WHO WILL LEAD AND WHO WILL FOLLOW? ASOCIAL PROCESS OF LEADERSHIP IDENTITY

CONSTRUCTION IN ORGANIZATIONS

D. SCOTT DERUESUSAN J. ASHFORD

University of Michigan

We propose that a leadership identity is coconstructed in organizations when indi-viduals claim and grant leader and follower identities in their social interactions.Through this claiming-granting process, individuals internalize an identity as leaderor follower, and those identities become relationally recognized through reciprocalrole adoption and collectively endorsed within the organizational context. We specifythe dynamic nature of this process, antecedents to claiming and granting, and anagenda for research on leadership identity and development.

Scholars have begun to question traditionalconceptualizations that position leadership astop-down, hierarchical, and equivalent to formalsupervisory roles in organizations (Ancona &Backman, 2008; Bedeian & Hunt, 2006). Whileholding a formal position within an institution-alized hierarchical structure clearly conveyssome meaning with respect to leadership, thishierarchical perspective does not explain whysome supervisors are not seen as leaders (Bede-ian & Hunt, 2006) or why some individuals areseen as leaders despite not holding “leader-like” positions (Charan, Drotter, & Noel, 2000;Spreitzer & Quinn, 2001).

Recently, theorists have begun to conceptual-ize leadership as a broader, mutual influenceprocess independent of any formal role or hier-archical structure and diffused among the mem-bers of any given social system (Bedeian & Hunt,2006; Collinson, 2005; Gemmill & Oakley, 1992;Gronn, 2002; Parry, 1998; Uhl-Bien, Marion, &McKelvey, 2007). Although references to this dis-tributed form of leadership date back to thework of Selznick (1957), this perspective is be-coming more prominent in contemporary lead-ership theories. For example, Quinn (1996) ar-gues that leadership is a state of being that

people can enter into irrespective of their formalrole or position within an organization. Simi-larly, recent research on team leadership con-ceptualizes it as a shared property of the groupsuch that all members of the group, regardlessof their formal role or position, participate in theleadership process (Carson, Tesluk, & Marrone,2007; Morgeson, DeRue, & Karam, 2010; Pearce &Conger, 2003).

If leadership is not simply prescribed becauseof one’s position in an institutionalized hierar-chy, then a fundamental question that remainsto be answered is how leadership and leader-follower relationships develop in organizations.What are the relational and social processesinvolved in coming to see oneself, and beingseen by others, as a leader or a follower? Thisarticle presents a theory explaining the devel-opment of a leadership relationship that is com-posed of reciprocal and mutually reinforcingidentities as leaders and followers, is endorsedand reinforced within a broader organizationalcontext, and is dynamic over time.

By illuminating the interplay of leader andfollower identities in the development of a lead-ership relationship, our theory makes severalnoteworthy contributions to the leadership liter-ature. Prior research on “leader” as a personalidentity (Day & Harrison, 2007; DeRue, Ashford, &Cotton, 2009) informs but does not fully explainthe leadership identity construction process.Our theory makes clear that leader and followeridentities are not only cognitions that residewithin an individual’s self-concept (Day & Har-rison, 2007; Day & Lance, 2004; DeRue et al., 2009);

We thank Blake Ashforth, Natalie Cotton, Jane Dutton,Ned Wellman, and the participants at the 2008 ExploringPositive Identities Conference and the May Meaning Meet-ing for their helpful comments and suggestions in the devel-opment of this paper. We also thank associate editor JeffLePine and the three anonymous reviewers for an extremelygenerative review process.

� Academy of Management Review2010, Vol. 35, No. 4, 627–647.

627Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyrightholder’s express written permission. Users may print, download, or email articles for individual use only.

they are also socially constructed and inher-ently related (e.g., granting one person a leaderidentity frequently instantiates a follower iden-tity for others). By equating “leaders” with thoseholding supervisory positions and “followers”with those reporting to others in an organiza-tion, the leadership literature and the emergingliterature on followership both underplay thesocially constructed and reciprocal relationshipbetween leaders and followers (Collinson, 2006;Hollander, 1993; Van Vugt, Hogan, & Kaiser,2008), whereas our theory foregrounds this con-sideration.

Our theorizing also reflects a dynamism thatis absent from much of the existing literature.Leader identities are generally portrayed as in-trapersonal, one-directional, and static. But ifleadership is a mutual influence process amongindividuals, then social interaction among thoseindividuals and various contextual factors cancause leader and follower identities to shift overtime and across situations. Most research onleadership and identity acknowledges thatidentities develop over time but then goes on totheorize about a leader identity that, once inter-nalized, becomes a static and enduring featureof the person (DeRue et al., 2009; Komives, Owen,Longerbeam, Mainella, & Osteen, 2005) or aboutleadership relationships (e.g., leader-memberexchange; Graen & Uhl-Bien, 1995) that arestatic and of a particular nature. A static identityis also presumed in the emerging literature onfollowership (e.g., Collinson, 2006; Kellerman,2008; Van Vugt et al., 2008), where the focus is onpersonal attributes that make individuals effec-tive followers. In contrast, we propose thatleader and follower identities can shift amonggroup members through a social constructionprocess.

By emphasizing the dynamic nature of leaderand follower identities, we also bring into focusthe antecedents that shape the construction of aleadership identity. Existing theories of howpeople come to be seen as leaders focus nar-rowly on the cognitive aspects of the process.For example, implicit theories of leadership andperceptions of group prototypicality can causepeople to see others as leaders (DeRue et al.,2009; Lord, 1985; van Knippenberg & Hogg, 2003).We extend prior theory by offering a broader andmore integrative framework for understanding theantecedents to the construction of a leadershipidentity and leader-follower relationships.

Developing insights into the leadership iden-tity construction process is important since indi-viduals’ identities as leaders and followers arethought to be significant drivers of their subse-quent thought, affect, motivation, and action(Day & Harrison, 2007; Gardner & Avolio, 1998;Shamir, House, & Arthur, 1993). Indeed, prior re-search suggests that seeing oneself as a leadernot only enhances one’s motivation to lead(Chan & Drasgow, 2001; Kark & van Dijk, 2007)and one’s engagement in the leadership process(Kempster, 2006) but also promotes the seekingout of leadership responsibilities and opportu-nities to develop leadership skills (Day, Harri-son, & Halpin, 2009). Additionally, understand-ing the mutual construction of leaders’ andfollowers’ respective roles and identities willhelp us explain and predict the relational out-comes associated with leader-follower relation-ships. For example, a strong leadership identityimplies that there is clarity in the leader-follower relationship and individuals’ identitiesas leader and follower. When this clarity exists,there is greater acceptance of the right of theperson constructed as leader to exert influenceover the person constructed as follower. Whenthis clarity is missing, we expect increased con-flict and tension in the relationship (Collinson,2005). In this sense, the construction of a leader-ship identity and the respective identities asleader and follower are inputs into the quality ofleader-follower relationships (Graen & Uhl-Bien,1995).

We present our theory as follows. First, wedefine the concept of leadership identity con-struction and highlight how our theory offersnew insights for the broader identity literature.We then describe an identity work process ofclaiming and granting whereby individuals co-create reciprocal and mutually reinforcing iden-tities as leaders and followers and, through thisprocess, develop a leader-follower relationship.From there we elaborate on the relationship be-tween claiming and granting by specifying theconditions under which claims are reciprocatedby grants and grants are reciprocated by claims.We conclude with a discussion of the anteced-ents that prompt individuals to claim and grantboth leader and follower identities, as well asan agenda for future research that would extendour theory in new directions.

628 OctoberAcademy of Management Review

LEADERSHIP IDENTITY CONSTRUCTION

Identity involves the meaning attached to theself (Gecas, 1982). Any particular identity can beconceptualized along three levels of self-construal: individual, relational, and collective(Brewer & Gardner, 1996). Because leadershipinvolves multiple individuals engaged in a pro-cess of interpersonal and mutual influence thatis ultimately embedded within some collective(Hollander, 1978; Parry, 1998), it is necessary tointegrate across these three levels to fully cap-ture the process of constructing a leadershipidentity. Therefore, in contrast to the existingliterature, we propose a conception of leader-ship identity that invokes all three levels of self-construal. Specifically, a leadership identitycomprises three elements: individual internal-ization, relational recognition, and collective en-dorsement.

Individual internalization is a state where in-dividuals come to incorporate the identity ofleader or follower as part of their self-concept(DeRue et al., 2009; Gecas, 1982). It involves “thecreation of new aspects of the self that relate tothe leader (or follower) role (e.g., growth in theleader sub-identity)” (Hall, 2004: 157). Prior the-ory suggests that the designation of these per-sonal attributes to the self is not simply a cog-nitive, intraindividual assessment but, rather, isembedded in specific contexts where an identityis asserted and ascertained in the course of so-cial interaction (DeRue et al., 2009; Snow &Anderson, 1987).

This embeddedness suggests a second, morerelational aspect of leadership identity con-struction—relational recognition. Individuals’identities are often tied to various roles (Stryker,1980; Stryker & Burke, 2000), and certain roles arereciprocally related (e.g., parent/child or leader/follower) such that individuals in the situationmutually recognize the role relationship (Ash-forth, 2001; Sluss & Ashforth, 2007). In this sense,leadership “is not something the leader pos-sesses” (Hollander, 1993: 29); rather, it expressesa recognized relationship among individuals(Shamir & Eilam, 2005). Relational identity pro-cesses suggest that, in addition to individuals’internalizing a leader or follower identity, theleadership identity will be stronger to the extentthat it is relationally recognized through theadoption of reciprocal role identities as leaderand follower (i.e., for leaders, when others take

on a reciprocal follower identity). This relationalrecognition can be, but is not necessarily, syn-onymous with the organizational hierarchy andindividuals’ positions in that hierarchy.

Collective endorsement is about being seenwithin the broader social environment as part ofa particular social group—for example, leadersor followers (Brewer & Gardner, 1996). We expectthat the more an individual is collectively en-dorsed as part of the group “leaders” or thegroup “followers,” the more those related iden-tities will be reinforced and the stronger andmore stable that particular identity constructionwill be. This collective endorsement might comefrom other individuals (e.g., an upper-level man-ager addressing one member of the group as theleader) or the social context more broadly. Forexample, an individual might not perceive himor herself as possessing the attributes of aleader (follower) or as being in a leader-like(follower-like) position, but the social contextwithin which that individual works might col-lectively endorse him or her as a leader (fol-lower) and thereby initiate the leadership iden-tity construction process.

By conceptualizing leadership identity acrossall three levels of self-construal, we are suggest-ing that leadership development and the con-struction of a leadership identity are about theconstruction of a relationship. As Kouzes andPosner note, “Leadership is a reciprocal rela-tionship. . . . any discussion of leadership mustattend to the dynamics of this relationship”(2003: 1). This recognition is in contrast to muchof the existing literature on leadership that fo-cuses on an individual and the static sense ofbeing a leader but misses how leadershipcomes to be and how it changes over time (Col-linson, 2005). As a result, current theory offerslittle insight into how individuals influenceeach other to collectively construct their respec-tive identities as leaders and followers and toconstruct the leader-follower relationship. Thisarticle moves the leadership field away from astatic and hierarchical conception of leadershipand toward a more dynamic, social, and rela-tional conception of the leadership developmentprocess.

Although our primary contribution is to leader-ship theory, this article makes several contribu-tions to theories of identity and identity construc-tion. First, the tripartite identity constructionprocess that we describe serves as a framework

2010 629DeRue and Ashford

for how scholars might integrate personal, rela-tional, and collective theories of identity. Con-sidering the calls for more integrative theorizingin the identity literature and leadership litera-ture (Avolio, 2007; Hogg, Terry, & White, 1995;Sluss & Ashforth, 2007), our theory showing howthese different levels of self-construal are com-plementary and how they help explain the con-struction of one’s identity is noteworthy.

Second, leadership as an identity may differfrom other more commonly studied identities(e.g., race, gender, or specific role identities).Leadership is ambiguous, with no clear defini-tion or meaning across people (Bass & Bass,2008; Pfeffer, 1977). What it takes to be a leader orfollower, as well as who is a leader or followerin any given social context, is ambiguous, dy-namic, and contextual. These attributes makethe leadership identity high in what Hoang andGimeno (2010) term identity complexity andstrongly suggest a role for social processes in itscreation. While the idea of social interactionismis not new to the identity literature (Goffman,1959; Mead, 1934), the identity literature is onlybeginning to explain when social interaction ismore or less important for identity construction(Ibarra & Barbulescu, 2010). Our theory suggeststhat the nature of the identity itself (in terms ofits ambiguity or other attributes) impacts theprocess by which that identity is constructed.Social mutual influence processes may be mostimportant for the construction of more ambigu-ous identities, such as leadership.

Third, the identity literature focuses primarilyon how individuals come to see themselves vis-à-vis their self-concept. Our theory emphasizesthat it is as important to understand the socialprocesses by which others attribute identities toan individual as it is the identities that an indi-vidual attributes to him or herself. In addition,our theory explains not only how an individualcomes to see him or herself in a particular way,but it also focuses on how a leadership relation-ship is socially constructed and, ultimately, howpatterns of influence form and evolve amongindividuals. In this sense, the target of our the-orizing is different from that in the identity liter-ature and prior treatments of leader identity(e.g., Day & Harrison, 2007; DeRue et al., 2009).

Finally, social interactionist perspectives onidentity construction implicitly assume that so-cial interaction is free, fluid, and without inter-ruption. These perspectives generally do not

model the impact of prior interactions or thenature of the actual claims and grants that occurduring the identity construction process. Ourtheory illustrates the limitations of such as-sumptions. For example, we specify how the na-ture of claims and grants themselves, alongwith the prior history among individuals, canfacilitate or impair the fluid and reciprocal na-ture of the identity construction process. By chal-lenging some of the implicit assumptions em-bedded in the identity literature, we hope thatthis article surfaces new insights about theidentity construction process.

THE WORK OF CONSTRUCTING ALEADERSHIP IDENTITY

If identities are inherently social (Mead, 1934)and both leader and follower identities areavailable to anyone (Day et al., 2009; Kempster,2006; Van Vugt, 2006), then the process by whichcertain people become socially constructed asleaders, and other people as followers, becomesparticularly important to understand. We rootour description of this process in what is called“identity work” in the literature (Pratt, Rock-mann, & Kaufmann, 2006; Snow & Anderson,1987). Based on social interactionism (Blumer,1969; Goffman, 1959), identity work refers to“people being engaged in forming, repairing,maintaining, strengthening or revising” theiridentities (Sveningsson & Alvesson, 2003: 1165).In this sense, identity work is seen as an indi-vidual undertaking aimed at creating, present-ing, and sustaining particular identities. For ex-ample, research has examined the identity workused to sustain a positive image (Ashforth &Kreiner, 1999; Snow & Anderson, 1987), balancedifferent identities (Kreiner, Hollensbe, & Sheep,2006), and customize identities to fit particularenvironments (Pratt et al., 2006).

In this article we offer a broader and moresocial conception of identity work. In particular,we draw from theories of symbolic interaction-ism (Goffman, 1959; Mead, 1934) to propose thatidentity work is undertaken both by an individ-ual projecting a particular image and by othersmirroring back and reinforcing (or not) that im-age as a legitimate identity (Hatch & Schultz,2002). We refer to this broader, multiparty pro-cess as identity construction and find reflectionsof it in Hatch and Schultz’s (2002) examination ofthe iterative and reciprocal process of identity

630 OctoberAcademy of Management Review

construction at the organizational level and inBartel and Dutton’s (2001) qualitative descriptionof how temporary workers engage in behaviorsaimed at establishing themselves as legitimatemembers of the organization. In this processother members of the organization meet thesemoves and acts with affirming or disaffirmingresponses, and through this “reciprocal” identitywork, the ambiguity of organizational member-ship is resolved.

As presented in Figure 1, we propose identitywork in which individuals “claim” an identityand others affirm or “grant” that identity as theunderlying process by which leader and fol-lower identities become socially constructedand form the basis of leader-follower relation-ships. Claiming refers to the actions people taketo assert their identity as either a leader or fol-lower. For example, consider Lebron James’statement to the press upon joining his NBAbasketball team, the Cleveland Cavaliers, as anineteen-year-old rookie: “I’m a leader. I am theleader of this team” (InsideHoops.com, 2004). Orconsider people in organizations who say, “I’mjust not the leader type.” Both statements areverbal assertions that represent claims to a

leader or follower identity in a particularcontext.

In contrast, granting refers to the actions thata person takes to bestow a leader or followeridentity onto another person. Grants can comefrom individuals actively involved in work withthe focal person (who then take on followerroles) or from people who simply notice andendorse a person as a leader (e.g., a colleaguefrom another department). Our focus is on theformer set of individuals. For example, grantsmight include publicly referring to someone asyour group’s leader or, in the case of a followeridentity, explicitly indicating that a personshould act in accordance with the direction ofanother. As shown in Figure 1, individuals cangrant a leader or follower identity by agreeingto a claimer’s assertion or by bestowing theidentity onto a person prior to any initial claim(i.e., the dashed box at the top of Figure 1). Thus,granting can occur in response to other individ-uals’ claiming behavior and/or it can be the mo-tivation for future claiming behavior. It isthrough the interplay of these claims and grantsthat leader and follower identities are, as Sven-ingsson and Alvesson state, “frequently in

FIGURE 1Leadership Identity Construction Process

Person A Person B

leader

LEADER

Claim leader/ Grant follower

identity

Claim follower/Grant leader

identity

Grant leader/ Claim follower

identityClaim leader/ Grant follower

identity

Grant leader/ Claim follower

identityClaim leader/ Grant follower

identity

follower

FOLLOWER

Individual internalization

Relational recognition

Collective endorsement

Individual internalization

Relational recognition

Collective endorsement

Clarity and acceptance of leader-follower relationship

Identity work

2010 631DeRue and Ashford

movement” (2003; 1165). This recursive propertyof the process is consistent with recent work onthe follower’s role in leadership (Howell &Shamir, 2005).

Drawing from the existing literature on iden-tity construction in leadership (DeRue et al.,2009, Gardner & Avolio, 1998; Luhrmann & Eberl,2007) and other contexts (Alvesson, 1994; Bartel &Dutton, 2001; Pratt et al., 2006; Snow & Anderson,1987), we theorize that claiming and grantingtactics vary on two basic dimensions: verbal/nonverbal and direct/indirect. Direct verbal actsaimed at claiming a leader identity might in-clude a person making statements that he or sheis a leader or statements consistent with beingleader-like, while direct verbal granting actsmight include referring to another person as aleader. Similar direct verbal acts can also beused to claim a follower identity, such as statingthat you are simply following the direction ofanother person or that you expect to follow thelead of others in a particular situation. In con-trast, people can also claim or grant a leaderidentity via direct nonverbal acts, such as ma-nipulating physical artifacts associated withleadership or followership (Gallo, 2006). In thiscase a person might claim leadership by dis-playing particular identity cues (e.g., looking thepart; Swann, 1990) or by sitting at the head of ameeting table. In the case of followership, aperson might claim a follower identity by choos-ing to speak in a meeting only when called on.Similarly, a person might grant leadership byoffering the head of a meeting table to anotherperson, or grant followership by not includingthat person in an important (direction-setting)conversation.

Claiming and granting leader and followeridentities can also be more indirect. Indirectclaiming tactics might include the invoking ofrelational ties that communicate and highlightcloseness with recognized authorities or otherleaders. Examples include dropping the name ofan influential organizational leader (in the caseof claiming) or acknowledging a person’s rela-tionship with other notable leaders in the orga-nization (in the case of granting). In the case offollowership, we often see groups in which indi-viduals actively refrain from taking initiativewithin the group. This form of inaction is anindirect claim of followership.

Figure 1 depicts the claiming and grantingprocess as iterative and generative. When a fo-

cal person claims a leader or follower identity,this stimulates other people in the social envi-ronment to consider seeing that focal person inaccordance with that particular identity. Theycommunicate their acceptance of this percep-tion by granting that particular identity to thefocal person through their words or actions (di-rectly or indirectly). Although this granting ofthe identity may not always occur immediatelyand may even require several claims before theidentity is granted, the relational recognition ofthe claim through a reinforcing grant is essen-tial to the identity construction process. For ex-ample, if a person claims leadership in a settingbut others do not reinforce that claim with sup-portive grants, the three aspects of leadershipidentity construction are insufficient for a leader-follower relationship to emerge. The leadershipidentity will not be fully internalized by the in-dividual, it will not be recognized in relationalties between individuals, and it will not be en-dorsed in the broader organization. In this casethe leadership identity and leader-follower rela-tionship do not become part of a “working con-sensus” defining the situation (Goffman, 1959).In contrast, if a person claims a follower identity(e.g., states explicitly or communicates throughactions that he or she expects someone else tolead) and other people reinforce the claim with asupportive grant (e.g., do not look to this personfor guidance, direction, or vision for the task),then that person’s follower identity becomes es-tablished in that particular context.

These reciprocal claims and grants promotethe individual internalization of leader and fol-lower identities and their relational recognitionin group members’ roles and relationships (de-picted by the boxes on the right and left in Fig-ure 1). As others in the organization come torecognize and understand this emerging rela-tional structure and pattern of influence, theleadership identity becomes collectively en-dorsed in the broader organizational context. Aleader-follower relationship is more or less es-tablished (i.e., is clearer and mutually accepted)to the degree that these three conditions aremet. As situations evolve, leader and followeridentities can shift among individuals throughthis same reciprocal process of claiming andgranting, thereby creating a revised structurefor the leader-follower relationship. It is throughthis iterative and generative claiming andgranting process that the leader-follower rela-

632 OctoberAcademy of Management Review

tionship becomes a social reality in organiza-tions and a leadership identity is constructed.

Proposition 1: The construction of aleadership identity occurs whenclaims and grants of leader and fol-lower identities are endorsed with re-ciprocal grants and claims.

The Reciprocal Nature of Claiming andGranting a Leadership Identity

Proposition 1 suggests that a leadership iden-tity is constructed when claiming and grantingmutually reinforce each other. Over time, thispattern forms “deviation-amplifying” loops (Ma-such, 1985; Weick, 1979), in which a deviation inone variable (e.g., more granting behavior)leads to a similar deviation in another variable(e.g., more claiming behavior), which, in turn,further amplifies deviation in the first variable.As the process unfolds, the cyclical nature of theclaiming-granting process is thought to result ineither positive or negative spirals (DeRue et al.,2009). A positive spiral occurs because grants ofa particular identity, leader or follower, conveyinformation about how others in that social en-vironment see the focal individual with respectto that identity. Thus, when individuals receivegrants supporting their claims of a leader orfollower identity, they are inclined to respondwith more frequent and stronger claims for thatidentity. In contrast, a negative spiral occurswhen claiming or granting behaviors are notpositively reinforced (e.g., claims are not rein-forced by grants), and, as a result, these behav-iors are less likely to be repeated in the future(Ferster & Skinner, 1957). The response to fewergrants of a leader identity will be that the focalindividual engages in fewer or weaker claimingbehaviors, which, in turn, will yield fewer sub-sequent grants.

Implicit in this discussion of positive and neg-ative spirals and in our description of the lead-ership identity construction process are severalassumptions about the reciprocal nature ofclaims and grants. First, a deviation-amplifyingpattern presumes that there is convergence inindividuals’ beliefs regarding how leadership isand should be structured in groups. However,we propose that individuals’ conceptions ofleadership range from hierarchical (only oneleader in a group) to shared (multiple leaders in

a group) and that differences in this conceptionhave important implications for how the processunfolds. Second, a deviation-amplifying patternpresumes that the claims and grants exchangedare of a sufficient quality to be perceived accu-rately and have influence on others. Yet bothclaims and grants can vary in their clarity andvisibility to others, and this variation will likelyaffect the process of constructing leader-follower identities. Finally, a description of theclaiming-granting process as deviation amplify-ing leaves unstated the role of history in itsunfolding, but it is likely that a prior history ofreinforcing claims and grants between individ-uals will carry forward and affect how likelyindividuals are to reciprocate the claims andgrants of others. To more carefully delineate theleadership identity construction process, wenow consider the implications of relaxing thesethree assumptions.

Leadership-structure schemas. Consistentwith the leadership literature in general (Bede-ian & Hunt, 2006), we propose that individualsrange from conceptualizing leadership as a pro-cess that can be shared and mutually enactedamong group members (e.g., Carson et al., 2007;Gemmill & Oakley, 1992) to one that is hierar-chically structured such that there is only oneleader in a group and leader and follower iden-tities are mutually exclusive. References to asingle-leader assumption date back to Bion’s(1961) psychoanalytic work on groups, and itspredominance in the management literaturehas been traced by Gemmill and Oakley (1992).Similar to the zero-sum/positive-sum assump-tions that shape negotiation behavior (Pruitt &Rubin, 1986; Thompson & Hastie, 1990), we ex-pect individual differences in leadership-structure schemas to shape when claims arereciprocated with grants and when grants arereciprocated with claims.

When a person holds a hierarchical leader-ship-structure schema, that person is morelikely to conceive of leadership as zero sum.Therefore, a grant of leadership to another indi-vidual implies the claiming of a follower iden-tity. Similarly, a claim of leader identity is likelyto be accompanied by a reciprocal grant of afollower identity. We posit that the degree towhich people converge around a common lead-ership-structure schema will influence the re-ciprocal nature of the claiming-granting pro-cess. For example, in Figure 2a we depict a

2010 633DeRue and Ashford

claiming-granting cycle where individuals ex-perience convergence around a hierarchicalleadership-structure schema. In this scenario aclaim of leadership that is then granted will beaccompanied by reciprocal claims and grants ofa follower identity. This occurs because the in-dividuals involved see leadership as reservedfor a single individual, and so once a claim ofleader identity has been granted, it is consistentwith the individuals’ leadership-structure sche-mas to follow that grant with the claiming andgranting of a reciprocal follower identity. A com-mon understanding and clarity about who is aleader and who is a follower in this particularsituation results, and the individuals experiencelittle tension over leadership. Individuals inter-nalize their identity as leader or follower, mutu-ally recognize their roles and relationships asleaders and followers, and, as a result, thebroader organizational context begins to en-dorse the leader-follower relationship.

Similarly, when individuals experience con-vergence around a shared leadership-structureschema, a more dynamic yet still well-definedleadership identity emerges. Given group mem-bers’ convergent beliefs that more than one

leader can emerge in a group, individuals willlikely grant another’s claim of leadership andaccede to his or her leadership by claiming afollower identity, but at the same time may claimleadership for themselves and receive reciprocal,supportive grants from others. As depicted in Fig-ure 2b, individuals can claim and grant a leaderidentity while also taking on a follower identity inrelation to others. In such situations there is claritywith respect to individuals’ leader and followeridentities, but in a way that involves a dynamicexchange of leadership and followership that isconstantly being renegotiated across time and sit-uations. In such contexts the boundaries betweenleader and follower identities are permeable; as aresult, few identity conflicts and little tension overleadership will emerge.

The reciprocal dynamics of the identity con-struction process become more complex whenindividuals have different leadership-structureschemas. For example, consider a person with ahierarchical leadership-structure schema (Per-son A in Figure 2c) who claims a leader identityin an interaction with a person holding a sharedleadership-structure schema (Person B). PersonB may grant the initial claim but may also con-

FIGURE 2Impact of Leadership-Structure Schemas on Claiming and Granting

 (a)  Successful construction: Reinforced hierarchical   leadership-structure schemas

leadership-structure schemas

  (c) Failed construction: Unreinforced claim

(b) Successful construction: Reinforced shared  (d) Failed construction: Unreinforced grant

Person A (hierarchical)

Person B (hierarchical)

Claim leader

Claim follower/Grant leader

Grant follower

Person A (hierarchical)

Person B(shared)

Claim leader

Grant leader/Claim follower

Denies person B’s claim

Claim leader

Person A(shared)

Person B(shared)

Claim leader

Grant leader/Claim follower

Grant leader/Claim follower

Claim leader

Person A (hierarchical)

Person B(shared)

Claim leader

Grant leader/Claim follower

Grant leader/Claim follower

Does not claim leader

2

1

3 4

1

2

3

1

2

3

4

1

2

3

4

634 OctoberAcademy of Management Review

tinue to claim leadership for him or herself. WithPerson A conceiving leadership as zero sum, heor she will likely resist Person B’s claim (“Whydoesn’t he just follow my lead?”) and continue togrant only a follower identity to Person B. Inturn, we would expect confusion and conflictover leader and follower identities to emerge,thereby resulting in less clarity around the lead-ership identity. Alternatively, as shown in Fig-ure 2d, if Person B holds a shared leadership-structure schema and claims a leader identity,Person A’s reaction may also be confusing. Per-son A might grant the leader identity; comfort-ably claim a follower identity, reflecting his orher hierarchical leadership-structure schema;and never initiate any subsequent claims for aleader identity. In this case Person B may bepuzzled by the lack of subsequent initiative andleadership exhibited by Person A, leading toidentity-based conflict in the development of aleader-follower relationship. In general, diver-gence in leadership-structure schemas will bringabout less individual internalization, relationalrecognition, and collective endorsement ofleader and follower identities and, as a result,will lead to less stable and ill-defined leader-follower relationships.

Proposition 2: When there is conver-gence in hierarchical leadership-structure schemas, once an initialclaim of leadership is granted to anindividual, a clear leadership identityis constructed based on a mutual under-standing that granting a leader identityto one implies the claiming of a fol-lower identity by others (Figure 2a).

Proposition 3: When there is conver-gence in shared leadership-structureschemas, leader and follower identi-ties flow back and forth within therelationship based on a mutual under-standing that granting a leader or fol-lower identity to one individual doesnot preclude the possibility that theidentity will be claimed by andgranted to others (Figure 2b).

Proposition 4: When there is diver-gence in leadership-structure sche-mas, the leadership identity construc-tion process will break down such thatafter an initial claim of leadership has

been granted (Figures 2c and 2d), (a)leaders with hierarchical leadership-structure schemas will not grant thecontinued claims of a leader identityby people with shared leadership-structure schemas and (b) leaders withshared leadership-structure schemaswill continue to grant a leader iden-tity to individuals with hierarchicalleadership-structure schemas whohave discontinued their claims of aleader identity.

Visibility, clarity, and credibility. To sustainthe reciprocal nature of the leadership identityconstruction process, the claims and grantsmust be of sufficient quality. The literature onsocial information processing suggests that so-cial information (e.g., claims and grants) influ-ences human judgment, thought, and actionwhen the information is clear and easy to un-derstand (Daft & Lengel, 1984; Petty & Cacioppo,1986), the saliency and visibility of the informa-tion are high (Fiske, Kenny, & Taylor, 1982), andthe information is credible (Chaiken & Ma-heswaran, 1994; Fisher, Ilgen, & Hoyer, 1979).Extending this perspective to the leadershipidentity construction process, we propose thatclaims and grants of a leader or follower iden-tity are more likely to promote reciprocal grantsand claims when they are clear, credible, andvisible within the broader social context.

For example, consider an individual who iselected to be the designated leader of a group ina public election process during a meeting. Thisgrant of a leader identity is clear, visible, andcredible. In this case credibility results becausethe grant is representative of the group withinwhich the leader identity is being constructed,but credibility could also be high if the granthad come from an expert or highly respectedgroup member.

Contrast this example with a situation inwhich an inexperienced group member pri-vately suggests to another individual in thegroup that he or she is really good at setting anagenda for the group and motivating the groupto accomplish its goals. Is setting an agenda forand motivating the group an aspect of leader-ship? Is this group member a credible reflectionof the group’s opinion or even skilled enough tomake this judgment? In contrast to the first ex-ample, this grant of leader identity is much less

2010 635DeRue and Ashford

visible to other group members, may be unclearin its meaning and intention, and is less credi-ble. As a result, the focal individual will be morelikely to respond with his or her own reciprocalclaim of a leader identity in the first examplethan the second.

Clear claims and grants create transparencyas to how individuals see themselves and howthey are viewed within the social context. Inaddition, highly visible grants of a leader orfollower identity should reduce the image riskassociated with claiming that particular iden-tity (because everyone saw the grant) and, inaddition, likely increase the felt pressure tocomply with a reciprocal claim of that identity.Such claims and grants, though, can be difficultin organizations. Consider, for example, a groupthat is connected primarily through virtualmeans. Lacking media richness (Daft & Lengel,1984), claims and grants made in this contextcan easily be misunderstood by others.

We propose that clarity, visibility, and credi-bility all enhance the likelihood that claims of aleader or follower identity will be reciprocatedwith supportive grants and that grants of aleader or follower identity will be reciprocatedwith supportive claims.

Proposition 5: The greater the clarity,visibility, and credibility of claimsand grants, the more likely thoseclaims and grants will be reinforcedvia reciprocal grants and claims.

History of claims and grants. According to so-ciological theories of role enactment, the way inwhich people have enacted their roles in rela-tion to others in the past strongly influences howthey will enact their roles and behave towardothers in the present and future (Turner, 1978). Inaddition, empirical evidence from social and ap-plied psychology shows that prior behavior in-fluences future behavior (Ouellette & Wood,1998) and that these behavioral response pat-terns can be conscious or unconscious (Bargh,1989; Wegner & Bargh, 1998). In fact, as long asthe situation and context do not differ dramati-cally, prior behavior can lead individuals to de-velop habitual responses that get enacted infuture situations with minimal thought and ef-fort (Ouellette & Wood, 1998; Proctor & Dutta,1993). Similarly, in the identity literature schol-ars have noted that people establish relation-ships between themselves and others and then

engage in behaviors to maintain consistencyand symmetry in those relationships over time(Gergen, 1968; Sampson, 1963, 1985). In thissense, a history of social interaction createsboth an expectation for and a consistent patternof behavior that influences future social interac-tion and behavior.

Drawing from this literature and recent theo-rizing on the role of personal history in leader-ship (Shamir, Dayan-Horesh, & Adler, 2005), wepropose that a prior history of reciprocal andreinforcing claims and grants between peoplewill carry forward and affect those individuals’future claims and grants of leader and followeridentities—especially when the situational con-text is relatively stable over time. The effect ofhistory can be vicarious or direct. A person mayhave a reputation as a leader that carries overinto a new situation, and even though peoplehave never worked with this individual before,his or her reputation will serve as a vicariousmechanism for increasing the likelihood thatclaims will be reciprocated with grants andgrants with claims. The direct effect of historyoccurs when an individual has granted a fol-lower identity to another person previously and,as a result, is more inclined to grant that persona follower identity in the future (especially ifthat person were to first claim a follower iden-tity). Similarly, if a person has claimed a leaderidentity in the past and been granted that iden-tity by others, it is likely this person will claim aleader identity again in the future. In fact, as thehistory of claims and grants between peopledevelops over time, we expect the reinforcingnature of these claims and grants to becomemore habituated and mindless (Langer, Blank, &Chanowitz, 1978). As the automaticity of theleadership identity construction process in-creases, a pattern of claiming and granting be-havior that is reciprocal and mutually reinforc-ing will emerge, leading to a more coherent andenduring leader-follower relationship.

Proposition 6: A prior history of recip-rocal and reinforcing claims andgrants between individuals will carryforward and increase the likelihoodthat current claims and grants will bereciprocated.

Thus far, we have delineated an identity workprocess explaining how leadership relation-ships get constructed through reciprocal and

636 OctoberAcademy of Management Review

mutually reinforcing claims and grants. Wehave also specified several mechanisms thatfacilitate or impair the reciprocal nature of theclaiming-granting process. We now turn our at-tention to predicting and explaining when indi-viduals will initiate a claim or grant of leaderand follower identities.

When Will People Claim and Grant Leader orFollower Identities?

We provide a general framework for identify-ing the antecedents that we believe will be es-pecially important predictors of claiming andgranting. In particular, we focus on (1) implicittheories of leadership that refer to individuals’beliefs about what makes someone an effectiveleader, (2) the motivational risks and rewardsassociated with claiming or granting leader andfollower identities, and (3) the institutionalstructures that can impose leader and followeridentities in group settings. This framework isnot intended to be exhaustive; rather, we iden-tify general categories of antecedents to illus-trate how the origins of claiming and grantingspan from individuals’ internal belief systems tothe organizational context within which theleadership identity is being constructed. Wechose to focus on these particular antecedentsfor two important reasons.

First, the predominant theories of identity de-velopment and maintenance highlight the im-portance of each of these antecedents. For ex-ample, self-categorization theory (Turner, Hogg,Oakes, Reicher, & Wetherell, 1987) explains howone’s implicit theory about the attributes asso-ciated with different social groups influencesthe self-concept and identity that one ultimatelyinternalizes. Likewise, research in social psy-chology (Higgins, 1987; La Guardia, 2009; Markus& Nurius, 1986; Waterman, 2004) points to theimportance of motivation and incentives for theexploration and internalization of particularidentities. Finally, classic theories in sociology(see Stryker & Burke, 2000) and social anthropol-ogy (Cohen, 1994) emphasize the role of socialstructure and context in shaping individuals’identities. In our theory we do not give preferen-tial treatment to any one of these perspectivesbut, rather, span across these different domainsto provide an integrative account of whatprompts someone to claim or grant a leader orfollower identity.

Second, each of these factors plays an impor-tant role in the broader leadership literature,especially given the emphasis in prior researchon cognitive, behavioral, and social constructiv-ist accounts of leadership (Chen & Meindl, 1991;Sjostrand, Sandberg, & Tyrstrup, 2001). However,until now, scholars have not fully articulated theprocess by which these factors enable leader-ship relationships and identities to develop(Day, Zaccaro, & Halpin, 2004). By focusing onthese antecedents, we situate our theory in thebroader nomonological network of leadershipresearch, while also extending prior research byspecifying how these factors shape the develop-ment of leader-follower relationships.

Implicit theories of leadership. By the timepeople begin working in organizations, theyhave developed varying assumptions and be-liefs that form an implicit theory about whatleaders and followers “look like” and how lead-ership unfolds in groups (DeRue et al., 2009;Lord, 1985; Lord & Alliger, 1985; Schyns & Meindl,2005). We propose that these implicit theories ofleadership and followership influence whetherpeople claim a leader or follower identity forthemselves and/or grant a leader or followeridentity to others.

Prior research suggests that individuals at-tribute leadership to others depending on howwell they correspond to or match the perceivers’implicit theory and that this cognitive processcan be conscious or unconscious (Lord, 1985;Lord & Maher, 1991; Schyns & Meindl, 2005). Theexisting literature also suggests that under con-ditions of high social identification with agroup, the standard for perceiving someone as aleader shifts from how prototypical that personis of an effective leader in general to how pro-totypical that individual is of the local group(Lord & Hall, 2005; van Knippenberg & Hogg,2003). In both cases, the more consistency be-tween the focal individual and one’s implicittheory of leadership, the more likely one willattribute the identity of leader to that individual.

Extending this perspective, we propose thatthis reliance on implicit theories of leadershipand followership not only creates a belief aboutwhether a person is a leader or follower but alsoprompts the granting of a leader identity to in-dividuals who match their implicit theory. Thatis, when an individual looks like, seems like,and acts like a leader (follower), people aremore likely to grant that person a leader (fol-

2010 637DeRue and Ashford

lower) identity. Granting can occur both in re-sponse to claiming acts and without a priorclaim by the focal person and can pertain toeither a leader or follower identity. For example,individuals’ implicit theories for what it meansto be a leader can differ. If the focal person’simplicit theory differs from that of others in thesocial environment, others may see leadershipattributes and characteristics in the focal personthat he or she does not yet see. Thus, the processof constructing a leadership identity may beginwith a granting act (e.g., an unexpected desig-nation as the leader of a group or task force).Similarly, when a focal person seems like andacts in line with perceivers’ implicit theory for afollower, perceivers grant that identity to thefocal person.

Not only can implicit theories affect the grant-ing of leader and follower identities, but wepropose that this same process can also promotemore frequent claiming of these identities. Thatis, just as individuals compare others’ attributesto their implicit theories about the prototypicalattributes and characteristics of leaders and fol-lowers (Kenney, Schwartz-Kenney, & Blascovich,1996; Lord & Alliger, 1985; Lord, Foti, & Devader,1984), so they do with their own personal at-tributes (e.g., their traits, behaviors, skills) aswell (DeRue et al., 2009). Given that individualsare motivated to act authentically in accordancewith their self-views (Foote, 1951) and to engagein acts designed to align others’ perceptions ofthem with their self-views (Swann, 1990; Swann,Rentfrow, & Guinn, 2002), if there is a matchbased on this internal-to-self comparison pro-cess, they will be more likely to claim a leader orfollower identity in social interactions (i.e., to actas a leader or to follow others’ leadership). Thisprocess, at times, may be quite automatic, oper-ating without a lot of conscious thought. At othertimes claiming behaviors might be the result ofa deliberate and conscious process whereby in-dividuals decide if the attributes of a leader orfollower are self-descriptive and then engage inclaiming behaviors based on that assessment.Based on these ideas, we put forth two proposi-tions related to individuals’ implicit theoriesand the likelihood of claiming or grantingleader and follower identities.

Proposition 7: The more consistencypeople see between their own at-tributes and their own implicit theory

of leadership (followership), the morethey will claim a leader (follower)identity.

Proposition 8: The more consistencypeople see between others’ attributesand their own implicit theory of lead-ership (followership), the more theywill grant others a leader (follower)identity.

Motivational risks and rewards. A well-established tenet in our understanding of hu-man motivation is that self-interest shapes hu-man behavior and action (Miller, 1999; Miller &Ratner, 1998; Schwartz, 1986). Acting leader-likeand being seen as a leader is a socially valuedand rewarded “ideal self” (Higgins, 1987;Markus & Nurius, 1986) in many organizationalsettings (Day et al., 2009; Kempster, 2006; VanVugt, 2006). It may lead to instrumental rewardssuch as promotions, interpersonal rewards suchas power or status, or image-based rewardssuch as a positive reputation. These rewardscreate a motivation to claim this identity. Inaddition, individuals are often simply motivatedto get things accomplished and claim a leaderidentity because it helps facilitate that accom-plishment (Quinn, 1996). Likewise, some individ-uals, such as those with a high need for power(McClelland & Burnham, 2003), might claim aleader identity because they derive personal,intrinsic satisfaction from influencing othersand being seen as a leader. When these instru-mental, interpersonal, and image-based re-wards are present and associated with seeingoneself and being seen as a leader, we expectindividuals to be motivated to claim a leaderidentity and grant others a follower identity.

For example, even if a discrepancy exists be-tween an individual’s self-view and his or herimplicit theory of leadership (i.e., an individualdoes not see him or herself as leader-like), he orshe may be motivated to experiment with or “tryout” a possible rendition of the self that mightbe particularly valued within the organizationalsetting (Kempster, 2006). Similar to Ibarra’s (1999)notion that individuals experiment with “provi-sional selves,” by taking small steps to act like aleader, individuals can explore where theystand with respect to a leader identity. Thus, inaddition to claiming an identity because it isbelieved to be authentic to oneself (Shamir &Eilam, 2005), individuals may claim a leader

638 OctoberAcademy of Management Review

identity within their social environment both asa way to gain instrumental outcomes that comefrom being seen as a leader and as a way todetermine for themselves whether or not theyview themselves as a leader. Ibarra’s researchon provisional selves (Ibarra, 1999; Ibarra,Snook, & Guillen Ramo, 2008) suggests thatgrants from others will further enhance this ex-perimentation process because they help createthe motivation for further experimenting withthe identity (e.g., “This person sees me as aleader; perhaps I can lead well in this group”).Mentors often grant a leader identity to a men-tee, for example, in the hopes of spurring thatperson’s leader identity and subsequent leaderbehaviors.

The instrumental (e.g., promotions, formalpower, and authority), interpersonal (e.g., infor-mal power and status), and image (e.g., beingseen by others positively) rewards associatedwith seeing oneself and being seen by others asa leader will also make the claiming of a fol-lower identity less likely. In settings where aleader identity is highly esteemed, the per-ceived rewards of leadership will motivate indi-viduals to try to be seen as a leader and, in turn,not “simply” a follower. However, taking onleadership responsibilities also entails consid-erable instrumental, interpersonal, and imagerisk (Gardner, 1995; Heifetz, 1994). While risk per-ceptions may be mitigated somewhat by a pasthistory of leadership success, psychoanalyticperspectives on leadership suggest that peopletake on the follower role in part as a defenseagainst the anxieties associated with the risksof leadership (Gemmill & Oakley, 1992). We ex-pect that individuals, consciously or uncon-sciously, assess the level of risk involved whendeciding whether or not to claim a leader iden-tity or whether to grant that identity to anotherperson.

The amount of instrumental risk involved inleadership depends in part on the likely fate ofthe group or organization, since groups’ suc-cesses and failures are frequently attributed tothe leader (Meindl, 1995; Meindl, Ehrlich, & Duk-erich, 1985). Thus, to the extent there is uncer-tainty regarding a group’s proper course of ac-tion, and the greater the complexity,uncertainty, and dynamism of forces affectingthe group’s performance, the more we expectindividuals to perceive greater instrumentalrisk in taking on leadership responsibilities. Ro-

mance-of-leadership ideas (Meindl, 1995) sug-gest that it is in these uncertain times that indi-viduals are especially motivated to grant aleader identity to other people, since leadershipis seen as an antidote to uncertainty.

Interpersonal and image risks stem from thesocial nature of the claiming-granting process.Individuals may fear the interpersonal awk-wardness that will arise if their claims for lead-ership go ungranted—for example, if they at-tempt to lead but no one follows. Further, aswith any proactive, extrarole behavior, such asvoice (Avery & Quinones, 2002) or issue selling(Dutton & Ashford, 1993), actively claiming lead-ership brings with it a myriad of concerns re-garding how the act of claiming might be per-ceived by others (e.g., as appropriate,presumptuous, or overly controlling). Proactiveactions are thought to be especially risky be-cause observers consider them to be expressiveof individuals’ true underlying nature and de-sires (Bem & Funder, 1978, Grant & Ashford,2008), and this risk is enhanced when individu-als are new to a situation or lack an establishedtrack record of success. In addition, the moredissension there is in a group regarding theproper enactment of leadership, the more riskthere is in claiming leadership because it is notclear how best to lead. For example, claimsbased on one style of leadership (e.g., a partic-ipative style) may be negatively received bythose advocating a different style of leadership(e.g., more authoritarian). These arguments sug-gest the following two propositions.

Proposition 9: The more individualsperceive instrumental, interpersonal,and image rewards associated withleadership, (a) the more they willclaim a leader identity and (b) themore they will grant a follower iden-tity to others.

Proposition 10: The more individualsperceive instrumental, interpersonal,and image risks associated with lead-ership, (a) the less they will claim aleader identity and (b) the more theywill claim a follower identity.

Institutional structures. The processes we aredescribing take place within an organizationwhose formal structures are themselves an in-stitutionalized form of leader/follower grants.

2010 639DeRue and Ashford

Consistent with the reciprocal nature of the pro-cess we are describing, these formal structuresmay be the residual effect of past cycles ofclaiming and granting, the resultant collectiveendorsement of leader and follower identities,and an antecedent to subsequent leadershipidentity construction. Thus, occupying a super-visory role represents a powerful institutionalgrant of a leader identity conveyed through aformal social structure that all group membersrecognize and operate within. Within such sys-tems people often hold expectations of a super-visory role that include leadership. Indeed,“leader” is one of the roles of a manager’s jobidentified by Mintzberg (1973). While leadershipis not synonymous with holding a supervisoryposition and many people in supervisory posi-tions do not embody a leader identity, the gen-eralized expectations of that role bias people’sobservations and interpretations of a supervi-sor’s behavior. As a result, individuals will bemore likely to grant a leader identity to peoplein supervisory positions than they will to indi-viduals who are not in these positions. Thisgreater likelihood of granting a leader identitywill continue until that individual is ineffectiveor acts in ways that are inconsistent with indi-viduals’ implicit theories of leadership.

Position incumbents also hold similar social-ized expectations. As such, they may feel in-creased responsibility to and comfort in claim-ing a leader identity. Incumbency also reducesthe potential image risks of such claiming, sinceindividuals feel particularly free to try leader-like acts because of their position. These leader-claiming behaviors are likely to be reinforcedand affirmed by subordinates as role senders,creating even greater freedom to experimentwith a leader identity over time. Thus, it is likelythat a person’s leader identity will be enhancedby being placed in a formal supervisory role,even though the two are not synonymous.

A similar argument can be made for theclaiming and granting of a follower identity. Thesame institutional structures in organizationsthat grant leadership to one person also advo-cate that those people formally reporting to thedesignated supervisor follow that person’s di-rection and guidance. In this sense, an institu-tional structure grants a follower identity tolower-level actors. At the same time, designatedfollowers hold socialized expectations abouthow they are supposed to act in relation to their

designated supervisor (e.g., often to follow andnot challenge his or her direction). As a result,these dyadic reporting relationships shape peo-ple’s behaviors in ways that help construct andmutually reinforce leader and follower identi-ties. For any given formal relationship in anorganization, the person in the supervisory po-sition is more likely to claim a leader identityand grant a follower identity to the subordinate.In parallel, the subordinate is more likely toclaim a follower identity and grant a leaderidentity.

Proposition 11: To the extent peoplehold formal positions of authority,those individuals will be more likelyto (a) claim a leader identity and (b)receive grants of a leader identityfrom others.

The antecedents that we have identified inde-pendently and in combination will influencewhen individuals claim and grant leader andfollower identities both initially and in responseto others’ claims and grants. In the next sectionwe build on these ideas to develop an agendafor future research that will hopefully serve as aspringboard for research on leadership identityand development.

AN AGENDA FOR FUTURE RESEARCH:EMPIRICAL TESTS AND

THEORETICAL EXTENSIONS

We have presented an identity-based processmodel of leadership development that explainshow leader-follower relationships become insti-tutionalized in the social fabric of organizations.To help build an agenda for future research, wefirst specify several methodological consider-ations relevant to the empirical testing of ourmodel and then offer several ideas for howscholars might extend our theory in new andinteresting directions.

Empirical Tests of the Model

The leadership development process that wepropose is decidedly social, occurs over time,and involves multiple levels of analysis. Assuch, a research program designed to empiri-cally test these ideas should have several char-acteristics. First, research must account for theindividual, dyadic, and organizational aspects

640 OctoberAcademy of Management Review

of the identity construction process. The ante-cedents to claiming and granting span multiplelevels of analysis, ranging from individual cog-nition to organizational hierarchy and socialstructure. As such, research testing our modelneeds to capture the individual, relational, andorganizational factors that influence the leader-ship identity construction process. Similarly, aleadership identity is internalized at the indi-vidual level, recognized in role relationshipsamong individuals, and collectively endorsed ina broader organizational context. Thus, futureresearch, beginning with measurement, mustaccount for these different levels of identity con-struction. One question that needs to be ad-dressed is related to the relative importance ofthese different levels of identity constructionand how the different levels complement or sup-plement each other in the development of lead-er-follower relationships. Is it possible that aleadership identity cannot be collectively en-dorsed until it is individually internalized orrelationally recognized, or might collective en-dorsement prompt individuals to internalize aleader or follower identity—and how would pro-cesses starting from these different points un-fold differently over time? Another question forfuture research is what happens when the threelevels of identity construction do not converge.For example, what are the implications for lead-ership development if a person fails to individ-ually internalize a leader identity that is bothrelationally recognized and collectively en-dorsed in the broader organizational context?

Second, to specify causality and model thereciprocal nature of claiming and granting, theleadership identity construction process needsto be examined over time. The importance ofdelineating the temporal aspects of the processis evident in research by Shamir and Eilam(2005), where behavioral acts that we wouldclassify as grants of a leader identity were notalways followed by the focal individual’s claim-ing leadership and did not lead to the construc-tion of a leadership identity. In our theory weposit that prior claims and grants can carry for-ward and influence future claims and grants.Longitudinal research testing these ideasshould seek to specify why claiming and grant-ing are sometimes reciprocal and create posi-tive spirals and why the process is sometimesdisrupted and breaks down. Another questionthat should be explored via longitudinal re-

search is how the content and scope of leaderand follower identities evolve over time. Leaderand follower identities will initially develop inrelation to specific situations, but through re-peated claiming-granting processes, those iden-tities can shift from situated to generalized iden-tities (Sluss & Ashforth, 2007). In addition, claimsor grants of a particular identity may be accept-able within a specific context or in regard tospecific issues, but in other contexts or regard-ing different issues, those same claims or grantsmay be outside the zone of acceptance and,thus, may be met with resistance.

Given the social and temporal aspects of ourtheory, several research methods are particu-larly well-suited to the study of leadership iden-tities. First, we propose that scholars considerusing a social relations approach (Kenny, 1994;Livi, Kenny, Albright, & Pierro, 2008) for model-ing the individual, dyadic, and group-level in-fluences on the claiming and granting of leaderand follower identities. Specifically, researcherscould have group members and/or observers as-sess the claiming and granting behaviors of allindividuals in the group and then use the socialrelations model to specify the amount of vari-ance in claiming and granting that is explainedby individual differences (e.g., motivation), rela-tional ties between actors (e.g., history of priorclaims and grants), and group-level factors (e.g.,institutional structures).

Second, experience-sampling methods (Csik-szentmihalyi & Larson, 1992; Wheeler & Reis,1991) could be used to capture in real time thepatterns of claiming and granting that lead tothe construction of leader-follower relation-ships. For example, researchers could examinenew groups where these relationships are notalready established and could use daily sam-pling of claims and grants to predict the emer-gence of leader-follower identities and relation-ships in the group over time. This particularmethod would also be ideal for studying otherdynamic factors that might prompt claiming andgranting, such as an individual’s performanceand personal track record of success as a leader.

Third, echoing Parry’s (1998) call for moregrounded theory research on leadership, we callfor more in-depth, qualitative studies to under-stand the form and nature of claiming andgranting in leader-follower relationships. Qual-itative methods that involve observational (e.g.,ethnography) and/or narrative (e.g., autoethnog-

2010 641DeRue and Ashford

raphy) techniques will be particularly valuablein capturing what may only be a semi- or uncon-scious process of acting and reacting to others.By gathering rich, in-depth accounts of the indi-vidual cognitive processes and relational pro-cesses that underlie the claiming and grantingprocess, this research would go a long way to-ward explaining the development and evolutionof leader-follower relationships.

Finally, given the early stage of theorizing inthis area, perhaps the greatest potential for fu-ture research lies in experimental studies thatenable researchers to control the presence andabsence of claiming and granting behaviors ingroups. These experiments will be particularlyvaluable for establishing causality and under-standing what factors govern the initiation ofclaims and grants or impede the reciprocation ofreinforcing claims and grants. By manipulatingthe presence of grants following claims or thepresence of claims following grants, researcherscan capture the reciprocal nature of the processand, by varying context, can explicitly model theeffect of contextual elements such as groupnorms on the leadership development process.

Theoretical Extensions and New Directions

There are several aspects of our theorizingthat provide the foundation for new and inter-esting directions in research on leadership iden-tity and development. First, while our theorizingdescribes how the identity construction processbetween individuals might unfold over time, fu-ture research needs to complicate this picturefurther by considering the process at the grouplevel. When there are numerous individuals en-gaged in the claiming-granting process, thegroup-level composition and dispersion of vari-ables such as implicit leadership theories, mo-tivation, and leadership-structure schemas willlikely influence how the process unfolds (DeRue,Hollenbeck, Ilgen, & Feltz, 2010; Harrison &Klein, 2007). For example, if everyone in a groupperceives high instrumental rewards for being aleader, the leadership identity construction pro-cess will likely feature more competitive claims,less granting, and, as a result, less emergence ofa well-defined leadership identity. In fact, theimpact of having numerous individuals moti-vated to lead may depend, importantly, on thepredominant leadership-structure schema in thegroup. For example, if group members who are

motivated to lead converge on a hierarchicalleadership-structure schema, the group willlikely experience a great deal of competitiveclaiming with little reciprocal granting. Accord-ing to our theory, this will create a situation inwhich leader and follower identities are notfully internalized, recognized, or collectively en-dorsed. It is important that future theorizingspecify how such competing claims get resolvedin groups.

Second, although we identify general catego-ries of antecedents to claiming and granting,there are complexities related to these anteced-ents that are not fully captured in our theorizingand, thus, necessitate further theory develop-ment. For example, we highlight different typesof motivational rewards and risks associatedwith claiming and granting either a leader orfollower identity, but in organizational contextsit is likely that these rewards and risks will notalways be aligned. Future research that ex-plains how people evaluate and make trade-offsbetween different rewards and risks during theleadership identity construction process wouldoffer a meaningful extension to our theory. Inaddition, although we highlight formal institu-tional structures as an important antecedent, weencourage scholars to extend our theory by alsoconsidering the impact of informal structures,such as social stratification and status hierar-chies (Blau & Duncan, 1967; Lenski, 1984) andsocial networks (Granovetter, 1985; Krackhardt,1990). For example, do men and women, becauseof status or stereotype differences, experiencethe leadership identity construction process indifferent ways? Or how does individuals’ posi-tion within informal networks influence theclaiming and granting of both leader and fol-lower identities, independent of their positionwithin the formal organizational hierarchy? Re-search investigating questions such as thesewould begin to model the intersection of theclaiming and granting processes specified inthis article with the underlying social structuresembedded within organizations.

Finally, there is a need for future research thatexplicitly models the role of organizational andcultural contexts on the leadership identity con-struction process. Our work provides specificsfor a nascent recognition of organizations as“identity workspaces” within which individualswork out who they are with respect to the con-text and each other (Petriglieri & Petriglieri,

642 OctoberAcademy of Management Review

2010). In some contexts acts of leadership mightbe discouraged by cultural norms and values(Lord, Brown, Harvey, & Hall, 2001), whereasother contexts (e.g., empowered, decentralizedorganizational cultures; Spreitzer, 1996) mightencourage individuals to take on leadership re-sponsibilities in groups (Howell & Shamir, 2005).In these contexts, claiming of a leader identity isexpected and normatively sanctioned, giving in-dividuals the space to experiment with a leaderidentity and assess the extent to which thatidentity reflects their true self.

It is also possible that some organizationshave norms that enable more rapid constructionof leadership identities—for example, in organi-zations where there is a common leadership-structure schema that is firmly held and widelyshared. Whether that leadership-structureschema is hierarchical or shared, its strength asa norm should facilitate reciprocal claiming andgranting and allow for the rapid development ofwell-defined leader-follower relationships. Incontrast, organizations without such norms ororganizations going through significantchanges might experience greater conflict overleadership and within leader-follower relation-ships (Kan & Parry, 2004), which, in turn, maydistract from effective work performance. Oneinteresting question that should be explored iswhat happens when individuals’ leadership-structure schemas converge (e.g., around ashared schema) but those schemas are inconsis-tent with the dominant leadership-structureschema in the broader organizational context.Scholars could extend our theorizing in new andinteresting directions by modeling how thesedifferent organizational contexts influence theleadership development process.

CONCLUDING REMARKS

To address recent calls to define leadershipas a social and mutual influence process, ourmodels of leadership must illuminate the pro-cesses by which leadership relationships andidentities are developed. In this article we buildon prior research to articulate a process bywhich individuals jointly construct a leadershiprelationship and explain why some individualscome to be seen as leaders and others as follow-ers. We describe a generative process wherebyindividuals cocreate their respective identitiesas leaders and followers, specify broad catego-

ries of tactics used to achieve this cocreation,and delineate some of the antecedents and con-ditions that govern its unfolding.

Understanding this process may be particu-larly important for understanding leadershipdevelopment in contexts increasingly character-ized by rotated (Erez, LePine, & Elms, 2002), dis-tributed (Gronn, 2002), and/or shared (Carson etal., 2007; Pearce & Conger, 2003) leadership. Themore members of these groups have internal-ized leadership identities that are mutually rec-ognized and collectively endorsed, the more suc-cessful these distributed and shared forms ofleadership will be. Our hope is that the presenttheory provides the foundation for expandingthe field’s conception of leadership and leader-ship development processes and that it stimu-lates future research on leadership develop-ment as a social and mutual influence process.

REFERENCES

Alvesson, M. 1994. Talking in organizations: Managing iden-tity and impressions in an advertising agency. Organi-zation Studies, 15: 535–563.

Ancona, D., & Backman, E. V. 2008. Distributed, shared orcollective leadership: A new leadership model for thecollaborative era? Paper presented at the annual meet-ing of the Academy of Management, Anaheim, CA.

Ashforth, B. E. 2001. Role transitions in organizational life: Anidentity-based perspective: Mahwah, NJ: LawrenceErlbaum Associates.

Ashforth, B. E., & Kreiner, G. E. 1999. “How can you do it?”:Dirty work and the challenge of constructing a positiveidentity. Academy of Management Review, 24: 413–434.

Avery, D. R., & Quinones, M. A. 2002. Disentangling the ef-fects of voice: The incremental roles of opportunity, be-havior, and instrumentality in predicting proceduralfairness. Journal of Applied Psychology, 87: 81–86.

Avolio, B. J. 2007. Promoting more integrative strategies forleadership theory-building. American Psychologist, 62:25–33.

Bargh, J. A. 1989. Conditional automaticity: Varieties of au-tomatic influence in social perception and cognition. InJ. S. Uleman & J. A. Bargh (Eds.), Unintended thought:3–51. New York: Guilford Press.

Bartel, C., & Dutton, J. 2001. Ambiguous organizational mem-berships: Constructing organizational identities in in-teractions with others. In M. Hogg & D. J. Terry (Eds.),Social identity processes in organizational contexts: 115–130. Philadelphia: Psychology Press.

Bass, B., & Bass, R. 2008. The Bass handbook of leadership:Theory, research, and managerial applications. NewYork: Free Press.

Bedeian, A. G., & Hunt, J. G. 2006. Academic amnesia and

2010 643DeRue and Ashford

vestigial assumptions of our forefathers. LeadershipQuarterly, 17: 190–205.

Bem, D. J., & Funder, D. C. 1978. Predicting more of peoplemore of the time—Assessing personality of situations.Psychological Review, 85: 485–501.

Bion, W. R. 1961. Experiences in groups. New York: BasicBooks.

Blau, P. M., & Duncan, O. D. 1967. The American occupationalstructure. New York: Wiley.

Blumer, H. 1969. Symbolic interactionism: Perspective andmethod. Englewood Cliffs, NJ: Prentice-Hall.

Brewer, M. B., & Gardner, W. 1996. Who is this“ we”? Levelsof collective identity and self representations. Journal ofPersonality and Social Psychology, 71: 83–93.

Carson, J. B., Tesluk, P. E., & Marrone, J. A. 2007. Sharedleadership in teams: An investigation of antecedent con-ditions and performance. Academy of ManagementJournal, 50: 1217–1234.

Chaiken, S., & Maheswaran, D. 1994. Heuristic processingcan bias systematic processing: Effects of source credi-bility, argument ambiguity, and task importance on at-titude judgment. Journal of Personality and Social Psy-chology, 66: 460–473.

Chan, K. Y., & Drasgow, F. 2001. Toward a theory of individ-ual differences and leadership: Understanding the mo-tivation to lead. Journal of Applied Psychology, 86: 481–498.

Charan, R., Drotter, S., & Noel, J. 2000. The leadership pipe-line: How to build the leadership-powered company.San Francisco: Jossey-Bass.

Chen, C. C., & Meindl, J. R. 1991. The construction of leader-ship images in the popular press: The case of DonaldBurr and People Express. Administrative Science Quar-terly, 36: 521–551.

Cohen, A. P. 1994. Self consciousness: An alternative anthro-pology of identity. New York: Routledge.

Collinson, D. 2005. Dialectics of leadership. Human Rela-tions, 58: 1419–1442.

Collinson, D. 2006. Rethinking followership: A post-structuralist analysis of follower identities. LeadershipQuarterly, 17: 179–189.

Csikszentmihalyi, M., & Larson, R. 1992. Validity and reliabil-ity of the experience sampling method. In M. W. de Vries(Ed.), The experience of psychopathology: Investigatingmental disorders in their natural settings: 43–57. Cam-bridge: Cambridge University Press.

Daft, R. L., & Lengel, R. H. 1984. Information richness—A newapproach to managerial behavior and organization de-sign. Research in Organizational Behavior, 6: 191–233.

Day, D. V., & Harrison, M. M. 2007. A multilevel, identity-based approach to leadership development. Human Re-source Management Review, 17: 360–373.

Day, D. V., Harrison, M. M., & Halpin, S. M. 2009. An integra-tive approach to leader development: Connecting adultdevelopment, identity, and expertise. New York: Psy-chology Press.

Day, D. V., & Lance, C. E. 2004. Understanding the develop-ment of leadership complexity through latent growthmodeling. In D. V. Day, S. J. Zaccaro, & S. M. Halpin (Eds.),Leader development for transforming organizations: 41–69. Mahwah, NJ: Lawrence Erlbaum Associates.

Day, D. V., Zaccaro, S. J., & Halpin, S. M. (Eds.). 2004. Leaderdevelopment for transforming organizations. Mahwah,NJ: Lawrence Erlbaum Associates.

DeRue, D. S., Ashford, S. J., & Cotton, N. C. 2009. Assuming themantle: Unpacking the process by which individualsinternalize a leader identity. In L. M. Roherts & J. E.Dutton (Eds.), Exploring positive identities and organi-zations: Building a theoretical and research foundation:213–232. New York: Taylor & Francis.

DeRue, D. S., Hollenbeck, J. R., Ilgen, D. R., & Feltz, D. 2010.Efficacy dispersion in teams: Moving beyond agreementand aggregation. Personnel Psychology, 63: 1–40.

Dutton, J. E., & Ashford, S. J. 1993. Selling issues to top man-agement. Academy of Management Review, 18: 397–428.

Erez, A., LePine, J. A., & Elms, H. 2002. Effects of rotatedleadership and peer evaluation on the functioning andeffectiveness of self-managed teams: A quasi-experi-ment. Personnel Psychology, 55: 929–948.

Ferster, C. B., & Skinner, B. F. 1957. Schedules of reinforce-ment. New York: Appleton-Century-Crofts.

Fisher, C. D., Ilgen, D. R., & Hoyer, W. D. 1979. Source credi-bility, information favorability, and job offer acceptance.Academy of Management Journal, 22: 94–103.

Fiske, S. T., Kenny, D. A., & Taylor, S. E. 1982. Structuralmodels for the mediation of salience effects on attribu-tion. Journal of Experimental Social Psychology, 18: 105–127.

Foote, N. N. 1951. Identification as the basis for a theory ofmotivation. American Sociological Review, 16: 14–21.

Gallo, C. 2006. Leaders must look the part. BusinessWeek.January 16: http://www.businessweek.com/smallbiz/content/jan2006/sb20060118_041997.htm.

Gardner, H. 1995. Leading minds: An anatomy of leadership.New York: Basic Books.

Gardner, W. L., & Avolio, B. J. 1998. The charismatic relation-ship: A dramaturgical perspective. Academy of Manage-ment Review, 23: 32–58.

Gecas, V. 1982. The self-concept. Annual Review of Sociol-ogy, 8: 1–33.

Gemmill, G., & Oakley, J. 1992. Leadership—An alienatingsocial myth. Human Relations, 45: 113–129.

Gergen, K. J. 1968. Personal consistency and the presentationof self. In C. Gordon & K. J. Gergen (Eds.), The self insocial interaction: 299–308. New York: Wiley.

Goffman, E. 1959. The presentation of self in everyday life.Garden City, NY: Doubleday.

Graen, G. B., & Uhl-Bien, M. 1995. Development of leader-member exchange (LMX) theory of leadership over 25years: Applying a multi-level multi-domain perspective.Leadership Quarterly, 6: 219–247.

Granovetter, M. 1985. Economic-action and social-struc-

644 OctoberAcademy of Management Review

ture—The problem of embeddedness. American Journalof Sociology, 91: 481–510.

Grant, A., & Ashford, S. J. 2008. The dynamics of proactivity atwork. Research in Organizational Behavior, 28: 3–34.

Gronn, P. 2002. Distributed leadership as a unit of analysis.Leadership Quarterly, 13: 423–451.

Hall, D. T. 2004. Self-awareness, identity, and leader devel-opment. In D. V. Day, S. J. Zaccaro, & S. M. Halpin (Eds.),Leader development for transforming organizations:153–176. Mahwah, NJ: Lawrence Erlbaum Associates.

Harrison, D. A., & Klein, K. J. 2007. What’s the difference?Diversity constructs as separation, variety, or disparityin organizations. Academy of Management Review, 32:1199–1228.

Hatch, M. J., & Schultz, M. 2002. The dynamics of organiza-tional identity. Human Relations, 55: 989–1018.

Heifetz, R. A. 1994. Leadership without easy answers. Cam-bridge, MA: Harvard University Press.

Higgins, E. T. 1987. Self-discrepancy—A theory relating selfand affect. Psychological Review, 94: 319–340.

Hoang, H., & Gimeno, J. 2010. Becoming a founder: Howfounder role identity affects entrepreneurial transitionsand persistence in founding. Journal of Business Ventur-ing, 25: 41–53.

Hogg, M. A., Terry, D. J., & White, K. M. 1995. A tale of twotheories: A critical comparison of identity theory withsocial identity theory. Social Psychology Quarterly, 58:255–269.

Hollander, E. P. 1978. Leadership dynamics: A practical guideto effective relations. New York: Free Press.

Hollander, E. P. 1993. Legitimacy, power, and influence: Aperspective on relational features of leadership. In M. M.Chemers & R. Ayman (Eds.), Leadership theory and re-search: Perspectives and directions: 29–48. San Diego:Academic Press.

Howell, J. M., & Shamir, B. 2005. The role of followers in thecharismatic leadership process: Relationships and theirconsequences. Academy of Management Review, 30: 96–112.

Ibarra, H. 1999. Provisional selves: Experimenting with im-age and identity in professional adaptation. Adminis-trative Science Quarterly, 44: 764–791.

Ibarra, H., & Barbulescu, R. 2010. Identity as narrative: Prev-alence, effectiveness, and consequences of narrativeidentity work in macro work role transitions. Academy ofManagement Review, 35: 135–154.

Ibarra, H., Snook, S., & Guillen Ramo, L. 2008. Identity-basedleader development. INSEAD working paper series No.32, Fontainebleau, France.

InsideHoops.com. 2004. Lebron James interview. http://www.insidehoops.com/lebron-james-interview-0104003.shtml, ac-cessed January 30, 2010.

Kan, M. M., & Parry, K. W. 2004. Identifying paradox: Agrounded theory of leadership in overcoming resistanceto change. Leadership Quarterly, 15: 467–491.

Kark, R., & van Dijk, D. 2007. Motivation to lead, motivation to

follow: The role of the self-regulatory focus in leadershipprocesses. Academy of Management Review, 32: 500–528.

Kellerman, B. 2008. Followership: How followers are creatingchange and changing leaders. Boston: Harvard BusinessSchool Press.

Kempster, S. 2006. Leadership learning through lived expe-rience: A process of apprenticeship? Journal of Manage-ment & Organization, 12(1): 4–22.

Kenney, R. A., Schwartz-Kenney, B. M., & Blascovich, J. 1996.Implicit leadership theories: Defining leaders describedas worthy of influence. Personality and Social Psychol-ogy Bulletin, 22: 1128–1143.

Kenny, D. A. 1994. Interpersonal perception: A social relationsanalysis. New York: Guilford Press.

Komives, S. R., Owen, J. E., Longerbeam, S. D., Mainella, F. C.,& Osteen, L. 2005. Developing a leadership identity. Jour-nal of College Student Development, 46: 593–611.

Kouzes, J. M., & Posner, B. Z. 2003. The leadership challenge(3rd ed.). San Francisco: Jossey-Bass.

Krackhardt, D. 1990. Assessing the political landscape—Structure, cognition, and power in organizations. Ad-ministrative Science Quarterly, 35: 342–369.

Kreiner, G. E., Hollensbe, E. C., & Sheep, M. L. 2006. Where isthe “me” among the “we”? Identity work and the searchfor optimal balance. Academy of Management Journal,49: 1031–1057.

La Guardia, J. G. 2009. Developing who I am: A self-determination theory approach to the establishment ofhealthy identities. Educational Psychologist, 44: 90–104.

Langer, E., Blank, A., & Chanowitz, B. 1978. Mindlessness ofostensibly thoughtful action—Role of placebic informa-tion in interpersonal interaction. Journal of Personalityand Social Psychology, 36: 635–642.

Lenski, G. E. 1984. Power and privilege: A theory of socialstratification. Chapel Hill: University of North CarolinaPress.

Livi, S., Kenny, D. A., Albright, L., & Pierro, A. 2008. A socialrelations analysis of leadership. Leadership Quarterly,19: 235–248.

Lord, R. G. 1985. An information processing approach tosocial perceptions, leadership perceptions, and behav-ioral measurement in organizational settings. Researchin Organizational Behavior, 7: 87–128.

Lord, R. G., & Alliger, G. M. 1985. A comparison of fourinformation processing models of leadership and socialperceptions. Human Relations, 38: 47–65.

Lord, R. G., Brown, D. J., Harvey, J. L., & Hall, R. J. 2001.Contextual constraints on prototype generation andtheir multilevel consequences for leadership percep-tions. Leadership Quarterly, 12: 311–338.

Lord, R. G., Foti, R. J., & Devader, C. L. 1984. A test of leader-ship categorization theory—Internal structure, informa-tion-processing, and leadership perceptions. Organiza-tional Behavior and Human Performance, 34: 343–378.

Lord, R. G., & Hall, R. J. 2005. Identity, deep structure and the

2010 645DeRue and Ashford

development of leadership skill. Leadership Quarterly,16: 591–615.

Lord, R. G., & Maher, K. J. 1991. Leadership and informationprocessing: Linking perceptions and performance. Bos-ton: Unwin Hyman.

Luhrmann, T., & Eberl, P. 2007. Leadership and identity con-struction: Reframing the leader-follower interactionfrom an identity theory perspective. Leadership, 3: 115–127.

Markus, H., & Nurius, P. 1986. Possible selves. AmericanPsychologist, 41: 954–969.

Masuch, M. 1985. Vicious circles in organizations. Adminis-trative Science Quarterly, 30: 14–33.

McClelland, D. C., & Burnham, D. H. 2003. Power is the greatmotivator. Harvard Business Review, 81(1): 117–126.

Mead, G. H. 1934. Mind, self and society. Chicago: Universityof Chicago Press.

Meindl, J. R. 1995. The romance of leadership as a follower-centric theory—A social constructionist approach. Lead-ership Quarterly, 6: 329–341.

Meindl, J. R., Ehrlich, S. B., & Dukerich, J. M. 1985. The ro-mance of leadership. Administrative Science Quarterly,30: 78–102.

Miller, D. T. 1999. The norm of self-interest. American Psy-chologist, 54: 1053–1060.

Miller, D. T., & Ratner, R. K. 1998. The disparity between theactual and assumed power of self-interest. Journal ofPersonality and Social Psychology, 74: 53–62.

Mintzberg, H. 1973. The nature of managerial work. NewYork: Harper & Row.

Morgeson, F. P., DeRue, D. S., & Karam, E. P. 2010. Leadershipin teams: A functional approach to understanding lead-ership structures and processes. Journal of Manage-ment, 36: 5–39.

Ouellette, J. A., & Wood, W. 1998. Habit and intention ineveryday life: The multiple processes by which pastbehavior predicts future behavior. Psychological Bulle-tin, 124: 54–74.

Parry, K. W. 1998. Grounded theory and social process: A newdirection for leadership research. Leadership Quarterly,9: 85–105.

Pearce, C. L., & Conger, J. A. 2003. Shared leadership: Re-framing the hows and whys of leadership. ThousandOaks, CA: Sage.

Petriglieri, G., & Petriglieri, J. 2010. Identity workspaces: Thecase of business schools. Academy of ManagementLearning & Education, 9: 44–60.

Petty, R. E., & Cacioppo, J. T. 1986. Communication and per-suasion: Central and peripheral routes to attitudechange. New York: Springer.

Pfeffer, J. 1977. The ambiguity of leadership. Academy ofManagement Review, 2: 104–112.

Pratt, M. G., Rockmann, K. W., & Kaufmann, J. B. 2006. Con-structing professional identity: The role of work andidentity learning cycles in the customization of identity

among medical residents. Academy of ManagementJournal, 49: 235–262.

Proctor, R. W., & Dutta, A. 1993. Do the same stimulus-response relations influence choice reactions initiallyand after practice? Journal of Experimental Psychology:Learning, Memory, and Cognition, 19: 922–930.

Pruitt, D. G., & Rubin, J. Z. 1986. Social conflict: Escalation,stalemate and settlement. New York: Random House.

Quinn, R. E. 1996. Deep change: Discovering the leaderwithin. San Francisco: Jossey-Bass.

Sampson, E. E. 1963. Status congruence and cognitive con-sistency. Sociometry, 26: 146–162.

Sampson, E. E. 1985. The decentralization of identity: Towarda revised concept of personal and social order. Ameri-can Psychologist, 40: 1203–1211.

Schwartz, B. 1986. The battle for human nature. New York:Norton.

Schyns, B., & Meindl, J. R. 2005. Implicit leadership theories:Essays and explorations. Charlotte, NC: InformationAge.

Selznick, P. 1957. Leadership in administration: A sociologi-cal interpretation. New York: Harper & Row.

Shamir, B., Dayan-Horesh, H., & Adler, D. 2005. Leading bybiography: Towards a life-story approach to the study ofleadership. Leadership, 1: 13–29.

Shamir, B., & Eilam, G. 2005. What’s your story?—Toward alife-story approach to authentic leadership. LeadershipQuarterly, 16: 395–418.

Shamir, B., House, R. J., & Arthur, M. B. 1993. The motivationaleffects of charismatic leadership: A self-concept basedtheory. Organization Science, 4: 577–594.

Sjostrand, S. E., Sandberg, J., & Tyrstrup, M. 2001. Invisiblemanagement: The social construction of leadership. Lon-don: Thomson Learning.

Sluss, D. M., & Ashforth, B. E. 2007. Relational identity andidentification: Defining ourselves through work rela-tionships. Academy of Management Review, 32: 9–32.

Snow, D. A., & Anderson, L. 1987. Identity work among thehomeless: The verbal construction and avowal of per-sonal identities. American Journal of Sociology, 92:1336–1371.

Spreitzer, G. M. 1996. Social structural characteristics of psy-chological empowerment. Academy of ManagementJournal, 39: 483–504.

Spreitzer, G. M., & Quinn, R. E. 2001. A company of leaders:Five disciplines for unleashing the power in your work-force. San Francisco: Jossey-Bass.

Stryker, S. 1980. Symbolic interactionism: A social structuralversion. Menlo Park, CA: Benjamin/Cummings.

Stryker, S., & Burke, P. J. 2000. The past, present, and future ofan identity theory. Social Psychology Quarterly, 63: 284–297.

Sveningsson, S. F., & Alvesson, M. 2003. Managing manage-rial identities: Organizational fragmentation, discourseand identity struggle. Human Relations, 56: 1163–1193.

646 OctoberAcademy of Management Review

Swann, W. B., Jr. 1990. To be adored or to be known. Motiva-tion and Cognition, 2: 408–448.

Swann, W. B., Jr., Rentfrow, P. J., & Guinn, J. S. 2002. Self-verification: The search for coherence. In M. Leary &J. Tagney (Eds.), Handbook of self and identity: 367–383.New York: Guilford Press.

Thompson, L., & Hastie, R. 1990. Social perception in negoti-ation. Organizational Behavior and Human DecisionProcesses, 47: 98–123.

Turner, J. C. 1978. Social categorization and social discrimi-nation in the minimal group paradigm. In H. Tajfel (Ed.),Differentiation between social groups: 235–250. London:Academic Press.

Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Weth-erell, M. S. 1987. Rediscovering the social group: A self-categorization theory. Oxford & New York: Blackwell.

Uhl-Bien, M., Marion, R., & McKelvey, B. 2007. Complexityleadership theory: Shifting leadership from the indus-trial age to the knowledge era. Leadership Quarterly, 18:298–318.

van Knippenberg, D., & Hogg, M. A. 2003. A social identity

model of leadership effectiveness in organizations. Re-search in Organizational Behavior, 25: 245–297.

Van Vugt, M. 2006. Evolutionary origins of leadership andfollowership. Personality and Social Psychology Review,10: 354.

Van Vugt, M., Hogan, R., & Kaiser, R. B. 2008. Leadership,followership, and evolution: Some lessons from the past.American Psychologist, 63: 182–196.

Waterman, A. S. 2004. Finding someone to be: Studies on therole of intrinsic motivation in identity formation. Iden-tity, 4: 209–228.

Wegner, D. M., & Bargh, J. A. 1998. Control and automaticityin social life. In D. Gilbert, S. T. Fiske, & G. Lindzey(Eds.), Handbook of social psychology (4th ed.): 446–496.New York: McGraw-Hill.

Weick, K. 1979. The social psychology of organizing. NewYork: McGraw-Hill.

Wheeler, L., & Reis, H. T. 1991. Self-recording of everyday lifeevents: Origins, types, and uses. Journal of Personality,59: 339–354.

D. Scott DeRue (dsderue@umich.edu) is an assistant professor of management andorganizations at the University of Michigan’s Stephen M. Ross School of Business. Hereceived his Ph.D. from Michigan State University. His research focuses on leadershipand team dynamics, with a particular focus on how leaders and teams in organiza-tions adapt, learn, and develop over time.

Susan J. Ashford (sja@umich.edu) is the Michael and Susan Jandernoa Professor ofManagement and Organization at the University of Michigan’s Stephen M. RossSchool of Business. Sue received her Ph.D. from Northwestern University. Her currentresearch interests include leader effectiveness, issue selling, feedback, self-management, and individual proactivity.

2010 647DeRue and Ashford

Copyright of Academy of Management Review is the property of Academy of Management and its content may

not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written

permission. However, users may print, download, or email articles for individual use.