Exploring Paradox: Toward a More Comprehensive Guide PDF

^ Academy of Management Review 2000. Vol. 25. No. 4. 760-776, EXPLORING PARADOX: TOWARD A MORE COMPREHENSIVE GUIDE MARIANNE W. LEWIS...

^ Academy of Management Review 2000. Vol. 25. No. 4. 760-776, EXPLORING PARADOX: TOWARD A MORE COMPREHENSIVE GUIDE MARIANNE W. LEWIS University of Cincinnati "Paradox" appears increasingly in organization studies, often to describe conflicting demands, opposing perspectives, or seemingly illogical findings. This article helps researchers move beyond labeling—to explore paradoxes and contribute insights more in tune with organizational complexity and ambiguity. I first develop a frame- work that clarifies the nature of paradoxical tensions, reinforcing cycles, and their management. Using this framework, I then review studies in which paradoxes spurred by change and plurality are investigated. I conclude by outlining strategies for identifying and representing paradox, addressing implications for research. Paradoxes... seem to smile ironically at our Increasing technological change, global com- nicely constructed theories with their clear-cut distinctions and point at an unthought-of possi- petition, and workforce diversity reveal and in- bility, a blind spot in oppositional thinking tensify paradox. Managers, for example, are (Ybema, 1996: 40). asked to increase efficiency and foster creativ- ity, build individualistic teams, and think glo- "Paradox" denotes contradictory yet inter- bally while acting locally. "It's a paradox," how- related elements—elements that seem logical ever, is rapidly becoming the management in isolation but absurd and irrational when ap- cliche of our time—overused and underspeci- pearing simultaneously. Over a decade ago, fied (Handy, 1994). Reviewing studies from 1990 Cameron and Quinn (1988) claimed that by ex- to 1997, Davis, Maranville, and Obloj (1997) ploring paradox, researchers might move be- found the term used in over 300 major publica- yond oversimplified and polarized notions to tions. Most often, researchers use paradox to recognize the complexity, diversity, and ambi- describe conflicting demands, opposing per- guity of organizational life. They praised para- spectives, or seemingly illogical findings. Yet, dox for offering a potentially powerful frame- labeling paradox does not necessarily foster un- work for examining the impacts of plurality and derstanding. Bouchikhi (1998) claims that while change, aiding understandings of divergent per- organization researchers continue to unveil par- spectives and disruptive experiences. Since adoxes, few explore them at greater depths. then, in a growing body of texts (e.g.. Handy, Such exploration requires answering several 1994; Kets de Vries, 1995; Koot, Sabelis, & Ybema, fundamental questions: What is paradox? How 1996) and studies (e.g.. Hatch & Ehrlich, 1993; might paradox serve as a framework, fostering Murnighan & Conlon, 1991; Vince & Broussine, insights into contradictions and complex inter- 1996), researchers have examined paradox. Such relationships? And how might researchers ac- exemplars depict individuals, groups, and or- ganizations as inherently paradoxical, em- tively seek out paradox and build concepts and broiled in tensions and reinforcing cycles at theories that reflect the intricacies of organiza- their very core. Abandoning the notion that tional life? change is a smooth, linear, and planned jour- In this article I address these questions to ney, these researchers have examined how con- help guide explorations of paradox. I begin by tradictions both hamper and encourage organi- building a framework that integrates extant un- zational development. derstandings of paradox from philosophy, psy- chology, and organization studies. The frame- work clarifies the nature of underlying tensions, reinforcing cycles, and their management. I then Many thanks to special issue editor Nilin Nohria, three review organization studies in which paradoxes anonymous reviewers, and Kristen Taylor for their insightful spurred by plurality and change are examined. comments on earlier versions of this article. This review serves two purposes: (1) it illustrates 760 2000 Lewis 761 the potential value of the framework for foster- quently simplify reality into polarized either/or ing creative and complex insights, and (2) it distinctions that conceal complex interrelation- offers a key to exemplars, which may serve as ships. Third, paradoxes become apparent models for future inquiry. In conclusion, I glean through self- or social reflection or interaction from reviewed studies strategies for identifying that reveals the seemingly absurd and irration- and representing paradox and suggest implica- al coexistence of opposites. tions ior research. Some researchers, however, claim that com- prehending paradox requires more than defin- ing its characteristics; it requires a guiding TENSIONS, REINFORCING CYCLES, AND framework: a tool to help researchers explore THEIR MANAGEMENT: A PARADOX paradoxical tensions, reinforcing cycles, and FRAMEWORK their management (e.g., Argyris, 1993; Cameron Paradox: some 'thing' that is constructed by indi- & Quinn, 1988; Smith & Berg, 1987). In response, I viduals when oppositiona! tendencies are weave together insights from literature in psy- brought into recognizable proximity through re- chology, philosophy, and organization studies to flection or interaction (Ford 8E Backoff, 1988: 89). detail key elements of paradox. The resulting "Paradox" engenders numerous and varied framework clarifies (1) how paradoxical ten- meanings. Philosophers from the ancient sions stem from polarized cognitive or social Greeks to Existentialists have viewed human constructions, (2) how actors' defensive reac- existence as paradoxical—grounded in ten- tions might fuel reinforcing cycles, and (3) how sions between life and death, good and evil, self actors can avoid becoming stuck in these para- and other (see Hampden-Turner, 1981, and lyzing and often vicious cycles via greater cog- Schneider, 1990). Similarly, psychologists have nitive and behavioral complexity. long stressed the cognitive nature of paradox, Figure 1 illustrates these three components; examining the impacts of tensions on creativity the arrows between components in the figure and mental health (e.g., Bateson, 1972; Rothen- denote the exploration process. By linking man- berg, 1979) or using paradoxical therapy to help agement back to tensions, the framework de- actors face their inner conflicts (e.g., Harris, 1996; picts exploration itself as paradoxical. Rather Watzlawick, Weakland, & Fisch, 1974). In or- than a linear progression marked by a distinct ganization studies some researchers define par- endpoint or resolution, exploring paradox is an adox as contradictions embedded within a ongoing and cyclical journey. As researchers statement (e.g., Murnighan & Conlon, 1991), hu- learn to comprehend paradox, they may dis- man emotions (e.g., Vince & Broussine, 1996), or cover other, potentially more intricate, contra-^ organizational practices (e.g., Eisenhardt & dictions. According to Rothenberg, Westcott, 1988). Others describe paradox as an observation that counters common beliefs (e.g., Sophistication in science and other fields allows Davis et al., 1997) or as an unintended conse- for greater and greater specification of meaning- quence (e.g., Sitkin & Bies, 1993). Koot et al. (1996) ful and appropriate opposites. As old opposites are overthrown, new ones arise in a never ending use Wittgenstein's notion of paradox as some- spiral of sell-generation paralleling the spiral of thing surprising, whereas other researchers increasing knowledge (1979: 244). avoid explicit definition altogether (e.g., West- enholz, 1993). Tensions: The Underlying Sources of Paradox Ford and Backoff's (1988) definition, quoted at the start of this section, offers a useful umbrella, Paradoxical tensions are perceptual—that is, encompassing varied views and highlighting cognitively or socially constructed polarities three overarching characteristics of paradox. that mask the simultaneity of conflicting truths. First, as some 'thing,' a paradox may denote a Unlike continua, dilemmas, or either/or choices, wide variety of contradictory yet interwoven paradoxical tensions signify two sides of the elements: perspectives, feelings, messages, de- same coin. Yet actors' perceptions often obscure mands, identities, interests, or practices. Sec- this relatedness. According to personal con- ond, paradoxes are constiucted. As actors at- struct theory (Kelly, 1955), most actors accentu- tempt to make sense of an increasingly intricate, ate contradictions by interpreting data (e.g., ambiguous, and ever-changing world, they fre- their own and others' feelings, organizational  762 Academy of Management Review October FIGURE 1 A Paradox Framework Tensions Reiniorcing Management cycles Cognitively and/or Paialyzing defenses, Attempts to explore tensions socially constructed which initially reduce and therehy tap the potential polaiities that obscure discomfort and anxiety, energy, insights, and power (he interielatedness of yet eventually intensify of paradox that enable the contiadictions: tensions; for example: dramatic change; for example: for example: - self-referential loops - splitting - acceptance - mixed messages - projection - conirontation - system contradictions - repression - transcendence - regression - reaction formation - ambivalence practices, environmental cues) through simple, ena into ever smaller and more disparate bipolar concepts, constructing logical, inter- pieces. Yet, formal logic is based on either/or nally consistent sets of abstractions that sepa- thinking, incapable of comprehending the intri- rate opposites. Such frames of references or cacies of paradox (Ford & Ford, 1994). schemes enable actors to make sense of com- Language feeds the tendency to polarize. Con- plex realities, but they are biasing and, once ceptual boundaries help actors distinguish entrenched, become highly resistant to change what belongs and what does not, thereby giving (Bartunek, 1988). Ford and Ford explain that dis- meaning to both sides of a polarity (Vince & tinctions "reside in the observer(s), not the ob- Broussine, 1996). Actors typically define phe- served" {1994: 760). Actors distinguish a phenom- nomena by explaining what it is not. For exam- enon by placing brackets or boundaries around ple, by differentiating trust from mistrust or dis- it, differentiating figure from ground to facilitate cord from harmony—sensations that often their understanding. But such distinctions be- coexist—paradoxical tensions "reveal the limi- come objectified over time, giving actors the im- tations of conventional grammar based on con- pression that their perceptions are distinct and ventional logic" (Weick & Westley, 1996: 455- immutable entities. 456). In organization studies distinctions are Hampden-Turner (1981) reviewed varied ex- apparent in such polar constructs as quality/ planations of how actors make sense of para- cost, differentiation/integration, stability/ doxical tensions, such as autonomy/dependence change, and cohesion/division. In contrast. East- and reason/imagination, and claimed that for- ern philosophies stress the need to avoid mal logic—the basis of traditional scientific in- simplistic distinctions. For instance, the T'ai-chi quiry— emphasizes polarization. Grounded in T'u—the Taoist symbol of Yin and Yang (used to the philosophies of Aristotle, Descartes, and illustrate tensions in Figure D^signifies a nat- Newton, formal logic requires parsing phenom- ural wholeness composed of contradictions. 2000 Lewis 763 When one force (e.g., Yin, femininity, intuition, recognizing their cognitive and social foibles dark) escalates to its extreme state, it retains (Harris, 1996). elements of its opposition (e.g., Yang, masculin- According to Freudian psychology, paradoxi- ity, rationality, light), eventually reversing the cal tensions endanger the ego, producing anxi- trend (see Hampden-Turner, 1981: 20-21; Mor- ety that naturally raises actors' defenses gan, 1997: 283-285; and Rothenberg, 1979: 140- (Schneider, 1990). Defenses denote "any policy or 142, for discussions of this symbol). action that prevents someone (or some system) Paradoxical tensions may appear in several from experiencing embarrassment or threat, and forms. Putnam (1986) summarizes three inter- simultaneously prevents anyone from correcting related types of tensions: self-referential loops, the causes of the embarrassment or threat" mixed messages, and system contradictions. (Argyris, 1993: 40). By suppressing the related- Self-referential loops operate when contradic- ness of contradictions and maintaining the false tions are embedded within a cohesive state- appearance of order, defenses may temporarily ment, concept, or process. For instance, the reduce anxiety. But suppressing one side of a Liar's paradox—"I am lying"-—is a circular polarity intensifies pressure from the other. The statement. If I am lying, then this statement is result is a "strange loop" (Hofstadter, 1979). In false, implying that I am telling the truth. Weick attempting to reduce the frustrations and dis- (1979) describes cognition as an inherently self- comfort of tensions, actors' defensive behaviors referential process: actors filter experiences initially produce positive effects but eventually through their extant cognitive frames, choosing foster opposite, unintended consequences that explanations that confirm their existence. Mixed intensify the underlying tension. messages denote inconsistencies between state- Researchers have detailed numerous de- ments or between verbal and nonverbal re- fenses operating at individual, group, and or- sponses that appear during social interactions. ganizational levels (e.g., Argyris, 1993; Schnei- According to Argyris (1993), such tensions stem der, 1990). In conjunction, Smith and Berg (1987) from the construction of clearly ambiguous mes- and Vince and Broussine (1996) catalogued six sages. For example, a supervisor calls for team- defensive reactions toward paradox. Splitting work but closely monitors individual perfor- entails further polarizing contradictions—for ex- mance. Over time, such tensions may become ample, by forming subgroups or artificial we/ objectified within system contradictions, en- they distinctions that mask similarities. Projec- trenched "within the goals, reward systems, re- tion signifies the transfer of conflicting source demands, and division of labour of an attributes or feelings, often onto a scapegoat or organization" (Putnam, 1986: 161). For instance, repository of bad feelings. Repression or denial an organization composed of a network of flex- denotes the blocking of awareness of tenuous ible, autonomous teams may use highly formal- experiences or memories, whereas regression ized and centralized procedures for employee involves resorting to understandings or actions control. that have provided security in the past. Reaction formation entails excessively manifesting the feeling or practice opposite to the threatening one. In contrast, ambivaience signifies the com- Reinforcing Cycles: The Negative Dynamics of promise of conflicting emotions within "luke- Paradox warm" reactions that lose the vitality of ex- tremes. The dynamics of paradox are often vicious. As actors seek to resolve paradoxical tensions, they may become trapped within reinforcing cycles Management: Tapping the Positive Potential of that perpetuate and exacerbate the tension, for Paradox paradox is a double-edged sword. Tensions might serve as "a trigger for change," spurring Managing paradox means capturing its en- actors to rethink existing polarities and recog- lightening potential. The goal is to journey be- nize more complicated interrelationships. Yet, yond reinforcing cycles, dramatically rethinking tensions simultaneously inhibit change. Actors' past perceptions and practices. According to more typical and often first reactions are defen- Farson, "Our natural inclination when con- sive, clinging to past understandings to avoid fronted with paradoxes is to attempt to resolve  784 Academy of Management Review October them, to create the familiar out of the strange, to Vince & Broussine, 1996). According to Ford and rationalize them" (1996: 13). Yet, in today's com- Ford (1994), by identifying and discussing their plex organizations, models based on linear and underlying logic, actors might subject their rational problem solving do managers a tremen- ways of thinking to critique, thereby raising dous disservice. Managers need to recognize, their chances of escaping paralysis. Hatch and become comfortable with, and even profit from Ehrlich describe humor as a low-risk means of tensions and the anxieties they provoke, for "the confrontation: "By reducing the social strain, hu- contribution of paradox to management think- mour may even permit some of these difficulties ing is the recognition of its power to generate to make their way into social awareness where creative insight and change" (Eisenhardt & they could eventually become the subject of the Westcott, 1988: 170). serious discourse of change efforts" (1993: 524). Tapping the power of paradox is difficult, For example, in joking about their concerns, however, because escaping reinforcing cycles mid-level managers recognized that they both requires seemingly counterintuitive reactions monitor their employees and are heavily scruti- (Cameron & Quinn, 1988). Paradox management nized by senior executives. entails exploring, rather than suppressing, ten- sions. Smith and Berg (1987) explain that to es- Lastly, transcendence implies the capacity to cape paralyzing cycles, actors must reclaim think paradoxically. Watzlawick et al. (1974) ex- emotions that have been polarized or projected plain that actors cannot break out of reinforcing elsewhere and immerse themselves within the cycles by using first-order thinking—slight al- tensions. According to Vince and Broussine, terations to the logic and behaviors they have "Staying with the paradox makes it possible to used in the past—for first-order thinking pro- discover a link between opposing forces and duces a solution that is part of (and thereby opens up the framework that gives meaning to exacerbates) the problem. In contrast, second- the apparent contradictions" {1996: 4). Hence, a order thinking entails critically examining en- paradox framework shifts the notion of "manag- trenched assumptions to construct a more ac- ing" from modern definitions based on planning commodating perception of opposites. Critical and control to coping—its original meaning self- and social reflection might help actors re- (Handy, 1994). Rather than management being frame their assumptions, learn from existing the sole responsibility of a few, managing par- tensions, and develop a more complicated rep- adox requires that local actors learn to cope ertoire of understandings and behaviors that with their tensions. Openly and critically exam- better reflects organizational intricacies (e.g., ining their polarized perceptions (e.g., of distinct Denison, Hooijberg, & Quinn, 1995; Quinn, Kahn, occupational groups, demands of customers ver- & Mandl, 1994). Such reframing marks a dra- sus employees) may enable more dramatic matic change in the meaning attributed to a changes in their understandings and behaviors. situation as paradoxical tensions become Researchers suggest three, often interrelated viewed as complementary and interwoven. means of managing paradox: acceptance, con- Rothenberg (1979), in examining the creative frontation, and transcendence, Schneider (1990) proposes that acceptance—learning to live with process of scientific and artistic geniuses, de- paradox—offers a sense of freedom. For exam- scribes this as the capacity for lanusian think- ple, Murnighan and Conlon (1991) found that ing: "In an apparent defiance of logic or of phys- members of successful string quartets are well ical possibility, the creative person consciously aware of their tensions (e.g., desires for personal formulates the simultaneous operation of anti- autonomy and strong group leadership). Accep- thetical elements and develops those into inte- tance helps members avoid debates, which grated entities and creations. It is a leap that might open Pandora's box and spark vicious transcends ordinary logic" (1979: 55). For exam- cycles, and "play through" paradox by focusing ple, Mozart and Beethoven linked harmony and on their intense tasks. discord to fuel their musical inspiration. Artists such as Picasso and van Gogh used jarring jux- Other researchers, however, call for actors to tapositions to portray the ambiguity, diversity, confront paradox, discussing their tensions to and complexity of perceptions. And Einstein en- socially construct a more accommodating un- visioned a man falling off a building at rest derstanding or practice (e.g.. Smith & Berg, 1987; relative to things falling beside him and moving 2000 Lewis 765 relative to sights he passed on the way down, analysis, focal topics, and empirical approach- forever altering understandings of physics. es—for example, some are literature reviews of organization theory (e.g.. Miller, 1993) and others are interpretive analyses of a specific organiza- PARADOXES, CHANGE, AND PLURALITY: tion (e.g.. Hatch, 1997). Yet they share themes of A REVIEW OF EXEMPLARS change and plurality. In these studies research- In this section I illustrate the framework- ers explore how disruptions expose tensions in-use and provide a guide to exemplars. In- within organizations and how ambiguity fosters cluded works represent an extensive review of multiple, often conflicting interpretations of organization studies in which researchers use phenomena. paradox as more than a label. Paradox is both To summarize such varied exemplars, I cate- the guiding framework and the subject of in- gorize works within three, frequently inter- quiry. Although in some works scholars exam- related paradoxes—learning, organizing, and ine all three components of paradox (e.g., Leo- belonging—that appear repeatedly across stud- nard-Barton, 1992; Westenholz, 1993), in others ies (see Table 1). Learning paradoxes revolve they focus primarily on tensions (e.g., Ybema, around processes of sensemaking, innovation, 1996), reinforcing cycles (e.g., Ropo & Hunt, 1995), and transformation that reveal interwoven ten- or their management (e.g., Denison et al., 1995). sions between old and new. Paradoxes of or- The exemplars also vary considerably in level of ganizing stress conflicting yet simultaneous de- TABLE 1 Exemplary Organization Studies of Paradox Tensions Reinforcing Cycles Management Paradoxes of Learning Learning requires using, critiquing, and often destroying past understandings and practices to construct new and more complicated fiames of reference. Old/new Defenses ol repression, projection, Social retraming Sensemaking (Argyris, 1993; Bartunek, 1988; and regression Shocking crises and Davis, Maranville, & Obloj, 1997; Westenholz, Cognitive sell-reference discrepancies 1993) Inertial actions and Open communications and Innovation (Leonard-Barton, 1992; Ropo & competencies experimentation Hunt, 1995) Simpliiication of systems, values, Paradoxical leadership Transformation (Davis, Maianville, 8f Obloj, and structures 1997; Hatch, 1997; Miller, 1993; Vmce & Broussine, 1996) Paradoxes of Organizing Organizing denotes an ongoing process of equiJibrating opposing forces that encourage commitment, trust, and creativity while maintaining efficiency, discipline, and order. Control/flexibility Defenses of repression and reaction Dynamic equilibration Performance (Bouchikhi, 1998; Denison, formation Superordinate goals Hooijberg, & Quinn, 1995; Murnighan & Intense mistrust Humor Conlon. 1991; Quinn, 1988) Escalating resistance and Behavioral complexity Empowerment (Eisenhardt & Westcott, 1988; alienation Klein, 1994; O'Connor, 1995; Sabelis, 1998) Extreme chaos or rigidity Formalization (Hatch & Ehrlich, 1993; Sitkin & Bies, 1993) Paradoxes of Belonging Groups become cohesive, influential, and distinctive by valuing the diversity of their members and their inferconnecdons with other groups. Self/other Defenses of ambivalence, projection, Social acceptance Individuality {Amason, 1996; Murnighan & and splitting Task focus Conlon, 1991: Smith & Berg, 1987) Destructive conflict Valuing difference Group boundaries (Leonard-Barton, 1992; Group polarization Reduced power Martin, 1992; O'Connor, 1995; Ybema, 1996) Tribalism discrepancies Globalization (Koot, 1996; Naisbitt, 1994)  766 Academy of Management Review October mands for control and flexibility examined in cognitive level when she recognized that, year studies of organizational performance, empow- after year, an organization she was studying erment, and formalization. In turn, paradoxes of kept grappling with the same problematic wage belonging signify complex relationships be- issue with no progress. Employing paradox as a tween self and other, highlighting the problem- framework, she became acutely aware of how atic nature of individuality, group boundaries, actors were making sense of their experiences and globalization. Discussion of each paradox through their existing frames of reference. Ac- demonstrates uses of the paradox framework tors chose interpretations that supported, rather and the resulting insights into organizational than challenged, their frames, inhibiting critical change and plurality. self- and social reflection and reframing. In other exemplars researchers portray ac- tions, routines, and skills following similar pat- Paradoxes oi Learning terns yet based more on inertia than cognition Many paradoxes are caused by the hangover of (e.g., Davis et al., 1997; Ropo & Hunt, 1995). For one set of assumptions or beliefs into a new age instance, Leonard-Barton (1992) found that or- or environment and proliferate when change is ganizations' core capabilities (e.g., ingrained dramatic or rapid. Paradoxes emerge when be- human skills and managerial systems) foster liefs or assumptions fail to keep up with external changes (Cannon, 1996: 110). and frustrate innovation. The more actors stress their core capabilities, the more they invoke Why do actors seem to ignore dramatic their flip side: core rigidities. Extant strengths changes in their environment? Why do they fail offer routines that may guide innovative efforts. to take action when they do perceive inconsis- Yet, clinging to core competencies might inhibit tencies between their understandings and the actors from considering more drastic changes. world around them, or take action that produces Applying a higher level of analysis. Miller results contrary to those intended? A common (1993) uses paradox to extend theories of organ- factor in these questions is learning: the ability izational transformation. He explains how the to frame new knowledge within understandings, focus that enables an organization to prosper routines, and structures that enable actors to initially might lead to its eventual demise. As comprehend and adjust to variations. Hofstadter organizations and their environments become (1979) explains the learning paradox as follows: more complicated and diverse, rather than ad- perceptions and actions are self-referential, re- here to the theory of requisite variety, actors' lying and building upon themselves as actors cumulative learning during times of prosperity attempt to change. Paradoxical tensions may may simplify and narrow organizational compe- reveal the need for learning, yet actors often use tencies, strategies, and cultural values. their extant cognitive and behavioral frames to By delving into paradoxical dynamics, re- construct new frames, trapping them within a searchers reveal how actors' defenses fuel self- double-blind (Bateson, 1972). In the reviewed ex- referential cycles, fostering incremental learn- emplars researchers explore these puzzles, ing at best and organizational paralysis or striving to develop theories of sensemaking, in- decline at worst. In particular, repression, pro- novation, and organizational transformation jection, and regression serve to protect actors that emphasize, rather than downplay, the intri- temporarily from recognizing that extant skills, cacies of learning (e.g., Argyris, 1993; Bartunek, routines, and understandings may be obsolete. 1988; Westenholz, 1993). According to Davis et Vince and Broussine (1996) note how actors focus al., in such studies scholars seek a deeper un- on the positive aspects of change to avoid dis- derstanding of the penultimate paradox: "The cussing their fears and frustrations. To bolster more things change, the more they remain the their confidence, actors might also attribute suc- same" (1997: 276). cesses to existing skills and practices and prob- A key source of learning paradoxes is tension lems to elements beyond their control (Argyris, between old and new—a struggle between the 1993; Miller, 1993). Projecting blame, often across comfort of the past and the uncertainty of the organizational levels or groups (e.g., workers future. In exemplars researchers depict this ten- blaming management and vice versa), limits sion operating at differing levels of analysis. discussion of disparate understandings and so- Westenholz (1993) began exploring paradox at a cial reframing (Westenholz, 1993). Yet, possibly 2000 Lewis 767 the most documented learning defense is ing, and often tearing down the old to build the regression to the comfort of past practices new. The paradox framework might guide such (e.g., Argyris, 1993; Ropo & Hunt, 1995). In her studies, helping researchers depict learning as cases Leonard-Barton (1992) found that inconsis- a self-referential cycle, and encourage actors to tencies between core capabilities and innova- socially reflect upon conflicting feelings, exist- tion demands frequently led to more intense use ing and comfortable routines, and the defenses of extant strengths, resulting in teams repeat- they provoke. edly missing opportunities for creative break- throughs. Similarly, Miller (1993) claimed that rising simplicity, in the face of growing environ- Paradoxes of Organizing mental complexity, defends organizations Any action, any act of cognition, definition, or against recognizing the need for potentially dis- concept is based on and establishes distinctions. ruptive and painful change. Thus, the very actions of organizing involve the drawing of distinctions; organization itself is a Exemplars indicate that managing learning source of tension (Ford & Backoff, 1988: 86-87). paradoxes requires one, or more likely a combi- nation, of three elements that aid social refram- How can organizations operate efficiently and ing: shock, open communication and experimen- adapt continuously? Why do some attempts to tation, and paradoxical leadership. Shock—an increase employee involvement and commit- experience that clearly violates a frame of ref- ment intensify resistance and mistrust? Why do erence—^may force actors to admit the limits of formal procedures aimed at assuring fair treat- extant understandings and practices (Bartunek, ment of employees often trigger claims of injus- 1988: Leonard-Barton, 1992; Miller, 1993; Ropo & tice? Such questions, addressing the conflicting Hunt, 1995; Westenholz, 1993). But shocking situ- aspects of organizational design, are becoming ations might be insufficient to provoke refram- a hallmark of organization studies. Yet Bouch- ing. Subtle means of confrontation might help ikhi (1998) notes that theories often downplay open communications, fostering discussion this complexity in favor of either/or frameworks, among actors with varied frames of reference to such as contingency theories in which the im- surface their divergent insights. Vince and pacts of organic versus mechanistic structures Broussine (1996) used drawings as a means for are distinguished or in which organizations are actors across levels to recognize and discuss categorized as stable oi in flux. In contrast, I their anxieties, whereas Leonard-Barton (1992) now review studies in which dynamic and op- found that experimentation reduced the risk of posing forces coexisting within organizations testing new frames and practices. Westenholz are accented. If we use a paradox framework, (1993) proposed that a forum conducive to open organizing appears to inherently juxtapose con- discussion and experimentation might aid so- tradictions. Exemplars depict the intricate na- cial reflection. In support, other exemplars illus- ture of performance, empowerment, and formal- trate how isolating a diverse group enabled ization as managers seek to encourage their developing a new identity as "resistors," commitment, participation, and trust and in- liberating them to battle organizational rigidi- crease efficiency, coordination, and security. ties (Leonard-Barton, 1992; Ropo & Hunt, 1995). Tensions between control and flexihility un- Finally, paradoxical leadership may be vital to derlie paradoxes of organizing, typically mani- managing learning paradoxes. A paradoxical fested as mixed messages and system contra- leader is an influential actor capable of thinking dictions. Quinn (1988; see also Denison et al., paradoxically and guiding social reflection, 1995), for instance, uses a paradox framework to helping others examine, rather than suppress, explore conflicting performance demands, such their tensions (Leonard-Barton, 1992; Westen- as concerns for morale, profitability, innovation, holz, 1993). and stability. Similarly, Bouchikhi defines or- In sum, the reviewed exemplars depict the ganizations as "social spaces continuously torn need for researchers to explore paradoxical ten- by members in multiple and contradictory direc- sions that inhibit and foster learning. In future tions" (1998: 224). studies scholars can extend these works by ex- Others note the rise of such tensions during amining the simultaneity of construction and attempts to foster empowerment or to increase destruction, ior learning requires using, critiqu- formalization. Sabelis (1996) turned to paradox  768 Academy of Management Review October when she became puzzled by system contradic- tion to achieve high performance. In the least tions during a case study. While implementing successful quartet examined, they found that as teams to foster employee autonomy, manage- rehearsals became chaotic and calls for leader- ment also introduced an automatic time regis- ship intensified, the first violinist sought greater tration system to track employees' comings and participation, feeding frustrations and impeding goings. Such contradictions are often inherent in performance. Similarly, Sitkin and Bies (1993) empowerment practices. For example, just-in- claim that legalization improves procedural jus- time (JIT; e.g., Eisenhardt & Westcott, 1988) and tice for highly subjective practices (e.g., hiring total quality management (TQM; e.g., Klein, and grievance-handling practices). Yet, legalis- 1994) practices require employee discretion and tic procedures might protect the powerful, limit problem solving as well as formal statistical human discretion, and create an adversarial cli- and monitoring controls. mate in organizations. As problems arise, man- Sitkin and Bies (1993) used paradox to examine agers often become increasingly reliant on for- legalization: formalization that mirrors legalistic mal methods, fueling further rationalization that procedures. They sought to explain why legaliza- ignores wider social implications and, in turn, tion helps ensure fair treatment of employees and reduces trust and productivity in a vicious offers those in power means of legitimating poten- spiral. tially unfair decisions. Hatch and Ehrlich (1993) Across reviewed works, use of superordinate detailed similar struggles with formalization. To goals, humor, and behavioral complexity ap- fulfill increasing demands for profit, managers peared to help actors manage organizing para- sought to bolster the commitment of subordinates doxes. Several exemplars show how common and protect corporate assets from pilfering. Yet, and expansive goals, which place contradic- implementing security mechanisms (symbols of tions within a wider context, help actors accept suspicion) contradicted their messages of trust paradox. For instance, Murnighan and Conlon and cooperation. (1991) note that quartets sharing a true love for, In several exemplars researchers examine de- or even obsession with, their music were more fenses, such as repression and reaction forma- likely to embrace conflicting needs for authority tion, to help comprehend the negative dynamics and democracy and to thrive. Other researchers of organizing paradoxes. Romanticizing empow- claim that successfully implementing empower- erment practices is a typical form of repression. ment practices such as TQM (Klein, 1994) and JIT For instance, Klein (1994) examined the problem- (Eisenhardt & Westcott, 1988) requires a focus atic nature of TQM, a practice inspired by Jap- upon ultimate goals—for example, continuous anese philosophies of discipline through em- improvement. Actors then might recognize au- powerment. Klein claims that managers often tonomy as vital for fostering creativity and inno- downplay the tensions of TQM to revel in its vation and control as key to tightening effi- controls, the feature most fitting Western organ- ciency standards. izational cultures, and neglect the need for high Hatch found that humor offered actors a employee commitment and autonomy. O'Connor means of neutralizing their "emotional denial so (1995) describes similar behaviors during the im- that contradictory understandings could be con- plementation of a participation program. Top fronted in a less-threatening, more-playful fash- management called for mid-level managers' in- ion" (1997: 287). Hatch and Ehrlich (1993) found volvement while setting binding limits on their that humor conveys powerful messages, helping discretion. Such mixed messages increased the actors express their frustrations with formaliza- resistance and alienation of the mid-level man- tion. As humorous remarks swirled around an agers as they increasingly questioned their image of a prison—for example, gun turrets and roles in such practices. search-and-seizure jokes—managers shared Other exemplars depict reaction formation: their feelings of being guards relative to their actors overreact to tensions of flexibility and subordinates, yet prisoners relative to their su- control by focusing on one pole, thereby spark- periors, who forced the responsibility of security ing a stronger pull from its opposite (e.g., Bouch- upon them. ikhi, 1998). Murnighan and Conlon (1991) found Quinn (1988), in contrast, claimed that truly that string quartets require an authoritarian transcending and prospering from organizing leader (first violinist) and democratic participa- paradoxes requires behavioral complexity: hav- 2000 Lewis 769 ing an expansive repertoire of actions to draw often seemingly absurd nature of membership. upon when responding to conflicting and fluctu- Paradoxical tensions arise because actors strive ating demands. Denison et al. (1995) tested this for both self-expression and collective affilia- premise. They discovered that highly effective tion. Seeking to comprehend their roles in a leaders display intricate and paradoxical be- group, organization, and/or community, mem- havioral repertoires that mirror tenuous condi- bers attempt to artificially distinguish them- tions in the environment. Such leaders are selves (e.g., differentiating personal competen- capable of fostering consistency, stability, cies, occupational practices, or ethnic values). and control, as well as passion, courage, and In the exemplars tensions between self and wonder. other are the crux of belonging paradoxes. Ac- In sum, the more complex, diverse, and dy- cording to Smith and Berg, the paradox of indi- namic organizations become, the more tradi- viduality is a self-referential cycle: groups be- tional either/or thinking oversimplifies manage- come strong and resourceful only if the ment practices and demands. At the extremes, individuality of their members is expressed organizational life appears chaotic or alienat- (1987: 102). Individual expression, however, en- ing. Yet, managing paradoxical tensions de- genders group conflict—that is, conflict capable notes not compromise between flexibility and of fostering more accommodating and novel un- control but awareness of their simultaneity. derstandings and disrupting group decision Exemplars offer both/and insights into organi- making and performance (Amason, 1996). Mur- zational characteristics and performance, em- nighan and Conlon (1991) used paradox as a phasizing the coexistence of authority and de- framework for analyzing such tensions in string mocracy, discipline and empowerment, and quartets. Actors joined quartets, rather than formalization and discretion. Using the paradox symphonies, to have a greater impact on the framework, in future studies researchers can ex- music. Quartets, however, achieved acclaim by plore organizing as an ongoing process of equil- integrating the talents of their players. ibrating opposing forces and detail its tensions, Within large organizations a related paradox cyclical dynamics, and management. appears in the construction of group boundaries, which define membership by differentiating one collective from another. Yet, we/they distinc- Paradoxes of Belonging tions are plentiful and ambiguous. Actors are Because of the connection... between the iden- members of the organization and of varied oc- tity of "seli" and the identity oi "other," and cupations and subcultures within and outside among the multiple frames that can be and are the organization, provoking feelings of inclusion applied to any social event, our experience of and exclusion simultaneously (Martin, 1992). contradiction is often an experience of paradox Ybema (1996) documented such tensions occur- (Smith & Berg, 1987: 61). ring in a Dutch amusement park. Actors shared How do actors become integral members of a a pride in the organization's success but empha- group and retain their individuality? Why do sized social distinctions between new profes- consensus and cohesion appear to coexist with sionals and established craftsmen. conflict and division in organizational life? As In multinational organizations globalization organizations become more global and intercon- appears to provoke similar tensions. Naisbitt nected, why are battles to retain local traditions (1994) used paradox to examine the rise of "uni- escalating? In contemporary organizations blur- versalism," as multinationals connect distant lo- ring hierarchical, occupational, and national cales through central goals and advanced tele- distinctions intensify the paradoxical nature of communications, and the rise of "tribalism," as social relationships. The following reviewed local groups seek to retain their ethnic and lin- works contribute fascinating insights into is- guistic distinctions. sues of individuality (e.g., Amason, 1996; Mur- Paradoxes of belonging often raise actors' de- nighan & Conlon, 1991), group boundaries (e.g., fenses, intensifying conflict, polarization, and Leonard-Barton, 1992; Ybema, 1996), and global- tribalism. According to Smith and Berg (1987), ization (e.g., Koot, 1996; Naisbitt, 1994). According defenses of ambivalence and projection might to Smith and Berg (1987), these issues revolve fuel destructive conflict as actors seek to ex- around paradoxes of belonging: the tenuous and press their differences yet remain valued mem-  770 Academy of Management fleview October bers of a group. Murnighan and Conlon (1991), Three methods appear to help actors manage for example, found that when debates erupted paradoxes of belonging: maintaining a task fo- during musical rehearsals, less successful quar- cus, valuing difference, and reducing power dis- tets often quickly agreed to bland compromises. crepancies. Amason (1996) explains that a task Such ambivalence helped mute opposing per- focus might tap the functional side of conflict— spectives but allowed strong feelings to fester cognitive conflict based on individuals' judg- and resurface in later rehearsals or, worse, in ments regarding how to achieve a common concert. At other times, members would project goal—and avoid the dysfunctional flip side— their frustrations by blaming other members for affective conflict based on emotional and per- poor performance, launching personal attacks sonal disputes. Likewise, Murnighan and Con- that paralyzed the group. lon (1991) found that successful string quartets Other exemplars illustrate the dysfunctional played through their conflicts, using their in- impacts of splitting. By exaggerating cultural tense task to foster individual and group impro- and occupational uniqueness, actors tempo- visation. rarily secure their identity but at the expense of Exemplars also depict the need to value dif- greater mutual understanding (Martin, 1992). ference (e.g., Leonard-Barton, 1992; Murnighan & O'Connor (1995) documented a growing polar- Conlon, 1991). From a paradoxical perspective, ization between managers deemed "in" and valuing difference means appreciating varied "out" of a new participation program. To bolster perspectives and capabilities, rather than ac- their egos, the team implementing the program centuating personal or ethnic distinctions. described those opposed to the change as igno- Valuing difference also might help actors rant of its value. Rather than hear their concerns lessen power discrepancies, which often fuel po- and recognize potential flaws in the program, larization and tribalism (Naisbitt, 1994). Koot dis- the team sent opposing managers away for covered that "if one wants to reduce cultural training, intensifying the managers' resistance distance—paradoxically—one should not and exclusion. mainly focus on the differences in culture, but Leonard-Barton (1992) found that splitting concentrate on the differences in power and con- also might reduce the performance of less flicting interests" (1996; 81). Similarly, Leonard- powerful groups. In the product development Barton (1992) found that when crises (e.g., miss- projects she studied, interdisciplinary cooper- ing deadlines) occurred during product ation was vital. Yet, as members of the domi- development projects, members of the dominant nant discipline (design engineering) sought discipline often began listening and giving cre- and received increasing praise for their skills, dence to views of the less dominant. In some the status and expectations of those in other companies this led to a rise in the status of disciplines (manufacturing and marketing) de- nondominant groups, reversing the negative re- clined. She described the vicious cycle as fol- inforcing cycle by raising the expectations and lows: the lower the expectations of lower- power of these groups. status disciplines, the lower their confidence In sum, as change pressures and growing and subsequent contributions to the project, workforce diversity complicate the social milieu, reinforcing and further lowering expectations. the ability to comprehend belonging paradoxes Other researchers note that in multinational becomes increasingly vital. Exemplars demon- firms, attempts to manage diversity, and strate that the stronger the push toward cohe- thereby to increase organizational cohesive- sion and consensus, the stronger the pull toward ness, often spark tribalism or splitting by minor- individuality and fragmentation. By applying ity groups (e.g., Naisbitt, 1994). Koot discovered the paradox framework, in future studies re- that "creating distance and accenting one's eth- searchers can extend understandings of how nic origin tend to be weapons—very strong groups become cohesive, influential, and dis- ones—in the struggle against the dominant tinctive by valuing the diversity of their mem- party" (1996: 80). By stressing their distinctions, bers and their links with other groups. A health- less powerful and marginalized ethnic groups ier sense of belonging in contemporary seek to retain a sense of local identity, often organizations requires a deep, social accep- subverting organizational performance in the tance of differences, aided by critical examina- process. tion of artificial distinctions. 2000 Lewis 771 IDENTIFYING AND REPRESENTING PARADOX: (Hatch & Ehrlich, 1993). Exemplars depict three EXPLORATION STRATEGIES potentially valuable approaches for iden- tifying paradox: narrative, psychodynamic, Cameron and Quinn note that "Paradoxes are and multiparadigm. These approaches share paiadoxical"—common and surprising, confus- two characteristics. First, they help research- ing and understandable (1998: 13). In the works ers focus on everyday organizational life and reviewed, scholars contribute valuable insights a search for taken-for-granted contradictions. into organizational paradoxes, offering founda- Farson (1996) compares this focus to the The- tions and exemplars for future inquiry. However, atre for the Absurd: a strategy, in direct con- studies explicitly focused on paradox remain trast to artistic renditions that romanticize life, sparse, despite the increasing appearance of which paradoxically serves the dignity of hu- the paradox label in organization studies man beings by revealing confusion and odd- (Bouchikhi, 1998; Davis et al., 1997). One expla- ity. As Hatch and Ehrlich explain, "The under- nation for this scarcity is the inadequacy of tra- noticed aspects of daily life are rich with clues ditional approaches for examining tensions. As to cultural meaning and processes of sustain- Teunissen notes, how can researchers "investi- ing social construction" (1993: 521). Second, gate an intangible phenomenon, like para- these approaches help researchers "read" doxes, with sophisticated research methods complexity (Morgan, 1997), offering techniques which are based on logic, rationality, and con- for exposing conflicting messages, feelings, sistency?" (1996: 18). Exploring paradox requires and perspectives. remaining acutely aware of contradictions and anomalies and expanding our strategies ac- In some exemplars researchers apply a narra- cordingly. In this light, paradoxes may offer tive approach, analyzing discourse to identify challenging, even frame-breaking experiences, paradox. Those using this strategy rely on the pushing researchers to question approaches premise that paradoxes are both recognizable that oversimplify and overrationalize complex and socially constructed through actors' rhetoric phenomena. and conversations. Hatch and Ehrlich (1993), for instance, examined actors' use of humor. To In this final section I extract from reviewed grasp the intangible nature of paradox, they exemplars potentially powerful strategies for searched for the most immediately recognizable identifying and representing paradox. Identi- manifestation, coding laughter as a sign of hu- fying paradox means recognizing and inter- mor a sign of paradox. Humor helped expose preting tensions, and representation signifies conflicting emotions, from playfulness to anxi- methods of conceptualizing, mapping, and ety, and the "duplicity of meaning" from which theorizing paradox. Regardless of whether re- paradoxes arise. For example, in analyzing nu- searchers approach paradoxes accidentally, merous remarks in which firm security was com- seeking to comprehend surprising findings in pared to a prison, these researchers recognized their studies (e.g., O'Connor, 1995; Sitkin & actors' awareness that "accepting a position as Bies, 1993), or purposefully, making paradox a guard makes one a prisoner" of a larger sys- the subject of their inquiry (e.g., Davis et al., tem of control, guarded by the executives of the 1997; Murnighan & Conlon, 1991), these strate- firm (1993: 517). gies might aid exploration. Since detailing these strategies would require a separate ar- Similarly, Hatch (1997) used irony as a clue to ticle, I contain the following discussion to out- paradoxical tensions operating in actors' cogni- lining the premises of each strategy, noting tive and emotional domains, defining irony as a specific techniques used in the reviewed ex- form of humor based on contradiction. For exam- emplars, and recommending sources for more ple, in a meeting with mid-level managers, a elaborate descriptions. general manager described firm executives' ex- pectations for a new inventory reduction pro- gram. He reported the extreme changes de- manded by the executives (e.g., a $32 million Identification: Narrative, Psychodynamic, and dollar drop in inventory in 60 days and the re- Multiparadigm Approaches scheduling of 12 million parts) and then said, A key challenge when exploring paradox is "Other than that, everything went great! [laugh- locating and bracketing the phenomena ter]" (1997: 280). This statement revealed con- 111 Academy of Management Review October flicts between negative and positive emotions. tions, such as "Why doesn't he understand; it's "Great" was both used and understood as ironic. so obvious?" and "He thinks I'm full of s and It represented the organization's emphasis on is telling me what I want to hear." As the actors positive thinking and the value of change while repeatedly stressed their main points, without stressing the painful implications oi the new discussing their varied underlying perceptions, program. Irony served a reflexive purpose: it they became increasingly wedded to their own enabled the managers to express and manage views. Paradoxically, the more cordial the dis- their own thoughts and emotions. Laughter cussion became, the more disagreements inten- helped neutralize the potential for emotion- sified. By feeding information back to the actors, based denial of contradictory feelings {despair/ the researcher sought to help them reflect on opportunity). their tensions and escape vicious cycles via O'Connor (1995) applied a third narrative more open communication. technique. She used textual analysis to iden- Imagery offers another psychodynamic tech- tify paradoxical tensions during the introduc- nique used in some exemplars {e.g., Ybema, tion of a participation program. Employing 1996). Vince and Broussine (1996) employed text—case studies written by the implementa- drawings to prompt actors to "say the unsaid" tion teams—as data, O'Connor searched for mixed messages. Such messages often be- during a disruptive organizational change. The came apparent by the lack of attention given majority of participants drew pictures that jux- to them, highlighting the typical undiscuss- taposed images of dread (e.g., a person pushing able nature of paradox. For instance, she ex- a boulder up a steep hill, a fortress under siege) amined how "inclusion" meant "exclusion," and hope {e.g., a "First Prize" gold cup, a dinghy because labeling those included in the pro- sailing toward a sunset). "The wide spectrum of gram necessarily defined those excluded. As feelings about oneself related to the change was resistance rose among those excluded from illustrated, literally" (1996: 16-17). Discussing the change process, the text increasingly jux- the drawings helped the actors distance them- taposed "resistance" with language denoting selves from the anxieties of work and to legiti- ignorance. Teams appeared to rationalize dis- mate and share their feelings. Actors were able agreement as a need for education. This form to stay with the uncertainty long enough to work of denial, in addition to the splitting of actors through their tensions by connecting surface into "in" and "out" groups, intensified the {actual work issues) and more latent {emotions teams' tensions between encouraging cooper- and assumptions) concerns. ation and applying coercion. A multiparadigm approach entails using op- A psychodynamic approach offers additional posing theoretical perspectives as sensitizing techniques for identifying paradox. Researchers devices {see Lewis & Grimes, 1999, and Poole & conduct research with, not on, actors to help Van de Ven, 1989, for more extensive discus- both the researchers and organizational actors sions). According to Teunissen, locating para- recognize inner conflicts. Argyris (1993; Argyris dox requires taking into consideration the diver- & Schon, 1996) detailed several psychodynamic sity of interpretations possible in order to detail techniques. For instance, one method requires contradictions "in absurdity and weirdness or in actors to track their conversations—writing brightness and curiosity" (1996: 29). A paradigm what they say in the right column and what they denotes a way of thinking about phenomena think in the left. At the end of the conversation, based on distinct epistemological and method- the researchers and actors compare the actors' ological assumptions. Exemplars depict alterna- right and left columns for mixed messages and tive paradigms as offering divergent yet com- signs of defensiveness. plementary lenses (e.g., radical/functionalist, In a seminar, for example, two actors debated subjectivist/objectivist), which may identify mul- the value of different strategic actions. Analyz- tiple perceptions of a situation. Martin (1992), for ing the actors' right and left columns, they real- example, used three perspectives to analyze an ized how each actor sought to avoid conflict by organizational culture. An integrative lens not expressing more bluntly why they felt the helped expose perceptions of a cohesive culture, other's plan would not work. Over time, the ac- a differentiation lens highlighted conflict and tors' left-hand columns indicated rising frustra- division, and a fragmentation lens revealed am- 2000 Lewis 773 biguity. By applying three lenses, she became conflict both enhances and destroys groups. acutely aware of how Cognitive conflict denotes task-oriented de- some things will be consistent, clear, and gener- bates that are focused on perceptual diversity^ ate organization-wide consensus. Simulta- differences in how actors perceive a situation— neously, other aspects of the culture will coalesce and might extend the scope and creativity of within subcultural boundaries and still other el- decision making. In contrast, affective conflict is ements of culture will be fragmented, in a state of emotional and aimed at personal disputes, in- constant flux, and infused with confusion, doubt, tensifying actors' defenses and tendency to and paradox (1992: 5). cling to extant frames. Similarly, Ybema (1996) collected thick anthro- Similarly, Quinn (1988) conceptualized perfor- pological data that enabled analysis from dis- mance along multiple dimensions: external and parate perspectives. By helping him examine internal, control and flexibility. Denison et al. language used in private conversations, an in- (1995), in turn, sought to test Quinn's model in a terpretivist lens exposed contrasting values study of leadership and behavioral complexity. held by new professionals and established Rather than forming a 2 x 2 matrix, they used craftsmen. Each subculture stressed its unique multidimensional scaling to portray leadership skills (e.g., technical know-how versus experi- as a circular and spatial pattern of eight con- ence) and berated the other's. In contrast, a pos- flicting roles that may coexist (e.g., innovator/ itivist lens unveiled their shared pride in the coordinator, director/mentor). The operation- organization and concern for its success by fo- alized construct expands understandings of cusing on formal mission statements and dis- distinct and mutually exclusive leadership cussions in large meetings. By using multiple styles, enabling examination of whether highly paradigms, Ybema recognized that "the com- effective leaders behave more paradoxically plexity of group membership, the ambiguity of than those who are less effective. rules and rituals, and the coexistence of com- Strategies for mapping paradox provide po- mon and contrasting interests and identities are tentially valuable means of representing ten- all reasons to expect unity and division to occur sions and defenses in action. Ropo and Hunt successively or simultaneously" (1996: 43). (1995), for instance, used grounded theory devel- opment to study entrepreneurship over time. An- alyzing a series of events that took place during Representation: Conceptualizing, Mapping, the deregulation of two Finnish banks, they and Theorizing Paradox mapped the complexity of reinforcing cycles. Exemplars indicate several strategies for rep- Unlike static and discrete models of entrepre- resentation—the daunting task of conveying the neurship, they posed a provocative image of intricate and seemingly absurd nature of para- a holistic and dynamic process. They mapped dox. These strategies for conceptualizing, map- vicious and virtuous cycles, depicting how the ping, and theorizing paradox share a conscious banks came to manage the pull between com- awareness of space and/or time, for effectively petitiveness/noncompetitiveness and change/ depicting paradox requires detailing its context stability by reframing actors' views of strategic and multiple dimensions, as well as providing a decision making, the organization, and the dynamic view of tensions, defenses, and their banking industry. management. Similarly, Argyris (1993) detailed a method of Conceptualizing paradox entails building mapping conversations. Documenting the con- constructs that accommodate contradictions. text, behavior, and consequence of successive Rather than polarize phenomena into either/or statements serves to organize conversational notions, researchers need to use both/and con- processes into an explanatory model of how de- structs for paradoxes, allowing for simultaneity fenses reinforce patterns of behavior. Such mod- and the study of interdependence. For example, els might illustrate how vicious cycles ensue to depict the paradoxical nature of conflict, when actors cling to existing frames of reference Amason (1995) moved to a higher level of ab- and avoid critical self-reflection and more open straction capable of encompassing its multiple discourse. Models are then testable by seeing if dimensions. By distinguishing between cogni- an actor's pattern persists in differing conversa- tive and affective conflict, he illustrated how tions, or if closely related others (e.g., members 774 Academy of Management Review October of the same group) share the pattern. The result the complications of organizational life—and is a rich image of the nature and dynamics of learning to explore the natural ebb and flow oi mixed messages. tensions. Theorizing paradox entails developing a Increasingly, organization theorists claim that frame that encompasses opposites, enabling a making sense of rising plurality and change more complicated understanding of their coex- might require alternative frames—frames that istence and interrelationships {see Bouchikhi, help researchers and practitioners explore par- 1988, and Poole & Van de Ven, 1989). According to adox {e.g.. Ford & Ford, 1994; Handy, 1994; Quinn Argyris (1993), a theory of paradox accounts for et al., 1994). Such frames are appearing in other the seemingly illogical, ideally by depicting ten- sions, defenses, and their management in dy- areas of management. For example, Csikszent- namic and multidimensional detail. Such theo- mihalyi (1997) used paradox to examine every- rizing requires theorists to think paradoxically day tensions. He explored how self-reflecting on themselves: to work toward tensions and the the often conflicting yet interwoven facets of anxieties they provoke in search of insightful professional life might help individuals find the interconnections. flow between work and family, theory and prac- tice, research and teaching. Lewis and Dehler In the reviewed works researchers theorize paradox through rich case studies (Leonard- (in press) applied a paradox framework to man- Barton, 1992; Westenholz, 1993), anthropological agement education, posing a strategy and exer- inquiry (e.g., Koot, 1996; Ybema, 1996), or litera- cises for examining tensions, the defenses they ture reviews {e.g.. Miller, 1993; Sitkin & Bies, provoke, and means of managing paradox in the 1993). Rather than parse organizational life into classroom. A "pedagogy of paradox" may help polar distinctions and rational prescriptions complicate students' perceptions of what it that mask complexity, they depict the chal- means to manage in a turbulent world. lenges of plurality and change embedded in This article furthers development of a compre- cognitive, emotional, and social processes. As hensive guide to exploring paradox in research. Poole and Van de Ven (1989) suggest, many use I advocate using paradox as much more than a notions of time and/or space to explain paradox. label; as a thought-provoking tool or perspec- For example, some depict how opposing ele- tive. The paradox framework serves as a lens for ments may emanate from different locales in the examining surprising findings and seemingly organization, such as in private or public con- absurd aspects of organizational life. This versations {Ybema, 1996), or from the percep- framework might help researchers address tions of managers or subordinates (Martin, 1992). Others depict contradictions operating at vari- what tensions exist, why they may fuel reinforc- ous points in time—for example, organizations ing cycles, and how actors may manage para- appear stable and calm at one point and dy- doxes to foster change and understanding. In namic and chaotic the next {e.g., Ropo & Hunt, the reviewed exemplars scholars offer models 1995). for future inquiry, illustrating the value of using paradox to examine multifaceted and puzzling phenomena. These studies provoke new in- CONCLUDING NOTE sights into learning, organizing, and belonging As dramatic changes in technology, competi- by offering what appear initially as counterin- tion, and globalization continue, "the Chinese tuitive premises but, on deeper reflection, seem curse, 'May you live in interesting times' seems inherent aspects of daily life. The overview of to have become almost too real" {Kets de Vries, research strategies broadly outlines means of 1995: xiii). Indeed, the rising intricacy, ambigu- identifying and representing paradox. In combi- ity, and diversity of organizations place a pre- nation, this still-emerging guide might help re- mium on researchers' abilities to think paradox- searchers actively seek out paradox or explain ically: to live and even thrive within the anomalies as they arise to extend understand- plurality and changes of organizational life and ings of contradictory emotions, demands, and help practitioners do likewise. Building this ca- practices—understandings more in tune with pacity requires confronting our own defenses— the paradoxical nature of individuals, groups, the desire to overrationalize and oversimplify and organizational life. 2000 Lewis 775 Life, disconcertingly and reassuringly, is bigger Farson, R. 1996. Management of the absurd: Paradoxes in than straight-line logic, it conforms with a kind of leadership. New York: Simon and Schuster. curved logic which turns things around and often, Ford. I. D., & Backoff, R. W. 1988. Organizational change in before you become aware of it, turns them into and out of dualities and paradox. In R. E. Quinn & K. S. their opposites. Cameron (Eds), Paradox and transformation: Toward a Pacifists become militants. theory of change in organization and management: 81- Freedom fighters become tyrants. Blessings become curses. 121. Cambridge, MA: Ballinger. Labor saving devices become intolerable Ford, J. D., & Ford, L. W. 1994. Logics of identity, contradiction, burdens. and attraction in change. Academy of Management Help becomes hindrance. Review, 19: 756-795. More becomes less (VanderBroek, quoted in Hampden-Turner, C. 1981. Maps of the mind. New York: Mac- Thompson, 1988). millan. Handy, C. 1994. The age of paradox. Cambridge, MA: Har- vard Business School Press. REFERENCES Harris, A. S. 1996. Living with paradox: An introduction to Amason, A. C. 1996. Distinguishing the effects of functional Jungian psychology. Albany, NY: Brooks/Cole. and dysfunctional conflict on strategic decision making: Resolving a paradox (or top management teams. Acad- Hatch. M. J. 1997. Irony and the social construction of contra- emy of Management Journal, 39: 123-148. diction in the humor of a management team. Organiza- tion Science, 8: 275-288. Argyris, C. 1993. Knowledge for action: A guide (o overcom- ing barriers to organizational change. San Francisco: Hatch, M. J., & Ehrlich. S. B. 1993. Spontaneous humour as an Jossey-Bass. indicator of paradox and ambiguity in organizations. Organization Studies, 14: 505-526. Argyris, C, & Schon, D. A. 1996. Organizational /earning II: Theory, method, and practice. Reading, MA: Addison- Hofstadter, D. R. 1979. Godel, Escher, Bach: An eternal golden Wesley. braid. New York. Vintage Books. Bartunek, J. M. 1988. The dynamics of personal and organi- Kelly. G. 1955. The psychology of personal constructs. New zational reframing. In R. E. Quinn & K. S. Cameron (Eds.), York: Norton. Paradox and transiormation: Toward a theory oi change Kets de Vries, M. F. R. 1995. Organizafiona/ paradoxes: Clin- in organization and management: 137-162. Cambridge, ical approaches to management. New York: Routledge. MA: Ballinger. Klein, J. A. 1994. The paradox of quality management: Com- Bateson, G. 1972. Steps to an ecology of mind. San Francisco: mitment, ownership, and control. In C. Heckscher & Chandler Publishing. A. Donnellon (Eds.), The post-bureaucratic organization: Bouchikhi, H. 1998. Living with and building on complexity: New perspectives on organizational change: 178-194. A constructivist perspective on organizations. Organiza- Thousand Oaks, CA: Sage. tion, 2: 217-232. Koot. W. 1996. The rhetoric of synergy and the practice of Cameron, K. S., & Quinn, R. E. 1988. Organizational paradox increasing ethnic rivalry in organization. In W. Koot, and transformation. In R. E. Quinn & K. S. Cameron I. Sabelis, & S. Ybema (Eds.), Contradictions in context: (Eds.), Paradox and transformation: Toward a theory of Puzzling over paradoxes in contemporary organizations: change in organization and management: 12-18. Cam- 63-86. Amsterdam: VU University Press. bridge, MA: Ballinger. Koot, W., Sabelis, I., & Ybema, S. 1996. Epilogue. In W. Koot, Cannon. T. B. 1996 Welcome (o the revolution: Managing I. Sabelis, & S. Ybema (Eds.), Contradictions in context: paradox in the 21st century. London: Pitman. Puzzling over paradoxes in contemporary organizations: Csikszentmihaiyi, M. 1997. Finding How: The psychology of 208-212. Amsterdam: VU University Press. engagement with everyday life. New York: Basic Books. Leonard-Barton. D. 1992. Core capabilities and core rigidi- Davis. A. S.. Maranville, S. J.. 8f Obloj, K. 1997. The paradox- ties: A paradox in managing new product development. ical process of organizational transformation: Proposi- Strategic Management Journal. 13: 111-125. tions and a case study. Research in OrganizationaJ Lewis, M. W.. & Dehler, G. E. In press. Learning through para- Change and Development. 10: 275-314. dox: A pedagogical strategy for exploring contradictions Denison, D., Hooijberg, R., & Quinn, R. E. 1995. Paradox and and complexity. Journal of Management Education. performance: Toward a theory of behavioral complexity in Lewis. M. W., & Grimes, A. J. 1999. Metatriangulation: Build- managerial leadership. Organization Science, 6: 524-540. ing theory from multiple paradigms. Academy of Man- Eisenhardt, K. M., & Westcott, B. J. 1988. Paradoxical de- agement Review, 24: 672-690. mands and the creation of excellence: The case of just- in-time manufacturing. In R. E. Quinn & K. S. Cameron Martin, J. 1992. Cultures in organizations: Three perspectives. (Eds.), Paradox and transiormation: Toward a theory of New York: Oxford University Press. change in organization and management: 169-194. Miller, D. 1993. The architecture of simplicity. Academy of Cambridge, MA: Ballinger. Management Review. 18: 116-138. 776 Academy of Management Review October Morgan, G. 1997. /mages of organization. Thousand Oaks, Sitkin. S.. & Bies. R. J. 1993. The legalistic organization: Def- CA: Sage. initions, dimensions, and dilemmas. Organization Sci- Murnighan. J. K.. & Conlon, D. E. 1991. The dynamics of ence: 4: 345-351. intense work groups: A study of British string quartets. Smith. K. K., & Berg. D. N. 1987. Paradoxes of group life. San Administrative Science Quarterly. 36: 165-186. Francisco: Jossey-Bass. Naisbitt, J. 1994. Global paradox. New York: Avon. Teunissen. J. 1996. Paradoxes in social science and re- O'Connor, E. S. 1995. Paradoxes of participation: Textual search. In W. Koot. L Sabelis, & S. Ybema (Eds.), Con- analysis and organizational change. Organization Stud- tradictions in context: Puzziing over paradoxes in con- ies, 16: 769-803. temporary organizations: 17-38. Amsterdam: VU Poole, M. S., & Van de Ven, A. H. 1989. Using paradox to build University Press. management and organization theories. Academy of Thompson. M. P. 1988. Being, thought, and action. In Management Review. 14: 562-578. R. E. Quinn & K. S. Cameron (Eds.), Paradox and Putnam, L. L. 1986. Contradictions and paradoxes in organi- transformation: Toward a theory of change in organi- zations. In L, Thayer (Ed.). Organization communica- zation and management: 123-136. Cambridge, MA: tions: Emerging perspectives: 151-167. Norwood, NJ: Ballinger. Ablex Publishing. Vince. R., & Broussine. M. 1996. Paradox, defense and attach- Quinn, R. E. 1988. Beyond rafionaJ management San Fran- ment: Accessing and working with emotions and rela- cisco: lossey-Bass. tions underlying organizational change. Organization Ouinn, R. E., Kahn, J. A.. & Mandl, M. J. 1894. Perspectives on Studies, 17: 1-21. organizational change: Exploring movement at the in- Watzlawick, P., Weakland, J. H.. 8f Fisch. R. 1974. Change: terface. In J. Greenberg (Ed.), Organizational behavior: Principles of problem formation and problem resolution. The state of science: 109-133. Hillsdale, NJ: Lawrence New York: Norton. Eribaum Associates. Weslenholz, A. 1993. Paradoxical thinking and change in Ropo, A., & Hunt, J. G. 1995. Entrepreneurial processes as frames of reference. Organization Studies, 14: 37-58. virtuous and vicious spirals in a changing opportunity structure: A paradoxical perspective. Entrepreneurship Weick, K. E. 1979. The social psychology of organizing. Read- Theory and Practice, 19(3): 91-111. ing, MA: Addison-Wesley. Rothenberg. A. 1979. The emerging goddess. Chicago: Uni- Weick. K. E.. & Westley. F. 1996. Organizational learning: versity of Chicago Press. Affirming an oxymoron. In S. R. Clegg, C. Hardy. & W. R. Sabelis, I. 1996. Temporal paradoxes: Working with cultural Nord (Eds.). Handbook of organization studies: 440-458. diversity in organizations. In W. Koot, I. Sabelis. & Thousand Oaks, CA: Sage. S. Ybema (Eds.). Contradictions in context: Puzzling over Ybema, S. 1996. A duck-billed platypus in the theory and paradoxes in contemporary organizations: 171-191. analysis oi organizations: Combinations of consensus Amsterdam: VU University Press. and dissensus. In W. Koot. I. Sabelis, & S. Ybema (Eds.). Schneider, K. J. 1990. The paradoxical self: Toward an under- Contradictions in context: Puzzling over paradoxes in standing of our contradictory nature. New York: Insight contemporary organizations: 39-61. Amsterdam: VU Uni- Books. versity Press. Marianne W. Lewis is currently an assistant professor of management at the Univer- sity of Cincinnati. She received her Ph.D. in management irom the University of Kentucky. In her research she explores tensions, conflicts, and paradoxes that impede and enable innovation, particularly in the areas of advanced manufacturing technol- ogy, product development, and organization theory.