overlap, identities share common elements.

At the same time, Mae1 and Ashorth (1992) note that different identities are often “loosely coupled” in that a change in one does not necessarily affect another. This point, too, is consistent with the model we describe here. To the extent that the identities are generally only loosely coupled, they retain a certain degree of independence. Thus, each identity region also has a unique area that represents incompatibility with or separation from other identities. As Christensen and Cheney (1994) suggest, multiple and competing identities (as well as the lack of consistent “grounding”) are what make identity crises possible and in fact necessary. An organizational trainer, for example, may find that increased class sizes are consistent with company concerns about lower- ing training costs, but may run very much counter to occupational iden- tifications that stress the need for more personalized interactions with trainees (facilitated in part through smaller class sizes). In a scenario such as this, the distinct regions of the organizational and occupational identities would be comparatively large given the lack of congruency between these two identities. Front and Back Regions. Whereas the characteristics of overlap and uniqueness describe relations between multiple identities, the front and back characteristics of a region begin to account for the ability of members to draw on the same identity during the expression of identifi- cation or disidentification with any one target. Although front and back distinctions can be ambiguous, we use this terminology specifically to distinguish between two key portions of any identity region: the region associated with “positive” identifications and the region associated with more “negative” disidentifications. Front regions, which are closely connected t o Goffman’s (1959) notion of face, are associated with what is sanctioned and official in a culture. In terms of identity regions

315

Communication Theory

specifically, the front regions are where we might find premises, beliefs, values, etc., in line with other similarly socialized individuals. Thus, identifications with a particular target would draw o n the front regions of the corresponding identity. Clearly, the front regions associated with identifications have received the most attention in the scholarly literature.

However, “back regions clearly often do form a significant resource that both the powerful and the less powerful can utilize reflexively to sustain a psychological distancing between their own interpretation of social processes and those enjoined by ‘official’ norms” (Giddens, 1984, p. 126). As Goffman (1959) notes, the back regions allow for more “re- gressive” behavior in situations of copresence, including elaborate grip- ing, inconsideration of others, and profanity. Thus, it is from these back regions of the identity that disidentifications, or negative identifications, are drawn.9 Giddens notes that these back regions are still a very impor- tant part of the region, and that they may serve as a way of relieving tension from the more tightly controlled front regions. Similar to Hirschman’s ( 1970) views on how complaints may still indicate loyalty, even though identity salience frequently results in strong identification with a target, complaints may also signify a strong disidentification (or at least an active perception of difference and distance). For example, during a n especially turbulent time period in one’s company, the organi- zational identity may remain very salient because of all the changes and discussions of them; however, that saliency may serve t o highlight disidentification with the organization if the employee resents the changes or the company’s attitudes toward workers. Though not considered as often, back regions and disidentification represent important counter- parts to the more widely addressed front regions and identifications. Identity Size and Position. Each identity region can also vary in size, relative to one another and relative t o the contribution each makes to the self. Size is an indication of the importance of that identity to the individual self as well as the number of features encompassed by that identity. For example, we might expect a thoroughly socialized organi- zational member t o have a relatively large organizational identity region, relating attachment to the whole organization. In such a case, organizational identification is especially likely given the prominence of the organizational identity. Over time, the influence of identification on identity may serve t o expand (or reduce) the size of certain identities, as our social interactions with others constantly adjust who we are as selves. As Giddens (1984) notes, larger regions may be associated with institu- tionalization; thus, our institutional identities may likely expand (in terms of size relative t o other identities and centrality t o the self) t o be among the largest of our identities as they are appropriated and reproduced repeatedly over time. For example, one’s set of role-related behaviors

316

Identification Theory

required by the bureaucratic organization may come to occupy a sub- stantial portion of that person’s identities with respect to both profes- sion and organization. Often these identities extend beyond a single or- ganization to encompass an industry or even to the market in general.

Identity size may also be an influence on the position of each identity relative to other identities in the self. The relative position of the identi- ties is a way of conceptualizing which identities are central and which become more peripheral over longer spans of time. Several commitment researchers have also theorized that some identities are conceptually more proximal or distant with respect to one another (see Gregerson, 1993; Hunt & Morgan, 1994), suggesting the identities’ positions relative to one another vary (e.g., occupational identity might be more proximal to union identity than to work-group identity). For simplicity, Figure 1 presents the identities as approximately equal-sized spheres and as ap- proximately equally central to the overall self. We, however, believe the centrality of certain identities may be paramount for shaping which iden- tity is most likely appropriated when one expresses identification (and recursively, our identifications can serve to alter the relative centrality of various identities over time). Identity Tenure. Each identity is also characterized by its duration, or tenure (the length of time that a specific identity has been part of the self). Although the spatial model does not illustrate time well,’O tenure seems to us an important characteristic of identity (even though mea- surement in chronological years may not be as meaningful as duration through trying times or periods of change). Tenure, of course, has gener- ally been found to be highly related to various forms of attachment (for recent examples, see Barker & Tompkins, 1994; Scott, 1997). The iden- tities with which we are born (e.g., gender, ethnicity, nationality) may have the longest tenure, but their involuntary nature may make them relatively small for many people in the absence of other influences. More voluntary identities, such as the types we address in this paper, may begin later in the self’s life, but they are still characterized by various degrees of tenure relative to one another. We suspect that tenure, at least for voluntary identities, contributes to larger and more central identities over time as well. After a considerable length of time working for an organization, there is often the sense of “sunk costs,” which can sustain a measure of identification for the individual even against some negative experiences. Additionally, the tenure of the identity in no way makes it less dynamic since the existence or presence of the identity is different (though not necessarily less important) than the salience of that identity. Furthermore, tenures do change as people join, leave,. and rejoin new work groups, occupations, organizations, etc. Summary. Although we do not wish to push the geographic metaphor too far, we do find it valuable in a heuristic ways and believe that it

317

Communication Theory

parallels folk understandings of this issue. Specifically, the regions meta- phor illustrates our contention that identities are in competition in their unique regions and are compatible in their overlap regions. Furthermore, the regions model addresses issues of identification and disidentification as well as relative position and size among the other identities composing the self. Finally, although the tenure of each identity is less visible in the diagram, it too seems necessary to capture the duration of each identity for the individual, which in turn may be strongly related to size and position. Thus, identities can be seen in diachronic as well as synchronic terms.

Furthermore, we see each of these characteristics as operationalizable. Compatibility o r competition might be assessed by correlations among identification scores with various targets. More qualitative techniques (e.g., eliciting accounts of decisions) could provide even clearer indica- tions of what identity o r identities are relevant in a given situation, thus helping t o indicate degree of overlap. Account analysis and other means of examining members’ talk can also reveal instances of identification and disidentification with a salient identity. Although current survey in- struments usually ignore the back regions of relevant identities, such assessment seems possible. Identification scores o n existing surveys have been used t o essentially get at the relative size, or importance, of a certain identity. Degrees of overlap and correlations among different identity regions may also indicate their positions relative to one another. Finally, tenure is most easily operationalized as length of time in a given social group or in association of some sort with it. However, we again emphasize that the number of chronological years in a given group is only one way t o operationalize duration. Additional means might include number of significant organizational changes or periods of envi- ronmental turbulence.

To extend the example with our associate professor John, these issues are made salient when a last-minute scheduling problem results in John’s being asked to cover a required course on interpersonal conflict in which his daughter is enrolled. As an educator, he wants to be sure he teaches his students everything they need to know to be effective negotiators and conflict managers. As a person who likes to be in control, he wor- ries that he will be “opening the playbook” for his daughter, who has traditionally been o n the receiving end of his strategies. John has also been supportive of university policy over the years, including rules and strong norms against nepotism. John, however, has always helped his daughter with difficult materials in other classes, seeing that as part of what a good father does. In this situation, as many as four long-standing and fairly central identities-professional, personal, organizational, and fatherly-are competing with each another. Thus, simultaneous identifi- cation with all of them seems unlikely, although disidentification with one o r more is not necessary either. Given this competition, John’s ex-

318

Identification Theory

pression of identification with his professorial identity may well pre- clude some fatherly behavior and may even override personal interests. Professional identification in this case, though, may be fairly compatible with organizational identification. Conversely, these identifications will further influence what it means to be a professor and a member of the university, as well as who he is personally and as a father. However, we know that John will not just stop being a father completely, nor will his personal interests be set aside permanently. Thus, a n understanding of when certain identities are more or less relevant seems essential. It is to such a n analysis that we turn next.

A Situated-Action View of Identification Although we have utilized the spatial metaphor to clarify differences between the front and back dimension, to address areas of overlap and uniqueness, and to mention relative size and centrality and duration, it is beyond the scope of our current efforts t o further explore the broader dimensions of the self in which these identity regions are situated. For us, the regions metaphor is useful because it helps visualize the relations between and among identities, but it is also limiting in that it tends to give less attention to issues of “time” and how identification takes place “in the moment.” With regard to this concern, perhaps the most impor- tant, though least addressed, issue in the attachment literature is when we identify with one o r more identification targets. We believe that a structurational model of identification can also begin to address this question by illustrating that identification varies from situation t o situa- tion. To explore this issue, we first review the current thinking on the matter. Second, we turn to a situated-action approach of identification, drawing o n structuration theory, and work on activity and communica- tion as a means of defining situations in which identification varies. Stability and Change in Identification Tompkins and Cheney (1985) summarize some of their work by point- ing out that identification varies contextually and changes over time. In spite of such claims, much identification research seems to make a tacit o r even explicit assumption that identification is stable and almost trait- like (see, for example, Bullis & DiSanza, 1995). Virtually all empirical studies of identification report a general and presumably stable measure of the construct that ignores specific contexts or changes over larger periods of time. Indeed, Larkey and Morrill (1995) criticize both the identification and commitment literatures for assuming essentially stable features. Kahn (1990, p. 693) charges that “these concepts [various forms of attachment] suggest that organization members strike and hold en- during stances (committed, involved, alienated), as if posing in still pho- tographs. Such photographs would show people maintaining average

319

Communication Theory

levels of commitment and involvement over time.” The general nature of most identification and commitment questionnaires may explain why many studies report moderately high levels of attachment. However, given the emphasis in the literature that people are not always identified with a given identification target and may well identify with some targets more so than with others, we are concerned that general, and by impli- cation stable, conceptualizations and operationalizations are inadequate for assessing more fleeting forms o f identification.

Although none of the existing identification research appears t o operationalize identification in a way that allows for a more changing (situationally dependent) view of the construct, several scholars suggest it should. Indeed, the various forms of fragmentation and integration that characterize advanced industrial society may require such instabil- ity: Both individuals and organizations find that stable identities become more elusive even as they are pursued more vigorously (Cheney & Christensen, in press). Interestingly, some of the earliest writing about organizations and communication suggests a view of identification as situationally changing. Writing about “willingness t o cooperate” and “loyalty,” Barnard (1968, p. 8 5 ) notes that “the willingness of any indi- vidual cannot be constant in degree. It is necessarily intermittent and fluctuating.” In more recent organizational identification work, reviews of social identity theory (Ash forth & Humphrey, 1993), self-categoriza- tion theory (Abrams & Brown, 1989), and the identity theory work of Peter Burke and colleagues (Burke, 1980,1991; Burke & Reitzes, 1981, 1991; Burke & Tully, 1977; Reitzes & Mutran, 1994) all suggest iden- tity, and thus identification, may vary situationally. Furthermore, Ashforth and Mael (1989) conclude that particular social identities are cued or activated by relevant settings. Carbaugh (1996) has also examined so- cial identities in the context of what he calls specific “scenes.” Bullis and Bach (1989) found that various turning points were related differently to identification with the organization; if one views turning points- which are labeled as events or occurrences associated with a change in a relationship (see Baxter & Bullis, 1986)-as comprising a certain situa- tion or occasion, then this research supports a more changing view of identification. Similarly, Bullis and Bach ( 1 991) found that varied topics of conversation were differently related to identification. To the extent that multiplexity, which is defined as the multiple bases for communica- tion between people, is analogous to varied contexts, this research is also in support of the argument being made here. Although referring specifically to service provision, Miller, Birkholt, Scott, and Stage (1995) have suggested a retreat from “gross categories” in favor of considering “the ebb and flow of identification” (p. 144). Identification may change when the collective’s identity is challenged (Dutton et al., 1994) or in crisis situations such as contract negotiations (Beauvais et al., 1991).

320

Identification Theory

Speaking directly about employment commitment, Banks and Henry (1993, p. 182) note that “it is not such a stable disposition concept as was once thought.”

Thus, we note an implicit assumption in much identification research t h a t it is a stable construct, especially with regard t o h o w it is operationalized. Furthermore, the lay notion of identity presumes and references stability or constancy in a sea of change. Indeed, there is an understandable felt need for such moorings of the “self,” both for indi- viduals and groups. At the same time, many researchers suggest identifi- cation is much more changing and fleeting. However, little has been offered to help explain the “changing” view of identification. Therefore, we believe a useful model of identification must be one that allows for some variation from situation to situation and a view that accounts for the enduring and shifting aspects of identity and identification. Thus, we are saying both that a single identification (say with one’s profes- sion) can vary over time (along with shifts in the economy) and even from one work activity to another (suggesting much more rapid changes during even short periods of work), and that movement across situa- tions can evoke a variety of identifications (including work team, orga- nization, and occupation). As Kahn (1990, p. 693) suggests, a n appro- priate approach for studying issues of attachment “requires deeply prob- ing people’s experiences a n d situations during the discrete moments that make u p their work lives.” To accomplish this, we turn to a situated action view of identification, ultimately linking activities to identification. Situated Action Perspective To help in conceptualizing identification as a more fluid construct, we turn to a situated-action view. By this we simply mean that we wish to highlight the importance of social contexts for identity formation (though such contexts do not completely constitute identity) and for expressions of identification. As we attempt to illustrate, situations may be defined largely by activities-and it is those activities that can then be related to the attachment process. To elaborate on this idea, we draw again on structuration theory. Structuration Theory. Giddens’s ( 1984) structuration theory explains that all social interaction is situated in time and space. Contextual issues of time and space are “one of the most distinctive emphases in Giddens’ structuration theory” (Bryant & Jary, 1991, p. 13), yet these features have been largely neglected by communication theorists. In this theory, structure is instantiated in social interactions (systems) situated contex- tually. Contextuality is defined largely by locales, which may range from a room or a street corner to the shop floor or a suburban mall. These locales are the settings in which “the routine activities of different indi- viduals intersect” (Giddens, 1984, p. 119). Organizational life is likely

Communication Theory

characterized by numerous locales, such as one’s workspace, the cafete- ria, customer sites, the assembly plant floor, the trade conference, the videoconference meeting, cyberspace, etc. It is true that organi- zations today increasingly operate in multiple places and are less tied t o specific locations; however, this makes the notion of “locales” even more relevant to organizational activities, not less so. Further- more, these locales both shape a n d are shaped by the content of people’s interactions.

In these locales, routine activities are the predominant forms of day- to-day social life. Giddens (1984) claims that the enactment of routines is one of the main ways in which we maintain who we are. He goes on to explain that the contextuality of interaction also includes what he refers t o as the copresence of actors. For him, it is the context of copresence that makes gatherings, social occasions, encounters, and routine inter- actions possible. As explained earlier, individuals encounter each other in their daily routines in contexts with copresent others. Applying these ideas to our structurational model of identification, we can say that iden- tification is expressed via narrative and other behaviors in varied con- texts, o r locales, of social interaction, usually to those and with those who are copresent. More relevant to our task here, we use Giddens to argue that it is one’s daily routines-and more precisely, one’s activi- ties-in a given locale that provide the context for identifications. This was demonstrated in Seashore’s ( 1954) examination of industrialized work-groups, where specific work situations, or contexts, were strongly related to group cohesiveness (and by implication, to identification). To develop our arguments about activity and identification, we turn next to a discussion of activities as defining features of situations. Activity, Communication, and Identification. Pervin ( 1978) argues that any situation is “defined by who is involved, including the possibility that the individual is alone, where the action is taking place, and the nature of the action or activities occurring” (p. 79). We concur in our belief that activity is a central characteristic of contexts that situate ex- pressions of identity.” Unfortunately, as Cody and McLaughlin (1985) note, the activity component has received very little attention from com- munication scholars, even though it is a defining feature of the situation. Few useful descriptions of the specific activities engaged in by organiza- tional members exist (see Russell, 1996, however, for a discussion of various concrete work activities and situations as they affect other orga- nizational practices), and most that do focus o n activities associated with role prescriptions o r extremely general dimensions of activity (see Hatch, 1987; Lau & Pavett, 1980; Mintzberg, 1971, 1975, 1980; Palmer and Beishon, 1970; Paolillo, 1981).

One recent theoretical effort that helps us relate identification to ac- tivity is Corman and Scott’s (1994) reticulation theory of communica-

322

Identification Theory

tion networks. This theory also takes a structuration approach, seeking to explaining how structural network relationships are produced and reproduced instances of observable communication. This duality between network structure and observed communication exists alongside a dual- ity between structural activity foci (Feld, 1981) and manifest systemic activities. In Corman and Scott’s theory, focused activities form the con- texts that situate reproduction of network relationships in communica- tive interaction (see also McPhee & Corman, 1995). The triggering of one set of activities (for example, strategic planning meetings) may lead to interaction between certain others (for example, increased contact between division heads in an organization). Conversely, that interaction may serve to reproduce not only underlying network structure, but the nature of those various activities. Thus, activity is clearly connected to communication in the theory of reticulation.

The link specified by Corman and Scott (1994) is important to our current efforts because it provides a basis for linking activity and identi- fication, which is itself very much concerned with communication. The left half of Figure 1 represents activity foci as structure and activity as system, forming a duality between the two in that activity foci both produce and are produced by the activities they organize. Our intent here is not to specify the activity-activity foci relationship (see Corman & Scott, 1994, for details), but rather to describe its relationship to the identity-identification duality. Activities influence the identities that are appropriated and reproduced in identification. Only in particular situa- tions, defined significantly by activity and activity foci, will a person identify in particular ways. For example, an engineer on a production team is most likely to identify with that team during group meetings or when an important team-set goal is met; yet, that same person may ex- perience more of an organizational identification during a budget meet- ing or when the company has just been awarded a major contract. Con- versely, the identification process also influences the very activities in which we are engaged. Engineers on a team of diverse professionals who view themselves as engineers more so than team members or organiza- tional employees can be expected to engage in more activities associated with their profession. Of course, in many instances we may not find such a simplistic one-to-one correspondence; indeed, research o n activ- ity foci suggests that they (like identity regions) also overlap and have varying degrees of compatibility with one another (see Corman, Stage, & Scott, 1997). Despite the complexity of the activity-identification re- lationship and the need for continued specification of these constructs, we believe that the work on activity and activity foci represents a useful way to begin talking about identification in situationally changing terms.

Mackenzie (1978), drawing on Burke’s work with identhcation, seems to confirm our proposed link between identification and activity, stating that:

323

Communication Theory

. . . a t least in complex societies, the use of “we” depends o n context; . . . certainly our individuality shifts a little according to context, chooses different words and gestures, gives priority to different interests, according to the “we” which temporarily has the upper hand in the social context. (p. 116)

Furthermore, Mottaz’s (1988) study of commitment concludes that the content of job tasks may be the most important influence on organiza- tional commitment, a finding we think provides support t o our links between activity and identification.

As we have briefly shown, the relationship between the activation and attachment modalities (see Figure 1) in structuration theory pro- vides a framework t o help illustrate how situated activities and activity foci shape and are shaped by our identifications and the identities drawn upon in that identification. Drawing on and extending the work of Corman and Scott (1994), this theory allows us to link activity, identity, and observable communication as it is used to express identification. Some activities will evoke one identity, while other activities may focus on another identity and support identification with an alternate target. More importantly for scholars of communication, the activities that de- fine our situations shape and are shaped by the social interaction that is so important in the identification process. The expression of identifica- tions that shapes and is shaped by our identities varies situationally. Thus, understanding communication in this process requires an appre- ciation of the ways that situated activities both enable and constrain identification.

Of course, it is unlikely that there are certain activities that univer- sally or exclusively link t o certain identifications. Rather, the activity- identification linkages must be specified a t much more local levels. For example, each organizational culture may be characterized by fairly unique employee-employer relations and distinct activities. Neverthe- less, there may well be consistencies across similar locales with regard to the identifications associated with certain activities or activity foci (for example, organizations in the same industry may mimic one another substantially; see Haveman, 1993). Importantly, activity may serve to provide the situational context in which perceived memberships (iden- tity) become instantiated in interaction with others (identification).

Perhaps our example of the associate professor and his competing identities can again be illustrative. Recall that John’s daughter is in his class, which has created potential conflict among personal, professional, organizational, and fatherly identities. Although John may not simulta- neously hold strong professional and personal identifications, he can certainly express different identifications in different situations. During class presentations, his professional identity may well be especially strong and certainly stronger than a personal identity that would like to main-

324

Identification Theory

tain control over his daughter by not sharing his conflict strategy se- crets! During grading activity, professional ethics and university rules against nepotism may be very compatible so that strong identification with both prevails. During a conflict a t home, the professional and uni- versity identifications may be thrown out completely as personal inter- ests and father identities become most salient. We could provide other specifics here, but the point is John’s identifications cannot be under- stood without considering the situations in which he is engaged. The activities that define those situations shape his identifications. Conversely, the identifications he holds also influence the activities in which he is likely t o find himself. His professional identification reinforces the do- ing of lecturing, evaluation, and attending conventions. Similarly, his fatherly identifications may be associated with a wide range of family activities.

Conclusions Ongoing changes in organizational forms and practices have made is- sues of attachment, specifically identification, to various aspects of the workplace one of the most salient issues in organizations today. In an attempt to provide the theoretical understandings necessary to address this important issue, we have developed a largely structurational theory that features organizational activity t o describe identification in the workplace. Our goal has been to offer a middle-range theoretical and heuristic framework for understanding how organizationally relevant identities relate to a variety of identifications in contexts of situated ac- tion. To review our key points, the attachment process (Figure 1 ) in- volves both identities (as structures) and identifications that can often be observed in interactions with others (either real or hypothetical). These two elements of the attachment process form a duality in that each shapes the other. The structural component of this model is composed of sev- eral possible identities conceptualized as regions that range in size or position and tenure, have front and back regions, and possess both unique and overlapping regions with one another. Additionally, we incorporate the idea of activity to help define the situation and thus account for the fact that organizational members may be identified with certain resources of identity as opposed to or along with others at various times. We believe that certain activities or activity foci are linked to a certain identity region (or set of overlapping regions) more so than to other regions (e.g., customer ser- vice activities may link to an organizational identity, but skill training ac- tivities may be more clearly linked to a n occupational identity). If a situ- ation is focused o n the unique portion of a n identity region, then only that identity is drawn upon; if the situation focuses o n overlapping iden- tities, all those common identities are relevant.

325

Communication Theory

Discussion We believe the situated-action view of identification presented here rep- resents an important addition to our current understanding of identifi- cation for several reasons. First, this model relates identity to identifica- tion via structurational terms. Such a move better allows us to examine the attachment process as a duality that recognizes identification as both drawing on and (re)creating identity, without unduly preferencing struc- tural or agency-based views of how identification occurs. This move also allows definitions based o n perceived membership to be joined with those based on more behavioral outcomes. Furthermore, it emphasizes the communicative aspects of identification by highlighting the social nature of identification. Second, it helps to account for the relationships between the multiplicity of identities and identifications that are at play in organizational life. Our model suggests that identities as such are usefully viewed as regions, which helps us discuss issues of compatibility and competition, identification and disidentification, and importance, centrality, and longevity relative to other identities. Each identity region is associated with one o r more identification targets, accounting for the multiple targets of attachment described in the literature. Third, we link activity to identification and communication. We see identification as somewhat changing and fluid in practice and as having an essential rela- tionship to activity. During certain situated activities, some identifica- tions are more likely than others. Given the relationships between activ- i t y and communication theorized by Corman and Scott (1994), we also have the theoretical tools to understand how acts of identification, shaping and shaped by activity, are related to instances of communication among organizational members.

Our encouragement about this situational model is bolstered in sev- eral ways. First, we see this model as providing reasonable answers to several issues raised in the identity-identification literature. For example, this approach addresses concerns about compatibility and competition among identification targets and offers an explanation for when we iden- tify with one o r more targets as opposed to others. Additionally, a structurational view also addresses the action-structure paradox. Second, Giddens (1991 b) has noted that structurational views fit a variety of orga- nizational phenomena where contextualized views are needed. Indeed, we see his writings about duality, regionalization, and situated action as rel- evant to identity-identification. Third, we see our arguments as being con- sistent with a range of scholarship regarding identity and identification (see Cheney, 1991; Cheney & Christensen, in press; Cheney & Tompkins, 1987; Mae1 & Ashforth, 1992, 1995). Although the structurational approach provides a different framework than other work in this area has empha- sized, we believe that our model can account for a wide range of views and findings in the organizational attachment literature.

326

Identification Theory

A Brief Application. Perhaps the utility, of this model is best seen when applied to an actual case of organizational restructuring through semi- autonomous work-teams. In an in-depth, critical-ethnographic analysis of a Japanese “transplant” in Indiana, Graham (1993, 1995) captures well the dynamics of control and resistance o n the shop floor of a Subaru- Isuzu factory. This case of team implementation is complicated by the very ambiguity surrounding the team concept, in that many U.S. em- ployees take it to be group focused while most of their Japanese manag- ers view it as company centered. According to Graham, in the course of team development, tensions develop within the context of specific work activities, resulting in some cases in resistance. The struggle for control o n the shop floor is initiated by what Graham sees as a top-down, highly circumscribed form of “employee empowerment,” where the pace of work is accelerated, workers’ responsibilities are expanded, and moni- toring of work activities is fostered through extensive reporting, close supervision, and peer pressure. Above all, worker compliance is sought through the internalization of the responsibilities of team membership and the adoption of self-discipline. However, resistance comes in a vari- ety of forms, ranging from some employees’ refusal to join morning exercises for the entire workforce to consideration of sabotage.

Although Graham (1993,1995) does not see her participant observa- tion of work experience in terms of organizational identification per se, we believe that the concept and our model of it can shed further light on this poignant case study. First, we observe the tensions between and among multiple identifications in the case. Second, we.note the shifts in identification for many employees, who, although initially resistant to the idea of working in teams, come to take the idea of team very seri- ously (but a t the group level). The newly formed team identification, grounded in shop-floor camaraderie, becomes a source of loyalty against the company. Third, we find differences across employees in how the multiple identifications are configured, in that, for example, some main- tain strong linkages between team and company as two complementary resources of loyalty and others d o not. Fourth, with even further infor- mation, we might speculate o n how “front” and “back” regions of iden- tity are interrelated for various employees, considering as well issues of unconscious tendencies, authenticity, and “gaming.” Fifth, we can see how in the context of specific work activities, both routine a n d nonroutine, various identifications are reflected, formed, and evoked. For one employee, a certain prescribed activity, such as the collective morning exercises, may have no symbolic significance whatsoever. For another, such a regimen could represent organizational control, thereby sparking a strong disidentification with the employer and encouraging resistance. Because of its vivid detail, its chronological perspective, and its emphasis on concrete aspects of work processes, Graham’s (1993)

327

Communication Theory

case affords us a glimpse of how organizational identification can be understood from a structurational, multilevel, and situation-sensitive perspective.

We also can envision ways in which research based on our model would provide insights the existing views of identification might over- look. Traditional (survey-based) empirical identification research would likely approach organizational members with a questionnaire and ask members to indicate their level of identification with their employing organization. Having done so, most researchers would report on the employees’ global identification with the company and possibly relate that to other variables measured as part of the study. Although the former approach is certainly valuable, research based o n the model we have proposed here would differ in several ways. First, it would look at mul- tiple relevant identities, attempting to assess the degree of compatibility and competition between and among various attachment targets and the relative importance of each. Second, an attempt might be made to measure manifest identification associated with each of those identities by looking a t decision making, citizenship behaviors, and especially com- municative interactions (including retrospective-prospective accounts) with others. Additionally, identity salience and identification could be assessed for each of several activities or activity foci, recognizing that the situation may evoke certain identifications ( o r combinations of iden- tifications). Ideally, research utilizing this model would tend t o be multimethod and longitudinal in that the ebb and flow of identifications would be evident in participants’ commentaries on the streams of their experience. Because of space limitations and our desire t o focus on theory development in this essay, we do not report any data in support of our arguments; however, we believe the sorts of methodological studies we suggest here are feasible, and we have already begun collecting data using this framework. Future Clarifications. Despite our enthusiasm for the view of identifica- tion we have begun t o develop here, we qualify that support with a clear recognition that much remains to be done in clarifying these views. For example, the boundary between identity and identification is not always clear, in part because language is to some extent constitutive of both. Furthermore, much greater specification of the activity-identification relationship is needed. Additionally, there is a need t o explore the emo- tional “side” of organizational identification more fully than does the present articulation of this model. Nevertheless, this model can have some rather practical payoffs. Beyond a recognition of the ways in which multiple identities or identifications relate, being able to better connect certain activities (and activity foci) with corresponding identities or iden- tifications may help organizational members to reflect more deeply on their attachments. Our ideas may also have import into other contexts

328

Identification Theory

where identification issues are of concern: in family tensions, in diverse communities, and in international negotiations.

Yet, perhaps the greatest value of our identification model lies in its ability t o help us think about identification in more complete and realis- tic ways. A structurational view, as we have presented here, should serve as a framing device that helps us to see issues of identification in new, and more appropriately dialectical, ways. As we have tried to suggest throughout this essay, the organizational identification literature is in need of a model that clearly incorporates notions of identity, addresses the action-structure issue, recognizes the relations between and within multiple identities, and accounts for situationally variant identifications based on activity. We believe the structurational theory presented here meets these needs more completely than d o existing models and opens up a range of issues for theorists and researchers to explore. In an age of changing loyalties and fleeting attachments to various aspects of our work, models that help us understand this complex set of issues are essential.

- Craig R. Scott is a n assistant professor in the Department of Speech Communication a t the Univer- sity of Texas at Austin. His research concerning identification and structurational approaches to communication phenomena has been published in Communication Theory and Management Com- munication Quarterly. T h e current manuscript is based o n his dissertation research conducted at Arizona State University. Steven R. Corman is an associate professor in the Department of Commu- nication a t Arizona State University. H e directed the dissertation o n which the current essay is based. His research on situationalist approaches to communication phenomena has appeared in such journals as Communication Theory a n d Communication Monographs. George Cheney is a professor in the Department of Communication Studies a t the University of Montana. H e has published extensively on the topic of identity in organizational life, as well as o n power, democracy, and corporate public discourse. H e is the author of Rhetoric in an Organizational Society: Manag- ing Multiple Identities (Columbia: University of South Carolina Press, 1991) a n d the forthcoming book, Managed Democracy (Ithaca, NY: Cornell University Press).

Authors

’ Although we recognize the limitations of the target metaphor, we find it useful for suggesting the more active dimensions of the identification process by which the individual directs himself or herself toward a n d appropriates resources from various elements of the world (even though other features of the identification process may be less conscious). Furthermore, “target” is consistent with the current attachment literature.

Several researchers (Ashforth & Mael, 1989; Barge, 1985; Sass &Canary, 1991) have noted the equation of identification and commitment in some organizational research, which is n o t surpris- ing since commitment is often defined as a process of identification (Gorden & Infante, 1 9 9 1 ) . As defined by Mowday, Steers, a n d Porter ( 1 979, p. 226), commitment is the “relative strength of a n individual’s identification with and involvement in a particular organization.” Sass a n d Canary ( 1 9 9 1 , p. 277) have made convincing arguments t h a t “attitudinal commitment is conceptually and empirically similar to the product form of identification . . .” (although they d o view the process of identification and behavioral commitment each as distinguishable), prompting them to call for a rethinking of the commitment-identification relationship. In o u r view, the focus here should n o t be o n separating or integrating the terms as it has been in past literature. Instead, w h a t is needed is a view of identification that recognizes process and product forms of identification as well as attitu- dinal a n d behavioral forms of commitment. Attempts to collapse and distinguish identification and commitment have in some ways obscured efforts to understand the relationships between them.

In addition to the boundaries mentioned here, w e qualify and clarify o u r views in three ways. First, we neither support nor wish to be seen as endorsing a unidirectional view of the ongoing

- Notes

329

Theory

construction of identity that works only from the individual "up." As we see it, the other must be assumed and incorporated into the analysis, in that the resources used by the individual person to construct and narrate identity are overwhelmingly social in nature (see, e.g., Burke, 1969). Ulti- mately, our use of social identities, o u r structurational view of identification, and the attention we give to identification as expressed in social interactions help maintain a strong awareness of those others. Second, the identification process, as a linkage between the individual "self" and elements of the social scene outside the self, can function so as to shape future actions in predictable ways. Once an identification, say, with the employing organization as a whole, is clearly established for an employee, the organization can count o n that employee to make decisions with the perceived interests of the firm uppermost in mind (cf., Simon, 1976; Tompkins & Cheney, 1983, 1985). Identification, in this sense, reduces our range of vision in confronting a multitude of options and in making decisions. Identification, especially as expressed in the form of accounts for decisions, pro- vides both a point of reference for the individual person and a guideline for "organizationally appropriate" behavior. Third, from our standpoint, language is central to identity formation be- cause it is language use that gives people an extraordinarily rich vocabulary for "telling their story," both a t a specific point in time and with respect to one's life narrative (see Shotter, 1983). Identity was first conceptualized in terms of narrative by Nietzsche (1982), who saw identity as the story that an individual (or a social group) tells about itself (cf., Cheney & Tompkins, 1988). Thus, identity functions as a response given (or prepared to be given) that explains the self.

In the same vein, we recognize the risks in positing stable or enduring forms of identification as identity. In many ways, identity and identification are dialectically interdependent; thus, there are several metaphors we find useful in avoiding the reification of identity: identity as an underlying principle (Sarbin & Scheibe, 1983), identity a s an organizing grammar of the self (Harri, 1984), and identity as a set of highly evocative and deeply held labels (Tajfel, 1981). Importantly, this is what we will treat throughout o u r essay as identification: the largely observable and expressible definition of the self (as composed of multiple identities) in terms of other social units to which the individual appeals (cf., Cheney & Tompkins, 1987).

We realize that some researchers may object to this division as being artificial and exclusionary, arguing, for example, that a notion of class-even in a modified form that acknowledges the types of stratification explicit or implicit in the workplace today-may be central and not peripheral to comprehending the dynamics of identification. Nevertheless, we are forced to begin somewhere with a less than exhaustive set of identities to consider.

Regarding individual agents, Giddens (1991a) takes self-identity as his main subject. This is "the self as reflexively understood by the person in terms of her or his biography" (p. 53, emphasis in original omitted). Self-identity is a phenomenon with a unitary impulse and is related to one's ability to "keep a particular narrative going" (p. 54, emphasis in original omitted). Self-identity presumes and produces trust relationships that form the basis of all social interaction. It develops in large part early in life from critical interactions with caregivers. For Giddens, self-identity serves as a point of reference that enables people t o relate to social institutions and other people in predict- able ways by indicating "who I am" and "what 1 do." However, such individualistic phenomena represent only part of the picture. Large-scale institutional processes are closely implicated in the development and maintenance of self-identity. For example, the caregiving so critical in the early development of self-identity is itself often institutionalized. Maintenance of a personal narrative in the modern industrialized world presumes a biographical picture painted on an institutional can- vas. At a more mundane level, institutions routinely reproduce social systems that we count on to provide a stable (or at least predictable) future for our self-identiries. They provide a context for our trust relationships with others, and we even trust in the abstract (secular or religious) to main- tain physical or social conditions presumed by our own narratives. For Giddens (1991a), self- identity is a unitary individual-centered process that operates in and interacts with a social-institu- tional context that is broad in scope and deeply subject to structural forces. Important as these phenomena might be, we believe that communicative interaction at intermediate levels of social organization is equally important to a full understanding of identity. As Poole and DeSanctis (1992) note, "local structures are produced and sustained through the development of intersubjective understandings and coordinated actions. Over long time frames and repeated reproduction, the structures that evolve in local systems ultimately generalize into institutions" (p. 6, emphasis added). The converse, of course, is also true. Institutions are made (be it intentionally or not) by adults who interact in groups and organizations, and we believe this fact is sometimes lost in Giddens's more broadly sociological work.

Giddens offers the notion of a modality as a way of better understanding the relationship be- tween structural and system-level elements. Essentially, it is a depiction of the duality between these

330

Identification Theory

two elements. Although “duality” or “process” are also used here, we too shall sometimes use modality to describe this relationship, which is consistent with structuration and with other writ- ings drawn on later in the essay (e.g., activation modality). ’ Though not the way we use the term here, we note that some “structuralist” explanations of identity emphasize the controlling influences of internal states, drives, or characteristics. Freud’s (1922) theory of identity is in this sense “structuralist.”

For example, identification can be seen as undesirable from a managerial or administrative point of view when it “prevents the organized individual from making (organizationally) correct decisions in cases where the restricted area of values with which he [or she] identified himself [or herself] must be weighed against other values outside that area” (Simon, 1976, p. 218). Tompkins and Cheney (1985) contend that in some instances identification with a department instead of with the larger company may have deleterious effects on the organization, for example, in hindering the development of broad perspectives o n the use of a scarce resource. Van Maanen (1976) notes that what may be a peripheral, and thus dysfunctional, behavior for one target can be a pivotal one for another target. Bullis and DiSanza (1995) also noted instances where members of the U. S. Forest Service identified more strongly with their unit than with the larger organization. From a broader sociological stance, clashes of identification are inevitable in a fragmented, complex, and changing society.

We distinguish disidentification as a “negative” form of identification from alienation, which assumes no substantial implication of self-hood. Thus, disidentification involves identity in that one may define himself or herself in opposition to some target person or group. ‘O It is possible to display tenure using something like tree rings around each identity, with each ring representing a year of tenure for that identity. However, such an indication unnecessarily com- plicates the simplicity of the regions design and may serve to overemphasize chronological years as the sole indicator of duration.

An emphasis o n activity can be understood in at least three theoretical and practical senses: (a) in terms of Marxist activity theory, which emphasizes how consciousness (and by extension, com- munication) is grounded or rooted in the physical activities that constitute a particular type of work; (b) the sheer identification of different types of action in which people are engaged and which they themselves recognize (e.g., work, play, education, politics, family life, etc.); and (c) the social matrix in which one is embedded at a particular moment and point in space. We prefer to think of activity (and situated action) in the third sense, including under this rubric such dimen- sions and terms as place, physical space, role, bounded practices, types of communication media employed, etc. Thus, our notion of activity would include the lay notions of the situated actions in which one is participating and would reference important material aspects of the experience. How- ever, we would take a broader perspective on activity or situated action than do either the activity theory or lay-action understandings of the term above. With our concept of “activity” we wish to capture the importance of the context for and type of action being performed (e.g., the writing of memos, the facilitation of meetings, the conduct of performance-appraisal interviews, etc.), while also keeping in view the whole of a person’s role experience while performing the specific action. In offering this view, though, we want to avoid a strictly Goffmanesque characterization of situations or “role-boundedness” that would allow little room for larger social-structural influences. Thus, to continue with one of our examples from above, we would include as a relevant piece of data for our memo writer the fact that the memo has a particular history in the employee’s organization- perhaps as a means of replacing or avoiding face-to-face interaction. In this case, then, the writing of memos would not only place the person in an immediate situation, through which she perhaps drew upon her identity as a professional manager, but would also place her within the rhetorical constraints for that genre of communication as revealed in the ongoing life of the organization.

- Abrahamson, M., & Anderson, W. P. (1984). People’s commitments to institutions. Social Psychol-

Abrams, D., & Brown, R. (1989). Self-consciousness and social identity: Self-regulation as a group

Albert, S., & Whetten, D. A. (1985). Organizational identity. In B. M. Staw & L. L. Cummings

Alonso, W. ( 1 995). Citizenship, nationality, and other identities. ]ournu1 of International Affairs,

References ogy Quarterly, 43, 371-381.

member. Social Psychology Quarterly, 52, 31 1-31 8.

(Eds.), Research in organizational behavior (Vol. 7 , pp. 263-295). Greenwich, CT: JAI.

48,585-599.

331

Communication Theory

Ashforth, B. E., & Humphrey, R. H. (1993). Emotional labor in service roles: The influence of identity. Academy o f Management Review, 18, 88-1 15.

Ashforth, B. E., & Mael, F. (1989). Social identity theory and the organization. Academy ofMan- agement Review, 14,20-39.

Banks, M. H., & Henry, P. (1993). Change and stability in employment commitment. ]ournu1 of Occupational and Organizational Psychology, 66, 177-1 84.

Barge, J. K . (1985, November). Organizational identification and social motivation: A field de- scriptive study in t w o organizations. Paper presented at the Speech Communication Associa- tion conference, Denver, C O .

Barker, J. R. (1995, May). Leadership in the unobtrusive organization: Issues and implications. Paper presented to third annual symposium o n work teams, Dallas, TX.

Barker, J. R., & Tompkins, P. K . (1994). Identification in the self-managing organization: Charac- teristics of target and tenure. Human Communication Research, 2 I , 223-240.

Barling, J., Wade, B., & Fullager, C. ( 1 990). Predicting employee commitment to company and union: Divergent models. Journal of Occupational Psychology, 63,49-6 1 .

Barnard, C. (1968). The functions o f t h e executive. Cambridge, MA: Harvard University Press. Bastien, D. T., McPhee, R. D., & Bolton, K . A. (1995). A study and extended theory of the

structuration of climate. Communication Monographs, 62, 87-109. Baxter, L., & Bullis, C. (1986). Turning points in developing romantic relationships. Human Com-

munication Research, 1 2 , 4 6 9 4 9 3 . Beauvais, L. L., Scholl, R. W., & Cooper, E. A. ( 1 991 ). Dual commitment among unionized faculty:

A longitudinal investigation. Human Rellztions, 44, 175-1 93. Becker, T. E. (1992). Foci and bases of commitment: Are they distinctions worth making? Academy

o f Management Journal, 35,232-244. Becker, T. E., Randall, D. M., & Riegel, C. D. (1995). T h e multidimensional view of commitment and

the theory of reasoned action: A comparative evaluation. Journal o f Management, 21, 617-638. Becker, T. E. & Billings, R. S. (1993). Profiles of commitment: An empirical test. Journal of Orga-

nizational Behavior, 1 4 , 177-190. Bemmels, B. (1995). Dual commitment: Unique construct o r epiphenomenon? Journal of Labor

Research, 1 6 , 4 0 1 4 2 2 . Bryant, C. G. A,, & Jary, D. (1991). Introduction: Coming to terms with Anthony Giddens. In C.

G. A. Bryant & D. Jary (Eds.), Giddens’ theory o f structuration: A critical appreciation (pp. 1- 3 1). London: Routledge.

Bullis, C. (1991). Communication practices a s unobtrusive control: An observational study. Com- munication Studies, 42, 254-271.

Bullis, C., & Bach, B. (1989). Socialization turning points: An examination of change in organiza- tional identification. Western Journal of Speech Communication, 53, 273-293.

Bullis, C., & Bach, B. (1991). An explication and test of communication network content and multiplexity as predictors of organizational identification. Western Journal o f Speech Commu- nication, 55, 180-197.

Bullis, C. A., & DiSanza, J. R. (1995, November). “Everybody loves Smokey the Bear ”: Employee responses to newsletter identification inducements at the U.S. Forest Service. Paper presented at the Speech Communication Association conference, San Antonio, T X .

Bullis, C., & Tompkins, P. K . (1989). T h e forest ranger revisited: A study of control practices and identification. Communication Monographs, 56,287-306.

Burke, K. (1937). Attitudes toward history. N e w York: N e w Republic. Burke, K. (1969). A grammar ofmotives. Berkeley: University of California Press. Burke, K. (1973). T h e rhetorical situation. In L. Thayer (Ed.), Communication: Ethical and moral

Burke, P. J. (1980). T h e self: Measurement requirements from an interactionist perspective. Social

Burke, P. J. (1991). Identity processes and social stress. American Sociological Review, 56, 836-

Burke, P. J., & Reitzes, D. C. (1981). T h e link hetween identity a n d role performance. Social Psy-

Burke, P. J., & Reitzes, D. C. (1991). An identity theory approach to commitment. Social Psychol-

Burke, P. J., & Tully, J. (1977). T h e measurement of rolelidentity. Social Forces, 55, 880-897. Carbaugh, D. (1996). Situating selves: The communication o f social identities in American scenes.

issues (pp. 263-275). London: Gordon & Breach.

Psychology Quarterly, 43, 18-29.

849.

chology Quarterly, 44, 83-92.

ogy Quarterly, 54, 280-286.

Albany: State University of New York.

332

Identification Theory

Carroll, C. (1995). Rearticulating organizational identity: Exploring corporate images and em- ployee identification. Management Learning, 26, 4 6 3 4 8 2 .

Cheney, G. (1982). Identification as process and product: A field study. Unpublished master’s the- sis, Purdue University.

Cheney, G. (1983a). T h e rhetoric of identification a n d the study of organizational communication. Quarterly Journal o f Speech, 69, 143-158.

Cheney, G. (1983b). O n the various a n d changing meanings of organizational membership: A field study of organizational identification. Communication Monographs, 50,342-362.

Cheney, G. (1991). Rhetoric in an organizational society: Managing multiple identities. Columbia: University of South Carolina Press.

Cheney, G., & Christensen, L. T. (in press). Identity a t issue: Linkages between “internal” a n d “external” organizational communication. In F. M. Jablin & L. L. Putnam (Eds.), N e w hand- book of organizational communication ( 2 n d ed.).

Cheney, G., & Tompkins, P. K. (1987). Coming to terms with organizational identification a n d commitment. Central States Speech Journal, 38, 1-15.

Chey, G., & Tompkins, P. K. (1988). O n the facts of the “text” as the basis of human communica- tion research. In J. A. Anderson (Ed.), Communication Yearbook, 1 1 (pp. 4 4 5 4 8 1 ) . Thousand Oaks, CA: Sage.

Christensen, L. T., & Cheney, G. (1994). Articulating identity in a n organizational age. In S. A. Deetz (Ed.), Communication Yearbook, 17 (pp. 222-235). Thousand Oaks, CA: Sage.

Clegg (1994). Weber a n d Foucault: Social theory for the study of organizations. Organization, I ,

Cody, M . J., & McLaughlin, M . L. (1985). T h e situation as a construct in interpersonal cornmuni- cation research. In M. L. Knapp & G. R. Miller (Eds.), Handbook of interpersonal communica- tion (pp. 263-312). Beverly Hills, CA: Sage.

Cohen, A. (1993). Age a n d tenure in relation t o organizational commitment: A meta-analysis. Basic and Applied Social Psychology, 14, 143-159.

Corman, S . R., & Scott, C. R. (1994). Perceived networks, activity foci, a n d observable communi- cation in social collectives. Communication Theory, 4 , 171-190.

Corman, S. R., Stage, C. W., & Scott, C. R. (1997, M a y ) . Communication-related activity systems: An empirical description and computational organization model of activity foci in a grocery store chain. Paper presented a t t h e International Communication Association conference, M o n t r i a l , Canada.

Czarniawska-Joerges, B. (1994). Narratives of individual a n d organizational identity. In S. A. Deetz (Ed.), Communication Yearbook, I7 (pp. 193-221). Thousand Oaks, CA: Sage.

Death of corporate loyalty (1993, April 3). Economist ( p p . 63-64). Devereaux, G. (1975). Ethnic identity: Its logical foundations and its dysfunctions. In G. De Vos &

L. Romanucci-Ross (Eds.), Ethnic identity: Cultural continuity and change (pp. 42-70). Palo Alto, CA: Mayfield.

De Vos, G. (1983). Adaptive conflict a n d adjustive coping: Psychocultural approaches t o ethnic identity. In T. R. Sarbin & K. E. Scheibe (Eds.), Studies in social identity (pp. 204-230). New York: Praeger.

Durkheim, E. (1933). The division of labor in society (G. Sirnpson, Trans.). New York: Free Press. Dutton, J. E., & Dukerich, J. M. (1991). Keeping a n eye o n the mirror: Image and identity in

Dutton, J. E., Dukerich, J. M., & Harquail, C . V. (1994). Organizational images a n d member

Feld, S. L. (1981). T h e focused organization of social ties. American Journal of Sociology, 86,

Foucault, M. (1984). The Foucault reader (P. Rabinow, Ed.). N e w York: Pantheon. Freud, S. (1922). Group psychology and the analysis o f the ego. N e w York: Norton. Fullager, C., & Barling, J . (1991). Predictors a n d outcomes of different patterns of organizational

a n d union loyalty. Journal of Occupational Psychology, 64, 129-143. Giddens, A. (1981). Agency, institution, and time-space analysis. In K. Knorr-Cetina & A. V. Cicourel

(Eds.), Advances in social theory and methodology: Toward an integration of micro- and macro- sociologies (pp. 161-174). London: Routledge & Keegan Paul.

149-178.

organizational adaptation. Academy o f Management Journal, 34,517-554.

identification. Administrative Science Quarterly, 39, 239-263.

1015-1035.

Giddens, A. (1984). The constitution of society. Berkeley: University of California Press. Giddens, A. (1991a). Modernity and self-identity. Stanford, CA: Stanford University. Giddens, A. (1991b). Structuration theory: past, present a n d future. In C. G. A. Bryant & D. Jary

(Eds.), Giddens’theory ofstructuration: A critical appreciation (pp. 201-221). London: Routledge.

3 3 3

Communication Theory

Goffman, E. (1959). The presentation o f s e l f in everyday life. N e w York: Anchor. Gorden, W. J., & Infante, D. A. (1991). Test o f a communication model of organizational commit-

Graham, L. (1993). Inside a Japanese transplant: A critical perspective. Work & Occupations, 20,

Graham, L. (1995). On the line at Subaru-lsuzu: The Japanese model. Ithaca, NY: Cornell Univer-

Gray, D. E. (1989). Gender a n d organizational commitment among hospital nurses. Human Rela-

Gregersen, H. B. (1992). Commitments to a parent company and a local work unit during repatria-

Gregersen, H . B. (1993). Multiple commitments at work and extrarole behavior during three stages

Haines, V. A. (1988). Social network analysis, structuration theory a n d the holism-individualism

Hall, D. T., & Schneider, B. (1972). Correlates of organizational identification as a function of

Hall, D. T., Schneider, B., & Nygren, H . T. (1970). Personal factors in organizational identification.

Harre, R. (1984). Personal being: A theory for individual psychology. Cambridge, MA: Harvard

Harre, R., & Gillett, G. (1994). The discursive mind. Thousand Oaks, CA: Sage. Hatch, M. J. (1987). Physical barriers, task characteristics, a n d interaction activity in research a n d

Haveman, H. A. (1993). Follow the leader: Mimetic isomorphism a n d entry into new markets.

Hecht, M. L. (1993). 2002-A research odyssey: Toward the development of a communication

Hirschman, A. 0. (1970). Exit, voice, and loyalty: Responses t o decline in firms, organizations,

Hunt, S. D., & Morgan, R. M. (1994). Organizational commitment: O n e of many commitments or

James, W. (1950). The principles ofpsychology, Vol. I . N e w York: Dover. Kahn, W. A. (1990). Psychological conditions of personal engagement a n d disengagement a t work.

Academy of Management Journal, 33, 692-724. Kaufman, H. ( 1 960). The forest ranger: A studv in administrative behavior. Baltimore: Johns Hopkins

Press. Kilrninster, R. (1991). Structuration theory a s a world-view. In C. G. A. Bryant & D. Jary (Eds.),

Giddens’ theory of structuration: A criticd appreciation (pp. 74-1 15). London: Routledge. Kleinfeld, N . R. (1996, February 5). The downsizing of workplace security. Austin-American States-

men (pp. A l , AS). Knorr-Cetina, K . D. (1981). Introduction: T h e micro-sociological challenge of macro-sociology:

Towards a reconstruction of social theory and methodology. In A. V. Cicourel & K. D. Knorr- Cetina (Eds.), Advances in social theory and methodology (pp. 1 4 7 ) . Boston: Routledge & Keegan Paul.

Koslowsky, M. (1990). Stafflline distinctions in job a n d organizational commitment. Journal of Occupational Psychology, 63, 167-1 73.

Larkey, L., & Morrill, C. (1995). Organizational commitment as symbolic process. Western Jour- nal of Communication, 59, 193-213.

Lau, A. W., & Pavett, C. M . (1980). T h e nature of managerial work: A comparison of public- and private-sector managers. Group 6 Organization Studies, 5,4531166.

Mackenzie, W. J. M. (1978). Political identity. N e w York: St. Martin’s Press. Mael, F., & Ashforth, B. E. (1992). Alumni and their a h a mater: A partial test of the reformulated

model of organizational identification. Journal o f Organizational Behavior, 13, 103-123. Mael, F. A., & Ashforth, B. E. (1995). Loyal from day one: Biodata, organizational identification,

and turnover among newcomers. Personnel psycho log^, 48, 309-333. March, J. G., & Simon, H. A. (1958). Organizations. N e w York: J o h n Wiley. May, S. K.. (1995). Flying the not-so-friendly skies: Organizational identification and resistance

during an airline labor dispute. Paper presented a t the Speech Communication Association conference, San Antonio, Texas.

ment. Communication Quarterly, 39, 1 4 4 1 5 5 .

147-1 73.

sity Press.

tions, 42,801-813.

tion. Personnel Psychology, 45, 29-54.

of organizational tenure. Journal o f Business Research, 26, 3 1 4 7 .

debate. Social Networks, 10, 157-182.

career path and organizational type. Administrative Science Quarterly, 17, 340-350.

Administrative Science Quarterly, 15, 176-1 90.

University Press.

development firms. Administrative Science Quarterly, 32, 387-399.

Administrative Science Quarterly, 38, 593-627.

theory of identity. Communication Monographs, 60, 76-82.

and states. Cambridge, MA: Harvard University Press.

key mediating construct? Academy o f Management Journal, 37, 1568-1587.

334

Identification Theory

McCaul, H. S., Hinsz, V. B., & McCaul, K. D. (1995). Assessing organizational commitment: An employee’s global attitude toward the organization. Journal o f Applied Behavioral Science, 3 I , 80-90.

McPhee, R. D. (1985). Formal structure and organizational communication. In R. D. McPhee & P. K. Tompkins (Eds.), Organizational communication: Traditional themes and new directions (pp. 149-177). Beverly Hills, CA: Sage.

McPhee, R. D., & Corman, S. R. (1995). An activity-based theory of communication networks in organizations, applied to the case of a local church. Communication Monographs, 62,132-151.

Mead, G . H. (1934). Mind, self and society (C. W. Morris, Ed.). Chicago: University of Chicago Press.

Meyer, J. P., & Allen, N. J. (1997). Commitment in the workplace. Thousand Oaks, CA: Sage. Miller, K., Birkholt, M., Scott, C., & Stage, C. (1995). Empathy and burnout in human service

work: An extension of a communication model. Communication Research. 22. 123-147. . . Mintzberg, H. (1971). Managerial work: Analysis from observation. Management Science, 18,

B97-B110. Mintzberg, H. (1975). The manager’s job: Folklore and fact. Harvurd Business Review, 5 3 , 4 9 4 1 . Mintzberg, H. (1980). The nature of managerial work (2nd ed.). Englewood Cliffs, NJ: Prentice-

Morrow, P. C., & McElroy, J. C. (1993). Introduction: Understanding and managing loyalty in a

Moskal, B. S. (1993,l March). Company loyalty dies, a victim of neglect. Indus2*y Week (pp. 11-12). Mottaz, C. J. (1988). Determinants of organizational commitment. Human Relations, 41,467482. Mottaz, C. J. (1989). An analysis of the relationship between attitudinal commitment and behav-

Mowday, R. T., Steers, R. M., & Porter, L. W. (1979). The measurement of organizational commit-

Nietzsche, F. (1982). Daybreak (R. J. Hollingdale, Trans.). Cambridge, UK: Cambridge University

Palmer, A. W., & Beishon, R. J. (1970). How the day goes. Personnel Management, 2(4), 3 6 4 1 . Paolillo, J. G. (1981). Manager’s self assessments of managerial roles: The influence of hierarchical

level. Journul of Management, 7,43-52. Patchen, M. ( 1 970). Participation, achievement, and involvement on the job. Englewood Cliffs,

NJ: Prentice-Hall. Pervin, L. A. ( 1 978). Definitions, measurements and classifications of stimuli, situations and inter-

action. Journal of Personality and Social Psychology, 3 4 , 4 6 5 4 7 4 . Poole, M. S. (1985). Communication and organizational climates: Review, critique, and a new

perspective. In R. D. McPhee & P. K. Tompkins (Eds.), Organizational communication: Tradi- tional themes and new directions (pp. 79-108). Beverly Hills, CA: Sage.

Poole, M. S., & DeSanctis, G. (1992). Microlevel structuration in computer-supported group deci- sion making. Human Communication Research, 1 9 , 5 4 9 .

Poole, M. S., & McPhee, R. D. (1983). A structurational analysis of organizational climate. In L. L. Putnam & M. E. Pacanowsky (Eds.), Organizational communication: An interpretive approach (pp. 195-220). Beverly Hills, CA: Sage.

Poole, M. S., Seibold, D. R., & McPhee, R. D. (1985). Group decision-making as a structurational process. Quurterly Journal of Speech, 71, 7 4 1 0 2 .

Pye, L. W. (1962). Politics, personality, and nation-building: Burma’s search f o r identity. New Haven, C E Yale University Press.

Reed, C. S., Young, W. R., & McHugh, P. P. (1994). A comparative look at dual commitment: An international study. Human Relations, 47, 1269-1293.

Reichers, A. E. (1985). A review and reconceptualization of organizational commitment. Academy of Management Review, 10,4654176.

Reitzes, D. C., & Mutran, E. J. (1994). Multiple roles and identities: Factors influencing self-esteem among middle-aged working men and women. Social Psychology Quarterly, 57,313-325.

Riley, P. (1983). A structurationist account of political culture. Administrative Science Quarterly, 28,414-437.

Robertson, P. J., & Tang, S. (1995). The role of commitment in collective action: Comparing the organizational behavior and rational choice perspectives. Public Administration Review, 55( l ) ,

Rotondi, T., Jr. ( 1975). Organizational identification: Issues and implications. Organizational Be-

Hall.

multi-commitment world. Journal of Business Research, 26, 1-2.

ioral commitment. Sociological Quarterly, 30, 143-158.

ment. Journal of Vocational Behavior, 4,224-247.

Press.

67-80.

havior and Human Performance, 13,95-109.

335

Communication Theory

Russell, R. (1996, November). Workplace democracy and organizational communication. Paper

Ruud, G. (1996). T h e symbolic construction of organizational identities a n d community in a re-

Sarbin, T. R., & Scheibe, K . E. (1983). Studies in social identity. New York: Praeger. Sass, J. S., & Canary, D. J. ( 1 991). Organizational commitment and identification: An examination

of conceptual a n d operational convergence. Western Journal of Communication 55,275-293. Scott, C . R. ( 1997). Identification with multiple targets in a geographically dispersed organization.

Management Communication Quarterly, 1 1 , 491-522. Seashore, S. (1954). Group cohesiveness in the industrial work group. Ann Arbor: University of

Michigan Institute for Social Research. Shorter, J. (1983). Duality of structure and intentionality in an ecological psychology. Journal for

the Theory of Social Behavior, 1 3 , 19-43. Shotter, J. ( 1984). Social accountability and selfl7ood. Oxford, UK: Blackwell. Simon, H. A. (1976). Administrative behavior ( 3 r d ed.). New York: Free Press. Stoecker, R. ( 1 995). Community, movement, organization: T h e problem of identity convergence in

Stohl, C., & Perrucci, R. (1997). Organizational commitment and community embeddedness.

Tajfel, H. ( 1981). Human groups and social categories. Cambridge, MA: Harvard University Press. Tett, R. P., & Meyer, J. P. (1993). J o b satisfaction, organizational commitment, turnover intention,

a n d turnover: Path analyses based on meta-analytic findings. Personnel Psychology, 46, 259- 293.

Tompkins, P. K . , & Cheney, G. (1983). Account analysis of organizations: Decision making a n d identification. In L. L. Putnam & M. E. Pacanowsky (Eds.), Communication and organizations: A n interpretive approach (pp. 123-146). Beterly Hills, CA: Sage.

Tompkins, P. K., & Cheney, G. (198.5). Communication and unobtrusive control in contemporary organizations. In R. D. McPhee S: P. K. Tompkins (Eds.), Organizational communication: Tra- ditional themes and new directions (pp. 179--210). Beverly Hills, CA: Sage.

Van de Ven, A. H., & Poole, M. S. (1988). Paradoxical requirements for a theory of organizational change. In R. E. Quinn & K. S . Cameron (Eds.), Paradox and transformation: Toward a theory of change in organization and management (pp. 19-80). Cambridge, MA: Ballinger.

Van Maanen, J . (1976). Breaking in: Socialization to work. In R. Dubin (Ed.), Handbook of work, organization, and society (pp. 67-130). Chicago: Rand-McNally.

Vandenberg, R. J., & Scarpello, V. (1994). A longitudinal assessment of the determinant relation- ship between employee commitments to the occupation a n d the organization. Journal o f Orga- nizational Behavior, 1 5 , 53.5-547.

Wallace, J. E. ( 1 993). Professional and organizational commitment: Compatible o r incompatible? Journal of Vocational Behavior, 4 2 , 333-349.

Weigert, A. J., Teitge, J. S., & Teitge, D. W. (1986). Society and identity. New York: Cambridge University Press.

Willis, P. (1977). Learning to labor: H o w working class kids get working class jobs. N e w York: Columbia University Press.

Winter, G. (1968). Religious identity: A study oireligious organization. New York: Macmillan. Yoon, J., Baker, M. R., & KO, J. (1994). Interpersonal attachment and organizational commitment:

presented a t the Speech Communication Association conference, San Diego, CA.

gional symphony. Communication Studies, 46,201-221.

collective action. Sociological Quarterly, 3 6 , 11 1-130.

Working paper, Purdue University.

Subgroup hypothesis revisited. Human Relations, 47, 329-35 1 .

336