Ambivalence in Organizations: A Multilevel Approach

This art icle was downloaded by: [ 149. 169. 136. 200] On: 17 Sept ember 2014, At : 09: 11 Publisher: Inst it ut e f or Operat ions Research and t he Management Sciences (INFORMS) INFORMS is locat ed in Maryland, USA Organization Science Publicat ion det ails, including inst ruct ions f or aut hors and subscript ion inf ormat ion: ht t p: / / pubsonline. inf orms. org Ambivalence in Organizat ions: A Mult ilevel Approach Blake E. Ashf ort h, Krist ie M. Rogers, Michael G. Prat t , Camille Pradies To cite this article: Blake E. Ashf ort h, Krist ie M. Rogers, Michael G. Prat t , Camille Pradies (2014) Ambivalence in Organizat ions: A Mult ilevel Approach. Organizat ion Science 25(5): 1453-1478. ht t p: / / dx. doi. org/ 10. 1287/ orsc. 2014. 0909 Full terms and conditions of use: http:/ / pubsonline. informs. org/ page/ terms-and-conditions This art icle m ay be used only for t he purposes of research, t eaching, and/ or privat e st udy. Com m ercial use or syst em at ic downloading ( by robot s or ot her aut om at ic processes) is prohibit ed wit hout explicit Publisher approval, unless ot herwise not ed. For m ore inform at ion, cont act perm issions@inform s.org. The Publisher does not warrant or guarant ee t he art icle’s accuracy, com plet eness, m erchant abilit y, fit ness for a part icular purpose, or non- infringem ent . Descript ions of, or references t o, product s or publicat ions, or inclusion of an advert isem ent in t his art icle, neit her const it ut es nor im plies a guarant ee, endorsem ent , or support of claim s m ade of t hat product , publicat ion, or service. Copyright © 2014, I NFORMS Please scroll down for article—it is on subsequent pages INFORMS is t he largest prof essional societ y in t he world f or prof essionals in t he f ields of operat ions research, management science, and analyt ics. For more inf ormat ion on INFORMS, it s publicat ions, membership, or meet ings visit ht t p: / / www. inf orms. org Organization Science Vol. 25, No. 5, September–October 2014, pp. 1453–1478 ISSN 1047-7039 (print) — ISSN 1526-5455 (online) http://dx.doi.org/10.1287/orsc.2014.0909 © 2014 INFORMS Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. Ambivalence in Organizations: A Multilevel Approach Blake E. Ashforth Department of Management, W. P. Carey School of Business, Arizona State University, Tempe, Arizona 85287, blake.ashforth@asu.edu Kristie M. Rogers Management Area, School of Business, University of Kansas, Lawrence, Kansas 66045, kristie.rogers@ku.edu Michael G. Pratt, Camille Pradies Management and Organization Department, Carroll School of Management, Boston College, Chestnut Hill, Massachusetts 02467 {prattmg@bc.edu, pradies@bc.edu} T he experience of simultaneously positive and negative orientations toward a person, goal, task, idea, and such appears to be quite common in organizations, but it is poorly understood. We develop a multilevel perspective on ambivalence in organizations that demonstrates how this phenomenon is integral to certain cognitive and emotional processes and important outcomes. Specifically, we discuss the organizational triggers of ambivalence and the cognitive and emotional mechanisms through which ambivalence diffuses between the individual and collective levels of analysis. We offer an integrative framework of major responses to highly intense ambivalence (avoidance, domination, compromise, and holism) that is applicable to actors at the individual and collective levels. The positive and negative outcomes associated with each response, and the conditions under which each is most effective, are explored. Although ambivalence is uncomfortable for actors, it has the potential to foster growth in the actor as well as highly adaptive and effective behavior. Keywords: ambivalence; hybrid identities; role conflicts; dualities; paradox; sensemaking; sensegiving; emotional contagion; avoidance; domination; compromise; holism; defense mechanisms; coping mechanisms; wisdom; multilevel History: Published online in Articles in Advance May 5, 2014. Introduction Consider the popular imagery of the leader in an organization. For some of the many below him in the hierarchy, he is secure, knowing, decisive, powerful, dynamic, threatening, driving, and altogether remote, acting in clear or obscure ways to affect the future of the organization he leads. At eye level, he is more often seen as filled with troubled doubts as he tries to deal with the ambivalences and contradictions of his status. And if his feet are made of a substance more solid than clay, it is because on his climb to the top and with the aid of those who help hold him there, he has learned to still the doubts, to live with the ambivalences, and to cope with the contradictions of his position. (Merton 1976, p. 73) As this quotation from Robert Merton suggests, ambivalence appears to be fairly common in organizational settings. Actors are buffeted by complex and dynamic work environments; play multiple and, at times, contradictory roles; and confront multifaceted issues, goals, and the like (Wang and Pratt 2008). However, as the quote goes on to suggest, actors are often able to “live with the ambivalence” and “cope with the contradictions.” The quote thus begs the question, how do actors accomplish this difficult feat—how do they respond to the vexing mixed feelings and thoughts that characterize ambivalence and somehow move forward with a clear focus and coherent action? 1453 Research in organizational studies is turning increasingly to this and related questions. Indeed, since the turn of this century, organizational scholars have linked ambivalence and ambivalent relationships with organizations to resistance to change (Piderit 2000), the propensity to commit corporate crimes (Vadera and Pratt 2013), and to an increase in escapist behaviors among service personnel (Pratt and Doucet 2000). However, during this same period, other research has linked cognitive and/or emotional ambivalence to better chief executive offier (CEO) decision making (Plambeck and Weber 2009), heightened creativity (Fong 2006), increased receptivity to alternatives resulting in improved judgment accuracy (Rees et al. 2013), and stronger organizational commitment (Pratt and Rosa 2003). Unfortunately, such research remains scattered, with few systematic attempts to either link or systematize the dizzying number of reactions to ambivalence in organizations. Specifically, a review of extant research on this topic suggests that we know much more about specific types of responses (e.g., nonconscious responses; Horney 1945) than the conceptual relationship between responses. Although we know at a general level that individuals might embrace or deny their ambivalence (Pratt and Pradies 2011)1 —and that specific responses range widely from paralysis to vacillation to strong and concerted Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. 1454 action (see Pratt and Doucet 2000, Weigert and Franks 1989)—we know relatively little about how to compare these various responses. As such, research may be overlooking key response types. As we explain in greater detail below, we argue that some types of responses have received a lot of attention (e.g., avoidance), whereas others have been largely ignored (e.g., holism). In addition, we know that various types of responses exist, but we know relatively little about the effectiveness of these responses. We define effectiveness as the degree to which a response reduces the intensity of ambivalence—given the situation—by dealing with the source directly and/or the symptoms that result. Finally, efforts to organize various responses to ambivalence have tended to focus on a single level of analysis—namely, the individual (cf. Pradies and Pratt 2010). In short, the field is lacking a comprehensive typology for building more systematic research on responses to ambivalence across levels of analysis in organizational settings. Prior to building a theoretical framework that provides clarity to such responses, it is important to grasp where ambivalence comes from—that is, the key triggers of ambivalence in organizations—and how the experience of ambivalence tends to diffuse across levels of analysis. Organizational research suggests that ambivalence is experienced and acted upon by both individuals (Fong 2006, Pratt 2000) and collectives (Peters et al. 2011, Pradies and Pratt 2010, Weick 2004). What might trigger ambivalence within the individual level, the collective level, or across the levels, and through which social psychological mechanisms might ambivalence diffuse across levels? Establishing a model linking the upstream processes that connect triggers of ambivalence to the experience itself at both the individual and collective levels of analysis will set the stage for examining what actors then do with the ambivalence experienced. In sum, our paper seeks to provide an organizing framework of responses to ambivalence: a framework that is applicable to both the individual and collective levels of analysis and that clarifies the conditions under which each response is relatively more effective. To build a foundation for these issues, we first discuss what ambivalence is. We then discuss the organizational triggers of ambivalence and how ambivalence at one level of analysis may diffuse to the other level. We then provide an indepth exploration of individual and collective responses to ambivalence. The result is an integrative multilevel framework of ambivalence in organizations that indicates how this phenomenon is central to certain cognitive/emotional processes and important outcomes. We specifically suggest that avoidance, domination, compromise, and holism are the major, more or less distinct, responses to ambivalence. We also discuss the conditions under which each response is most effective and the positive and negative outcomes of each response. We conclude by exploring implications for future research. Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS What Is Ambivalence? Ambivalence literally refers to the experience of two (ambi) opposing forces (valences) and is derived from the Latin ambo, or “both,” and valere, which means “to be strong” (Meyerson and Scully 1995). Although there have been some differences in how ambivalence has been defined (see Conner and Sparks 2002 for a review), there remains a high degree of convergence across definitions in psychological research. Specifically, we define ambivalence as simultaneously positive and negative orientations toward an object. “Orientation” refers to the actor’s alignment or position with regard to the object, where a positive orientation means attraction or a pull toward it and a negative orientation means repulsion or a push away from it. Ambivalence includes cognition (“I think about X”) and/or emotion (“I feel about X”); ambivalence occurs when cognitions clash, emotions clash, or cognitions and emotions clash. Thus, ambivalence is often described as having “mixed feelings,” being “torn between conflicting impulses,” and being “pulled in different directions.” Four important points regarding our definition of ambivalence should be clarified. First, following research on the psychodynamic roots of ambivalence (Freud 1950/1920, Horney 1945, Sincoff 1990), individuals may not be conscious of their ambivalence. However, even if nonconscious, ambivalence may nonetheless affect individuals (cf. Meyerson and Scully 1995). Second, we follow the lead of others (e.g., Horney 1945) and focus on substantive (or, as we describe later, high intensity) rather than superficial (low intensity) ambivalence, as the latter (1) is likely to be ignored, (2) has little impact on behavior, and by extension (3) is of relatively little concern to management. Third, although some conceptualizations of ambivalence also include a behavioral component, we focus exclusively on ambivalence as a cognitive-emotional construct.2 Ambivalence, as it was originally coined by Bleuler in the early 1900s, focused only on cognition and emotion (Sincoff 1990). In addition, abundant research on the attitude–behavior relationship (see Sheeran 2002 for a review) indicates that various factors may dampen the link between the two, such as situational constraints and impression management concerns. Thus, we regard behavioral tendencies and behavior itself as probabilistic outcomes of ambivalence rather than as part of its definition. Finally, with some notable exceptions, the literature tends to view ambivalence as more or less dysfunctional for individuals (e.g., Bowlby 1982, Horney 1945, Merton 1976). However, we will argue that ambivalence can be functional, dysfunctional, or perhaps—not ironically—both. Table 1 contrasts our definition of ambivalence with definitions of related constructs in the organizational literature—cognitive dissonance, emotional dissonance, hypocrisy, ambiguity, and equivocality. As the table suggests, the seminal differences between ambivalence Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach 1455 Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS Table 1 Construct Definitions and Relationships to Ambivalence at the Individual Level Construct Ambivalence Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. Cognitive dissonance Emotional dissonance Hypocrisy Ambiguity Equivocality Definition Simultaneously oppositional positive and negative orientations toward an object. Ambivalence includes cognition (“I think about X”) and/or emotion (“I feel about X”). “When a person has two beliefs or items of knowledge that are not consistent with each other” (Kantola et al. 1984, p. 417). There is conceptual overlap between dissonance and ambivalence (Baek 2010), but this definition suggests that dissonance arises when there is inconsistency between thoughts (e.g., I want to hire candidate A, but I also want to hire candidate B), whereas ambivalence requires opposite orientations, which is a more extreme contrast than inconsistency, toward a single object (e.g., I want to hire candidate A and I want to not hire candidate A). Additionally, cognitive dissonance is exclusively cognitive, whereas ambivalence can be cognitive and/or affective. “The discrepancy between emotions felt and those required by the job role is commonly referred to as emotional dissonance” (Diestel and Schmidt 2011, p. 643). As with the distinction above regarding cognitive dissonance, ambivalence requires opposition and not simply discrepancy. Thus, a role may require one to smile when one does not feel like it (emotional dissonance), but this is different than simultaneously feeling happy and not happy (ambivalence). Moreover, the inconsistency in emotional dissonance is between a feeling and a behavior that one’s role demands. Ambivalence is not necessarily emotional and does not include a behavior. “Clear inconsistency between word and deed” (Fassin and Buelens 2011, p. 587). Hypocrisy is a contradiction between a statement and action, which is generally perceived and labeled by an observer. Ambivalence involves cognition and/or emotion rather than behavior, although ambivalence may cause an actor to behave in ways that could be perceived by others as hypocritical. Thus, hypocrisy can be an outcome of ambivalence. “The degree of uncertainty inherent in perceptions of the environmental state” (Carson et al. 2006, p. 1059). Ambiguity is concerned with uncertainty or a lack of clarity, whereas ambivalence is the experience of two clear but opposing thoughts and/or feelings toward an object. “The multiplicity of meaning conveyed by information about organizational activities” (Daft and Macintosh 1981, p. 211). Equivocality captures the potential for multiple meanings and interpretations of a message. If these meanings are oppositional, there is potential for the equivocality to trigger ambivalence. and these other constructs are that ambivalence involves an individual’s oppositional orientations toward an object rather than mere uncertainty or inconsistency, and it involves cognition and/or emotion but not behavior. Collective Ambivalence Researchers such as Smith and Berg (1987), Weick (2004), Pradies and Pratt (2010), and Peters et al. (2011) have suggested the existence of ambivalence in collectives. For example, in referring to a network of medical specialists working to identify the West Nile Virus, Weick (2002, p. s13, emphasis added) notes that “[a]mbivalence can be a property of networks and distributed cognition, as well as a property of individuals.” However, beyond indicating the possibility that collective ambivalence can exist, little work has been done to formally define and differentiate collective-level ambivalence (cf. Pradies and Pratt 2010).3 Ambivalence at the collective level aligns with our definition presented earlier: simultaneously positive and negative orientations toward an object, whereby ambivalence includes cognition (“we think about X”) and/or emotion (“we feel about X”) experienced by a collective. Also, just as ambivalence at the individual level can be said to be more or less intense and more or less conscious, so too can ambivalence at the collective level. Finally, mirroring what we did at the individual level, Table 2 contrasts this definition with related constructs at the collective level—namely, duality, paradox, organizational hypocrisy, dialectics, and organizational ambidexterity. The major differences are that duality, paradox, and organizational hypocrisy refer to external conditions that when internalized by a collective may give rise to ambivalence, and that dialectics and organizational ambidexterity refer to means of managing tensions rather than to the actual experience of ambivalence. Organizational Triggers and the Diffusion of Ambivalence What situational factors cue ambivalence at the individual and collective levels of analysis, and how does ambivalence experienced at the individual level spread to the collective level, and vice versa? Figure 1 summarizes the arguments to follow. The multilevel dynamism of the figure is modeled after the approach of Kozlowski et al. (2009) to organizational learning across multiple levels of analysis, along with research on specific socialpsychological mechanisms through which cognition and emotion may spread in organizations. Organizational Triggers of Ambivalence Sociologists have tended to view norms and roles as sources of ambivalence, whereas psychologists have tended to view individual differences and relationships as sources (Wang and Pratt 2008). Although psychological sources of ambivalence are relatively well understood, the contextual sources of ambivalence, particularly in organizations, “have often been elusive or highly contingent” (Plambeck and Weber 2010, p. 705). Given our Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach 1456 Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS Table 2 Construct Definitions and Relationships to Ambivalence at the Collective Level Construct Ambivalence Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. Duality Paradox Organizational hypocrisy Dialectics Organizational ambidexterity Definition Simultaneously oppositional positive and negative orientations toward an object. Ambivalence includes cognition (“we think about X”) and/or emotion (“we feel about X”). “The twofold character of an object of study without separation 0 0 0 it retains the idea of two essential elements, but it views them as interdependent [complementary and contradictory], rather than separate and opposed” (Farjoun 2010, p. 203). Whereas ambivalence is internal to the actor, a duality exists external to the actor perceiving the duality. A duality may trigger ambivalence if the two essential elements produce opposite orientations. “ ‘Paradox’ denotes contradictory yet interrelated elements—elements that seem logical in isolation but absurd and irrational when appearing simultaneously” (Lewis 2000, p. 760). Paradox refers to situations that are external to actors. A paradox may trigger ambivalence if the contradictory elements foster opposite orientations to a particular object. Brunsson (1993) refers to organizational hypocrisy to describe occasions when an organization acts in contradiction to what it says. As with sociological ambivalence more generally, such a hypocritical “event” may trigger collective-level ambivalence if it is experienced by multiple members of the organization. However, organizational hypocrisy in and of itself does not denote how such hypocrisy is experienced—either as a univalent response (e.g., disillusionment, disgust) or an ambivalent response. Tracy (2004) explains how dialectics is a tension management technique that enables organizational actors to not view tensions as tensions per se but, via reframing, as complementary. However, dialectics is more about managing tensions or even contradictions than the experience of these tensions. As such, dialectics is more akin to the response we labeled “holism” than to ambivalence. “[F]irms are ambidextrous—aligned and efficient in their management of today’s business demands while simultaneously adaptive to changes in the environment 0 0 0 0 To be ambidextrous, organizations have to reconcile internal tensions and conflicting demands in their task environments” (Raisch and Birkinshaw 2008, p. 375). As with dialectics, organizational ambidexterity focuses on organizational actions to manage tension and conflicting demands, rather than the experience of ambivalence. interest in understanding how individuals and collectives respond to ambivalence, we focus on the most common potential organizational triggers of ambivalence. Each of the four major types of triggers discussed below can operate at the collective or individual level of analysis, as our examples will illustrate. It should be noted that these triggers, like any situational factor, are a matter of perception and are more or less amenable to framing by organizational leaders (Smircich and Morgan 1982). The major contextual roots of ambivalence appear to be complexity and dynamism in the environment and the organization itself (cf. Weigert and Franks 1989). Organizations typically face complex and dynamic environments, where the meaning of events is ambiguous or equivocal, the future is fundamentally uncertain, and the numerous and often conflicting demands of diverse stakeholders must be addressed. Such environments frequently foster the more proximal contextual root of oppositions, where a demand for A is juxtaposed with a demand for “not-A.” Oppositions appear to manifest in four major ways: (1) hybrid identities, contradictory goals, and role conflicts; (2) dualities; (3) multifaceted objects; and (4) temporal factors. Hybrid Identities, Contradictory Goals, and Role Conflicts. According to Merton and Barber (1976, p. 6), “[I]ncompatible normative expectations of attitudes, beliefs, and behaviors” are the essence of sociological ambivalence. Specifically, such oppositions may manifest in hybrid identities, contradictory goals, and role conflicts (Adler 2012, Albert and Whetten 1985, Brunsson 1989, Katz and Kahn 1978, Pratt and Foreman 2000). To illustrate, at the collective level, Glynn (2000) shows how the Atlantic Symphony Orchestra played out its competing “artistry” and “utility” identities during a musicians’ strike—with the former identity focusing on normative claims of excellent music and the latter focusing on economic concerns for profit. At the individual level, Fong and Tiedens (2002) note the emotional ambivalence experienced by women in high-status positions. On one hand, attainment of professional goals produces positive affect; on the other hand, pressure to conform to gender stereotypes (that contradict normative behaviors in high-power roles) produces negative affect. Organizational Dualities. Oppositional tendencies often manifest in the form of organizational dualities and related concepts such as paradoxes, dilemmas, double binds, oxymorons, and ironies (Clegg 2002, Hatch 1997, Smith and Lewis 2011, Tracy 2004). Dualities are particularly likely to provoke ambivalence because of their simultaneous injunction to do A and to do the opposite of A. Well-documented examples, particularly at the collective level, include continuity versus change, a global versus local focus, and competition versus cooperation. Although ostensibly opposites on a continuum, where more of one necessarily means less of the other, a dualities perspective views them as facets of a natural wholeness (Graetz and Smith 2008). That is, the dualistic facets are complementary even though each—as noted in Table 2 regarding Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach 1457 Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS Organizational Ambivalence Across Levels Organizational triggers of ambivalence Experience of ambivalence Responses to ambivalence Oppositions: Demand for A and not-A Social psychological diffusion mechanisms (See Figure 2 for expanded treatment) Collective level • Hybrid identities, contradictory goals, role conflicts • Dualities • Multifaceted objects • Temporal factors Collective triggers (e.g., hybrid organizational identity, continuity vs. change) 1 Individual level Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. Figure 1 Individual triggers (e.g., role conflict, membership dualities) 2 4 • Sensemaking • Sensegiving • Emotional contagion • Emotions-as-social-information, emotional comparison Collective ambivalence (i.e., simultaneously positive and negative orientations toward an object experienced by a collective) 5 3 • Avoidance • Domination • Compromise • Holism 8 Individual ambivalence (i.e., simultaneously positive and negative orientations toward an object experienced by an individual) paradox—may “seem logical in isolation but absurd and irrational when appearing simultaneously” (Lewis 2000, p. 760). However, dualities and related concepts may nonetheless trigger ambivalence toward a given action by fostering seemingly oppositional imperatives (e.g., competition versus cooperation). Perhaps the most common dualities faced by individuals involve memberships in various collectives, including the organization itself. Smith and Berg (1987) describe 12 membership dualities under the labels of “paradoxes of belonging” (e.g., maintaining individuality and maintaining commonality within a group), “paradoxes of engaging” (e.g., incorporating past experiences while breaking with past experiences), and “paradoxes of speaking” (e.g., accepting dependence on group members in order to act independently). Accordingly, even the most effective collectives are laced with many, at least tacit, dualities for the individual. Moreover, the more strongly that one dualistic quality is emphasized, the greater the countervailing need for the other (Erikson 1976, Smelser 1998). For instance, following Smith and Berg, a strong effort to identify with a group may spawn a counter-desire to establish one’s individuality. Multifaceted Objects. Oppositions often manifest in multifaceted objects, such as policies and cultures at the collective level and jobs and relationships at the individual level (Merton 1976, Pratt 2001). Although an individual’s or collective’s stance toward particularly salient facets may 6 Collective responses (i.e., conscious or nonconscious effort to reduce intensity; may involve cognition, affect, and/or behavior) 9 7 Individual responses (i.e., conscious or nonconscious effort to reduce intensity; may involve cognition, affect, and/or behavior) exert a positive or negative “halo” (e.g., Feeley 2002) over other facets, it remains that the individual or collective is unlikely to have a common stance toward every facet, partly because positive and negative aspects tend to surface over time (Wilson and Hodges 1992). Specifically, the more familiar an actor is with an object, the richer the store of information and the greater the probability of having encountered the object’s multiple facets and imperfections. In short, as Brooks and Highhouse (2006, p. 105) put it, “Familiarity breeds ambivalence.” For example, research on personal relationships indicates that ambivalence typically characterizes even the most intimate of bonds; indeed, Thompson and Holmes (1996, p. 502) conclude that “a moderate level of ambivalence may actually indicate a balanced, realistic assessment of a partner.” Moving to organizational contexts, Oglensky (2008) concludes that even robust mentor–protégé relationships are inherently ambivalent, and Kreiner and Ashforth’s (2004, p. 20) study suggested that ambivalent identification with the organization was “not a rare existential experience” among employees (see also Brief 1998, Ziegler et al. 2012). Temporal Factors. Finally, oppositional tendencies are often introduced by various temporal factors (HillcoatNallétamby and Phillips 2011, Pratt and Doucet 2000). As noted, organizational contexts are typically dynamic; indeed, Lewin (1951) argues that seemingly stable states are really temporary equilibria held in place by opposing forces. Dynamism may introduce change, inconsistencies, Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. 1458 ambiguities, and lags in objects at the individual and collective levels of analysis such that, similar to multifaceted objects, they generate oppositions. An otherwise supportive manager may act in a surprisingly unsupportive manner, and an organization that strongly espouses integrity may succumb to temptation and act in a highly unethical way. Furthermore, the use of developmental feedback at the individual level and continuous improvement at the collective level necessitates that even under conditions of success, actors need to acknowledge both the pros and cons (e.g., “The project was a success, but we can do better next time”). And research on counterfactual thinking indicates that reflecting on “what could have been” (Roese 1997, as cited in Fong 2006, p. 1017) may introduce, for individuals and collectives alike, an oppositional stance into otherwise positive events (“it could have been better”) and negative events (“it could have been worse”). For example, Olympic silver medalists were found to be less happy with their achievement than bronze medalists because they “just missed the gold” (Medvec et al. 1995). It is evident, then, that the complexity and dynamism of organizational life engender a number of potential oppositions. Moreover, as indicated by the bidirectional arrow 1 in Figure 1, oppositions at one level of analysis may affect oppositions at the other. For example, prisons are charged with the somewhat oppositional goals of rehabilitating inmates and punishing inmates, which have “trickled down” (Tracy 2004, p. 124) to foster role conflict among individual correctional officers (e.g., “respect versus suspect”). Conversely, the individual-level paradoxes of belonging noted above (e.g., maintaining individuality, maintaining commonality within a group) may foster a collective that is riven with oppositions (e.g., we are one versus we are many; see Smith and Berg 1987). Because these oppositions are so commonplace in organizations, we suspect that ambivalence is a very common experience in organizations. Not surprisingly, as indicated by arrows 2 and 3 in Figure 1, oppositions at the collective or individual level of analysis tend to trigger the experience of ambivalence at the same level. For instance, Das and Teng (2000) report that strategic alliances involve certain dialectical tensions (e.g., cooperation versus competition, rigidity versus flexibility, short-term versus long-term orientation), which are likely to provoke ambivalence among top management teams (e.g., we should cooperate and not cooperate). And Ashforth et al. (2013) find that role conflict experienced by telemarketers is associated with ambivalence toward their occupation. Furthermore, following the theorizing of Kozlowski et al. (2009), cross-level effects from collective triggers to individuals’ experience of ambivalence are likely common (see arrow 4 in Figure 1). For instance, Piderit (2000) argues that the most prevalent reaction to organizational change—a temporal factor in our model—is Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS employee ambivalence, as individuals simultaneously support and resist change efforts. Conversely, crosslevel effects from individual-level triggers to collective ambivalence are likely to be more indirect. We argue in the following section that such cross-level effects are largely mediated by collective sensemaking and emotional processes (hence, we do not depict a cross-level arrow in Figure 1 from individual triggers to collective ambivalence). Diffusion of Ambivalence Whereas the previous section established the organizational triggers of ambivalence and their role in fostering individual and collective ambivalence (arrows 1–4 in Figure 1), this section focuses on the social-psychological mechanisms through which ambivalence at the individual level may spread to the collective level (i.e., bottomup emergent effects; arrow 5) and ambivalence at the collective level may spread to the individual level (i.e., top-down cascade effects; also arrow 5). The upshot is a recursive loop where expressed ambivalence is likely to have a “ripple effect” (Barsade 2002, p. 644) such that “I think/feel” becomes “we think/feel,” and vice versa. Bottom-Up Emergent Effects. Morgeson and Hofmann (1999, p. 257) note that “the investigation of constructs at the collective level could begin with an understanding of the interaction of organizational members.” To that end, Kozlowski and Klein (2000, p. 55) observe that “[a] phenomenon is emergent when it originates in the cognition, affect, behaviors, or other characteristics of individuals, is amplified by their interactions, and manifests as a higher-level, collective phenomenon.” Research on collective-level cognitions and emotions provides examples of collective cognitive phenomena such as team mental models (Klimoski and Mohammed 1994, Lim and Klein 2006), collective interpretations (Isabella 1990), and intersubjectivity (Ashforth et al. 2011, Wiley 1988) and collective emotional phenomena such as group affective tone (George 1996), emotional culture (Ashforth and Saks 2002), and group emotion (Barsade and Gibson 1998). These collective-level constructs are instructive for understanding the collective-level nature of ambivalence and provide several pathways whereby ambivalence may emerge via individual interactions (Pradies and Pratt 2010). The sensemaking perspective (Gioia and Chittipeddi 1991, Maitlis 2005, Weick 1995, Weick et al. 2005) is a useful mechanism for articulating the cognitive nature of the vertical links between individual and collective experiences of ambivalence (arrow 5 in the upward direction in Figure 1). According to Stigliani and Ravasi’s (2012, p. 1241) process model of collective sensemaking, individual sensemaking and cognitions become shared by the collective through “the interplay between conversational and material practices.” Stigliani and Ravasi Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS found that design consultants alternated between working on their own and holding group meetings where they shared their ideas. Specifically, individuals noticed and bracketed information on their own, they articulated it to the group, and the group elaborated on it through explicit conversations (e.g., “storybuilding”), cognitive subprocesses (e.g., “building on each other’s ideas”), and material artifacts (e.g., “sketches”), which resulted in shared cognition among the collective. As an illustration regarding ambivalence, Gutierrez et al. (2010, p. 673) describe how revelations that the Catholic Church had covered up the sexual abuse of minors by priests fostered ambivalence toward the church among some individuals: “A few dozen grieving laity began gathering to discuss their reactions in a suburban church basement. Within five months, the group grew into a 20,000-member organization” called Voice of the Faithful; its motto is “Keep the faith, change the church.” Regarding the emotional processes through which ambivalence may emerge from the individual level (arrow 5 in the upward direction in Figure 1), one major pathway is via emotional contagion, defined as “a process in which a person or group influences the emotions or behavior of another person or group through the conscious or unconscious induction of emotion states and behavioral attitudes” (Schoenewolf 1990, as cited in Barsade 2002, p. 646). With contagion, emotions can spread to a collective via nonconscious and conscious interactions such as synchronization (Neumann and Strack 2000) and the social sharing of emotions (Peters and Kashima 2007). A second major pathway is suggested by Van Kleef’s (2008) emotions-as-social-information model, where emotional cues are “read” more indirectly via cognitive processing (“why does she feel that way?”). Similarly, individuals may engage in “emotional comparison” (Ashforth and Saks 2002, p. 350) with others to assess the meaning and appropriateness of their own arousal (“why do I feel this way?”; see, e.g., Bartel and Saavedra 2000). Rothman and Wiesenfeld (2007) confirm that ambivalent emotions can be read by others and thus may be suitable for contagion and social information effects. The upshot of collective sensemaking and the above emotional conduits is that individual ambivalence may foster collective ambivalence (arrow 5 in the upward direction in Figure 1). In the multilevel terminology of Kozlowski and Klein (2000, p. 16), these “forms of emergence” result in “composition” effects, that is, a convergence in the perspectives/experiences of individuals and the collective such that the ambivalence becomes shared.4 Top-Down Cascade Effects. In organizations, collective ambivalence is likely to spread from the collective to individuals, and particularly from leaders to followers, through both cognitive and emotional mechanisms. 1459 Because senior managers act on behalf of subunits and organizations, their experience of ambivalence implicates entire collectives (cf. Staw and Sutton 1993). Sensegiving enables managers to shape the way that individuals interpret and think about the context, which is perhaps the most important role that leaders play in times of change (Gioia and Chittipeddi 1991, Maitlis and Lawrence 2007). Sensegiving is generally thought to be a deliberate and even strategic process of communicating meaning downward (e.g., framing, decoupling; see Fiss and Zajac 2006). There are two mechanisms through which managers may sow ambivalence downward: intentional and unintentional. First, following the notion of “strategic ambiguity,” managers may intentionally convey mixed or at least equivocal messages as a means of stimulating change by implicitly challenging the status quo in a relatively nonthreatening manner (Eisenberg 1984, Gioia et al. 2012). This literature argues that equivocality enables actors to perceive “the” message they personally favor, thereby producing ideographic rather than holographic ambivalence across actors with different preferences (see endnote 3 and, e.g., Sonenshein 2010). However, we will argue later in our discussion section that managers might intentionally sow ambivalence to provoke actors into thinking more dialectically and acting more wisely. Second, managers may sow ambivalence by unintentionally conveying mixed or at least equivocal messages. Even where managers attempt to buffer followers from the managers’ own experience of ambivalence by delivering clear and unequivocal messages (Smircich and Morgan 1982, Smith et al. 2010), managers’ ambivalence may nonetheless “leak out” through various verbal and/or nonverbal cues. For instance, Larson and Tompkins (2005) found that senior managers of an aerospace company sent mixed messages about the desirability of a shift in corporate values from technical excellence to efficiency. Specifically, the managers continued to reward employees for technical excellence while propounding the overriding need for efficiency (at the expense of technical excellence), thereby sowing their ambivalence throughout the company and implicitly encouraging employee resistance to change. Collectives and, again, particularly leaders, may also spread ambivalence through emotional processes. Research clearly indicates that, whether intentionally or not, leaders serve to “model emotions” (Pescosolido 2002, p. 593), such that their expressed affect influences the affective experience of followers (e.g., Ashakanasy [sic] and Humphrey 2011, Erez et al. 2008). As with bottom-up emergence, the modeling may occur directly via emotional contagion or more indirectly through social information processing and comparison of emotional cues. In the above example of the aerospace company, Larson and Tompkins (2005, p. 13) note how management’s expressions of “collective pride” in the company’s technical excellence helped fuel employees’ ambivalence regarding the change from technical excellence to efficiency. Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach In sum, as arrows 1–5 in Figure 1 indicate, there are four major organizational triggers of ambivalence affecting both the individual and collective levels of analysis. In turn, the individual’s experience of ambivalence may spread to the collective level via sensemaking and emotional processes (arrow 5 in the upward direction), and the collective’s experience of ambivalence may spread to the individual level via sensegiving and emotional processes (arrow 5 in the downward direction). Accordingly, the experience of ambivalence at one level is likely to be positively correlated with the experience of ambivalence at the other level. How Do Individuals and Collectives Respond to Ambivalence? Consistency theories such as cognitive dissonance theory (Festinger 1957) and balance theory (Heider 1958) indicate that ambivalence tends to be aversive—that is, that the sense of being simultaneously attracted to and repulsed by an object is typically discomforting. Thus, Harrist (2006) finds the experience of ambivalence to be characterized by “disorientation,” described as a sense of disequilibrium, confusion, apprehension, and loss of control, where it feels wrong to have more than one orientation toward an object. We define a response as a conscious or nonconscious effort to reduce the intensity of ambivalence experienced, whether through addressing the underlying causes or the symptoms. A given response may involve cognition, affect, and/or behavior. Although our model of responses can be applied to cases of low- or high-intensity ambivalence, our discussion—as noted—focuses on the latter precisely because they are more problematic for the actor. Intensity refers to the extent to which the actor experiences opposing orientations as strong. Intensity is associated with issues that matter to the actor. The more intense ambivalence becomes, the more the actor will be motivated to take action to reduce the discomfort. Generally speaking, intense ambivalence is salient to the actor, where salience is the extent to which the actor is consciously aware of the opposing orientations. Intensity tends to trigger salience (the discomfort provokes awareness), and to a lesser extent, salience may reinforce intensity (awareness of a contradiction is itself disturbing). Most ambivalence research in psychology—and the goal in lab studies—focuses on at least moderately intense and highly salient ambivalence. However, it is possible to experience ambivalence that is intense yet not be aware of it. This is the primary focus of Freudian theories of defense mechanisms, where actors struggle nonconsciously to keep intense but socially undesirable thoughts and feelings from breaking into consciousness (Baumeister et al. 1998, Cramer 2006). Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS Horney 1945, Katz and Glass 1979, Tracy 2004; see Pratt and Doucet 2000 for a review). However, what is sorely lacking is a clear conceptual framework for organizing these disparate responses to ambivalence and applying them systematically to organizational contexts. Such a framework will clarify how various responses are (or are not) related and will identify gaps in what has been studied and, accordingly, opportunities for future research. In this section, we offer an integrative framework to explain when and why certain responses are most likely to occur in organizations (see Figure 2). Prior work that was most instrumental to our framework includes Horney’s (1945) very basic typology of moving toward/away from/against the object of ambivalence (see also Pratt and Doucet 2000) and research on interpersonal conflict-handling styles (Blake and Mouton 1964, Rahim 1985, Thomas 1992). Figure 2 decomposes the response to ambivalence into a focus on the positive orientation (i.e., attraction) toward the object and a focus on the negative (i.e., repulsion) orientation, as reflected by the two axes. The axes are partitioned into “low” and “high” for pedagogical purposes, suggesting four more or less distinct responses to ambivalence. First, avoidance, similar to “moving away,” involves a low focus—that is, little emphasis—on each orientation. Second, compromise involves a moderate focus on each orientation. Third, domination, similar to “moving toward” and “moving against,” involves a high focus—that is, much emphasis—on one orientation and a low focus on the other; thus domination appears twice in the figure. Fourth, holism involves a high focus on each. Not surprisingly, collective-level ambivalence tends to motivate collective responses and individual-level ambivalence tends to motivate individual responses (see arrows 6 and 7 in Figure 1). That is, responses are generally more likely to arise at the level where the ambivalence occurs, and such responses are generally more effective because they address the commensurate triggers. Figure 2 Negative orientation toward X Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. 1460 High Actor Responses to Ambivalence Holism Domination Compromise Low Avoidance Domination Low High Positive orientation toward X A Framework of Responses to Ambivalence Research in psychology and, to a lesser extent, sociology offers many responses to ambivalence (e.g., Coser 1966, Note. The axes reflect the focus or emphasis that the actor places on the orientations in responding to ambivalence (e.g., in avoidance, the actor places little emphasis on either axis). Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach 1461 Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS However, collective ambivalence may also motivate the cross-level effect of individual responses (arrow 8 in Figure 1) if (1) collective responses are absent or ineffective, or (2) individuals prefer an idiosyncratic approach (e.g., engaging in avoidance while one’s peers engage in holism). Conversely, individual ambivalence on its own is unlikely to motivate a collective response precisely because the ambivalence is confined to the individual level. As noted, collective responses will be most likely to occur if individual ambivalence becomes diffused to the collective level. Finally, collective and individual responses likely interact as collective responses provide a context for any subsequent individual responses (e.g., an individual experiencing residual ambivalence may enact an additional response), and individual responses to collective ambivalence may facilitate, through social learning processes (Bandura 1977), a more collective response (bidirectional arrow 9 in Figure 1). Each of the four responses—avoidance, dominance, compromise, and holism—is considered below. It should be noted that the authors of the studies we cite did not use our terminology; we are inferring the existence of specific responses, as well as, at times, ambivalence itself, from the descriptions the authors provide. Our treatment of holism will be more detailed because it is less well understood and more complex. We consider the nature of each response to ambivalence and likely positive and negative outcomes for the individual and collective experiencing ambivalence. Additionally, we consider how other actors may view any behaviors that accompany the response. At first blush, it would appear that holism is the optimal response as it seems to signify a “win–win” approach in the sense that both orientations become fully integrated. However, in addition to arguing that there are both positive and negative outcomes associated with each response, we argue that each response is more effective under certain circumstances (cf. Rahim 1985). Furthermore, because conscious and nonconscious processes tend to be intertwined, it should be noted that the responses may be enacted consciously, nonconsciously, or both. To embed the discussion in organizational contexts, we focus on cases where the ambivalence is attributable to the organizational triggers discussed earlier. We assume that the triggers lack a ready pragmatic solution or that the actor opts out of such a solution (e.g., quits the job). Given the triggers, many issues are inherently difficult to resolve (e.g., awareness of an object’s multiple facets). We also assume, in view of the proximal contextual root of opposition that underlies the triggers discussed above and based on our reading of diverse examples of ambivalence in organizations, that a given trigger is amenable to a variety of responses. That is, specific triggers (e.g., dualities, multifaceted jobs) do not necessitate specific responses. That said, we will return to these assumptions in our discussion section. Avoidance Nature of the Response. If intense ambivalence is by definition strong, how is it possible that an individual or collective can maintain a low focus on both the positive and negative orientations to the object? The answer, we contend, is through the use of defense mechanisms and coping mechanisms. Following Cramer (1998, p. 920; 2006), defense and coping mechanisms are means by which actors protect themselves from the “emotional consequences of adversity.” Defense mechanisms are nonconscious (in terms of awareness and effort) and unintentional, whereas coping mechanisms are conscious and intentional.5 Coping mechanisms focus on resolving the problem (what Lazarus and Folkman 1984 refer to as “problem-focused coping”) and/or alleviating tension (“emotion-focused coping”). Because defense mechanisms and the emotion-focused coping mechanisms utilized in this cell of Figure 2 enable the actor to avoid directly dealing with the ambivalence and perhaps even acknowledging it, this cell is labeled “avoidance.” As such, avoidance is typically reactive rather than proactive. Perhaps the most commonly used avoidance defense mechanisms for warding off ambivalence are denial and splitting. Denial protects the actor from threat by actively but nonconsciously rejecting, reinterpreting, forgetting, or minimizing disagreeable information (Baumeister et al. 1998, Sincoff 1992). Argyris (1990, p. 30) asserts that most organizational cultures actually facilitate and reward defensiveness because norms favor care for people within the organization over facing tough issues that could upset or offend them, such that “defense routines are protected and reinforced by the very people who prefer that they do not exist.” In splitting, the actor nonconsciously separates the positive and negative orientations so that opposition is not perceived. For example, Rowe et al. (2005) note that poor team performance may induce sports teams and their fans to be ambivalent about the team (“I love my team but I hate them for doing so poorly”). This ambivalence is often avoided by splitting the admired qualities of the team from the poor performance and attributing the latter to other causes, often the coach (cf. scapegoating; see Wexler 1993). The Boston Red Sox collapse at the end of the 2011 season, and the subsequent dismantling of the general management team, provides a good illustration of such dynamics (Brown 2011). Perhaps the most commonly used coping mechanisms in organizations for shielding the actor against the discomfort associated with marked ambivalence are suppression, escape or distraction, and emotional release. Because these mechanisms are commonly associated with all manner of stressors, our discussion will be brief. Suppression involves a conscious effort to dismiss anxiety-provoking thoughts about the issue in question. A manager facing ambivalence about his work role because of work–home conflicts (be at work versus be at home) may willfully Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. 1462 disattend to the conflicts. Escape or distraction involves diverting one’s attention from the problematic issue. Finally, emotional release involves venting the aversive feelings arising from ambivalence. For example, Rosen (1988) describes how the members of an advertising agency vented the negative emotions associated with their ambivalence toward the firm by performing skits at the firm’s Christmas party; the skits lampooned senior management and the firm. Positive and Negative Outcomes. Avoidance may appear to be a dysfunctional response to ambivalence in organizations. Because avoidance reduces the immediate tension associated with ambivalence, it may prevent the actor from even recognizing the ambivalence, thereby inhibiting learning and problem solving (Argyris 1993). Also, avoidance may prove futile in any event because unwelcome thoughts are not easily denied or suppressed (Wegner 1989), and focusing on unwelcome emotions through emotional release may only render them more salient (e.g., Brown et al. 2005). Thus, the defense mechanisms of denial and splitting have been labeled “immature” defenses (e.g., Andrews et al. 1993, p. 252), and emotion-focused coping mechanisms have been found to predominate when an actor believes that a stressor must be endured rather than constructively managed (Folkman and Lazarus 1980). However, there appears to be a functional side to avoidance in organizations. Although avoidance does not address the root cause of the ambivalence, it may nonetheless be effective under two conditions. The first is where the actor has low discretion or agency and thus cannot forge a more integrative response, if warranted, as in compromise and holism. The second condition is where immediate action is not required or the opposing orientations are not actively in conflict. Keeping tensions at bay through avoidance may be all that is truly needed. For example, nurses who smoked were ambivalent about their responsibility for advising patients against smoking. When a patient’s need for such advice appeared low (e.g., the malady was unrelated to smoking), some nurses tended to avoid advising patients against smoking (Radsma and Bottorff 2009). Defense and coping mechanisms help reduce tension to a more tolerable level, enabling an actor to persevere more effectively. Indeed, it appears that most actors continue to perform effectively even though ambivalence is more the norm than the exception in various domains of life, presumably including organizations (Smelser 1998). Furthermore, defense and coping mechanisms may facilitate subsequent compromise and holism by reducing debilitating tension while preserving the ambivalent qualities. For instance, Schneider (2003, p. 33) argues that splitting, by preventing “premature closure,” creates a “generative space” where actors have an opportunity to confront and work through the tension. Finally, because avoidance is largely signified by the absence of overt behaviors, other actors who are unaware Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS of the issue prompting the ambivalence will have little reason to make attributions about the focal actor. When other actors are aware of the issue and regard it as important, avoidance—as the absence of action in the face of threat—is likely to prompt negative attributions (e.g., “that group is in denial”). Domination Nature of the Response. Domination is a defense mechanism (i.e., a nonconscious process) and/or coping mechanism (i.e., a conscious process) through which actors bolster one orientation so that it overwhelms the other. Domination is thus characterized by a high focus on one orientation and a low focus on the other, and therefore it appears in two quadrants of Figure 2. Domination does not mean that all of the conflicting thoughts and feelings necessarily dissipate, only that the actor is able to choose one orientation—positive or negative thoughts/feelings— over the other (Harrist 2006). Similar to avoidance, given the apparent emphasis on defense mechanisms over coping mechanisms, domination is typically reactive rather than proactive. Response amplification (cf. reaction formation) is the primary nonconscious form of domination, where one essentially exaggerates either the positive or negative orientation to the object so as to override or “dominate” the other orientation and thereby relieve ambivalence (Baumeister et al. 1998, Bell and Esses 2002, Katz and Glass 1979). For example, ombudspersons handle the organizational ambivalence of (1) advocating for the best interests of employees who file complaints while (2) maintaining the status quo of organizational operations. Ombudspersons studied by Kolb (1987) tended to embrace one set of norms or the other, resolving the ambivalence by being either a “helping” ombudsperson or a “fact-finding” ombudsperson. It is important to note that domination via response amplification does not necessarily mean that the actor has permanently erased the opposite pole; rather, it may persist nonconsciously (Kets de Vries and Miller 1984). Response amplification may also be part of a nonconscious macro-oscillation that occurs over time, representing a shift from one domination cell in Figure 2 to the opposite domination cell, such that one experiences an entire paradigm shift. Kets de Vries and Miller (1984) provide an example. An artistic director and an administrative director of an opera company seemed to be inseparable, never differed in opinions, and consistently presented a united front in board meetings. This zealous relationship persisted for two years, until the administrative director violated the authority of the artistic director. Despite the interdependence of their roles, the two men quit interacting, and the artistic director vilified his colleague until the board fired him. The authors suggest that the relationship between the executives was not perfect prior Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS to the falling out, implying some ambivalence toward the relationship. However, the negative feelings were forcefully held in abeyance, eventually gaining enough momentum to shift the ostensibly positive relationship to its opposite. The change from extremely positive to extremely negative represents a paradigm shift around the fulcrum of ambivalence. As the saying goes, there is often a thin line between love and hate. Intense ambivalence may also lead to conscious domination processes—that is, to coping mechanisms. Having acknowledged both orientations prior to making a decision, conscious domination appears to be a way of ignoring the importance of one orientation rather than ignoring its existence. Conscious domination can be as simple as actively deciding to commit to one extreme and dismiss the other, if only for the sake of relieving discomfort. For instance, Bridge and Baxter (1992) note the oppositional norms involved when one is both a supervisor and a friend toward the same individual, including impartiality versus favoritism, evaluation versus acceptance, and confidentiality versus openness. A supervisor may cope with the resulting ambivalence by privileging one role over the other. A supervisor interviewed by Zorn (1995, p. 136) stated, “What Sally and I would always do is I’d say ‘All right, I’m talking to you Sally-friend not Sally-supervisor.’ ” An example at the collective level of analysis is provided by Tracy’s (2004, p. 125) study of a correctional facility, where the norms “create ongoing ambivalence—between kindness and control, flexibility and rigidity, solidarity and autonomy.” As a result, the facility’s leaders encouraged officers to manage their ambivalence by downplaying some dimensions (e.g., kindness) and emphasizing others (e.g., control). Research indicates that ambivalence may prompt information processing that is biased toward whatever orientation is initially stronger or more salient (Clark et al. 2008, van Harreveld et al. 2009a). Thus, although domination is expedient, it may short-circuit detailed consideration of the foregone orientation. Indeed, research suggests that ambivalent attitudes are more susceptible than nonambivalent attitudes to a variety of (often specious) primes, including mood, recent cues about the object, and consensus information (i.e., what others think) (Bell and Esses 1997, Hass et al. 1991, Hodson et al. 2001). It is as if the actor is looking for a way out, however specious the reasoning. Furthermore, it seems likely that conscious domination is very prone to rationalization because the actor may need a reassuring justification to bolster the choice (cf. neutralization theory; see Sykes and Matza 1957)—particularly if the actor is accountable to others (Tetlock 1985). Through rationalization, actors can convince themselves that one orientation is superior to the other or trivialize an orientation until the favored one can be enacted with greater ease (van Harreveld et al. 2009b). We acknowledge that domination is not the only process in Figure 2 where 1463 there may be a felt need to rationalize, but the need here appears particularly acute. In a study of “necessary evils” (i.e., tasks that cause harm to others to serve a greater purpose), Margolis and Molinsky (2008) observed the behaviors of ambivalent employees, such as police officers tasked with evicting tenants from their homes. Faced with the tension between what they considered as irreconcilable opposites—the obligation to do their job and the aversion toward causing hardship for those who could not pay their bills—some officers disengaged from the tenants, rationalizing the need for evictions (e.g., “Well, they [the evictees] put themselves in this situation. And what about the landlord 0 0 0?”; p. 856). Positive and Negative Outcomes. We argue that, in general, domination is most effective under one or more of the following conditions. The first is when the core of both the positive and negative orientations does not need to be preserved (e.g., having two equally desirable but mutually exclusive potential locations for a new plant, which creates ambivalence toward each alternative). In this case, decisive action is appropriate as it enables actors to relieve the intense ambivalence and align their thoughts and feelings with their behavior. The second condition is when the actor must choose between mutually exclusive orientations (e.g., hiring or not hiring a candidate for which there are mixed feelings)—and that is precisely when ambivalence becomes particularly intense (van Harreveld et al. 2009b). The third condition is when one orientation is actually counterproductive. For example, intense ambivalence could be caused by a need to complete a work-related project coupled with a fear of failure. Focusing on task completion while neutralizing the fear would provide an effective response to the ambivalence. That said, the rejected orientation may rear its head in unexpected ways (even when chased out of conscious awareness); as Lewis (2000, p. 763) notes, rejecting “one side of a polarity intensifies pressure from the other.” The final condition is when, as with avoidance, the actor has low discretion or agency and is thus less able to forge a more integrative response, if desired. Conversely, domination can be very dysfunctional in situations that require maintenance of both orientations. For instance, tension in the mission of a natural food cooperative required the simultaneous enactment of pragmatic business values and idealistic values that set the co-op apart from other grocers (Ashforth and Reingen 2014). To stay afloat, neither the pragmatic nor idealistic values could dominate over time. Domination would be a dysfunctional response that either destroys the co-op’s mission (if pragmatic values prevail) or bankrupts the coop (if idealistic values prevail). Additionally, domination may be dysfunctional when the decision is biased toward whatever orientation is initially stronger or salient, as noted above, or the decision is based solely on relieving discomfort. Gray (1999) suggests that choices made under Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. 1464 negative emotional states are often biased toward the alternative that will cause the least discomfort at the time. Taking the path of least resistance may seem desirable in the moment but has negative long-term consequences, particularly if the relative benefits of the chosen response are artificially exaggerated through response amplification. Because domination tends to be signified by decisive action (choosing one orientation over the other), other actors who are unaware of the issue prompting the ambivalence are likely to draw positive conclusions (e.g., “she provides clear direction”). However, where other actors are aware of the issue and see merit in both orientations, this positive conclusion is likely to be tempered and perhaps even reversed by concerns over the focal actor’s judgment (e.g., “she’s overly simplistic in her thinking”). This tempering or reversal is very significant because it suggests that what appears quite positive to the uninformed appears quite the opposite to those who better understand the situation. Compromise Nature of the Response. Compromise is associated with a moderate focus on both the positive and negative orientations (see Figure 2). Compromise involves reciprocal concessions between the opposing orientations such that they are mutually accommodated. It can be characterized as a coping mechanism in that it is typically a conscious and intentional process, whereby one acknowledges the simultaneous existence of the orientations and recognizes the desirability of partially honoring each. In this way, compromise—especially the form we refer to below as “black and white”—is more akin to “problem-focused coping” than “emotion-focused” as the focus is on responding to ambivalence by dealing with the cause rather than regulating the associated tension (Lazarus and Folkman 1984). Thus, although compromise can be either proactive or reactive, it tends to be more proactive than avoidance or domination. As a discrete action, compromise can take one of two forms. First, actors can “average” or combine the black and white into gray such that a middle ground is found that is not really true to either black or white. Such “gray compromises” are common and diverse in organizations; two examples will suffice. For one, individuals experiencing ambivalence about a task may self-handicap when they want to succeed but doubt their capability. Self-handicapping enables one to externalize failure by deliberately acting in ways that inhibit performance (Berglas and Jones 1978). An employee preparing for a presentation may procrastinate, making it impossible to prepare thoroughly. By telling coworkers that she prepared the slides just minutes before the meeting, she has an excuse if the presentation does not go well. Ambivalence was thus reduced by neither completing the task as well as possible (black) nor delegating the task to a coworker Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS (white). As another example, Weick (1979) explains a fire captain’s choices in responding to a call from an area noted for false alarms. If the captain compromises by sending all the firefighters, but at a slower speed than usual, the response is inappropriate whether there is a fire or not. If there is a fire, the firefighters will not arrive soon enough; if there is not a fire, all the resources have been needlessly sent to the false alarm. As these examples illustrate, gray compromises tend to be problematic when at least the core of each orientation needs to be preserved. A second type of compromise is where neither orientation is sacrificed, nor are the orientations wholly honored to the point where they are incompatible. This can be termed a “black and white compromise.” To illustrate, the fire captain can act as if there is a fire and as if there is not by very promptly sending only some crew members. If there is a fire, the crew has arrived as quickly as possible; if there is not a fire, all of the resources were not wasted (Weick 1979). As this example illustrates, the actor simultaneously respects the core of each orientation, albeit in a give-and-take form. Black and white compromise may also occur sequentially in the form of vacillation, where the actor alternates orientations based on the situation (Pratt and Doucet 2000, Tracy 2004). These short-term fluctuations differ from the macro-oscillation discussed under “domination,” where one experiences an entire paradigm shift. The short-term fluctuations, considered as a set, respect both orientations, even if each alternation seems to abandon one orientation for the other. Pratt and Dutton (2000, p. 117) found that librarians were ambivalent about homeless people in their library, feeling “compassion, sympathy, and empathy as well as anger, fear, and disgust.” The library’s director gave money to homeless patrons at times and asked them to leave at other times; one librarian would sometimes call social services to request help for the homeless and sometimes call the police to remove them from the library. As an example at the collective level, Wells (1988) notes that the staff at a Girl Scout camp experienced ambivalence because they identified with the organization but were put off by the wholesomeness of the Girl Scout identity. As a result, when working in public they strove to display a professional image, but behind closed doors they distanced themselves from that identity by resorting to humor and derision about the Girl Scout symbols. We posit that it is precisely because the librarians and Girl Scout staff opted for a negative orientation at one point that they were able to opt for a positive orientation at a later point, and vice versa (Ashforth and Mael 1998; see also Kosmala and Herrbach 2006, Kunda 1992). In this case, long-term compromise was attained via short-term vacillation. Positive and Negative Outcomes. As suggested above, compromise is generally most effective when neither the positive nor negative orientation alone is sufficient (gray or black and white compromise) or the core of each Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS orientation needs to be preserved (black and white compromise). However, opposites often do not lend themselves readily to negotiation; thus the process of compromise may prove difficult and actual resolutions may be hard to discern and enact. Compromise is therefore also more effective when the actor has at least moderate discretion or agency. For example, some Benedictine sisters identified with the Catholic Church but felt excluded by its male-dominated rituals, creating tension between feelings of belonging and exclusion. In response to the resulting ambivalence toward the church, the sisters chose to modify their practice of certain church-mandated rituals so that they were more inclusive of women (Hoffman and Medlock-Klyukovski 2004). As further noted, black and white compromise can also occur over time by vacillating between two orientations. However, as Merton (1976, p. 105) puts it, “This delicate process of alternating phases is a little like walking a tightrope, with all its attendant risks and successes.” If the core of black and/or white matter and are lost, the compromise becomes dysfunctional. To recognize a compromise, other actors must be aware of the issue that prompted the ambivalence. Thus, actors who are unaware of the issue will have little reason to draw inferences about the focal actor. Where other actors are aware of the issue, we posit that compromise is sufficiently normative in organizational settings that they are likely to draw positive inferences (e.g., “the team is realistic”). However, given organizational norms for consistency (Staw 1980), the vacillating form of compromise may induce others to view the actor as inconsistent, indecisive, and perhaps even deliberately hypocritical, particularly if the vacillation is prompted by nonconscious—and therefore difficult to justify— impulses (cf. Rothman and Wiesenfeld 2007). Rothman and Wiesenfeld (2007) add that such inferences may elicit frustration in others and inhibit liking and trust. Holism Nature of the Response. Unlike compromise, where the actor negates (gray) or diminishes (black and white) the integrity of each orientation, or domination, where only one orientation is given expression, holism involves the complete, simultaneous, and typically conscious acceptance of both opposing orientations.6 As such, holism can usually be characterized as a coping mechanism. Holism is in the same spirit as black and white compromise but represents less of a win–lose trade-off between the orientations and more of a win–win embracing of both orientations. Accordingly, holism tends to be proactive and generally more proactive than compromise. The notion of holism raises the question: Given intense ambivalence, how can an actor consciously retain it and thus “tolerate” it? The answer, we contend, is through the exercise of wisdom (Kessler and Bailey 2007). 1465 Weick (1998, 2004) writes of an “attitude of wisdom” where actors balance confidence with doubt, ready to act as if they know and yet as if they do not know. Cast in the terminology of ambivalence, “knowing” and “doubting” are opposite orientations held simultaneously. For example, an attitude of wisdom may involve treating “past experience as a guide as well as a trap” (Weick 2004, p. 662). More generally, we view wisdom as the capacity, in part, to simultaneously acknowledge and embrace opposing orientations, and thereby strive for a course of action that honors both (cf. Clayton 1982, Martin 2009, Weick 1998). This suggests that holding onto the whole of ambivalence may actually be functional. This view of wisdom shows the promise of ambivalence, but it sheds little light on how wisdom facilitates a holistic response to ambivalence. As with the other responses to ambivalence, we argue that holism can be enacted at both the individual and collective level. However, because wisdom as a holistic response has received the least attention in the literature relative to other responses, we address separately the individual responses of “wise actors” and the collective responses of “wise systems.” Once delineated, we then show how the two levels of responses may interact (bidirectional arrow 9 in Figure 1). Individual responses of wise actors: How, specifically, do wise actors facilitate a holistic approach to ambivalence? Building on our reading of the literature, we suggest three related ways: (1) mindfulness, (2) “both/and” thinking, and (3) informed choice. Dane (2011, p. 1000; cf. Shapiro and Carlson 2009) defines mindfulness as “a state of consciousness in which attention is focused on present-moment phenomena occurring both externally and internally.” Fiol et al. (2009, p. 44) add that “[m]indfulness results from a reluctance to simplify”—a willingness to embrace complexity. Most important for our purposes, Fiol et al. (2009, p. 44) further note that “[m]indfulness has three key characteristics: (1) the creation of new categories of meaning, (2) openness to new information, and (3) an implicit awareness of multiple perspectives.” We noted that holism tends to be the most proactive of the responses to ambivalence. A mindful actor is more likely to be actively aware of the external and/or internal stimuli that have fostered or made salient opposing orientations along with the experience of the ambivalence itself. For example, through a series of experiments, Rees et al. (2013, p. 367) found that emotional ambivalence increased accuracy in various judgments by increasing participants’ receptivity to others’ perspectives, concluding that emotional ambivalence can produce a “wise mind.” Mindfulness facilitates both/and thinking, described in detail below, by making the issue conscious or at least attuning actors to what may otherwise remain nonconscious (Dane 2011) and by inhibiting the often reflexive narrowing of attention (Fiol and O’Connor 2003, Weick and Putnam 2006) that is the hallmark of domination. Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. 1466 In addition, mindfulness may facilitate actions that address the negatively viewed aspects of the object in question so that the opposing orientations are better able to be meshed. Meyerson (2001) describes a senior vice president of a financial firm who was ambivalent about the privileges of office, enjoying her own but perceiving that the distribution of the privileges was unfair to other women and minorities. Unlike other, presumably less mindful, executives, she responded to the entreaties of working parents by offering more flexible work arrangements to accommodate their family obligations, and the firm eventually institutionalized such arrangements. Consistent with the awareness enabled by mindfulness, Meyerson describes various behaviors through which actors may approach ambivalence holistically, from “turning personal threats into opportunities” to “leveraging small wins” to “organizing collective action” (p. 8). The second way that wisdom facilitates a holistic approach to ambivalence is through both/and rather than either/or thinking. Either/or thinking refers to a tendency to separate the orientations, creating a dichotomy (e.g., a selection committee glorifies a job candidate for the high depth of her industry experience or demonizes the candidate for the low breadth of her experience across industries), whereas both/and thinking refers to a tendency to juxtapose the orientations, potentially resulting in perceived complementarity and even synergy (e.g., the committee recognizes that the low breadth made the high depth possible; see, e.g., Cameron et al. 2006, Chen 2002, Jubas and Butterwick 2008). Both/and thinking involves developing a more comprehensive understanding of the opposition at the heart of ambivalence—a synthesis of sorts of the “thesis/antithesis” that the opposition represents, a transcendence of the particulars. However, in holism the thesis and antithesis do not disappear; holism thus enables the actor to appreciate that the aspects of the object that engender the opposing orientations are integral to the object and perhaps even desirable (cf. Lewis 2000; cf. Janusian thinking, Rothenberg 1979). For example, the literature on organizational ambidexterity suggests that when actors embrace the apparent contradiction between the need for exploration and the need for exploitation— and the resulting ambivalence toward each—more nuanced and jointly optimizing actions can result (e.g., Raisch and Birkinshaw 2008, Smith and Tushman 2005). As F. Scott Fitzgerald (1945, p. 69) famously stated, The test of a first-rate intelligence is the ability to hold two opposed ideas in the mind at the same time, and still retain the ability to function. One should, for example, be able to see that things are hopeless and yet be determined to make them otherwise. Research suggests that ambivalence motivates heightened information processing (Jonas et al. 1997, Maio et al. 2001). Through both/and thinking, the actor recognizes the integrity underlying the aspects of the object that give Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS rise to ambivalence. That is, the actor may respond to the whole object, not just to its favored aspects (as in domination). Furthermore, the actor does so in a manner that honors that integrity, rather than betrays animus toward its constituent parts. Plambeck and Weber (2010, p. 694) find that CEOs who were the most ambivalent about the enlargement of the European Union engaged in “strategic ambidexterity,” where the orientation of their firms was simultaneously offensive and defensive, increasing the spectrum of information considered in strategic actions (see also Tang et al. 2010). Similarly, Lüscher and Lewis (2008) describe an intervention for production managers at Lego Company in the wake of a major restructuring from a traditional hierarchy to self-managing teams. Rather than be paralyzed by the resulting dilemmas that invited ambivalence (e.g., empowering teams versus controlling them), managers were encouraged to embrace the dilemmas and forgo simplistic either/or thinking in favor of both/and thinking. Lüscher and Lewis conclude, “Such awareness was empowering, reducing tendencies to blame executives [for sending mixed messages] and shifting responsibility to the managers to find means of living with tensions” (p. 234). As these examples suggest, both/and thinking in the context of holism does not mean that the negative orientation is necessarily converted into a positive orientation, only that the actor is able to make sense of the object in such a way that the totality of the object and the ambivalence it spawns is accepted and perhaps even prized. This also does not mean that the discomfort that attends ambivalence is necessarily reduced, given that the opposing orientations are actively sustained. Rather, it speaks to the actor coming to terms with the experience of ambivalence. To illustrate, Wright (2009) describes the ambivalent organizational identification of internal organizational consultants. The consultants simultaneously felt both attached as employees of the organization and detached as objective evaluators of the organization. Some came to regard their unique position of being “outsiders within” as a “positive source of strength and distinctiveness” (p. 319). The third way that wisdom facilitates holism is through informed choice. Research on commitment (Brickman 1987, Pratt and Rosa 2003) and trust (Pratt and Dirks 2006) illustrates holism as a choice made by wise actors. Brickman (1987) suggests that commitment is the “antidote” to ambivalence. How he views commitment, however, is much different from behavioral/escalation approaches (Salancik 1977) or attitudinal approaches to commitment (see Pratt and Rosa 2003). In Brickman’s view, actors must first know the positive and negative “elements” (p. 7) associated with forming an attachment to a person, idea, or other object. These positive and negative elements are not divisible from the object. Pratt and Rosa (2003, p. 395), for instance, describe how network marketing firms “harvest” work–family conflict to transform Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS individuals’ emotional ambivalence about the firm into commitment. They deliberately recruit individuals who are at risk for work–family conflict and encourage them to frame work in terms of helping their family (e.g., “you’re doing this for your children”), to recruit their own family and friends to help with bookkeeping and distribution, and to avoid those who would “steal their dreams” (p. 411). For Brickman, commitment involves the binding of the positive (helping one’s family) and negative (sacrifice) elements through choosing a course of action. Indeed, Pratt and Rosa report that newcomers experiencing little emotional ambivalence ultimately had weak commitment to the organization. Pratt and Dirks (2006) make a similar argument about ambivalence and choice when discussing the concept of trust in organizations. Trust involves being vulnerable so as to achieve something positive in a relationship in the future (Rousseau et al. 1998). They suggest that trust involves choosing both a positive (future reward) and negative (vulnerability) element simultaneously. To be sure, sometimes the negative elements will be more salient (similar to “being committed” to an asylum) while at other times the positive elements will be more salient. But at its core, both positive and negative elements always remain. Thus, commitment as a holistic action differs from compromise or domination where one orientation is diluted or discarded. Furthermore, by choosing to accept the opposition, the actor transforms what may have been outside his or her control—such as opposing norms— into something that is experienced as more controllable (Harrist 2006, Meyerson 2001). In short, an actor may not choose to be exposed to ambivalence but can choose how to respond to it—to accept that ambivalence as an essential element for commitment or trust. For instance, Ashcraft (2001) describes the “organized dissonance” of a feminist bureaucracy that assisted victims of domestic violence. Faced with ambivalence about relying on a traditional hierarchy (hierarchy both facilitates and impedes the mission), members opted to embrace the ambivalence and utilize a hierarchy in a nontraditional manner with informal, egalitarian practices (e.g., consensual decision making, eschewing heavy-handed supervision). Collective responses via wise systems: A wise system is a set of practices, as embodied in informal (e.g., culture) and formal structure, for actively embracing and honoring positive and negative orientations toward an object. Wise systems keep ambivalence in play. Under both/and thinking, we quoted Lüscher and Lewis (2008, p. 234) that wise actors “find means of living with tensions.” Similarly, a wise system connects opposing orientations in a more or less ongoing manner to keep alive “creative tension” (Cameron 1986, p. 549) and thus the potential to honor the orientations in adaptive and novel ways as emerging circumstances warrant. Research on organized dissonance (Ashcraft 2001), organizational ambidexterity (Raisch and Birkinshaw 2008), organizational dualities (Ashforth and Reingen 2014), and 1467 organizational mindfulness (Farjoun 2010) provide examples of how organizations may eschew one-time responses to ambivalence in favor of an ongoing juxtaposition of the orientations fueling ambivalence. For example, Ashcraft (2001, p. 1315) writes of the need for periodic “group reflection as a safety valve that keeps contradiction alive.” Culture refers to a set of values, beliefs, norms, and assumptions that typify an organization or subunit (cf. Schein 2010). A culture conducive to holism is one that privileges certain values (e.g., openness, inquiry), beliefs (e.g., progress results from a full airing of both sides of an issue, the complexity of the decision making process should reflect the complexity of the issue), norms (e.g., encourage wide participation, have secondchance meetings to reconsider decisions), and assumptions (e.g., ambivalence is normal, reasonable group members can disagree). For instance, we mentioned Ashforth and Reingen’s (2014) study of a natural food cooperative, where ambivalence over business values and idealistic values was kept simmering. Numerous meetings were held to allow discussions of ambivalence-provoking issues, and certain rituals (e.g., appeals to collective interest, apologies for rudeness) helped regulate the resulting conflicts. Members described their decision-making process as decidedly “messy,” a term that we speculate generally describes wise systems. Structure refers to the design of the organization or subunit. Wise systems appear to facilitate holism through either differentiation (complemented with integration; cf. Lawrence and Lorsch 1967) or nondifferentiation. We noted earlier that organizations frequently embody hybrid identities and pursue contradictory goals, prompting ambivalence toward any one identity or goal. In differentiation, these identities and goals are assigned to (or emerge within) separate actors. For example, Raisch and Birkinshaw’s (2008) review of the literature on organizational ambidexterity suggests that one way organizations address the efficiency–innovation trade-off, and the ambivalence it triggers toward each, is by assigning each to separate subunits. Differentiation thus allows each subunit to pursue a particular goal, devoid of ambivalence. Similarly, a group may designate a specific member to play the role of devil’s advocate and represent the misgivings of other group members, thereby releasing the other members from the burden of mixed feelings (Smith and Berg 1987). As Lawrence and Lorsch (1967) argue, differentiation creates a need for integration to then reknit the actors together so that the collective can move forward as a more or less unitary body. Thus, differentiated organizations tend to create integration mechanisms, such as having senior management provide strategic integration of the efficiency- and innovation-oriented subunits (Raisch and Birkinshaw 2008) or having a matrix structure that intentionally juxtaposes a functional focus with a product focus (Galbraith 2009). In terms of ambivalence, then, differentiation and integration facilitate a holistic approach Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. 1468 by fostering an ongoing creative tension between the identities/goals. In these examples, the underlying cause of ambivalence at one level is “resolved” by activities at another level; indeed, what appears to be domination at the lower level (e.g., within a given subunit) may help foster holism at the higher level. In nondifferentiation, the identities/goals are jointly allocated to subunits or individuals so that actors must each struggle with the oppositions and hence the ambivalence toward any given identity or goal. As such, structures predicated on nondifferentiation tend to be less complex than those predicated on differentiation and integration. An example is for-profit hospitals where the medical staff is expected to jointly weigh the sometimes oppositional goals of patient service and cost considerations (Shore 1998). Note, however, that the structure per se does not actually facilitate actors’ responses to ambivalence; it simply makes ambivalence that much more intense and salient. Thus, to qualify as a holistic approach, nondifferentiation must be complemented with the other aspects of a wise system. Wise actors ←→ wise systems: The notions of wise actors at the individual level and wise systems at the collective level raise the question of how wisdom may play out across levels of analysis. What wise actors do may ultimately become institutionalized in the culture and structure of the wider organization. Wise actors, in short, can give rise to wise systems (arrow 9 in the upward direction in Figure 1; see Bierly and Kolodinsky 2007, Bierly et al. 2000). Martin (2009) describes how a sharp increase in research and development spending at Procter & Gamble (P&G) led to disappointing results and thus to ambivalence about the desirability of such spending (is it a good investment or a bad investment?). Helping to fuel the negative orientation to research and development spending was a belief among some senior executives that P&G was primarily a marketing company. CEO A. G. Lafley facilitated an effective approach dubbed “Connect & Develop,” which derived synergies from the positive and negative orientations, thereby “reconciling the irreconcilable” (p. 58). Specifically, Lafley set a goal that 50% of P&G’s innovations come from outside the firm, with P&G leveraging those innovations through its marketing and distribution muscle, thereby institutionalizing his own wisdom. By the same token, wise systems can promote and reinforce the wisdom of individuals. In short, wise systems can give rise to wise actors (arrow 9 in the downward direction in Figure 1). Following Sternberg (1998; cf. Bigelow 1992), wisdom is acquired by gaining tacit knowledge over time and is best developed through dialectical processes; clearly, the more ambivalent experiences actors have, the more opportunities they have to wrestle with such processes (in the form of oppositional tendencies). This suggests that a wise system, where ambivalence is actively kept alive, may provide the seedbed for actors Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS to become wise(r). For instance, Plambeck and Weber (2010) found that organizational characteristics that supply and facilitate diverse perspectives on a given issue, such as ambidextrous strategies and the functional diversity of the top management team, were more likely to have CEOs who reported ambivalence toward strategic issues. Although it is not clear what is cause and what is effect here, wise actors and wise systems can reinforce each other. Even though wisdom appears far more likely to result from a combination of wise actors in wise systems, our interpretation of examples of ambivalence in the literature suggests that it is possible for wisdom to result from wise actors working within an “unwise” system or from “unwise” actors working within a wise system. In short, the wisdom of the actors or of the system can compensate for its absence in the other. Whereas the popular press has many stories of heroic and often wise innovators, whistleblowers, and leaders salvaging faulty organizations, the notion of unwise actors in a wise system is less obvious. In Ashforth and Reingen’s (2014) study of a natural food co-op, advocates of the business values and advocates of the idealistic values tended to stubbornly deny and resist the views of their opponents. The authors liken the dynamics to the oppositional politics of Democrats and Republicans in the American political system (cf. Maybury-Lewis and Almagor 1989). In the co-op, wisdom was not a function of the actors but of the system that forcibly juxtaposed their views. Thus, as Bierly et al. (2000, p. 597) note, “[O]rganizations can act wisely even though it may not be possible to ascribe wisdom to any individual actor within the organization.” Can unwise actors in unwise systems nonetheless enact holism in response to ambivalence? It seems very unlikely. Given that wisdom is in part the capacity to simultaneously acknowledge and embrace oppositional orientations, it is difficult to see how an unwise actor in an unwise system would have the wherewithal to truly engage such orientations. That said, unwise actors in an unwise system may nonetheless be induced to attempt the processes that typify wisdom, as when external consultants try to restore an organization’s health (e.g., Lüscher and Lewis 2008), or may, through trial and error, discover such processes. Thus, unwise actors in an unwise system need not be a permanent condition. In sum, wise actors and wise systems are mutually reinforcing, as a holistic approach to ambivalence reinforces and further develops wisdom. Positive and Negative Outcomes. The discussion above suggests that holism is generally most effective when the full integrity of both orientations needs to be preserved, coupled with an actor who has sufficient discretion or agency—or a system with sufficient resources—to enact both the demanding process and the (typically) complex “resolution” of the underlying causes of the ambivalence. But how can individuals or collectives know when the full Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach 1469 Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS integrity of opposing orientations needs to be preserved? They may have to engage in holistic processes to even discern whether there is indeed an integrity that should and can be honored; therefore holism represents something of a leap of faith. That said, the desirability of full integrity is likely to be signaled before the fact by (1) the issue’s apparent significance (e.g., an issue that deeply affects an actor’s goals and functions, a persistent or recurring issue, an issue with key stakeholders aligned with each orientation) and (2) potential synergies that may be realized from simultaneously holding or even integrating the orientations (e.g., perceiving an opportunity to learn and innovate, recognizing that the positive orientation needs to be tempered by the concerns fueling the negative orientation) (cf. Martin 2009, Thomas 1992). Our discussion above also suggests the following positive outcomes of holism: (1) neither orientation is diminished; (2) growth in the actor’s understanding and appreciation of the object, fostering commitment to and trust in the object; (3) adaptability, that is, retaining the capacity to act in multiple and seemingly inconsistent ways, depending on the situation; and (4) acting as a change agent, that is, mindfully nudging the object (or system) in positive ways even while expressing the actor’s belief in the integrity of the object. However, a problematic feature of holism is that the requisite wisdom is very difficult to cultivate and sustain, and it is often unclear what specific behaviors should flow from wisdom in any event. Not surprisingly, then, Weick (1998) acknowledges that the pursuit of an attitude of wisdom—ambivalence that tempers knowing with doubting—can undermine confidence and impair action (see also Pratt and Pradies 2011). Furthermore, by actively keeping both orientations alive, holism may not relieve the discomfort that attends ambivalence. Finally, as with compromise, to recognize a holistic response, other actors must be aware of the issue that prompted the ambivalence. As noted, observers expect consistency and decisiveness from actors rather than expressions of ambivalence (Rothman and Wiesenfeld 2007). Thus, as with the vacillating form of black and white compromise, holism may result in behaviors that appear to others to be inconsistent in the short term— even if, when considered as a series over time, they are actually quite nuanced and respect the totality of the object. In short, because the effectiveness of holism is often only revealed over time, holism may lead to the actor being perceived as indecisive, inconsistent, or hypocritical, at least in the short term. For example, Cha and Edmondson (2006) studied a company whose CEO espoused employees’ best interests and being a family as key values. When the company more than doubled in size over three years, employees made hypocritical attributions about the CEO’s actions, viewing this growth as threatening their best interests (e.g., loss of camaraderie). However, in an interview with the authors, the CEO said he actually viewed growth, in part, as a vehicle for employee development and wage growth. This cautionary tale implies that the negative attributions arising from seemingly inconsistent actions can be mitigated and even reversed if the actor is able to clearly articulate a compelling need for holism. To summarize, Figure 2 depicts a conceptual framework for organizing actors’ responses to ambivalence in organizational settings and underscores that each response is associated with certain trade-offs that may make it, depending on the situation, functional or dysfunctional. The main arguments are summarized in Table 3. Discussion Ambivalence is defined as simultaneously positive and negative orientations toward an object. We developed a multilevel perspective on ambivalence in organizations that demonstrates how this phenomenon is integral to cognitive/emotional processes and important outcomes. Specifically, we argued that ambivalence results from oppositions (a demand for A and not-A) manifested in organizational contexts as hybrid identities, contradictory goals, role conflicts; dualities; multifaceted objects; and temporal factors. As displayed in Figure 1, these oppositions can trigger ambivalence at the individual level or collective level of analysis, which may diffuse to the other level through cognitive (sensemaking, sensegiving) and emotional (emotional contagion, emotions-as-socialinformation/emotional comparison) processes and prompt responses at each level (i.e., avoidance, domination, compromise, holism). Although previous research has identified various defense and coping mechanisms for responding to ambivalence, it focused on a limited set of the potential responses and lacked a conceptual framework for organizing these responses. Crossing two dimensions—a focus on the positive orientation toward the object in question and a focus on the negative orientation—enabled us to organize the better-known responses under the labels of avoidance (the focus on each is low), compromise (each is moderate), and domination (one is low and one is high) and to identify an intriguing but little-known response that we dubbed holism (each orientation is high). We argued that each response is applicable at the individual and collective levels of analysis and provided examples from the literature, although the original authors, of course, had not employed our framework. As summarized in Table 3, each response has positive and negative outcomes and is most effective under certain conditions. However, holism tends to be the most difficult to attain as it requires embracing both orientations simultaneously, suggesting the need for wise actors and/or wise systems. An image of the ambivalent and wise actor starkly contrasts with the popular image of an ambivalent actor as confused and indecisive (i.e., weak; see Rothman and Wiesenfeld 2007). Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach 1470 Table 3 Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS Actor Responses to Ambivalence: Description, Examples, Outcomes, and Conditions Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. Avoidance Domination Compromise Holism Description • The nonconscious and/or • The nonconscious and/or conscious evasion of the conscious bolstering of ambivalence caused by one orientation so that it the opposing overwhelms the other. orientations. • Typically reactive rather • Typically reactive rather than proactive. than proactive. Examples • Arribas-Ayllon et al. (2009) • Librarians felt humanistic • Internal organizational • Nurses who smoked consultants felt ambivalent discuss the ambivalence concern toward were ambivalent about because they felt among professionals homeless individuals their responsibility for involved in genetic who occupied their advising patients against simultaneously felt attached library but also felt smoking. When a need counseling who could see as insiders of the responsible for for such advice organization and detached the benefits but also appeared low, some as objective evaluators. eliminating disruptions to sensed moral danger in nurses tended to avoid Wright (2009, p. 319) found patrons. In response to the practice. In response that some consultants the resulting advising patients against to parents seeking a ambivalence, librarians smoking (Radsma and viewed their unique position genetic test for their child Bottorff 2009). vacillated in their of being both “insiders” without considering the • Tracy (2004) found that behavior (e.g., at times (employees) and “outsiders” ethical implications, the calling police to remove correctional facilities (roving consultants) as a professionals alleviated propounded “positive source of strength their ambivalence by citing the homeless, at other times calling social contradictory “extreme case scenarios” and distinctiveness.” agencies to provide (p. 183) that painted the • A natural food cooperative organizational norms help) (Pratt and Dutton regarding the behavior of parents as unreasonable. experienced tension 2000). officers toward inmates • Bowles (1991), interpreting between its idealistic values (e.g., nurture versus Kriegler (1980), describes • Strategic alliances (which supported the discipline). However, the the ambivalence that a involve certain dialectical organization’s mission) and group of shipyard its business values (which facilities’ cultures made tensions (e.g., managers felt from kept the organization cooperation– the complex tensions inflicting terrible work financially sound). competition) that are seem straightforward The organization kept the likely to provoke conditions on workers with and strongly resulting ambivalence in whom they empathized. To ambivalence (e.g., we discouraged officers reconcile the demeaning play by meetings that tacitly should cooperate and from asking questions, work they demanded of surfaced the tension, not cooperate). thereby promoting complemented with certain good people, the Accordingly, strategic avoidance of the rituals that helped regulate managers demeaned the alliances may opt for resulting ambivalence. the resulting conflicts (e.g., workers, allowing the “balancing” the opposing appeals to collective managers to see both the tensions (Das and Teng interest, apologies for work and workers 2000, p. 84). rudeness) (Ashforth and negatively. Reingen 2014). • The simultaneous and • A typically conscious typically conscious give-and-take between the opposing orientations acceptance of both opposing orientations. — Gray compromise • Typically proactive rather — Black and white than reactive. compromise. • Proactive or reactive. • Outcomes: Positive • Reduces tension to a • Maintains each • Alleviates tension, aligns more tolerable level, orientation. thoughts and feelings with enabling performance • May lead to positive behavior. (e.g., realistic, practical) and the other responses • May lead to positive (e.g., attributions by other to ambivalence. decisive, clear) attributions actors who are aware of by other actors who are the issue prompting unaware of the issue ambivalence. prompting ambivalence. • • Causes both orientations to be fully embraced, facilitating growth in understanding and appreciation of the object as well as commitment to the object. Facilitates adaptability. Enables acting as a change agent. • Reinforces wisdom. • May lead to positive (e.g., prudent) attributions by other actors if the actor articulates a compelling need for holism. Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach 1471 Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS Table 3 (cont’d) Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. Avoidance Domination Outcomes: Negative • May prevent recognition • Sacrifices integrity of both orientations. of ambivalence, • May be biased toward inhibiting learning and whatever orientation is problem solving. initially stronger or more • May prove futile; may salient, or is the least backfire and render uncomfortable, rather than ambivalence more the most effective. salient. • May lead to negative (e.g., • May lead to negative simplistic, reckless) attributions by other attribtuions by other actors actors if the issue is who are aware of the issue regarded as important. and see merit in both orientations. Conditions under which it is most effective • Immediate action is not required. • The actor has low discretion or agency. Compromise Holism • May be difficult to • May be difficult to implement the process, cultivate and sustain an and the outcomes may attitude of wisdom. be elusive. • Is often unclear what • May lead to negative behaviors should result. (e.g., inconsistent, • May not relieve hypocritical) attributions discomfort. about vacillation by other • May lead to negative (e.g., inconsistent, actors, absent hypocritical) attributions awareness of the issue by other actors, absent prompting ambivalence. an articulated need. • The core of both orientations • Neither orientation alone • The full integrity of both is sufficient (gray or does not need to be orientations needs to be black and white preserved. preserved. compromise), or the core • The actor has high • The actor must choose of both orientations between mutually exclusive discretion or agency (or, needs to be preserved orientations. in the case of wise systems, the system has (black and white • One orientation is actually sufficient resources). compromise). counterproductive. • The actor has low discretion • The actor has at least or agency. moderate discretion or agency. Implications for Future Research Our analysis suggests numerous possibilities for future research. We see the following five as particularly important. First, we argued that the specific organizational triggers of ambivalence are amenable to various responses, subject to the contingencies in Table 3 (see the “Conditions under which it is most effective” row). Future research should explore this argument. For example, perhaps certain triggers are more controllable by actors (e.g., hybrid identities versus a multifaceted job), encouraging more proactive and substantive responses. Or perhaps stakeholders hold normative expectations regarding the appropriate ways of addressing specific triggers, thereby limiting actors’ response repertoire. For instance, investors with a long-term stake in a corporation may have little tolerance for the use of avoidance in the face of the change–continuity, global–local, or quality–efficiency dualities. Second, our discussion of the four responses did not distinguish between the individual and collective levels of analysis. Indeed, each response has analogs at both levels of analysis, with the exception of specific tactics such as structural differentiation and integration, which cannot be enacted at the individual level. Importantly, though, it appears that ambivalence is easier to respond to (albeit not necessarily more effectively) at the collective level than at the individual level because members of the collective can help socially construct and sustain a reality that objective observers would disavow (e.g., Choo 1996, Hardin and Higgins 1996). Prime examples include the avoidance responses of denial and splitting (e.g., Smith and Berg 1987, Wexler 1993). Indeed, responses may become institutionalized—even if they are not particularly effective—such that they become taken-for-granted aspects of organizational life. Argyris (1985, 1990), for instance, writes of the reflexive defense mechanisms in organizations that routinely block desirable change. Future research should determine the applicability and boundary conditions of the various techniques for managing ambivalence at each level of analysis. Third, we identified holism as an intriguing but littleresearched phenomenon. Future research should focus on assessing and fleshing out our description of holism. In particular, we argued that wise actors embrace opposing orientations toward an object via mindfulness, both/and thinking, and informed choice. What synergies exist among these processes? What contextual factors are most conducive and least conducive to each practice? Are there additional processes through which wisdom is enacted? To what extent are holistic responses to ambivalence trainable and what specific practices might help cultivate wisdom? For instance, we noted that wise systems may help foster wise actors by actively keeping ambivalence alive and stimulating dialectical thinking. We described a wise system as a set of practices embodied in culture and structure for embracing opposing orientations. Are there particular sets of values, beliefs, norms, and assumptions that are necessary for a wise system, or is wisdom more of a “fuzzy set” of cultural properties? Differentiation and nondifferentiation suggest diametrically different Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. 1472 structural approaches for institutionalizing wisdom; under what conditions is each approach preferable? For example, because differentiation essentially removes the burden of ambivalence from single actors (by distributing the opposing orientations across actors), it is likely more accommodating of unwise actors than is nondifferentiation. Finally, we discussed how wise actors and wise systems are mutually reinforcing. Future research might also explore means of fostering and sustaining the wisdom of both actors and systems, such as through the institutionalization of external reviews, devil’s advocates, and training in counterfactual thinking (cf. remedies for groupthink and escalation of commitment; see, e.g., Ku 2008). Might it be possible to create reinforcing systems that create virtuous cycles of wisdom? Fourth, whereas we have treated the responses to ambivalence separately, future research should examine how they relate to one another. For instance, we mentioned that avoidance may reduce debilitating tension and thereby facilitate compromise and holism. Similarly, compromise may serve as a fallback if holism fails. Although the responses appear to generally serve as substitutes (e.g., domination may obviate the need for holism), a given response may prove inadequate at rendering the ambivalence less intense, thus provoking another response. In particular, one general response trajectory may be to move “up the ladder” toward typically more conscious, proactive, and effortful responses in the search for a resolution of the intense ambivalence; that is, avoidance → domination → gray compromise → black and white compromise → holism. For example, Gilbert (2006) discusses how a newspaper organization initially framed the emergence of the Internet as an opportunity but, in the face of setbacks, it responded to the presumable ambivalence by reframing it as a major threat (domination). However, with continued experience, the organization came to see the Internet as both an opportunity (for its online division) and a threat (for its print division), with the corporate level providing the strategic integration (holism). Conversely, an actor may be overwhelmed by the tension caused by a failed response; may dismiss more conscious, proactive, and effortful responses as untenable; or may not even consider the possibility of such responses. In such cases, the actor may move “down the ladder.” Future research exploring such trajectories will add a welcome dynamic element to the model. Finally, the integrative framework of responses to ambivalence may help shed new light on a variety of organizational studies’ topics across levels of analysis, from decision making to motivation, groupthink to group conflict, and organizational learning to organizational design. A prime example is leadership. We noted earlier that leaders may express and thereby sow their ambivalence in the organization. Although the literature typically regards this as quite dysfunctional (e.g., Larson and Tompkins 2005, Rothman and Wiesenfeld 2007), Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS our analysis suggests that there are times when leaders should actively sow ambivalence as a means of fostering personal growth, nuanced decision making, adaptation, and creativity. Thus, perspectives on transformational leadership, spiritual leadership, and servant leadership may be enriched by considering the potential role of ambivalence in fomenting actor change. Furthermore, given that each of the four responses to ambivalence is most effective under certain conditions, and that there are various ways that a given response can be enacted, leaders can serve as potent role models for when and how to enact the responses. Finally, leaders likely act as important conduits for institutionalizing responses to ambivalence in the culture and structure of organizations. Our larger point, then, is that the model of responses to ambivalence has the potential to inform a variety of traditional organizational topics in novel ways. Conclusion Far from indicating a rare and negative experience, ambivalence appears to be ubiquitous in organizational settings and—although uncomfortable—ambivalence has the potential to help foster growth and richly nuanced behavior. The discussion helps inform our understanding of how organizational triggers foster ambivalence at the collective and individual levels of analysis, how ambivalence diffuses across levels, and how actors are likely to respond to ambivalence and with what positive and negative effects. Acknowledgments The authors thank David Sluss, Lu Wang, senior editor Stuart Bunderson and three anonymous reviewers, along with audience members at presentations at Australian National University, Boston College, Chinese University of Hong Kong, Hong Kong Polytechnic University, Hong Kong University of Science and Technology, London Business School, Rice University, University of Alberta, University of California at Berkeley, University of Melbourne, University of Southern California, University of Texas at Austin, University of Western Australia, and Virginia Commonwealth University for their very helpful comments on earlier drafts of the paper. An earlier version of the paper was also presented at the 2010 meeting of the Academy of Management in Montreal. Endnotes 1 Pratt and Pradies (2011), who are primarily interested in “positive” responses to ambivalence, suggest that responses may vary in terms of their likelihood of leading to narrowed or expanded avenues for future action. This shifts the basis of comparison to response potentials. Because of our interest in understanding how responses relate to each other in terms of their orientations toward the same target (rather than in the future responses they might engender), we arrive at a very different way to organize responses. 2 Given that cognition and emotion are typically intertwined (Elfenbein 2007)—particularly in cases of intense ambivalence— we will not explore differences between cognitive ambivalence and emotional ambivalence (see Lavine et al. 1998 for a Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS treatment of affective–cognitive ambivalence; Conner and Armitage 2008 for attitudinal ambivalence, which combines cognition and affect; and Rothman and Wiesenfeld 2007 for emotional ambivalence). 3 Building off work by Smith and Berg (1987) at the group level and Weick (2002) at a broader network level, Pradies and Pratt (2010) argue that collectives can be conceptualized as ambivalent when they are composed of actors who (1) are each ambivalent (holographic ambivalence) or (2) are not ambivalent themselves but who hold thoughts or feelings that directly oppose those of other actors such that the collective can be said to be ambivalent (ideographic ambivalence). Because of our interest in a multilevel model, and because there is not a direct individual-level correlate with ideographic ambivalence (which requires multiple actors), we will focus mainly on holographic ambivalence. 4 Kozlowski and Klein (2000, p. 16) contrast composition effects with “compilation” effects—that is, when members of the collective have divergent perspectives/experiences but, when juxtaposed like pieces of a jigsaw puzzle, form a meaningful pattern. Compilation effects are more relevant to arrow 2 in Figure 1, as members of a collective socially construct a mutual understanding of a given organizational trigger. 5 We recognize that for a given example of defense mechanisms and coping mechanisms it is often impossible to know the degree to which the response was indeed nonconscious or conscious. In the examples below, we are inferring the level of consciousness from the authors’ descriptions. 6 We say only “typically conscious” because holistic approaches may be the product of “unwise” actors in a “wise” system (as explained below) or may, through repeated enactment, become at least somewhat automatic. References Adler PS (2012) The sociological ambivalence of bureaucracy: From Weber via Gouldner to Marx. Organ. Sci. 23(1):244–266. Albert S, Whetten DA (1985) Organizational identity. Cummings LL, Staw BM, eds. Research in Organizatonal Behavior, Vol. 7 (JAI Press, Greenwich, CT), 263–295. Andrews G, Singh M, Bond M (1993) The defense style questionnaire. J. Nervous Mental Disease 181(4):246–256. Argyris C (1985) Strategy, Change and Defensive Routines (Pitman, Boston). Argyris C (1990) Overcoming Organizational Defenses: Facilitating Organizational Learning (Allyn and Bacon, Needham Heights, MA). Argyris C (1993) Knowledge for Action: A Guide to Overcoming Barriers to Overcoming Change (Jossey-Bass, San Francisco). Arribas-Ayllon M, Sarangi S, Clarke A (2009) Professional ambivalence: Accounts of ethical practice in childhood genetic testing. J. Genetic Counsel. 18:173–184. Ashakanasy [sic] NM, Humphrey RH (2011) A multi-level view of leadership and emotion: Leading with emotional labor. Bryman A, Collinson D, Grint K, Jackson B, Uhl-Bien M, eds. The Sage Handbook of Leadership (Sage, Los Angeles), 365–379. Ashcraft KL (2001) Organized dissonance: Feminist bureaucracy as hybrid form. Acad. Management J. 44(6):1301–1322. Ashforth BE, Mael FA (1998) The power of resistance: Sustaining valued identities. Kramer RM, Neale MA, eds. Power and Influence in Organizations (Sage, Thousand Oaks, CA), 89–119. 1473 Ashforth BE, Reingen PH (2014) Functions of dysfunction: Managing the dynamics of an organizational duality in a natural food cooperative. Admin. Sci. Quart., ePub ahead of print May 23. Ashforth BE, Saks AM (2002) Feeling your way: Emotion and organizational entry. Lord RG, Klimoski RJ, Kanfer R, eds. Emotions in the Workplace: Understanding the Structure and Role of Emotions in Organizational Behavior (Jossey-Bass, San Francisco), 331–369. Ashforth BE, Rogers KM, Corley KG (2011) Identity in organizations: Exploring cross-level dynamics. Organ. Sci. 22(5):1144–1156. Ashforth BE, Joshi M, Anand V, O’Leary-Kelly AM (2013) Extending the expanded model of organizational identification to occupations. J. Appl. Soc. Psych. 43(12):2426–2448. Baek YM (2010) An integrative model of ambivalence. Soc. Sci. J. 47(3):609–629. Bandura A (1977) Social Learning Theory (Prentice-Hall, Englewood Cliffs, NJ). Barsade SG (2002) The ripple effect: Emotional contagion and its influence on group behavior. Admin. Sci. Quart. 47(4):644–675. Barsade SG, Gibson DE (1998) Group emotion: A view from top and bottom. Gruenfeld DH, Neale MA, Mannix EA, eds. Research on Managing Groups and Teams, Vol. 1 (JAI Press, Stamford, CT), 81–102. Bartel CA, Saavedra R (2000) The collective construction of work group moods. Admin. Sci. Quart. 45(2):197–231. Baumeister RF, Dale K, Sommer KL (1998) Freudian defense mechanisms and empirical findings in modern social psychology: Reaction formation, projection, displacement, undoing, isolation, sublimation, and denial. J. Personality 66(6):1081–1124. Bell DW, Esses VM (1997) Ambivalence and response amplification toward Native peoples. J. Appl. Soc. Psych. 27(12):1063–1084. Bell DW, Esses VM (2002) Ambivalence and response amplification: A motivational perspective. Personality Soc. Psych. Bull. 28(8):1143–1152. Berglas S, Jones EE (1978) Drug choice as a self-handicapping strategy in response to noncontingent success. J. Personality Soc. Psych. 36(4):405–417. Bierly PE III, Kolodinsky RW (2007) Strategic logic: Toward a wisdombased approach to strategic management. Kessler EH, Bailey JR, eds. Handbook of Organizational and Managerial Wisdom (Sage, Los Angeles), 61–88. Bierly PE III, Kessler EH, Christensen EW (2000) Organizational learning, knowledge and wisdom. J. Organ. Change Management 13(6):595–618. Bigelow J (1992) Developing managerial wisdom. J. Management Inquiry 1(2):143–153. Blake RR, Mouton JS (1964) The Managerial Grid: Key Orientations for Achieving Production Through People (Gulf, Houston). Bowlby J (1982) Attachment and Loss: Attachment, 2nd ed., Vol. 1 (Basic Books, New York). Bowles ML (1991) The organization shadow. Organ. Stud. 12(3): 387–404. Brickman P (1987) Commitment, Conflict, and Caring (Prentice-Hall, Englewood Cliffs, NJ). Bridge K, Baxter LA (1992) Blended relationships: Friends as work associates. Western J. Comm. 56(3):200–225. Brief AP (1998) Attitudes in and Around Organizations (Sage, Thousand Oaks, CA). Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. 1474 Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS Brooks ME, Highhouse S (2006) Familiarity breeds ambivalence. Corporate Reputation Rev. 9(2):105–113. Eisenberg EM (1984) Ambiguity as strategy in organizational communication. Comm. Monogr. 51(3):227–242. Brown SP, Westbrook RA, Challagalla G (2005) Good cope, bad cope: Adaptive and maladaptive coping strategies following a critical negative work event. J. Appl. Psych. 90(4):792–798. Elfenbein HA (2007) Emotion in organizations: A review and theoretical integration. Acad. Management Ann. 1:315–386. Brown T (2011) Red Sox turn Francona into scapegoat. Yahoo! Sports (September 30) http://sports.yahoo.com/news/red-sox-turnfrancona-scapegoat-202000713- -mlb.html. Brunsson N (1989) The Organization of Hypocrisy: Talk, Decisions and Actions in Organizations (John Wiley & Sons, Chichester, UK). Brunsson N (1993) Ideas and actions: Justification and hypocrisy as alternatives to control. Accounting, Organ. Soc. 18(6):489–506. Erez A, Misangyi VF, Johnson DE, LePine MA, Halverson KC (2008) Stirring the hearts of followers: Charismatic leadership as the transferal of affect. J. Appl. Psych. 93(3):602–615. Erikson KT (1976) Everything in Its Path: Destruction of Community in the Buffalo Creek Flood (Touchstone, New York). Farjoun M (2010) Beyond dualism: Stability and change as a duality. Acad. Management Rev. 35(2):202–225. Cameron KS (1986) Effectiveness as paradox: Consensus and conflict in conceptions of organizational effectiveness. Management Sci. 32(5):539–553. Fassin Y, Buelens M (2011) The hypocrisy-sincerity continuum in corporate communication and decision making: A model of corporate social responsibility and business ethics practices. Management Decision 49(4):586–600. Cameron KS, Quinn RE, DeGraff J, Thakor AV (2006) Competing Values Leadership: Creating Value in Organizations (Edward Elgar, Cheltenham, UK). Feeley TH (2002) Comment on halo effects in rating and evaluation research. Human Comm. Res. 28(4):578–586. Carson SJ, Madhok A, Wu T (2006) Uncertainty, opportunism, and governance: The effects of volatility and ambiguity on formal and relational contracting. Acad. Management J. 49(5):1058–1077. Cha SE, Edmondson AC (2006) When values backfire: Leadership, attribution, and disenchantment in a values-driven organization. Leadership Quart. 17(1):57–78. Chen M-J (2002) Transcending paradox: The Chinese “middle way” perspective. Asia Pacific J. Management 19(2/3):179–199. Choo CW (1996) The knowing organization: How organizations use information to construct meaning, create knowledge and make decisions. Internat. J. Inform. Management 16(5):329–340. Festinger L (1957) A Theory of Cognitive Dissonance (Row, Peterson and Company, Evanston, IL). Fiol CM, O’Connor EJ (2003) Waking up! Mindfulness in the face of bandwagons. Acad. Management Rev. 28(1):54–70. Fiol CM, Pratt MG, O’Connor EJ (2009) Managing intractable identity conflicts. Acad. Management Rev. 34(1):32–55. Fiss PC, Zajac EJ (2006) The symbolic management of strategic change: Sensegiving via framing and decoupling. Acad. Management J. 49(6):1173–1193. Fitzgerald FS (1945) The crack-up. Wilson E, ed. The Crack-Up (New Directions, New York), 69–84. Clark JK, Wegener DT, Fabrigar LR (2008) Attitudinal ambivalence and message-based persuasion: Motivated processing of proattitudinal information and avoidance of counterattitudinal information. Personality Soc. Psych. Bull. 34(4):565–577. Folkman S, Lazarus RS (1980) An analysis of coping in a middle-aged community sample. J. Health Soc. Behav. 21(3):219–239. Clayton V (1982) Wisdom and intelligence: The nature and function of knowledge in the later years. Internat. J. Aging Human Development 15(4):315–321. Fong CT, Tiedens LZ (2002) Dueling experiences and dual ambivalences: Emotional and motivational ambivalence of women in high status positions. Motivation Emotion 26(1):105–121. Clegg SR, ed. (2002) Management and Organization Paradoxes (John Benjamins, Amsterdam). Freud S (1950/1920) Beyond the Pleasure Principle, translated by Strachey J (Liveright, New York). Conner M, Armitage CJ (2008) Attitudinal ambivalence. Crano WD, Prislin R, eds. Attitudes and Attitude Change (Psychology Press, New York), 261–288. Galbraith JR (2009) Designing Matrix Organizations That Actually Work: How IBM, Procter & Gamble, and Others Design for Success (Jossey-Bass, San Francisco). Conner M, Sparks P (2002) Ambivalence and attitudes. Eur. Rev. Soc. Psych. 12:37–70. George JM (1996) Group affective tone. West MA, ed. Handbook of Work Group Psychology (John Wiley & Sons, Chichester, UK), 77–93. Coser RL (1966) Role distance, sociological ambivalence, and transitional status systems. Amer. J. Sociol. 72(2):173–187. Cramer P (1998) Coping and defense mechanisms: What’s the difference? J. Personality 66(6):919–946. Cramer P (2006) Protecting the Self: Defense Mechanisms in Action (Guilford Press, New York). Daft RL, Macintosh NB (1981) A tentative exploration into the amount and equivocality of information processing in organizational work units. Admin. Sci. Quart. 26(2):207–224. Dane E (2011) Paying attention to mindfulness and its effects on task performance in the workplace. J. Management 37(4):997–1018. Das TK, Teng B-S (2000) Instabilities of strategic alliances: An internal tensions perspective. Organ. Sci. 11(1):77–101. Diestel S, Schmidt K-H (2011) Costs of simultaneous coping with emotional dissonance and self-control demands at work: Results from two German samples. J. Appl. Psych. 96(3):643–653. Fong CT (2006) The effects of emotional ambivalence on creativity. Acad. Management J. 49(5):1016–1030. Gilbert CG (2006) Change in the presence of residual fit: Can competing frames coexist? Organ. Sci. 17(1):150–167. Gioia DA, Chittipeddi K (1991) Sensemaking and sensegiving in strategic change initiation. Strategic Management J. 12(6):433–448. Gioia DA, Nag R, Corley KG (2012) Visionary ambiguity and strategic change: The virtue of vagueness in launching major organizational change. J. Management Inquiry 21(4):364–375. Glynn MA (2000) When cymbals become symbols: Conflict over organizational identity within a symphony orchestra. Organ. Sci. 11(3):285–298. Graetz F, Smith ACT (2008) The role of dualities in arbitrating continuity and change in forms of organizing. Internat. J. Management Rev. 10(3):265–280. Gray JR (1999) A bias toward short-term thinking in threat-related negative emotional states. Personality Soc. Psych. Bull. 25(1):65–75. Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. Gutierrez B, Howard-Grenville J, Scully MA (2010) The faithful rise up: Split identification and an unlikely change effort. Acad. Management J. 53(4):673–699. Hardin CD, Higgins ET (1996) Shared reality: How social verification makes the subjective objective. Sorrentino RM, Higgins ET, eds. Handbook of Motivation and Cognition, Vol. 3 (Guilford, New York), 28–84. 1475 Kozlowski SWJ, Chao GT, Jensen JM (2009) Building an infrastructure for organizational learning: A multilevel approach. Kozlowski SWJ, Salas E, eds. Learning, Training, and Development in Organizations (Routledge, New York), 363–403. Kreiner GE, Ashforth BE (2004) Evidence toward an expanded model of organizational identification. J. Organ. Behav. 25(1):1–27. Harrist S (2006) A phenomenological investigation of the experience of ambivalence. J. Phenomenologic. Psych. 37(1):85–114. Kriegler RJ (1980) Working for the Company: Work and Control in the Whyalla Shipyard (Oxford University Press, Melbourne, VIC, Australia). Hass RG, Katz I, Rizzo N, Bailey J, Eisenstadt D (1991) Crossracial appraisal as related to attitude ambivalence and cognitive complexity. Personality Soc. Psych. Bull. 17(1):83–92. Ku G (2008) Learning to de-escalate: The effects of regret in escalation of commitment. Organ. Behav. Human Decision Processes 105(2):221–232. Hatch MJ (1997) Irony and the social construction of contradiction in the humor of a management team. Organ. Sci. 8(3):275–288. Kunda G (1992) Engineering Culture: Control and Commitment in a High-Tech Corporation (Temple University Press, Philadelphia). Heider F (1958) The Psychology of Interpersonal Relations (John Wiley & Sons, New York). Larson GS, Tompkins PK (2005) Ambivalence and resistance: A study of management in a concertive control system. Comm. Monogr. 72(1):1–21. Hillcoat-Nallétamby S, Phillips JE (2011) Sociological ambivalence revisited. Sociology 45(2):202–217. Hodson G, Maio GR, Esses VM (2001) The role of attitudinal ambivalence in susceptibility to consensus information. Basic Appl. Soc. Psych. 23(3):197–205. Hoffman MF, Medlock-Klyukovski A (2004) “Our creator who art in heaven”: Paradox, ritual, and cultural transformation. Western J. Comm. 68(4):389–410. Horney K (1945) Our Inner Conflicts: A Constructive Theory of Neurosis (W. W. Norton, New York). Isabella LA (1990) Evolving interpretations as a change unfolds: How managers construe key organizational events. Acad. Management J. 33(1):7–41. Jonas K, Diehl M, Brömer P (1997) Effects of attitudinal ambivalence on information processing and attitude-intention consistency. J. Experiment. Soc. Psych. 33(2):190–210. Jubas K, Butterwick S (2008) Hard/soft, formal/informal, work/learning: Tenuous/persistent binaries in the knowledge-based society. J. Workplace Learn. 20(7/8):514–525. Kantola SJ, Syme GJ, Campbell NA (1984) Cognitive dissonance and energy conservation. J. Appl. Psych. 69(3):416–421. Katz D, Kahn RL (1978) The Social Psychology of Organizations, 2nd ed. (John Wiley & Sons, New York). Katz I, Glass DC (1979) An ambivalence-amplification theory of behavior toward the stigmatized. Austin WG, Worchel S, eds. The Social Psychology of Intergroup Relations (Brooks-Cole, Monterey, CA), 55–70. Kessler EH, Bailey JR, eds. (2007) Handbook of Organizational and Managerial Wisdom (Sage, Los Angeles). Kets de Vries MFR, Miller D (1984) The Neurotic Organization: Diagnosing and Changing Counterproductive Styles of Management (Jossey-Bass, San Francisco). Klimoski R, Mohammed S (1994) Team mental model: Construct or metaphor? J. Management 20(2):403–437. Kolb DM (1987) Corporate ombudsman and organization conflict resolution. J. Conflict Resolution 31(4):673–691. Kosmala K, Herrbach O (2006) The ambivalence of professional identity: On cynicism and jouissance in audit firms. Human Relations 59(10):1393–1428. Kozlowski SWJ, Klein KJ (2000) A multilevel approach to theory and research in organizations: Contextual, temporal, and emergent processes. Klein KJ, Kozlowski SWJ, eds. Multilevel Theory, Research, and Methods in Organizations: Foundations, Extensions, and New Directions (Jossey-Bass, San Francisco), 3–90. Lavine H, Thomsen CJ, Zanna MP, Borgida E (1998) On the primacy of affect in the determination of attitudes and behavior: The moderating role of affective-cognitive ambivalence. J. Experiment. Soc. Psych. 34(4):398–421. Lawrence PR, Lorsch JW (1967) Differentiation and integration in complex organizations. Admin. Sci. Quart. 12(1):1–47. Lazarus RS, Folkman S (1984) Stress, Appraisal, and Coping (Springer, New York). Lewin K (1951) Field Theory in Social Science: Selected Theoretical Papers, edited by Cartwright D (Harper & Brothers, New York). Lewis MW (2000) Exploring paradox: Toward a more comprehensive guide. Acad. Management Rev. 25(4):760–776. Lim B-C, Klein KJ (2006) Team mental models and team performance: A field study of the effects of team mental model similarity and accuracy. J. Organ. Behav. 27(4):403–418. Lüscher LS, Lewis MW (2008) Organizational change and managerial sensemaking: Working through paradox. Acad. Management J. 51(2):221–240. Maio GR, Greenland K, Bernard M, Esses VM (2001) Effects of intergroup ambivalence on information processing: The role of physiological arousal. Group Processes Intergroup Relations 4(4):355–372. Maitlis S (2005) The social processes of organizational sensemaking. Acad. Management J. 48(1):21–49. Maitlis S, Lawrence TB (2007) Triggers and enablers of sensegiving in organizations. Acad. Management J. 50(1):57–84. Margolis JD, Molinsky A (2008) Navigating the bind of necessary evils: Psychological engagement and the production of interpersonally sensitive behavior. Acad. Management J. 51(5):847–872. Martin RL (2009) The Opposable Mind: Winning Through Integrative Thinking (Harvard Business School Press, Boston). Maybury-Lewis D, Almagor U, eds. (1989) The Attraction of Opposites: Thought and Society in the Dualistic Mode (University of Michigan Press, Ann Arbor). Medvec VH, Madey SF, Gilovich T (1995) When less is more: Counterfactual thinking and satisfaction among Olympic medalists. J. Personality Soc. Psych. 69(4):603–610. Merton RK (1976) Sociological Ambivalence and Other Essays (Free Press, New York). Merton RK, Barber E (1976) Sociological ambivalence. Sociological Ambivalence and Other Essays by Merton RK (Free Press, New York), 3–31. Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach 1476 Meyerson DE (2001) Tempered Radicals: How People Use Difference to Inspire Change at Work (Harvard Business School Press, Boston). Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. Meyerson DE, Scully MA (1995) Tempered radicalism and the politics of ambivalence and change. Organ. Sci. 6(5):585–600. Morgeson FP, Hofmann DA (1999) The structure and function of collective constructs: Implications for multilevel research and theory development. Acad. Management Rev. 24(2):249–265. Neumann R, Strack F (2000) Approach and avoidance: The influence of proprioceptive and exteroceptive cues on encoding of affective information. J. Personality Soc. Psych. 79(1):39–48. Oglensky BD (2008) The ambivalent dynamics of loyalty in mentorship. Human Relations 61(3):419–448. Pescosolido AT (2002) Emergent leaders as managers of group emotion. Leadership Quart. 13(5):583–599. Peters K, Kashima Y (2007) From social talk to social action: Shaping the social triad with emotion sharing. J. Personality Soc. Psych. 93(5):780–797. Peters MK, Rothman NB, Northcraft GB (2011) Beyond valence: Effects of group emotional tone on group negotiation behaviors and outcomes. Mannix EA, Neale MA, Overbeck JR, eds. Research on Managing Groups and Teams, Vol. 14 (Emerald Group Publishing, Bingley, UK), 163–185. Piderit SK (2000) Rethinking resistance and recognizing ambivalence: A multidimensional view of attitudes toward an organizational change. Acad. Management Rev. 25(4):783–794. Plambeck N, Weber K (2009) CEO ambivalence and responses to strategic issues. Organ. Sci. 20(6):993–1010. Plambeck N, Weber K (2010) When the glass is half full and half empty: CEOs’ ambivalent interpretations of strategic issues. Strategic Management J. 31(7):689–710. Pradies C, Pratt MG (2010) Ex uno plures: Toward a conceptualization of group ambivalence. Acad. Management Proc. August 2010(1):1–6. Pratt MG (2000) The good, the bad, and the ambivalent: Managing identification among Amway distributors. Admin. Sci. Quart. 45(3):456–493. Pratt MG (2001) Social identity dynamics in modern organizations: An organizational psychology/organizational behavior perspective. Hogg MA, Terry DJ, eds. Social Identity Processes in Organizatonal Contexts (Psychology Press, Philadelphia), 13–30. Pratt MG, Dirks KT (2006) Rebuilding trust and restoring positive relationships: A commitment-based view of trust. Dutton JE, Ragins BR, eds. Exploring Positive Relationships at Work: Building a Theoretical and Research Foundation (Lawrence Erlbaum Associates, Mahwah, NJ), 117–136. Pratt MG, Doucet L (2000) Ambivalent feelings in organizational relationships. Fineman S, ed. Emotion in Organizations, 2nd ed. (Sage, London), 204–226. Pratt MG, Dutton JE (2000) Owning up or opting out: The role of emotions and identities in issue ownership. Ashkanasy NM, Härtel CEJ, Zerbe WF, eds. Emotions in the Workplace: Research, Theory, and Practice (Quorum, Westport, CT), 103–129. Pratt MG, Foreman PO (2000) Classifying managerial responses to multiple organizational identities. Acad. Management Rev. 25(1):18–42. Pratt MG, Pradies C (2011) Just a good place to visit? Exploring positive responses to psychological ambivalence. Cameron KS, Spreitzer GM, eds. The Oxford Handbook of Positive Organizational Scholarship (Oxford University Press, Oxford, UK), 924–937. Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS Pratt MG, Rosa JA (2003) Transforming work-family conflict into commitment in network marketing organizations. Acad. Management J. 46(4):395–418. Radsma J, Bottorff JL (2009) Counteracting ambivalence: Nurses who smoke and their health promotion role with patients who smoke. Res. Nursing Health 32(4):443–452. Rahim MA (1985) A strategy for managing conflict in complex organizations. Human Relations 38(1):81–89. Raisch S, Birkinshaw J (2008) Organizational ambidexterity: Antecedents, outcomes, and moderators. J. Management 34(3):375–409. Rees L, Rothman NB, Lehavy R, Sanchez-Burks J (2013) The ambivalent mind can be a wise mind: Emotional ambivalence increases judgment accuracy. J. Experiment. Soc. Psych. 49(3):360–367. Roese NJ (1997) Counterfactual thinking. Psych. Bull. 121(1):133–148. Rosen M (1988) You asked for it: Christmas at the bosses’ expense. J. Management Stud. 25(5):463–480. Rothenberg A (1979) The Emerging Goddess: The Creative Process in Art, Science, and Other Fields (University of Chicago Press, Chicago). Rothman NB, Wiesenfeld BM (2007) The social consequences of expressing emotional ambivalence in groups and teams. Mannix EA, Neale MA, Anderson CP, eds. Research on Managing Groups and Teams, Vol. 10 (Emerald Group Publishing, Bingley, UK), 275–308. Rousseau DM, Sitkin SB, Burt RS, Camerer C (1998) Not so different after all: A cross-discipline view of trust. Acad. Management Rev. 23(3):393–404. Rowe WG, Cannella AA Jr, Rankin D, Gorman D (2005) Leader succession and organizational performance: Integrating the commonsense, ritual scapegoating, and vicious-circle succession theories. Leadership Quart. 16(2):197–219. Salancik GR (1977) Commitment and the control of organizational behavior and belief. Staw BM, Salancik GR, eds. New Directions in Organizational Behavior (St. Clair Press, Chicago), 1–54. Schein EH (2010) Organizational Culture and Leadership, 4th ed. (Jossey-Bass, San Francisco). Schneider JA (2003) Working with pathological and healthy forms of splitting: A case study. Bull. Menninger Clinic 67(1):32–49. Schoenewolf G (1990) Emotional contagion: Behavioral induction in individuals and groups. Modern Psychoanalysis 15(1):49–61. Shapiro SL, Carlson LE (2009) The Art and Science of Mindfulness: Integrating Mindfulness into Psychology and the Helping Professions (American Psychological Association, Washington, DC). Sheeran P (2002) Intention-behavior relations: A conceptual and empirical review. Eur. Rev. Soc. Psych. 12(1):1–36. Shore K (1998) Managed care and managed competition: A question of morality. Small RF, Barnhill LR, eds. Practicing in the New Mental Health Marketplace: Ethical, Legal, and Moral Issues (American Psychological Association, Washington, DC), 67–102. Sincoff JB (1990) The psychological characteristics of ambivalent people. Clinical Psych. Rev. 10(1):43–68. Sincoff JB (1992) Ambivalence and defense: Effects of a repressive style on normal adolescents’ and young adults’ mixed feelings. J. Abnormal Psych. 101(2):251–256. Smelser NJ (1998) The rational and the ambivalent in the social sciences. Amer. Sociol. Rev. 63(1):1–16. Smircich L, Morgan G (1982) Leadership: The management of meaning. J. Appl. Behav. Sci. 18(3):257–273. Smith AD, Plowman DA, Duchon D (2010) Everyday sensegiving: A closer look at successful plant managers. J. Appl. Behav. Sci. 46(2):220–244. Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS 1477 Smith KK, Berg DN (1987) Paradoxes of Group Life: Understanding Conflict, Paralysis, and Movement in Group Dynamics (JosseyBass, San Francisco). Wegner DM (1989) White Bears and Other Unwanted Thoughts: Suppression, Obsession, and the Psychology of Mental Control (Viking, New York). Smith WK, Lewis MW (2011) Toward a theory of paradox: A dynamic equilibrium model of organizing. Acad. Management Rev. 36(2):381–403. Weick KE (1979) The Social Psychology of Organizing, 2nd ed. (Addison-Wesley, Reading, MA). Smith WK, Tushman ML (2005) Managing strategic contradictions: A top management model for managing innovation streams. Organ. Sci. 16(5):522–536. Sonenshein S (2010) We’re changing—or are we? Untangling the role of progressive, regressive, and stability narratives during strategic change implementation. Acad. Management J. 53(3):477–512. Staw BM (1980) Rationality and justification in organizational life. Staw BM, Cummings LL, eds. Research in Organizatonal Behavior, Vol. 2 (JAI Press, Greenwich, CT), 45–80. Staw BM, Sutton RI (1993) Macro organizational psychology. Murnighan JK, ed. Social Psychology in Organizations: Advances in Theory and Research (Prentice-Hall, Englewood Cliffs, NJ), 350–384. Weick KE (1995) Sensemaking in Organizations (Sage, Thousand Oaks, CA). Weick KE (1998) The attitude of wisdom: Ambivalence as the optimal compromise. Srivastva S, Cooperrider DL, eds. Organizational Wisdom and Executive Courage (Lexington Press, San Francisco), 40–64. Weick KE (2002) Puzzles in organizational learning: An exercise in disciplined imagination. British J. Management 13(S2):S7–S15. Weick KE (2004) Mundane poetics: Searching for wisdom in organization studies. Organ. Stud. 25(4):653–668. Weick KE, Putnam T (2006) Organizing for mindfulness: Eastern wisdom and Western knowledge. J. Management Inquiry 15(3):275–287. Sternberg RJ (1998) A balance theory of wisdom. Rev. General Psych. 2(4):347–365. Weick KE, Sutcliffe KM, Obstfeld D (2005) Organizing and the process of sensemaking. Organ. Sci. 16(4):409–421. Stigliani I, Ravasi D (2012) Organizing thoughts and connecting brains: Material practices and the transition from individual to group-level prospective sensemaking. Acad. Management J. 55(5):1232–1259. Weigert A, Franks DD (1989) Ambivalence: A touchstone of the modern temper. Frank DD, McCarthy ED, eds. The Sociology of Emotions: Original Essays and Research Papers (JAI Press, Greenwich, CT), 205–227. Sykes GM, Matza D (1957) Techniques of neutralization: A theory of delinquency. Amer. Sociol. Rev. 22(6):664–670. Tang Z, Dickson P, Marino L, Tang J, Powell BC (2010) The value of organizational ambivalence for small and medium size enterprises in an uncertain world. British J. Management 21(4):809–828. Tetlock PE (1985) Accountability: The neglected social context of judgment and choice. Cummings LL, Staw BM, eds. Research in Organizational Behavior, Vol. 7 (JAI Press, Greenwich, CT), 297–332. Thomas KW (1992) Conflict and negotiation processes in organizations. Dunnette MD, Hough LM, eds. Handbook of Industrial and Organizational Psychology, 2nd ed., Vol. 3 (Consulting Psychologists Press, Palo Alto, CA), 651–717. Thompson MM, Holmes JG (1996) Ambivalence in close relationships: Conflicted cognitions as a catalyst for change. Sorrentino RM, Higgins ET, eds. Handbook of Motivation and Cognition, Vol. 3 (Guilford Press, New York), 497–530. Tracy SJ (2004) Dialectic, contradiction, or double bind? Analyzing and theorizing employee reactions to organizational tension. J. Appl. Comm. Res. 32(2):119–146. Vadera AK, Pratt MG (2013) Love, hate, ambivalence, or indifference? A conceptual examination of workplace crimes and organizational identification. Organ. Sci. 24(1):172–188. van Harreveld F, van der Pligt J, de Liver YN (2009a) The agony of ambivalence and ways to resolve it: Introducing the MAID model. Personality Soc. Psych. Rev. 13(1):45–61. van Harreveld F, Rutjens BT, Rotteveel M, Nordgren LF, van der Pligt J (2009b) Ambivalence and decisional conflict as a cause of psychological discomfort: Feeling tense before jumping off the fence. J. Experiment. Soc. Psych. 45(1):167–173. Van Kleef GA (2008) Emotion in conflict and negotiation: Introducing the emotions as social information (EASI) model. Ashkanasy NM, Cooper CL, eds. Research Companion to Emotion in Organizations (Edward Elgar, Cheltenham, UK), 392–404. Wang L, Pratt MG (2008) An identity-based view of emotional ambivalence and its management in organizations. Ashkanasy NM, Cooper CL, eds. Research Companion to Emotion in Organizations (Edward Elgar, Cheltenham, UK), 589–604. Wells PA (1988) The paradox of functional dysfunction in a Girl Scout camp: Implications of cultural diversity for achieving organizational goals. Jones MO, Moore MD, Snyder RC, eds. Inside Organizations: Understanding the Human Dimension (Sage, Newbury Park, CA), 109–117. Wexler MN (1993) The negative rite of scapegoating: From the ancient tribe to the modern organization. Internat. J. Group Tensions 23(4):293–307. Wiley N (1988) The micro-macro problem in social theory. Sociol. Theory 6(2):254–261. Wilson TD, Hodges SD (1992) Attitudes as temporary constructions. Martin LL, Tesser A, eds. The Construction of Social Judgments (Lawrence Erlbaum Associates, Hillsdale, NJ), 37–66. Wright C (2009) Inside out? Organizational membership, ambiguity and the ambivalent identity of the internal consultant. British J. Management 20(3):309–322. Ziegler R, Hagen B, Diehl M (2012) Relationship between job satisfaction and job performance: Job ambivalence as a moderator. J. Appl. Soc. Psych. 42(8):2019–2040. Zorn TE (1995) Bosses and buddies: Constructing and performing simultaneously hierarchical and close friendship relationships. Wood JT, Duck S, eds. Under-Studied Relationships: Off the Beaten Track (Sage, Thousand Oaks, CA), 122–147. Blake E. Ashforth is the Horace Steele Arizona Heritage Chair at the W. P. Carey School of Business, Arizona State University. He received his Ph.D. from the University of Toronto. His research concerns the ongoing dance between individuals and organizations, including identity and identification, socialization and newcomer adjustment, and the links among individual-, group-, and organization-level phenomena. Kristie M. Rogers is an assistant professor of organizational behavior at the University of Kansas School of Business. She received her Ph.D. from Arizona State University. Her research interests include social interactions, relationships, and the cognitive processes that shape perceptions of oneself and Ashforth et al.: Ambivalence in Organizations: A Multilevel Approach Downloaded from informs.org by [149.169.136.200] on 17 September 2014, at 09:11 . For personal use only, all rights reserved. 1478 one’s organization. She is particularly interested in the topics of respect, identity, and identification. Michael G. Pratt is the O’Connor Family Professor in the Management and Organization Department at Boston College. He earned his Ph.D. from the University of Michigan. His research interests include how individuals connect with the work that they do, as well as with the organizations, professions, and occupations in which they find themselves. Theoretically, Organization Science 25(5), pp. 1453–1478, © 2014 INFORMS his research draws heavily from theories of identity and identification, meaning, emotion (e.g., ambivalence), intuition, and culture (e.g., artifacts). Camille Pradies is a doctoral student in the Management and Organization Department of the Carroll School of Management at Boston College. Her current research focuses on ambivalence and identity—and on organizational and institutional conditions that influence them.