AN INTEGRATIVE MODEL OF LEGITIMACY JUDGMENTS...legitimacy, situate the construct of legitimacy in...

AN INTEGRATIVE MODEL OFLEGITIMACY JUDGMENTS

LEIGH PLUNKETT TOSTUniversity of Washington

I develop a theoretical framework that specifies the content underlying legitimacyjudgments and a model of the process by which these judgments develop and change.I argue that individual-level legitimacy judgments are based on evaluations that fallalong three dimensions (instrumental, relational, and moral). I specify three stages ofthe legitimacy judgment process and two modes by which judgments may be devel-oped or revised (evaluative and passive). I end by discussing implications for thestudy of institutional change.

The critical role of legitimacy in determiningthe development and endurance of organiza-tions and other social systems has been docu-mented by sociologists and strategy researchersfor decades. For example, Pollock and Rindova(2003) showed that perceptions of organizationallegitimacy shape investor behavior, and Bansaland Clelland (2004) demonstrated that organiza-tions with high levels of legitimacy are insu-lated from unsystematic variations in their stockprices. Indeed, legitimacy seems to provide or-ganizations with a “reservoir of support” thatenhances the likelihood of organizational sur-vival (Dowling & Pfeffer, 1975; Rao, 1994) andperpetuates organizational influence by in-creasing individuals’ loyalty to the organizationand willingness to accept organizational ac-tions, decisions, and policies (Tyler, 2006; Tyler& Blader, 2000, 2005). Similarly, Thomas, Walker,and Zelditch (1986) demonstrated that legiti-macy judgments lead to the persistence of ineq-uitable social structures, and political scientistshave long argued that legitimacy facilitates ef-fective governance (Gibson, 2004; Weatherford,1992).

Given this pivotal role that it plays in thesurvival of social systems, legitimacy has beendescribed as “perhaps the most central conceptin institutional research” (Colyvas & Powell,

2006). Legitimacy is critical in institutional re-search because it is a necessary component ofinstitutionalization, which occurs as an emerg-ing social entity gains a taken-for-granted qual-ity that leads it to be perceived as an objectiveand natural reality. Consequently, institutionaltheorists increasingly specify illegitimacy as acritical driver of the pursuit of institutional andorganizational change (e.g., Greenwood,Suddaby, & Hinings, 2002; Suchman, 1995). Thatis, changes in organizational forms, practices,and policies require that new arrangements beviewed as more legitimate than existing ones(Oliver, 1991; Suddaby & Greenwood, 2005).

The process of institutional change necessar-ily involves shifts in individuals’ judgments ofthe legitimacy of existing social entities and,consequently, shifts in individuals’ behaviorswith respect to those entities. Recognizing thisimplication, in research on institutional change,scholars have recently begun to focus more at-tention on the microlevel processes involved ininstitutional change (e.g., Phillips, Lawrence, &Hardy, 2004; Reay, Golden-Biddle, & GermAnn,2006; Suddaby & Greenwood, 2005; Zilber, 2002).Research in this area has focused on under-standing how interactions among individualsconstitute social reality and determine what isdeemed acceptable within social systems.

However, institutional theorists have paid rel-atively little theoretical or empirical attention tothe intraindividual dynamics of legitimacyjudgments (i.e., the content, formation, andchange of the judgments themselves). While le-gitimacy is ultimately a collective-level phe-nomenon, an understanding of the microleveldynamics of legitimacy judgments is crucial be-

I am grateful to Dan Ariely, Bruce Avolio, Chris Bauman,Drew Carton, Adam Grant, Rick Larrick, Allan Lind, SimSitkin, Scott Sonenshein, Tony Tost, Kim Wade-Benzoni, andBennet Zelner for comments on previous versions of thisarticle. I am also grateful to associate editor Roy Suddabyand three anonymous reviewers for their detailed and con-structive feedback throughout the review process.

� Academy of Management Review2011, Vol. 36, No. 4, 686–710.http://dx.doi.org/10.5465/amr.2010.0227

686Copyright 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.

cause individuals’ judgments and perceptionsconstitute the “micro-motor” (Powell & Colyvas,2008) that guides their behavior, thereby influ-encing interactions among individuals, which,in turn, coalesce to constitute collective-levellegitimacy and social reality. Therefore, an un-derstanding of the individual-level dynamics oflegitimacy judgments can help scholars to bet-ter understand not only the dynamics of institu-tional change but also the critical role that indi-viduals play in those change processes.

The lack of attention to individual-level judg-ments of legitimacy does not stem from a lack ofinterest in the individual-level dynamics of in-stitutional change. Indeed, calls for more inte-gration of microlevel and macrolevel researchon legitimacy have become commonplace in theinstitutional literature (e.g., DiMaggio & Powell,1991; Powell & Colyvas, 2008; Thornton & Ocasio,2008; Zucker, 1991). However, the development ofa model of the individual-level dynamics of le-gitimacy judgments requires an integration ofthe social psychological research on legitimacywith institutional theory, and a number of bar-riers impede such an endeavor. Specifically, thetwo fields of research use different definitions oflegitimacy, situate the construct of legitimacy indifferent nomological networks, tend to examinedifferent types of targets of legitimacy judg-ments, and obviously differ with respect to levelof analysis.

My goal in this article is to overcome thesebarriers in order to make two key contributionsto the study of institutional change. First, I inte-grate social psychological and institutional the-ories of legitimacy to specify the content of le-gitimacy judgments. By “content,” I refer to thesubstantive perceptions and beliefs that under-lie the judgment of an entity as legitimate orillegitimate. An understanding of the content oflegitimacy judgments helps scholars to answerthe question, “What does it mean substantivelyfor an individual to judge an entity, such as anorganization or a leader, to be legitimate?” Sec-ond, I integrate social psychological and insti-tutional theories of the process of legitimation inorder to construct a model of how legitimacyjudgments develop and change over time. Anunderstanding of the process of legitimacy judg-ment formation, use, and change can help re-searchers to understand when, how, and why anindividual’s judgment of the legitimacy of anentity changes from a judgment of legitimacy to

one of illegitimacy (or vice versa) and, conse-quently, leads the individual to seek change (orto preserve the status quo). Thus, my central aimin this article is to develop a better understand-ing of the individual-level dynamics of legiti-macy judgments with respect to both the contentof those judgments and the process by whichthey are developed and changed.

I proceed as follows. First, I explore the waysin which the construct of legitimacy has beendefined and used in institutional theory and so-cial psychological research. Next, I examinework from both fields on the content of legiti-macy judgments, and I specify a typology of thecontent underlying legitimacy judgments. I thenbuild on this typology, integrating work on theprocess of legitimation from institutional theoryand research on judgment formation andchange from social psychology to develop amodel of how legitimacy judgments developand change over time. I conclude with a discus-sion of the implications of this typology andmodel for research on institutional change.

DEFINING LEGITIMACY

While institutional theorists primarily haveexamined the construct of legitimacy in the con-text of the institutionalization of organizationsand organizational fields, social psychologistshave examined the construct primarily in thecontext of group inequality and support for rulesand procedures. In this section I review the waysin which institutional theorists and social psy-chologists have defined and used the term legit-imacy, and I situate my definition of the termwithin the broader nomological networks ofboth fields. In Table 1 I provide an overview ofhow the two fields have defined, used, and spec-ified the content of legitimacy.

Defining Legitimacy: Institutional Theory

Scott explains that institutions “consist of cog-nitive, normative, and regulative structures andactivities that provide stability and meaning tosocial behavior” (1995: 33). Similarly, Greif de-fines an institution as “a system of rules, beliefs,norms and organizations that together generatea regularity of (social) behavior” (2006: 30). Thus,institutions are social conventions that are self-enforcing (Jepperson, 1991; Phillips et al., 2004).

2011 687Tost

In this context, early definitions of organiza-tional legitimacy from institutional theoristsviewed legitimacy as a function of the congru-ence or conformity of an organization to socialnorms or laws (Meyer & Rowan, 1977; Parsons,1956, 1960; Weber, 1978; see Deephouse & Such-man, 2008, for a review of how legitimacy hasbeen conceptualized in organizational institu-tionalism). Meyer and Scott presented a moreextensive definition of organizational legiti-macy as “the degree of cultural support for anorganization—the extent to which the array ofestablished cultural accounts provide expla-nations for its existence, functioning, and ju-risdiction, and lack or deny alternatives” (1983:201). They emphasized this cognitive functionof legitimacy, further arguing that “a com-pletely legitimate organization would be one

about which no question could be raised”(1983: 201). This idea of legitimacy as the pres-ence or absence of questions became a criticalaspect of neoinstitutional theorists’ views oflegitimacy, wherein legitimacy is associatedwith a quality of taken-for-grantedness. Build-ing on these and other previous definitions,Suchman proposed a broad-based and inclu-sive definition of legitimacy as “a generalizedperception or assumption that the actions ofan entity are desirable, proper, or appropriatewithin some socially constructed system ofnorms, values, beliefs, and definitions” (1995:574). Thus, institutional theorists hold that en-tities are judged to be legitimate when theyare seen as appropriate for their social con-text. This is the definition I adopt in this arti-cle.

TABLE 1Overview of the Use of Legitimacy As a Construct in Institutional Theory and Social Psychology

Aspects ofLegitmacy Institutional Theory Social Psychology

Overlaps, Conflicts, andReconciliations

Current Model ofLegitimacy Judgments

Targets oflegitimacyjudgments

Primarily organizationsand organizationalforms

Actors (e.g., leaders),social hierarchies(e.g., status beliefs),group procedures andrules

There is substantialoverlap; researchersin both fields haveexamined targets oflegitimacy outsidetheir traditionalspheres of focus

The model presentedhere could apply tothe approaches andtargets characteristicof both fields (e.g.,judgments of thelegitimacy of leaders,policies,organizations, etc.)

Definition oflegitimacy

“A generalizedperception orassumption that theactions of an entityare desirable, proper,or appropriate withinsome sociallyconstructed system ofnorms, values, beliefs,and definitions”(Suchman, 1995: 574)

“The belief thatauthorities areentitled to be obeyed”(Tyler, 1997: 323), or,alternatively,“Subjectiveperceptions of thefairness or justice ofthe distribution ofsocially distributedoutcomes” (Major &Schmader, 2001: 180)

Voluntary deference isan outcome, ratherthan the substance, oflegitimacy;fairness representsonly one of the threedimensions of thecontent of legitimacyjudgments

Consistent with theinstitutional theoryperspective: thejudgment that anentity is appropriatefor its social context

Content oflegitimacyjudgments

Focus on instrumental(pragmatic) andmoral; also discussionof cognitive andregulative

Instrumental, relational,and moral

Instrumental and moraldimensions overlap;relational dimensionunique to socialpsychology

Institutional, relational,and moral asdimensions of thecontent of legitimacyjudgments; cognitivelegitimacy is viewedas the essence oflegitimacy; regulativelegitimacy representsauthorization and is,thus, a validity cue

688 OctoberAcademy of Management Review

It is critical to distinguish the construct oflegitimacy from the related construct of insti-tutionalization. Institutionalization is both anoutcome and a process (Colyvas & Jonsson,2011). Consistent with the view of institutionsas self-reinforcing and taken-for-granted so-cial conventions, an entity can be said to beinstitutionalized when it obtains both a taken-for-granted status (i.e., a particular variant oflegitimacy, which institutional theorists termcognitive legitimacy) and the capacity tomaintain itself (a capacity distinct from legit-imacy). Thus, legitimacy is a necessary but notsufficient condition for reaching the outcomeof institutionalization. That is, an entity can besaid to be legitimate but not institutionalizedif it has not obtained a capacity to self-reinforce (Jepperson, 1991). In terms of process,therefore, legitimation (the acquisition of le-gitimacy) is only one component of the processof institutionalization.

It is also important to distinguish betweenindividual-level legitimacy and legitimacy atthe collective level. Legitimacy at the collectivelevel is what Weber (1978) termed validity. Asocial order is considered valid, according toWeber’s theory, when two conditions are met: (1)the norms, beliefs, and values that guide thesocial order are perceived as legitimate by somepeople, and (2) even those people who do notperceive the order as legitimate at least knowthat others perceive it as legitimate and under-stand that it governs behaviors. For example,some individuals may not view a particular or-ganizational policy as appropriate, but if othersview it as appropriate and act accordingly, thenthose individuals who do not see it as appropri-ate will perceive that others view it as appropri-ate and will therefore permit it to govern theirbehavior. Therefore, such a policy can be legit-imate at the collective level (i.e., have validity)but may not be viewed as appropriate (i.e., aslegitimate) by all individuals in the group. Inthis way individual-level judgments of legiti-macy can differ from the collective-level validityof an entity. Dornbush and Scott (1975) labeledthis individual-level form of legitimacy propri-ety. In essence, propriety, or individual-level le-gitimacy, refers to an individual’s own judgmentof the extent to which an entity is appropriate forits social context, while validity refers to theextent to which there appears to be a generalconsensus within a collectivity that the entity is

appropriate for its social context. While the is-sue of how individual-level judgments of propri-ety coalesce to constitute collective-level valid-ity is an important consideration, that issue isoutside the focus of this article.

Thus, because the focus of this article is onindividual-level judgments of legitimacy, I fo-cus on propriety and adopt the definition ofindividual-level legitimacy judgments as indi-viduals’ judgments of the extent to which anentity is appropriate for its social context. Inthe following section I explain how legitimacyhas been defined in social psychology, high-lighting areas of divergence relative to insti-tutional theory.

Defining Legitimacy: Social Psychology

Social psychologists have used the constructof legitimacy to explain the stability of and be-havioral reactions to a broad range of socialentities, including individuals (e.g., leaders),group procedures, rules, norms, and social hier-archies. Although sociological psychologistshave adopted definitions of legitimacy that areconsistent with the definitions used by institu-tional theorists (see Johnson, Dowd, & Ridge-way, 2006, for an excellent review of sociologicalpsychologists’ research on legitimacy), in themajority of other work on legitimacy in socialpsychology, researchers have taken a divergentapproach that differs in two key ways from thedefinitions used by institutional theorists andfrom the one I adopt here.

First, in a substantial portion of research inthis area, scholars have defined legitimacy asdeference or obedience to authorities or rules.For example, Tyler defines legitimacy as “thebelief that authorities are entitled to beobeyed” (1997: 323). This approach to legiti-macy derives from French and Raven’s (1959)concept of legitimate power, which refers to aform of power that stems from a subordinate’ssense that an authority is entitled to rule. How-ever, while this type of power—and the feel-ings of obligation to obey that accompany it—may be an outcome of positive legitimacyjudgments, the feelings of desire or obligationto obey or provide support do not themselvesconstitute the legitimacy judgment. In otherwords, the perceptions and beliefs that under-lie the judgment that a leader (to take an ex-ample from this level of analysis) is legitimate

2011 689Tost

produce a perception that the leader is enti-tled to his or her power. This perception ofentitlement to power, in turn, produces a feel-ing of obligation to comply with the leader’srequests. I therefore conceive of the feeling ofobligation to comply with the leader’s requestas an outcome of the legitimacy judgment—not as the content of the judgment itself. Thisis an important distinction, because such afeeling of obligation can come from sourcesother than legitimacy, as when an individualfeels an obligation to comply with a leader’srequest not because he or she views the leaderas legitimate but because noncompliancewould produce negative outcomes for others.

Second, some social psychologists have con-flated the concepts of legitimacy and fairness.For example, Major and Schmader define legit-imacy as “subjective perceptions of the fairnessor justice of the distribution of socially distrib-uted outcomes” (2001: 180). Similarly, Weber,Mummendy, and Waldzus define illegitimacy as“the violation of group entitlements to certainoutcomes or a certain status position” (2002: 451),while Hornsey, Spears, Cremers, and Hogg de-fine illegitimacy as “the degree to which groupsperceive their status relations to conflict withvalues of justice or equity” (2003: 217). This ten-dency to conflate the constructs of legitimacyand justice likely stems from social psychologi-cal research specifying fairness as the key de-terminant of legitimacy judgments (e.g., Tyler,1997; Tyler & Lind, 1992). However, as I explainbelow, fairness is only one dimension of thecontent that underlies individual-level legiti-macy judgments. Because other dimensions oflegitimacy exist, it is critical that scholars dif-ferentiate the construct of legitimacy from theconstruct of fairness.

THE CONTENT OF LEGITIMACY JUDGMENTS

In this section I review social psychologists’and institutional theorists’ research to uncoverthe three dimensions of content underlying indi-vidual-level legitimacy judgments. The contentof legitimacy judgments consists of the substan-tive beliefs and perceptions that influence anindividual’s assessment of the extent to whichan entity is appropriate for its social context. Ibegin with social psychologists’ articulations ofthe instrumental, relational, and moral dimen-sions of legitimacy judgments. I then discuss

research from institutional theory on two ofthose dimensions—instrumental and moral—and I explain why two types of legitimacy com-monly discussed in institutional theory are notincluded within this typology.

Social Psychology: Instrumental, Relational,and Moral Dimensions

Social psychologists have proposed two mod-els to specify the content of legitimacy judg-ments at the individual level. Instrumentalmodels hold that individuals react to the instru-mental aspects of their experiences with socialentities and authorities (e.g., Hollander, 1980;Hollander & Julian, 1970; see also Tyler, 1997). Aninstrumental perspective on legitimacy predictsthat entities will be judged as legitimate whenthey are perceived as promoting the materialinterests of the individual. In contrast, rela-tional models of legitimacy hold that legiti-macy emerges from the extent to which a so-cial entity communicates to the individual thathe or she is accorded respect, dignity, andstatus within the group context and throughgroup membership (Tyler, 1997; Tyler & Lind,1992). From a relational perspective, an entityis seen as legitimate when it affirms individ-uals’ social identities and bolsters their senseof self-worth.

Previous social psychological research on in-strumental and relational models of legitimacyhas proceeded by contrasting the two models todetermine which better explains individuals’behaviors (e.g., Tyler, 1997). The implication ofthis approach, of course, is that the content oflegitimacy judgments is entirely (or at least pri-marily) derived either from instrumental or fromrelational concerns (but not both). Tyler (1997)has conducted the primary work in this area,examining the impact of instrumental and rela-tional concerns on voluntary deference to au-thorities when a conflict emerges betweenauthorities and subordinates. His work demon-strates that in cases of conflict between au-thorities and subordinates, the impact of rela-tional concerns is larger than the impact ofinstrumental concerns, and he therefore con-cludes that the content of legitimacy judg-ments derives from individuals’ identity con-cerns. However, the empirical evidence that hepresents indicates a significant, though some-what smaller, impact of instrumental concerns


as well. Thus, the empirical evidence suggeststhat both instrumental and relational concernshave some degree of impact on individuals’legitimacy judgments.

Based on this observation, I take a differentapproach and conceive of instrumental con-cerns and relational concerns as the bases fortwo separate dimensions of perceptions or be-liefs that underlie the content of legitimacyjudgments. For example, rather than examiningwhether an entity is supported primarily on in-strumental or relational bases, I instead advo-cate examining the independent and interactiveeffects of both bases of legitimacy. Viewing in-strumental and relational concerns as the basesof different dimensions of perception that cansimultaneously impact overall legitimacy judg-ments, rather than as separate models of legit-imacy judgments, permits researchers to con-sider how aspects of the social context orcharacteristics of the evaluators may moderatewhen one or the other comes to dominate in thelegitimacy judgment process. For example, Ty-ler’s (1997) analyses appear to indicate that re-lational concerns dominate legitimacy judg-ments in cases of conflict between supervisorsand subordinates, but there may be a number ofother situations in which instrumental judg-ments would predominate.

As another example, consider Reay and col-leagues’ (2006) account of the introduction of anew work role—nurse practitioner—into an es-tablished health care system in Alberta, Can-ada. Some individuals in the health care systemmay have viewed the new work role as legiti-mate, feeling that the new role promoted or pro-tected instrumental needs at either the individ-ual or group level. They may have perceivedthat the integration of nurse practitioners intothe system would benefit them personally (per-haps by increasing the chance they would findemployment in the nurse practitioner role—thatis, individual-level instrumental concerns), orthey may have believed that the change wouldpromote the organizational goal of more effec-tive and efficient health care provision (i.e.,group-level instrumental concerns). At the sametime, individuals may have viewed the changeas legitimate because the new work role pro-moted or protected relational needs at the indi-vidual or group level. For example, individualnurse practitioners may have felt personallyvalidated by being granted new status within

the health care system, and nurse practitionersin general may have felt that their social iden-tity as a group was gaining in status and re-spect as well.

However, these two dimensions of legitimacyjudgments are not mutually exclusive. For ex-ample, an individual may view an entity as le-gitimate on both instrumental and relationalgrounds. Alternatively, an entity may be viewedas legitimate from an instrumental standpointand as illegitimate from a relational standpoint.Thus, a given entity may be viewed as legiti-mate on one ground, both grounds, or neitherground. Viewing instrumental and relational di-mensions not as separate models of legitimacybut instead as separate bases of legitimacy per-mits researchers to consider the circumstancesunder which one or another basis of legitimacywill have greater or lesser influence on the over-all legitimacy judgment and, consequently, thelargest impact on behavior.

While the majority of their previous researchon legitimacy has focused on examining the rel-ative explanatory power of the instrumental andrelational models, social psychologists have re-cently begun to espouse a moral dimension tolegitimacy as well. Skitka, Bauman, and Lytle(2009) demonstrated that individuals’ degrees ofmoral conviction about an issue on which theSupreme Court had recently ruled predicted per-ceptions about the Supreme Court’s legitimacy.In addition, social psychologists have arguedthat morality is an important general dimensionof evaluation of social entities (e.g., Leach, Elle-mers, & Barreto, 2007). Leach and colleagues ar-gued that while both moral and relational con-cerns can be viewed as consistent with a singleconcept of benevolence, the two types of con-cerns are conceptually distinct. In a series ofstudies they demonstrated that instrumental, re-lational, and moral concerns constitute distinctfactors of evaluation and that in many circum-stances morality concerns are actually more im-portant in evaluations than are instrumentaland relational concerns.

Thus, research on the social psychology oflegitimacy judgments points to three dimen-sions underlying legitimacy: instrumental, rela-tional, and moral. In the next section I discussthe dimensionality of legitimacy from the per-spective of institutional theory.

2011 691Tost

Institutional Theory: Instrumental, Moral,Cognitive, and Regulative

Institutional theorists also recognize an in-strumental dimension to legitimacy. This basisof legitimacy has been termed pragmatic legit-imacy (Suchman, 1995) and is viewed as rootedin the self-interested calculations of individualsand groups. In this view, individuals or constit-uencies may support an entity because its con-tinued existence entails a higher expectedvalue than its absence, or because the entity isseen as being responsive to the individual orconstituency’s larger interests.

While institutional theorists have not ex-plored the relational dimension of legitimacy, amoral dimension has been studied extensively.Indeed, as Scott (2001) explains, the predomi-nant view of the substantive content of legiti-macy among sociologists, including many insti-tutional theorists, is one in which the primarydeterminant of legitimacy is the moral status ofthe entity, or the extent to which the entity con-forms to moral values and ethical principles.Suchman describes the moral dimension of le-gitimacy as grounded in “a prosocial logic” andconcerned with whether the entity in questionpromotes “social welfare, as defined by the au-dience’s socially constructed value system”(1995: 579). In this way the basis of moral legiti-macy differs fundamentally from the self-interested orientation involved in the instru-mental dimension.

Thus, there is considerable overlap betweensocial psychologists’ views of the instrumentaland moral bases of legitimacy and institutionaltheorists’ views of pragmatic and moral legiti-macy. Institutional theorists have also identifiedtwo other types of legitimacy that merit consid-eration in this discussion: cognitive legitimacyand regulative legitimacy. I argue that thesetwo constructs do not constitute substantivebases of the content of legitimacy judgments.Instead, cognitive legitimacy represents the ab-sence of substantive content in the legitimacyjudgment (i.e., taken-for-grantedness), whereasregulative legitimacy represents social cues in-dicating the validity of an entity (i.e., indicatorsof collective-level legitimacy) but does not rep-resent a substantive domain of judgment con-tent in itself.

The construct of cognitive legitimacy is basedon the early neoinstitutionalist definitions of le-

gitimacy involving the absence of questions orchallenges regarding an entity. Thus, Suchman(1995) explains that cognitive legitimacy is fun-damentally different from moral and pragmaticlegitimacy. Specifically, he contends that legit-imacy can entail either active or passive sup-port: “Legitimacy may involve either affirmativebacking for an organization or mere acceptanceof the organization as necessary or inevitablebased on some taken-for-granted cultural ac-count” (1995: 582). Whereas the instrumental, re-lational, and moral dimensions of legitimacyinvolve active affirmative backing on the basisof instrumental, relational, and moral concerns,respectively, cognitive legitimacy entails pas-sive support. Cognitive legitimacy is the ab-sence of questions about or challenges to anentity. In the absence of such questions or chal-lenges, there is no need for affirmative backing.Indeed, the provision of any affirmative instru-mental, relational, or moral account for an entitywith a high level of cognitive legitimacy maybackfire (Ashforth & Gibbs, 1990). As Suchmanexplains:

Both pragmatic and moral legitimacy rest ondiscursive evaluation, whereas cognitive legit-imacy does not: Audiences arrive at cost-benefit appraisals and ethical judgmentslargely through explicit public discussion, andorganizations often can win pragmatic andmoral legitimacy by participating vigorously insuch dialogues; in contrast, cognitive legitima-tion implicates unspoken orienting assump-tions, and heated defenses of organizationalendeavors tend to imperil the objectivity andexteriority of such taken-for-granted schemata(1995: 585).

Thus, cognitive legitimacy does not representa dimension of the substantive content of le-gitimacy judgments. Instead, it represents theabsence of content. Indeed, this absence ofcontent is its power: “for things to be otherwisebecomes literally unthinkable” (Zucker, 1983:25). Organizations (or other social entities forthat matter) with a high level of cognitive le-gitimacy require no justification, so there is noneed for content to underlie a justification. Forthis reason Suchman contends that the taken-for-granted nature of cognitive legitimacy“represents the most subtle and powerfulsource of legitimacy identified to date” (1995:583).

This description may seem to imply that cog-nitive legitimacy only applies to fully institu-


tionalized entities (Henisz & Zelner, 2005). How-ever, a key insight of institutional theory is thatisomorphism legitimates (Deephouse, 1996). Inother words, to the extent that a new entity con-forms to the expectations carved by existing in-stitutions, that new entity is not subjected toactive evaluations but, instead, is passively ac-cepted and unquestioned. In this way the powerof cognitive legitimacy can be applied to theemergence of the rules, norms, and organiza-tions that Greif (2006) calls “institutional ele-ments.” For example, Glynn and Abzug (2002)demonstrate that organizations that adoptnames that conform to the conventional struc-tures and styles of names within their institu-tional field benefit from greater legitimacy. Sim-ilarly, Deephouse (1996) demonstrates thatisomorphism in commercial banks is positivelyrelated to legitimacy. This phenomenon mayalso apply to emergent institutions. Henisz andZelner describe emergent institutions as institu-tions that are newly created and, thus, “stillsubject to evaluation” (2005: 363). However, to theextent that an emergent institution can be con-structed in such a way as to be compatible withexisting institutions, the evaluation (and im-plied challenge) that Henisz and Zelner refer tois less likely to occur, and so the emergent in-stitution can begin to accrue the immunity toquestioning that is cognitive legitimacy. In es-sence, isomporphism legitimates because itleads to the absence of questions or challengesand thereby holds substantive, content-basedevaluation at bay. Thus, while cognitive legiti-macy is typically conceived of as a property offully institutionalized (i.e., self-reproducing andtaken-for-granted) entities, emergent organiza-tions, institutions, and institutional elementscan tap into the power of cognitive legitimacyby conforming to cultural expectations andnorms.

Regulative legitimacy also emphasizes con-formity, but rather than conformity or congru-ence with cultural expectations (as in the case ofcognitive legitimacy), regulative legitimacyemerges from conformity with law or other formsof collective regulation (Greenwood et al., 2002;Greve, 2005; Scott, 1995). In this sense regulativelegitimacy is highly related to cognitive legiti-macy: organizations (the key target of consider-ation in research on regulative legitimacy) areexpected to conform to regulations, and failureto do so raises questions about the nature of the

organization that the organization would haveotherwise avoided. However, regulative legiti-macy is also distinct from cognitive legitimacybecause regulative legitimacy involves an ac-tive external validation of the organization bysome agent (e.g., a government agency or a pro-fessional association).

Validity, as explained above, refers not to in-dividual-level judgments of legitimacy but tocollective-level legitimacy. Researchers haveidentified two types of social cues that canemerge regarding the validity of a social entity:endorsement of the entity by peers and authori-zation of the entity by authorities (Dornbush &Scott, 1975). Regulative legitimacy represents aform of authorization. Authorization does not,however, establish a particular basis on whichto judge an entity to be legitimate; rather, itmerely provides evidence that others havejudged it to be legitimate. Importantly, themeaning of regulative legitimacy can varyacross contexts. In a democracy, authorizationin the form of formal laws supporting an insti-tutional arrangement indicates fairly broad-based endorsement of that arrangement. Thesame would not be the case in a dictatorship.Thus, while cognitive legitimacy is excludedfrom the typology of the content of legitimacyjudgments because cognitive legitimacy repre-sents the absence of substantive judgment con-tent, regulative legitimacy is excluded becauseit represents a special case of the use of others’evaluations as heuristic substitutes for individ-ual-level evaluation.

Summary

In summary, I argue that there are three di-mensions of content underlying legitimacy judg-ments: instrumental, relational, and moral. Spe-cifically, an entity is viewed as legitimate oninstrumental grounds when it is perceived tofacilitate the individual’s or group’s attempts toreach self-defined or internalized goals or out-comes. Examples of perceptions or beliefs thatconstitute the content of the instrumental di-mension of legitimacy judgments include per-ceptions or beliefs related to the effectiveness,efficiency, or utility of the entity. Second, anentity is viewed as legitimate on relationalgrounds when it is perceived to affirm thesocial identity and self-worth of individuals orsocial groups and to ensure that individuals or

2011 693Tost

groups are treated with dignity and respectand receive outcomes commensurate withtheir entitlement. Examples of perceptions orbeliefs that constitute the content of the rela-tional dimension of legitimacy judgments in-clude perceptions or beliefs related to the fair-ness, benevolence, or communality thatcharacterizes the entity. Finally, an entity isperceived as legitimate on moral groundswhen it is perceived to be consistent with theevaluator’s moral and ethical values. Thus,examples of perceptions or beliefs that consti-tute the content of the moral dimension of le-gitimacy judgments include perceptions or be-liefs related to the morality, ethicality, orintegrity of an entity.

It is important to highlight that these threedimensions are not mutually exclusive; entitiesmay be evaluated simultaneously on all threedimensions or on some subset of the dimen-sions. In addition, the three domains may alsooverlap—that is, the specific beliefs and percep-tions that underlie any given legitimacy judg-ment may fall into one or more categories. Forexample, the observation that a particular insti-tutional practice is highly efficient would cer-tainly fall into the instrumental dimension.However, in the context of an organizational cul-ture that places a high value on efficiency (e.g.,Wal-Mart), that observation may fall into themoral dimension as well (i.e., the observationthat a practice is efficient may constitute bothinstrumental and moral grounds for maintain-ing it). In other types of organizational or groupcultures, there may be a significant overlap be-tween the relational and moral dimensions.Similarly, in groups that do not have a strongculture or value system and do not recognizerelational practices as promoting efficiency,there may be relatively little overlap across thethree dimensions. Thus, the degree of overlapamong the three dimensions may be moderatedby a number of variables, such as group or or-ganizational culture and individual value orien-tations.

In the next section I build on this typology toconstruct a model of the legitimacy judgmentprocess, and in doing so I explain that the na-ture of the relationships between the three di-mensions of legitimacy judgments and an over-all legitimacy judgment depends on the stage ofthe legitimacy judgment process.

THE LEGITIMACY JUDGMENT PROCESS

While institutional scholars view legitimacyas the key driver of institutional change, verylittle research has examined how individual-level legitimacy judgments develop and changeover time. Institutional theorists have recentlypaid increasing attention to microprocesses ofinstitutional change (e.g., Phillips et al., 2004;Reay et al., 2006; Suddaby & Greenwood, 2005;Zilber, 2002). However, much of the research inthis area has focused not on individual-leveljudgments of legitimacy but, rather, on how in-teractions among individuals constitute socialreality and guide institutional change (e.g.,Reay et al., 2006; Zilber, 2002; Zucker, 1977) orhow rhetoric and discourse are used as tools ofinfluence in the process of institutional change(e.g., Phillips et al., 2004; Suddaby & Greenwood,2005; Vaara, Tienari, & Laurila, 2006). In otherwords, institutional theorists have not examinedhow individuals come to judge existing institu-tional arrangements as legitimate or illegiti-mate or how those judgments emerge to moti-vate individuals to work for change or maintainthe status quo.

In this section I build on research from insti-tutional theory and social psychology to developa model of the legitimacy judgment process thatcan speak to these issues and, therefore, cancontribute to research on the microprocesses ofinstitutional change (e.g., by helping to specifythe circumstances under which individuals be-come motivated to engage in activities forchange, as well as the forms and content ofrhetoric that are likely to resonate with differentaudiences). Specifically, I argue that the legiti-macy judgment process is a three-stage cyclicalprocess that is characterized by two judgmentstages (judgment formation and judgment reas-sessment) and an additional stage in which thejudgment is used (see Figures 1 and 2). In thejudgment formation stage an initial legitimacy

FIGURE 1The Legitimacy Judgment Cycle

Judgmentuse

Judgmentformation

Judgmentreassessment


judgment is formed either through an evaluativeor passive judgment mode. In the use stage theexisting judgment is deployed to guide behaviorand is bolstered through processes of affirma-tion and cognitive assimilation. The use stagemay continue in perpetuity, or instead individu-als may engage or reengage in the evaluativemode in the judgment reassessment stage. Inthe sections below I explain each of thesestages in more detail.

Judgment Formation

In the judgment formation stage individualsengage in either an evaluative or passive modeof information processing, which leads to a gen-eralized legitimacy judgment that representsthe entity as either appropriate (i.e., legitimate)or inappropriate (i.e., illegitimate) for its socialcontext. The two modes of the legitimacy judg-ment process differ with respect to the sources ofinformation used, the extent of cognitive effortemployed, and the effects on the generalizedlegitimacy judgment reached.

Evaluative mode. In the evaluative modejudgments of the overall legitimacy of an entityare constructed on the basis of evaluations ofthe entity along instrumental, relational, and/ormoral dimensions. In addition, the evaluativemode involves effortful attempts at judgmentcreation. In this mode the individual is activelymotivated to construct an evaluation of the en-

tity. It is important to note, however, that thiseffortful nature characterizing the evaluativemode does not mean that the evaluative mode isimmune to cognitive biases. Indeed, extensiveresearch documents that individuals can pro-duce biased judgments despite a motivation foraccuracy (Gilovich, Griffin, & Kahneman, 2002;Kahneman, Slovic, & Tversky, 1982; Kahneman &Tversky, 2000). The effortful nature of the evalu-ative mode does not imply that judgmentsare not biased; rather, it simply indicates thatindividuals are actively engaged in a consciousattempt to construct a judgment. Thus, in theevaluative mode of the legitimacy judgmentprocess, instrumental, relational, and moralevaluations drive judgments of generalized le-gitimacy.

Consequently, to understand the dynamics ofthe evaluative mode, it is necessary to under-stand the circumstances under which one or theother of the three dimensions is likely to beprioritized in the judgment process. Previous so-cial psychological research has demonstratedthat the relative prioritization of the three di-mensions is driven at least in part by evalua-tors’ social identification with the group that isassociated with the entity under evaluation. So-cial identification with the group refers to theextent to which group members form their iden-tities around their group membership and inte-grate the group into their self-concepts (Tajfel &Turner, 1979). In his research Tyler (1997) found

FIGURE 2Model of the Legitimacy Judgment Process

Validitycues

Instrumentalevaluations

Instrumentalevaluations

Relationalevaluations

Relationalevaluations

Moralevaluations

Moralevaluations

Mode:passive vs.evaluative

Generalizedlegitimacy

Generalizedlegitimacy

Extrinsic vs.intrinsic

orientation togroup

Judgment formation and reassessment stages

+

++

+

+

–

–+

–

–

–

Use stage

Support for(or resistance to)

change

+ +

+

++

+

+

2011 695Tost

that social identification with the group leads toa greater prioritization of the relational dimen-sion of legitimacy judgments. Specifically, indi-viduals who draw more heavily on the group fortheir personal sense of identity and who havecloser bonds with the group tend to placegreater emphasis on relational concerns in de-termining their evaluations of the legitimacy ofauthorities and group policies.

This effect likely emerges because individu-als having high levels of social identificationwith the group tend to have an intrinsic orienta-tion to the group such that engagement with thegroup is a source of identity, joy, and meaning(Amabile, Hill, Hennessey, & Tighe, 1994). Forthese individuals, the relational and moral le-gitimacy of group policies and practices arehighly personally relevant—these policies andpractices construct central aspects of their self-concept and personal meaning. Relative to thedeeply meaningful implications of the rela-tional and moral status of group-based entitiesfor these individuals, instrumental concerns arelikely to be less critical. In contrast, individualswho have a low level of identification with thegroup tend to have an extrinsic orientation tothe group, engaging in group activities not be-cause the group functions as a source of iden-tity, joy, or meaning but because the group canprovide valued outcomes (Amabile et al., 1994).For these individuals, the relational and moralstatus of group-based policies and practices areonly meaningful to the extent that these consid-erations impact the personal outcomes the indi-vidual desires (i.e., to the extent that these con-siderations are instrumental to the individual’spersonal interests). Therefore, individuals withan extrinsic orientation to the group (e.g., indi-viduals with low group identification) are likelyto place primary emphasis on the instrumentaldimension and very little emphasis on the rela-tional and moral dimensions in determiningtheir generalized legitimacy judgments in theevaluative mode.

Passive mode. In the passive mode, however,rather than engage in effortful information pro-cessing, individuals either use validity cues ascognitive shortcuts to reach a legitimacy judg-ment or passively assume the legitimacy of en-tities that conform to cultural expectations (orsome combination of the two). In the former pro-cess individuals observe authorizations or en-dorsements from others and base their own

judgments entirely on those observations ratherthan on their own evaluations of the instrumen-tal, relational, and moral status of the entity. Inthe latter process individuals simply passivelyaccept entities that conform to their expecta-tions. This latter process is consistent with thediscussion of how new entities can tap into thepower of cognitive legitimacy by merely con-forming to cultural expectations. As Johnson andcolleagues explain, sometimes legitimacy maybe acquired “simply by not being implicitly orexplicitly challenged” (2006: 60). Thus, in thepassive mode of the legitimacy judgment pro-cess, validity cues and/or mere acceptancedrives judgments of generalized legitimacy.

Relationship between the two modes. Thesetwo modes correspond to modes of reasoningidentified in a wide range of areas of socialpsychological research. Dual-process models insocial psychology generally distinguish be-tween two modes of cognitive operations: onethat is effortful, controlled, and self-aware—thatis, an evaluative mode—and another that is ef-fortless, automatic, and quick—that is, a passivemode (e.g., Chaiken & Trope, 1999; Kahneman &Frederick, 2002). Social psychological researchfurther suggests that because individuals tendto approach judgment tasks in ways that con-serve cognitive energy, the passive mode pre-dominates unless it becomes necessary or desir-able for the evaluative mode to intervene(Gilbert, 2002; Kahneman & Frederick, 2002;Lieberman, 2003).

Thus, the passive mode is likely to predomi-nate in the judgment formation stage unless theindividual deems it necessary or desirable touse more effort in the judgment process. Thereare therefore two factors likely to impact whichof the two modes predominates in the judgmentformation stage: the availability of validity cuesand the extent to which the entity that is thetarget of judgment conforms to cultural expecta-tions. If validity cues are unavailable, as may bethe case for a newly proposed entity, the evalu-ative mode is more likely to be engaged. Inaddition, if the entity conflicts with some aspectof the individual’s culturally based expecta-tions, the evaluative mode is more likely to beengaged. Hence, in most circumstances (i.e.,when validity cues are available or when thereis a high level of congruence with existing insti-tutional arrangements), the passive mode pre-dominates.


An important implication of the predominanceof the passive mode in most circumstances ofinitial judgment formation is that initial judg-ments are likely to be biased in a positive direc-tion. Specifically, Ridgeway and Berger (1986)argue that when peers or authorities merely actas though an entity is legitimate, their behavioris sufficient to produce validity cues. This dy-namic is referred to as weak validation becausemere behavioral compliance with the dictates ofan institutional arrangement implies collective-level legitimacy, regardless of the actual indi-vidual-level legitimacy judgments of the indi-viduals engaging in compliance. The power ofweak validation in the form of mere behavioralcompliance means that when individuals pre-suppose a consensus that an entity is legiti-mate, and they then act on that supposition,their act itself functions as confirmation of theirpresupposition for other actors, and the fact thatother actors do nothing to oppose the initial ac-tion provides further confirmation. This dynamiccan produce a strong bias in favor of positiveperceptions of validity, which, in turn, have apositive impact on individual-level generalizedlegitimacy judgments.

Regardless of whether the passive mode orevaluative mode predominates in the judgmentformation stage, the outcome of the judgmentformation stage is a generalized legitimacyjudgment that represents the entity as legiti-mate or illegitimate to some degree. Once sucha generalized legitimacy judgment is estab-lished, the individual moves into the use stageof the legitimacy judgment process.

Use Stage

The generalized legitimacy judgment formedin the judgment formation stage is carried overinto the use stage, where it guides behavior withrespect to the entity. In the use stage legitimacyjudgments come to function as pivotal cogni-tions (Lind, 2001) that can move people betweentwo very different types of behaviors: on the onehand, to the extent that an entity is viewed aslegitimate, it is supported, and attempts tochange it are resisted; on the other hand, to theextent that an entity is viewed as illegitimate,people actively seek to change it.

Thus, in the use stage the entity is no longerjudged; instead, the existing judgment is de-ployed. Consequently, in the use phase cogni-

tive energy is no longer geared toward judg-ment formation and is instead focused onassimilating incoming bits of information andstimuli to conform to the initial generalized le-gitimacy judgment in a process characterized bymotivated reasoning (Kunda, 1990). In this waythe generalized legitimacy judgment thatemerges from the judgment stage acts as ananchor that guides interpretations of new legit-imacy-relevant experiences such that new infor-mation is viewed as consistent with the existinggeneralized legitimacy judgment.

The assimilation process occurs for two rea-sons (Tost & Lind, 2010). First, this process helpsindividuals manage the uncertainty associatedwith their social worlds (Lind, 2001; Tost & Lind,2010). If each new legitimacy-related experiencerequired individuals to evaluate anew their ex-isting legitimacy judgments, the legitimacy ofmultitudes of social entities would constantly becalled into question. Using the initial general-ized legitimacy judgment as an anchor to guidethe interpretation of new information ensuresthat this type of potentially incapacitating am-bivalence and uncertainty is minimized. Thesecond reason for the assimilation process isrelated to the first. Specifically, assimilationminimizes the cognitive energy that must beallocated to legitimacy judgments (Lind, 2001;Tost & Lind, 2010). If assessing the legitimacy ofthe social environment required individuals toconstantly monitor their environments for evi-dence of illegitimacy, very little could be accom-plished. Such a high level of monitoring wouldrequire too much attention and cognitive energyand would leave individuals unable to engagein other judgment tasks or activities. Thus, theassimilation process characterizing the usestage reduces the cognitive resources that arenecessary for individuals to assess their secu-rity within their social environments.

There are three important implications emerg-ing from the nature of the assimilation that oc-curs in the use stage. First, to the extent thatsome form of substantive content (i.e., instru-mental, relational, or moral concerns) was usedto justify the initial generalized legitimacy judg-ment (i.e., if the evaluative mode was employedin the judgment formation stage), this contentdissipates in the use stage. The content dissi-pates because it is no longer needed—the entityis no longer a target of evaluation along instru-mental, relational, or moral dimensions. In-

2011 697Tost

stead, it is simply viewed as legitimate (or ille-gitimate, depending on the outcome of thejudgment formation stage), and new informationrelevant to its instrumental, relational, or moralstatus is interpreted in a manner to be consis-tent with the generalized legitimacy judgment.As Zilber explains, once institutionalization oc-curs, the relevant social entities “acquire arealitylike status, and their social origin is for-gotten” (2002: 234).

Thus, as the content dissipates, cognitive le-gitimacy emerges. The idea that more substan-tive justifications precede the development ofcognitive legitimacy is consistent with numer-ous models of and empirical findings on theprocess of institutionalization. For example, intheir model of the stages of institutional change,Greenwood and colleagues (2002) indicate thatpragmatic and moral legitimacy are assessedbefore cognitive legitimacy emerges. Similarly,Colyvas (2007) demonstrates how substantive le-gitimacy preceded cognitive legitimacy in theinstitutionalization of technology transfer atStanford University (see also Baum & Powell,1995, and Hoffman, 1999).

As another example, consider an individualwho engages in the evaluative processing modein the judgment formation stage. The individualmay conclude that an entity is legitimate oninstrumental but not on moral grounds and mayfurther conclude that the instrumental dimen-sion is the most important for consideration inthe present circumstance (e.g., the individualmay have an extrinsic orientation to the group).In this case the generalized legitimacy judg-ment will be positive. For some time after theestablishment of the judgment, the individualmay recall the moral qualms he or she initiallyheld about the entity such that those initialmoral evaluations continue to have an impacton subsequent moral evaluations, regardless ofthe nature of the generalized legitimacy judg-ment. However, as time passes, the process ofassimilation will bias and neutralize that recol-lection such that the negative moral implica-tions of the entity will be consistently and pro-gressively minimized (as in the process ofethical fading described by Tenbrunsel & Mes-sick, 2004) and the entity will come to be viewedmore favorably on the moral dimension. Thus,over time, the substantive grounds of the origi-nal judgment will become blurred and replacedby a noncritical acceptance of the entity. In this

way the assimilation process that characterizesthe use stage has the effect of nullifying or neu-tralizing the initial instrumental, relational, andmoral evaluations (or validity cues) that pro-duced the generalized legitimacy judgment inthe first place.

This example also points to a second impor-tant implication of the process of assimilation:assimilation reverses the causal direction be-tween the generalized legitimacy judgment andthe three dimensions of legitimacy judgments.In contrast to the evaluative mode of the judg-ment stages, in which instrumental, relational,and moral evaluations produce a generalizedlegitimacy judgment, in the use stage the gen-eralized legitimacy judgment produces the in-strumental, relational, and moral evaluations ofthe entity. In other words, the generalized legit-imacy judgments established in the judgmentformation stage can actually influence passiveperceptions of the entity with respect to instru-mental, relational, and moral considerationsduring the use stage.

This contention that the cognitive legitimacythat emerges in the use stage has a causal im-pact on the dimensions of evaluation is con-sistent with extensive research in social psy-chology. For example, research on systemjustification theory (for reviews see Jost, Banaji,& Nosek, 2004; Jost, Glaser, Kruglanski, & Sullo-way, 2003; Jost & Hunyady, 2002, 2005) shows thatindividuals have an innate motivation to viewhighly institutionalized entities as fair and just.Furthermore, researchers in this area haveshown that individuals rationalize institutional-ized entities by subjectively enhancing theirperceptions of the desirability of current institu-tional arrangements and outcomes (e.g., Kay etal., 2009; Kay, Jimenez, & Jost, 2002).

Thus, the stage and mode of the legitimacyjudgment process determine the causal direc-tion of the relationship between the generalizedlegitimacy judgment and the three domains ofcontent. In the judgment stages (judgment for-mation and judgment reassessment), when theevaluative mode is employed, the instrumental,relational, and moral evaluations of the entitydetermine the generalized legitimacy judgment.If the passive mode is used, then validity cuesdetermine the generalized legitimacy judgment.However, once the individual enters the usestage, this dynamic reverses. In the use stagethe generalized legitimacy judgment that was


formed in the judgment stage biases evalua-tions of the entity along the three dimensionsand also colors perceptions of the extent of theentity’s validity. The implications of this rever-sal of causal direction are not substantial incircumstances in which a judgment formed inthe evaluative mode did not involve conflictsacross dimensions. That is, if evaluations acrossall three dimensions point consistently to eitherpositive or negative evaluations, there is littlesubstantive consequence of the reverse ofcausal directions. However, if there is conflictacross the dimensions in the evaluative mode,as when the entity is viewed as legitimate oninstrumental but not on moral grounds, then thereversal of causal direction that occurs in theuse stage can result in a shift in perceptionsover time, leading the individual to view asmoral what was previously viewed as immoral(or to view as instrumental what was previouslyviewed as noninstrumental).

The third implication of the assimilation pro-cess is that the initial legitimacy judgment willbe perpetuated throughout the use stage. Thisimplication is particularly important given thepositive bias that characterizes the judgmentformation stage. This persistence of initial pos-itive judgments is a dynamic familiar to socialpsychologists (e.g., Klauer & Stern, 1992). Indeed,much of the social psychological research rele-vant to institutional change has examined thetopic of change indirectly, by exploring its ab-sence. Support for the status quo is a key depen-dent variable in research on system justificationtheory (Jost et al., 2004; Jost et al., 2003; Jost &Hunyady, 2002, 2005) and the just world hypoth-esis (which holds that individuals are motivatedto perceive their social environments as charac-terized by a high degree of fairness or justice;see Lerner, 1980), as well as in many studiesexamining the denial of injustice (Crosby, 1984)and victim derogation (e.g., Kay, Jost, & Young,2005). The majority of these theories and ap-proaches share the notion that individuals tendto resist viewing their social systems as illegit-imate and, thus, tend to support the status quo.Findings in this area of research provide sup-port for the contention that individuals gener-ally exhibit positive legitimacy judgments of ex-isting institutional arrangements and resist theperception that existing institutions and socialsystems are lacking in legitimacy.

However, it is clear that there are circum-stances in which individuals do come to viewexisting social entities as illegitimate—circum-stances where individuals do not resist changeand instead desire and promote institutionalchange. Thus, any theory of the legitimacy judg-ment process must not only account for the ten-dency to justify existing social entities but alsospecify the circumstances that mitigate this ten-dency and lead instead to a more critical con-sideration of the legitimacy of existing institu-tions and social arrangements. In the followingsection I describe the judgment reassessmentstage of the legitimacy judgment process andthen address a critical question in the study ofinstitutional change: Once an individual has en-tered the use stage, what are the factors that canmotivate the individual to reconsider the legiti-macy of a social entity?

Judgment Reassessment Stage

While individuals use motivated reasoning inthe form of assimilation processes to bolster ini-tial legitimacy judgments in the use stage, thejudgment reassessment stage is dominated by amotivation to make effortful and considered per-sonal assessments of the legitimacy of the en-tity. This motivation to engage in effortful as-sessments does not imply that individuals in thejudgment reassessment stage are free of percep-tual biases or are able to be more objective intheir judgments. Instead, they are simply moremotivated to engage in the process of judgmentformation. In addition, entering the judgmentreassessment stage does not necessarily meanthat the judgment itself will ultimately be re-vised; an individual may reassess the judgmentand deem that it does not require revision. Thekey characteristic of the judgment reassessmentstage is that the individual is motivated to ac-tively reconsider the existing legitimacy judg-ment. Thus, in the reassessment stage the eval-uative mode predominates, and individualsengage in active attempts to evaluate the entityalong the dimensions of instrumental, rela-tional, and/or moral legitimacy, which onceagain drive judgments of generalized legiti-macy. Individuals may also incorporate validitycues into consideration in the evaluative pro-cess in the reassessment stage, but the primaryemphasis is on their own assessments of theinstrumental, relational, and moral status of the

2011 699Tost

entity because the motive to form a personaljudgment becomes paramount.

As the individual engages in effortful consid-eration of the instrumental, relational, andmoral legitimacy of a social entity, he or shecreates a new generalized legitimacy judgment.Once a new generalized legitimacy judgment isformed, the individual reenters the use stage,where the newly formed generalized legitimacyjudgment will again function as a pivotal cog-nition, guiding behavior with respect to the en-tity, and as a heuristic, influencing the interpre-tation of additional information related to theentity. Moreover, because the evaluative mode(rather than the passive mode) predominates inthe judgment reassessment stage, the positivebias that characterizes the judgment formationstage is not present. Consequently, it is at thisstage in the legitimacy judgment process thatjudgments of illegitimacy (and, hence, supportfor change) are most likely to emerge. A criticalissue, therefore, is to determine what leads in-dividuals to transition from the use stage intothe judgment reassessment stage. I tackle thisissue in the following section.

Switching from the Use Stage to the JudgmentReassessment Stage

The process of switching from the use stage tothe judgment reassessment stage taps into fun-damental issues in both institutional theory andsocial psychology. Specifically, institutionaltheorists grapple with an issue that has beentermed the paradox of embedded agency (Batti-lana, Leca, & Boxenbaum, 2009; Seo & Creed,2002), which refers to the tension between theidea of individuals as active shapers of theirinstitutional environments and the view of indi-viduals’ behaviors as determined by the institu-tions in which they are embedded. How canindividuals change institutions “if their actions,intentions, and rationality are all conditioned bythe very institution they wish to change” (Holm,1995: 398)? In other words, how can an individualextract him or herself from the grips of cognitivelegitimacy? The paradox, thus, is between indi-vidual agency and institutional determinism,and part of the challenge is to determine whatmakes cognitive legitimacy erode at the individ-ual level. If cognitive legitimacy is the absenceof questions, what is it that leads individuals tobegin to actively interrogate an existing social

entity and to imagine possible alternatives? Toput the issue in the framework of the model ofthe legitimacy judgment process presentedhere, what are the factors that lead people tomove out of the use stage, where institutionalarrangements are passively accepted, and intothe judgment reassessment stage, where insti-tutional arrangements are actively interro-gated?

At the same time, a critical concern in socialpsychological research is identifying the cir-cumstances under which individuals will en-gage in effortful and reflective information pro-cessing rather than conserve cognitive energyand resources. Recent research in the area ofsocial cognitive neuroscience provides an inter-esting response to this issue. Specifically, thisresearch has identified a “neural alarm system”that appears to switch individuals between pas-sive and active judgment processes (Eisen-berger & Lieberman, 2004; Eisenberger, Lie-berman, & Williams, 2003; Lieberman &Eisenberger, 2004; Ullsperger, Volz, & Von Cra-mon, 2004). Research on the activity of this neu-ral alarm system indicates that the system isactivated when the potential for errors in judg-ments or outcomes is perceived to be high(Carter et al., 1998, 2000). For example, the Strooptask, which requires individuals to identify thecolor of ink in which a word is written, althoughthe word itself specifies a different color (e.g.,the word “red” presented in green ink), has beenshown to activate the neural alarm system(Lieberman, 2007). Thus, social psychological re-search suggests that individuals will move fromthe use stage to the judgment reassessmentstage when this mental alarm is activated.

Integrating insights from social psychologyand institutional theory, I argue that a mentalalarm is triggered when individuals detect andthen examine questions that can be raisedabout existing social entities. Institutional the-orists have posited three sources of such ques-tions: jolts, contradictions, and reflexivity.

Jolts. Institutional theorists have argued thatmajor events, such as technological changes,social upheaval, actions of competitors, or reg-ulatory changes, can act as jolts to the institu-tional field, disturbing the functioning of thefield and thereby prompting consideration of thepotential for institutional change (Battilana etal., 2009; Greenwood et al., 2002). In disruptingthe functioning of the institutional field, this


type of event produces a violation of the expec-tations that are based on the generalized legit-imacy judgment—social entities are no longerable to function as they did before the jolt. Thisviolation of expectations alerts individuals thattheir current judgments about existing entitiesare no longer reliable, and (to situate this dy-namic in the context of the present discussion)the mental alarm is triggered.

Institutional theorists, however, have said rel-atively little about the necessary features thatmust characterize an event in order for thatevent to constitute a jolt. I argue that for anevent to act as a jolt triggering the mentalalarm, the event must provide new informationor outcomes that preclude assimilation into ex-isting expectations. In particular, this event ornew piece of information must be sufficientlyoutside the realm of expectation that it cannotbe assimilated into existing legitimacy judg-ments without active and effortful considerationof the challenges the jolt presents. In this waythe jolt disrupts the assimilation process, pre-venting it from proceeding, and thereby acti-vates the mental alarm. Consequently, minorexpectation violations are not likely to triggerthe mental alarm. The contention that violationsof expectations can lead people to question theirexisting legitimacy judgments is also consistentwith previous social psychological researchdemonstrating that violations of expectationsproduce discontent with existing social entities(e.g., Rasinski, Tyler, & Fridkin, 1985).

Consistent with the notion that expectationviolation is necessary for the activation of themental alarm, I expect that the mental alarm ismore likely to be activated when the valence ofunexpected events suggests that a switch be-tween positive and negative legitimacy judg-ments may be in order. I view legitimacy as acontinuous variable, with values above a neu-tral point constituting positive legitimacy judg-ments and values below a neutral point consti-tuting negative legitimacy judgments (i.e.,illegitimacy). While dramatic unexpectedevents may indicate that an existing positivelegitimacy judgment should be increased or anexisting negative legitimacy judgment shouldbe decreased, events that suggest a reversal ofthe valence of the legitimacy judgment are par-ticularly likely to activate the alarm because thecontrast in valence between the existing judg-ment and the event’s implications is particularly

likely to catch the individual’s attention and dis-rupt the assimilation process. This suggestiondoes not imply, however, that these alarm acti-vations will necessarily elicit a change in thelegitimacy judgment; they merely elicit an entryinto the judgment reassessment stage, where adeliberate reevaluation of the legitimacy judg-ment takes place.

In addition, it is important to note that whileprevious institutional theory research has con-ceived of the jolt at the macro level (e.g., at thelevel of the institutional field or organization),the model presented here suggests that the joltcan also occur at the individual level. For exam-ple, major life changes, such as the loss of a job,a personal illness, or the death of a loved one,can alter the individual’s position within andperspective on an institutional field. This al-tered perspective may create new expectationsthat may go unmet or may render the currentinstitutional arrangements incapable of meet-ing preexisting expectations. Similarly, if a re-spected friend or colleague explicitly chal-lenges the legitimacy of existing social entitiesthat had previously been taken for granted, thischallenge can act as an exogenous jolt thatis not easily assimilated into the existing judg-ment. The assimilation process is therefore in-terrupted, which leads the individual to activelyconsider the challenge to legitimacy, leading toan erosion of cognitive legitimacy. In such acircumstance the mental alarm will be activatedfor the individual without any substantialchange at higher levels of analysis.

Thus, the experience of a dramatic violation ofexpectations can function as a trigger for themental alarm, alerting the individual to theneed to reconsider existing legitimacy judg-ments in a more effortful and reflective fashion.In this sense the function of jolts is to expose theformerly invisible assumptions underlying gen-eralized legitimacy judgments and to motivatethe individual to actively interrogate those as-sumptions previously passively accepted. Inthis way jolts cause cognitive legitimacy to dis-solve or dissipate. In doing so they lead individ-uals to switch from the use stage of the legiti-macy judgment process to the reassessmentstage, where they actively question the legiti-macy of social entities and, if they reach a judg-ment of illegitimacy, come to actively pursuechange.

2011 701Tost

Contradictions. Institutional theorists havealso specified an additional mechanism thatcan lead individuals to reconsider existing le-gitimacy judgments. Seo and Creed (2002) haveargued that contradictions in institutional logics(the underlying assumptions that shape ways ofviewing and thinking about the social worldwithin an institutional field) can lead individu-als to question the legitimacy of existing insti-tutional arrangements. This approach recog-nizes that actors are simultaneously embeddedin multiple institutional fields and that conflictsand contradictions can arise between andamong those fields (Greenwood & Suddaby,2006; Hoffman, 1999; Seo & Creed, 2002). Fromthis perspective, when these institutional fieldsproduce contradictions, these contradictionstrigger the mental alarm, alerting actors thattheir existing judgments may be unreliable. Atthis point the individuals exercise agency byentering the judgment reassessment stage andevaluating (or reevaluating) existing institu-tions and attributing problems to one or moreexisting institutions (i.e., specification; seeGreenwood et al., 2002).1 Thus, when an individ-ual detects contradictions among institutionallogics, the mental alarm is activated and cogni-tive legitimacy (at the individual level) begins toerode.

However, given that individuals are embed-ded in a multitude of institutional arrangementsand that such contradictions are pervasive, itremains unclear which contradictions are likelyto attract the attention of which actors. I arguethat for contradictions across institutional fieldsto trigger the mental alarm, they must interferewith an individual’s goal pursuits. If a contra-diction does not have meaningful implicationsfor the individual’s ability to pursue valuedgoals, such as the achievement of desired out-comes or the promotion of closely held values,then the individual is not likely to expend thecognitive energy necessary to engage in thejudgment reassessment stage. However, when

contradictions interfere with goal pursuit, theyhave the effect of revealing and calling intoquestion the nature of existing institutional ar-rangements and motivating individuals to reex-amine their existing legitimacy judgments, be-cause doing so can serve the individuals’ goals.Thus, just as jolts can occur at either the macro-institutional or individual levels (or any level inbetween), contradictions can emerge for singleindividuals or for collectivities (e.g., organiza-tions), and different contradictions will be evi-dent to different actors depending on the goalsand values the actors pursue.

Reflexivity. Reflexivity refers to the ability ofindividuals to consciously reflect on institu-tional arrangements. In order to do so, individ-uals must distance themselves from the institu-tional arrangements in which they areembedded by making deliberate efforts to inter-rogate those arrangements and consider possi-ble alternatives (Emirbayer & Mische, 1998). Ofcourse, this is precisely the type of reflectiveconsideration of institutional arrangements thatoccurs in the reassessment stage. However,some theorists have suggested that there maybe certain personality types or traits that willlead some individuals to be particularly predis-posed to engage in this type of reflection. Forexample, Mutch (2007) explored Archer’s (2003)concept of the autonomous reflexive, which re-fers to a type of individual who monitors thesocial environment and engages in internal de-bates that serve to challenge arrangements thatconflict with individual-level concerns. This ap-proach suggests that there may be individualswho are predisposed by the nature of their per-sonality or personal experience to question ex-isting institutional arrangements. These indi-viduals may have personal tendencies ormotivations that function as internal triggers ofthe mental alarm, thus moving them from theuse stage to the reassessment stage without theneed for an external jolt or contradiction.

Summary

In summary, the legitimacy judgment processis characterized by three stages. In the judgmentformation stage the individual forms legitimacyjudgments using either a passive or evaluativemode of information processing. The passivemode involves the use of validity cues as a basisfor a generalized legitimacy judgment, whereas

1 Of course, the individual can also further exerciseagency at this point by formulating a vision for institutionalchange and mobilizing to achieve that change. However, Ifocus here on the aspect of agency involved in the legiti-macy judgment formation process and leave for future workthe task of exploring the role of agency in determining howindividuals act on these judgments in pursuit of institutionalchange.


the evaluative mode involves an effortful con-sideration of the entity along instrumental, rela-tional, and moral dimensions. The generalizedlegitimacy judgment that emerges from thejudgment formation stage functions as a heuris-tic in the use stage, guiding perceptions andbehavior relevant to the entity. The use stage ischaracterized by a process of assimilation,which leads to the accumulation of cognitivelegitimacy. The use stage persists until an ex-ogenous jolt, contradictions in the institutionalfield, or reflexivity at the individual level triggerthe mental alarm and motivate the individual tomove into the judgment reassessment stage. Inthe reassessment stage the legitimacy judgmentprocess involves a more effortful and deliberateapproach to evaluating the legitimacy of thesocial entity along instrumental, relational, andmoral dimensions. It is therefore in the reassess-ment stage of the legitimacy judgment processthat judgments of illegitimacy are most likely toemerge. Those judgments of illegitimacy, inturn, produce the desire for institutional change.Upon forming a new positive or negative legiti-macy judgment, the individual reenters the usestage.2

DISCUSSION AND CONCLUSION

The theoretical framework of legitimacy judg-ments I have developed here integrates legiti-macy research from institutional theory and so-cial psychology. Social psychologists haveconceived of legitimacy as a function of instru-mental and relational considerations and haveonly recently come to examine the moral dimen-sion of legitimacy. At the same time, institu-tional theorists have conceptualized pragmaticlegitimacy (i.e., the instrumental dimension),moral legitimacy, and cognitive legitimacy(Suchman, 1995) but have generally neglected toconsider the importance of the relational dimen-sion and have not examined the formation andchange of these judgments at the individuallevel. The theoretical framework presented hereintegrates this previous work, highlighting theways in which research from both fields caninform and complement one another. Moreover,the model of legitimacy judgments presented

here can be applied broadly to help scholarsunderstand legitimacy evaluations of a varietyof social entities, including organizations, socialstructures, organizational policies, procedures,and leaders.

Importantly, this integrative theoreticalframework empowers researchers to exploreboth the content and process dynamics of legit-imacy judgments. With respect to the content oflegitimacy judgments, I argue that rather thanviewing the instrumental, relational, and moraldimensions as competing models for under-standing the content underlying legitimacyjudgments, these three types of perceptions, be-liefs, and concerns should be viewed as threedifferent dimensions of judgment that may si-multaneously impact individuals’ judgments ofthe generalized legitimacy of a social target.Viewing them in this way permits scholars toexplore the ways in which each of the types ofconcerns contributes to (or is guided by, as in theuse stage) generalized legitimacy judgments. Inaddition, this perspective empowers research-ers to consider the factors that influence whichof the three dimensions is prioritized in the eval-uative mode. An understanding of the contentunderlying legitimacy judgments can thereforecontribute meaningfully to scholars’ under-standing of the factors that impact individuals’judgments of the legitimacy of organizations,groups, social structures, policies, procedures,and leaders.

At the same time, with respect to the legiti-macy judgment process, the distinction betweenthe use stage and the judgment stages (judg-ment formation and judgment reassessment) ofthe legitimacy judgment process providesmuch-needed conceptual clarity to the relation-ships between legitimacy and power, on the onehand, and legitimacy and fairness, on the otherhand. First, the distinction between the judg-ment stages and the use stage helps to eluci-date when legitimacy is a source of power thatproduces deference to organizational authori-ties and rules and when it instead represents acontingent judgment that is under development.Specifically, in the use stage legitimacy judg-ments guide people’s interpretations of informa-tion related to social entities and determine in-dividuals’ behavioral orientations to thoseentities. Consequently, a positive legitimacyjudgment provides the entity with a cushion ofsupport that promotes deference to institutional

2 The cyclical nature of this process is similar to punctu-ated equilibrium models of change (Gersick, 1991).

2011 703Tost

constraints. In this sense, a positive legitimacyjudgment is a source of power for organizations,institutions, and institutional authorities whenevaluators are in the use stage. However, whenan individual enters the evaluative mode, legit-imacy is contested and can no longer functionas a cushion of support; it is instead a develop-ing judgment that is contingent on the individ-ual’s evaluations of the entity along instrumen-tal, relational, and/or moral dimensions.

Second, the distinction between the use stageand the judgment stages also helps scholarsconceptualize the distinctions between legiti-macy and fairness. Specifically, while muchprevious social psychological research has con-flated legitimacy and fairness, there has alsobeen some confusion within the field of socialpsychology as to whether legitimacy should betreated as an antecedent to fairness perceptions(e.g., Tyler, 2006; Tyler & Degoey, 1995) or as anoutcome of fairness perceptions (e.g., Hegtvedt &Johnson, 2000). The process model presentedhere suggests that, in fact, fairness is both anantecedent to and an outcome of legitimacy,depending on the stage of the legitimacy judg-ment process. Specifically, in the evaluativemode of the judgment stages, I would expectthat procedural and interactional fairnesswould contribute positively to relational evalu-ations and that fairness in general (e.g., Am-brose & Schminke, 2009) would have a positiveimpact on both relational and moral evalua-tions. However, in the use stage the generalizedlegitimacy judgment would be expected toguide judgments of the fairness of the entity justas it guides other judgments related to the en-tity. Therefore, in the use stage legitimacy func-tions as an antecedent to fairness perceptions,but in the evaluative mode fairness is an ante-cedent to legitimacy.

Implications for Institutional Change

In addition, an understanding of how legiti-macy judgments develop and change over timecan contribute substantially to scholarly under-standing of the individual-level dynamics ofsupport for and resistance to institutionalchange. Specifically, because legitimacy func-tions as a pivotal cognition that impacts indi-viduals’ inclinations to support a social entity orwork for change, understanding how and whylegitimacy judgments change can help re-

searchers understand how and why individuals’behavioral orientations to social entities mayshift and cause them either to support or to re-sist institutional change.

As explained above, most models of institu-tional change conceive of the deinstitutionaliza-tion process as being preceded by a precipitat-ing or destabilizing jolt to the social ororganizational system (e.g., Greenwood et al.,2002; Meyer, Brooks, & Goes, 1990). However, theway in which such a jolt would initiate changeprocesses at the level of individual behaviorand the circumstances under which it would doso are rarely addressed in this literature. Thepresent model has implications regarding boththe nature of the jolt and its impact. According tothe model presented here (but in contradiction toa common assumption in institutional research),the jolt need not occur at the macro level. Themental alarm can indeed be activated by radi-cal environmental changes, such as crises, thatoccur at the macro level (assuming such envi-ronmental changes produce unexpected out-comes). However, the mental alarm can also betriggered by unexpected outcomes at the indi-vidual level, leading a single individual to re-consider the legitimacy of existing social enti-ties and, if a judgment of illegitimacy is formed,to take on a change leadership role and worktoward change at the group or organizationallevels.

Furthermore, the view presented here of thetransition between the use stage and the judg-ment reassessment stage further specifies thecircumstances under which institutional contra-dictions act as a precursor to institutionalchange. Previous work in this area has exploredhow an actor’s social position may influence thelikelihood of detecting institutional contradic-tions (see Battilana et al., 2009, for a review). Iargue that in addition to considering the likeli-hood of detection, it is also necessary to con-sider the motivation to examine the questionsraised by any contradictions that are detected. Ifa contradiction is encountered but it does notinterfere with the pursuit of desired outcomes orthe promotion of personal values, then the indi-vidual is not likely to explore the questions thatsuch a contradiction raises about existing insti-tutional arrangements (and, consequently, thecognitive legitimacy of those arrangements re-mains intact). In order to enter the judgmentreassessment stage in response to institutional


contradictions, an individual must have both theopportunity to detect contradictions (e.g., socialposition) and the motivation to examine thequestions arising from those contradictions.

Thus, this model also has important implica-tions for the study of institutional entrepreneur-ship. Institutional entrepreneurs are individualswho take on leadership roles in institutionalchange efforts (Battilana et al., 2009; Greenwood& Suddaby, 2006), using what Fligstein (2001)calls “social skill” to induce others to cooperatein the pursuit of change. Accordingly, institu-tional entrepreneurs are individuals who haveformed a judgment of existing social entities asillegitimate and therefore seek change. Becauseof their critical role in initiating change andpersuading others to support change, Dacin,Goodstein, and Scott (2002: 47) call institutionalentrepreneurs “agents of legitimacy.” That is,institutional entrepreneurs use influence to per-suade others of the illegitimacy of existing so-cial arrangements and of the legitimacy of al-ternatives, thereby recruiting others to join themin institutional change efforts. The model of thelegitimacy judgment process presented herehas important implications for understandingthe determinants of institutional entrepreneurs’success in their role as agents of legitimacy.

First, the model suggests that institutional en-trepreneurs will be most effective in their per-suasion attempts if the targets of their influenceare in t

AN INTEGRATIVE MODEL OF LEGITIMACY JUDGMENTS...legitimacy, situate the construct of legitimacy in...

Documents

Transcript of AN INTEGRATIVE MODEL OF LEGITIMACY JUDGMENTS...legitimacy, situate the construct of legitimacy in...