HOW DYNAMIC CAN ORGANIZATIONAL … · Strategic Management Journal Strat. Mgmt. ... How Dynamic Can Organizational Capabilities Be? 915 ... tional capability is also a historical

Strategic Management JournalStrat. Mgmt. J., 28: 913–933 (2007)

Published online 3 April 2007 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/smj.613

Received 19 January 2005; Final revision received 10 October 2006

HOW DYNAMIC CAN ORGANIZATIONALCAPABILITIES BE? TOWARDS A DUAL-PROCESSMODEL OF CAPABILITY DYNAMIZATION

GEORG SCHREYOGG* and MARTINA KLIESCH-EBERLInstitute of Management, Freie Universitat Berlin, Berlin, Germany

The recent discussion in the field of strategic management broadly favors the idea of dynamiccapabilities in order to overcome potential rigidities of organizational capability building. Themajor question addressed in this paper is whether capabilities can actually be conceived asbeing in flux—and if so, to what extent and in which way? After briefly recapitulating thedistinguishing features of organizational capabilities, path dependency, structural inertia, andcommitment are identified as the main capability-rigidity drivers causing a managerial dilemma.In the search for a resolution of this dilemma different approaches of dynamic capabilities areidentified and discussed. The analysis shows that the approaches suffer from inherent conceptualcontradictions: the dynamization runs the risk of dissolving the original idea and strengthof organizational capability building. Ultimately, capabilities would lose the strategic powerattributed to them in the resource-based view. The last section of this paper therefore aims todevelop an alternative approach, which aims at preserving the original merits of organizationalcapability and solving the rigidity issue not by integrating a dynamic dimension into thecapability construct but rather by establishing a separate function (‘capability monitoring’).The suggestions mount up to a tier solution. Its logic builds on the dynamics of countervailingprocesses and second-level observation. Copyright 2007 John Wiley & Sons, Ltd.

INTRODUCTION

The concept of organizational capability hasattracted a lot of interest primarily in the fieldof strategic management. In the resource-basedview (RBV) organizational capabilities have beenidentified as one major source for the gener-ation and development of sustainable competi-tive advantages (Barney, 1991; Wernerfelt, 1984).Incomplete factor markets allow for heterogeneityamong firms in terms of resources and capabilities.

Keywords: resource-based view; evolutionary economics;organizational capabilities; dynamic capabilities; organi-zational rigidities; commitment; monitoring*Correspondence to: Georg Schreyogg, Institute of Manage-ment, Freie Universitat Berlin, Garystr. 21, 14195 Berlin, Ger-many. E-mail: [email protected]

Upon these differences firms can build competitiveadvantages and rent differentials (Amit and Schoe-maker, 1993; Barney, 1997; Peteraf, 1993). Thusthe strategic position of a firm varies systematicallywith the availability and allocation of resourceswhich are rare and superior in use, relative to oth-ers. Central to the generation of a (sustainable)competitive advantage is the capability of an orga-nization to create more value than the least efficientcompetitor (Peteraf and Barney, 2003: 314).

The identification of social and behavioralfeatures of resources and capabilities has beeninformed by insights from evolutionary economics(Helfat and Peteraf, 2003; Nelson and Winter,1982; Winter, 2003). Capabilities are developed inthe context of organizational resource allocationwhich is embedded in idiosyncratic social

Copyright 2007 John Wiley & Sons, Ltd.

914 G. Schreyogg and M. Kliesch

structures. On this basis capabilities are conceivedas distinct behavioral patterns, which are complexin nature involving both formal and informalprocesses (Dosi, Nelson, and Winter, 2000; Hoferand Schendel, 1978; Sanchez and Mahoney, 1996).Capabilities represent a repository of historicalexperiences and organizational learning (Winter,2000). In case of superior performance and aunique historical development, capabilities areassumed to build the foundation for sustainablecompetitive advantage.

Recently in the capability debate, the issues ofvolatile markets, environmental uncertainty, andchange have come to the fore. Building on theobservation that markets and superior market posi-tions have increasingly become subject to erosionprocesses, the reliance on a specific set of nur-tured capabilities has been called into question.Instead the emphasis has shifted to the ability tochange and quickly develop new organizationalcapabilities as a critical prerequisite for sustain-ing competitive advantages. The salient conceptsin this debate are ‘dynamic capabilities’ (Teece,Pisano, and Shuen, 1997; Kusunoki, Nonaka, andNagata, 1998; Eisenhardt and Martin, 2000; Zolloand Winter, 2002; Winter, 2003) or ‘dynamic corecompetencies’ (Danneels, 2002; Lei, Hitt, and Bet-tis, 1996), both call for a profound dynamiza-tion of organizational capabilities. The notion of‘dynamic’ is devoted to addressing the continu-ous renewal of organizational capabilities, therebymatching the demands of (rapidly) changing envi-ronments. The concept of dynamic capabilitiesrevises the RBV insofar as not only the marketsbut also the organizational capabilities are concep-tualized as being dynamic and flexible (Helfat andPeteraf, 2003: 998).

From our point of view, however, the postula-tion of continuous renewal on the one hand and thepatterned architecture of organizational capabilitieson the other constitutes a serious disparity imply-ing far-reaching theoretical and practical conse-quences. So far, this deep-seated discrepancy hasnot gained noticeable attention. This paper there-fore aims to elaborate on this disparity.

We start by asking the question of whether adynamic dimension can be included in the conceptof organizational capability without sacrificing the-oretical coherence—and if so, to what extent?The analysis will show that aside from undis-putable merits, the idea of dynamizing capabilitiesis prone to ‘throwing out the baby with the bath

water.’ The suggested dynamization is likely tocrowd out the genuine essence of an organiza-tional capability. In reaction to this conclusion wesuggest an alternative conception, which aims topreserve the genuine strengths of organizationalcapabilities and assure the dynamization of corpo-rate capability management in a different way byestablishing a separate function (‘capability mon-itoring’). After explaining its main elements andthe underlying logic, the closing section outlinespractical implications and discusses conclusionsfor future research on organizational capabilitiesand the requirements of a dynamic resource-basedtheory.

As the conceptions of organizational capabilityhave varied broadly it seems imperative to beginthe discussion by clarifying what organizationalcapabilities are supposed to mean—thereby estab-lishing a solid basis for the subsequent analyses.

WHAT ARE ORGANIZATIONALCAPABILITIES?

In strategic management organizational capabili-ties are depicted as critical success factors andthese days nearly every organization wants to beperceived as being capable of doing something inan outstanding manner. But what exactly does a‘capability’ mean and what are its essential fea-tures? A closer look at the literature reveals thatthe conception has often been left vague (Col-lis, 1994). Some authors address it as a well-known colloquial expression, while others empha-size particular dimensions only. It therefore seemsadvisable to start the discussion by clarifying themeaning of organizational capabilities. There isa huge variety in the literature on the labeling:some authors call it (core) competence; others callit collective skills, complex routines, best prac-tices, or organizational capabilities. In the contextof dynamics and change the term ‘capability’ hasgained predominance. We therefore use this termthroughout the article without denying the meritsof the other constructs.

There seems to be a consensus that a capabilitydoes not represent a single resource in the concertof other resources such as financial assets, tech-nology, or manpower, but rather a distinctive andsuperior way of allocating resources. It addressescomplex processes across the organization suchas product development, customer relationship, or

Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 913–933 (2007)DOI: 10.1002/smj

How Dynamic Can Organizational Capabilities Be? 915

supply chain management. In contrast to rationalchoice theory and its focus on single actor deci-sions, organizational capabilities are conceived ascollective and socially embedded in nature. Theyare brought about by social interaction and repre-sent a collectively shared ‘way of problem-solving’(Cyert and March, 1963). Accordingly, organiza-tional capabilities can be built in different fieldsand on different levels of organizational activity,for instance at departmental, divisional, or corpo-rate level.

From a conceptual point of view three charac-teristics seem to stand out:

Problem-solving and complexity

Capabilities are conceptualized in the context ofcollective organizational problem-solving. Capa-ble firms are assumed to solve emerging prob-lems effectively. A capability, however, is notattributed unless outstanding skills have provedto have solved extraordinary problems (otherwisecompetitive advantages could not be built). In mostcases extraordinary tasks and skills are understoodin terms of complexity (Levinthal, 2000).

The notion of complexity refers to the charac-teristics of problem situations and decision makingunder uncertainty (Dosi, Hobday, and Marengo,2003; Duncan, 1972), addressing ambiguous, ill-structured tasks (March and Olsen, 1976; Marchand Simon, 1958). Solving complex tasks requiressophisticated abilities with a broad absorptivecapacity (Cohen and Levinthal, 1990; Zahra andGeorge, 2002). As is well known from cybernetics,complexity needs complexity (Ashby, 1965). Thecomplexity of a capability therefore reflects theinternal requirements for mastering complex tasks.Problem-solving can be defined as a sequenceof generating complex combinations of cognitiveand habitual acts (Dosi et al., 2003: 170). Theseacts focus primarily on finding all the relevantresources needed and combining them effectively(Kogut and Zander, 1992). Due to its complexity,the organization may effectively solve challengingproblems without understanding the inherent logicof its capability; its internal functioning is likelyto remain opaque. As organizational capabilitiesare not the result of planned corporate conductbut emerge incrementally from daily interactionthey are often considered as ‘somewhat mysterioussocial phenomena’ (Dosi et al., 2000: 1).

Practicing and success

Capabilities are close to action; conceptually theycannot be separated from acting or practicing. Atthe same time, embedding organizational capabil-ities in practicing or doing means that capabilityrepresents more than explicit knowledge; it cov-ers more dimensions of an action: emotions, tacitknowing, and bodily knowledge (Polanyi, 1958,1966). Practicing a capability therefore meansa ‘generative dance’ (Cook and Brown, 1999)between explicit and tacit elements.

Furthermore, capabilities are bound to perfor-mance; they are conceived as doing somethingthat ‘must be recognized and appreciated’ (Gher-ardi and Nicolini, 2002: 421; Weinert, 2001). Theyare only recognized and attributed to a performingsocial entity in the case of a success (as comparedto other organizations, which are less capable atreaching such effective solutions). Finally, a singlecase of successfully mastering a problem situationdoes not on its own amount to an organizationalcapability. Actually, the notion of capability refersto habitualized action patterns. Some authors there-fore refer to the concept of routines as the buildingblocks of organizational capabilities (Dosi et al.,2000; Nelson and Winter, 1982; Winter, 2000).

Reliability and time

‘At a minimum, in order for something to qualifyas a capability, it must work in a reliable manner’(Helfat and Peteraf, 2003: 999). Capabilities rep-resent a reliable pattern: a problem-solving archi-tecture composed of a complex set of approvedlinking or combining rules. In other words, a set ofproblem-solving activities is not called a capabilityunless it has proved to be successful across varioussituations and organizations are able to reproduceit. As a consequence for an organization the abilityto transform an accidentally successful coordina-tion effort into a reliable problem-solving patterngains critical importance (Hannan and Freeman1977, 1984). A singular success can trigger thebuilding of a capability but a capability is notactually constituted unless a reliable ‘practice’ hasevolved over time. By implication, an organiza-tional capability is also a historical concept by itsvery nature, integrating past experiences with thepresent problem-solving activities and a prospectfor future direction of resource allocation.

Stressing the historical nature of organizationalcapabilities refers to the fact that time is a basic



dimension of capabilities. Capability developmentcomes close to a chain of reactions triggered byan initial event, thereby establishing a capabilitytrajectory. Capability development takes time andthe specific way in which time has been taken (i.e.,the intensity, frequency, and the duration of socialinteractions) is relevant for the gestalt of a capa-bility. At the same time the particular importanceof this process means that there are no time com-pression economies (Dierickx and Cool, 1989). Itshould be reiterated here that it is exactly this time-intensive and not fully understandable evolvementthat makes up the non-imitable essence of thestrategic relevance of organizational capabilities(Barney, 1991; Leonard-Barton, 1992).

Overall, any organizational capability is the resultof an organizational learning process, a processin which a specific way of ‘selecting and link-ing’ resources gradually develops. Although orga-nizational capabilities apply to various problemsituations, they do not apply everywhere. Theyhave been formed through successful responses tospecific historical challenges and are thus boundto specific types of constellations (Winter, 2003).Problem-solving is embedded in organizationaldesign, information procedures, micropolitics andcommunication channels, as well as other orga-nizational characteristics (culture, control-regimes,etc.). All of these features shape organizationalcapabilities (Henderson and Clark, 1990) and thusdefine their distinctiveness.

THE PARADOX OF ORGANIZATIONALCAPABILITIES

Nowadays organizational capabilities are highlyvalued attributes of firms; organizations want tobe perceived as possessing salient capabilities. Thecompetent and capable organization has becomea new ideal. From this perspective organizationswould be well advised to invest further in theircurrent capability set and to build their strategiesupon them.

A closer look at capability-based behavior andcompetition reveals, however, a much moreambiguous reality than the hymns of praise lead usto expect. The replication of successful and com-plex selection and linking patterns has its darksides too. This becomes evident particularly involatile environments and dynamic competitionwith changing rules of the competitive game. In

all of these cases organizational capabilities mayeasily invert from a strategic asset into a strategicburden. The strengths of capability-based behav-ior and its recursive reproduction can add up toa barrier to adaptation and a burden with respectto flexibility and change. The critical focus is onthe inability of organizations to change their famil-iar ‘ways of doing’ when confronted with newdevelopments. This inherent tendency to inertiaforms the very basis of the recent capability debateresulting in the call for ‘dynamic capabilities.’ Thisparadoxical persistence in the face of a changingenvironment needs further explanation. From ourpoint of view three main causes for the paradoxcan be identified.

Path-dependency and lock-in: One reason whyorganizations are often overly persistent in theirstrategic orientation is path dependence incapability-based activity. Path dependency meansfirst of all that ‘history matters’ (David, 1985),i.e., that a company’s current and future decisioncapabilities are imprinted by past decisions andtheir underlying patterns (Arthur, 1989; Cowanand Gunby, 1996). In many cases path depen-dency means, however, more than mere historicalimprinting: it refers rather to forceful dynamicscalled ‘increasing returns’ (Arthur, 1983). Thatis, once successful combinatorial activities gen-erate positive feedback loops, thereby emergentlyconstituting self-reinforcing processes. As is wellknown from empirical studies (Burgelman, 2002a;David, 1985; Helfat, 1994) such self-reinforcingprocesses may establish strategic paths which areprone to dramatically narrowing the scope ofstrategic management. In the worst case a specificorientation becomes locked, i.e., any other strategicalternative is excluded.

The same is true for capability building wherepositive feedback-processes are likely to bringabout path dependency in capability-based prac-tices. In other words, organizational capabilitiesor core competencies are prone to become fixedto the constellations in which they proved to besuccessful. If the constellations do not change sig-nificantly, this latent fixation does not add up toa problem. In the reverse case, however—and wedoubtless live in a world of change (in terms ofcompetitors, market structure, rules of competition,etc.)—new parameters determine competitive suc-cess, and the capability-driven action patterns are



likely to bind the organization to the past (and notto the future).

Apart from path dependency, other causes pro-vide additional insight into the paradox of capabil-ities.

Structural inertia: In their evolutionary frame-work, Hannan and Freeman stress the importanceof the ‘unusual capacity to produce collective out-comes of a certain . . . quality repeatedly’ (Hannanand Freeman, 1984: 153) for the survival and sus-tainable success of an organization, insofar as theyconsider ‘organizational inertia’ as a preconditionfor organizational success. Inertia is needed inorder to make an organization reliable and identifi-able as a distinct unit. It is therefore a requirementfor guaranteeing survival.

But, paradoxically, exactly this inertia bringsabout the risk of maladaptation. In the face ofa changing environment, organizations are boundto their stabilized structures and action patterns.Central to survival is the ability to overcome orga-nizational inertia.

Other approaches locate organizational inertiaprimarily in other mechanisms such as change-inhibiting organizational cultures (for instance, a‘kill-the-messenger-of-bad-news culture’), ormicro-political processes (Beatty and Ulrich, 1991;Markides, 1998).

Another stream of literature addresses the capa-bility paradox in the context of organizationallearning. The basic findings are that focusingon improvements of existing capabilities makesexperimentation with alternatives less attractive(Benner and Tushman, 2003; Henderson, 1993;Levitt and March, 1988; Repenning and Sterman,2002). By exploiting current strengths, there is atendency to crowd out explorative activities whichgo beyond the beaten track: ‘As organizationsdevelop greater competence in a particular activ-ity, they engage in that activity more, thus furtherincreasing competence and the opportunity costof exploration’ (Levinthal and March, 1993: 106).These exploitation processes not only lead to a fix-ation to existing capabilities but also prevent thedeveloping of new capabilities or, put differently,‘the pitfall is that this learning increases the rigid-ity of the firm’ (Kogut and Kulatilaka, 2001: 755).Thus capability development resides in the well-known trade-off between exploitation and explo-ration processes in organizational learning (March,

1991), emphasizing the dysfunctional dynamics ofexploitative learning processes.

With the ‘Icarus Paradox’ Miller points to asimilar dynamic thereby referring to the fact thatorganizations facing a long period of (outstand-ing) success inherently develop the fatal tendencyto (over)simplify their operational procedures andto blind the organization to discrepant feedback(Miller, 1993, 1994). A once successful patternmutates into its opposite: a pattern of failure. Thecause of failure paradoxically resides in what wasonce the source of success.

Commitment: Further insight into the texture of thecapability paradox is provided by the commitmentliterature, which highlights the binding effects ofinvestments and the resulting persistence of orga-nizational strategies (Ghemawat, 1991). Commit-ment to a particular strategic thrust is consideredthe prerequisite for sustained competitive advan-tage. The argument is advocated from both aneconomic and a psychological point of view.

The economic dimension focuses on resourceinvestments. On the one hand, firm-specific (andtherefore sticky) investments are needed to builtheterogeneity and superior performance, i.e., togenerate high quality, economies of scale, etc.(Ghemawat and Del Sol, 1998). On the other hand,investments in firm-specific resources are likely tobe irreversible and rigid because the cost of sepa-rating and abandoning such sticky resources is toohigh. In consequence, resource commitment tendsto restrict an organization’s options and flexibil-ity (Bercovitz, de Figueiredo, and Teece, 1996).The more dynamic the environment, the higheris the implied flexibility risk (Winter, 2003). Wedo not go into more detail here because, as wasshown at the beginning, capabilities do not actuallyrepresent a resource; they focus rather on the com-bination and linking of resources. Although thereare interactions between them, resources and capa-bilities represent two different conceptual levelswith their own commitment dynamics. The com-mitment to resources resulting from specific invest-ment should be clearly differentiated from commit-ments evolving when practicing capabilities. Thisdifferentiation accordingly implies a separation ofresource-based inertia and capability-based rigidity(Gilbert, 2005). Since this paper is concerned withthe dynamization of capabilities, we only addressthe issue of capability rigidities without denying inany way the relevance of resource inertia.



For understanding commitment dynamics incapability-based conduct, social psychological pro-cesses form the main bulk of explanations. Socialpsychological research stresses the tendency toact in favor of the consented current thoughtand to avoid confrontation with deviating neg-ative feedback and signals (Miller and Nelson,2002; Tripsas and Gavetti, 2000). One of the best-known effects in this context is groupthink (Esser,1998; Janis, 1982), which is likely to commit thegroup to their perspectives once developed. Thecommitment driver is the cohesion of the groupand the willingness to protect the group againstdisturbances and disharmonious themes. Anotherwell-known effect fostering an ultra-stabilizationof capabilities is ‘escalating commitment’ (Staw,1976). The argument draws on psychological sunkcosts or the phenomenon that people sometimes‘throw good money after bad’. Although there is agreat deal of theoretical controversy concerning thebasic causes of escalating commitment, it is pre-dominantly explained as being the outcome of self-justification processes (Festinger, 1957; Staw andRoss, 1978). The tendency to become entrappedwithin a failing course is explained by the decision-maker’s unwillingness to admit that their priorinvestment (resource allocation) was in vain. Thestrong urge to ‘save face’ in their own and other’seyes (Brockner and Rubin, 1981; Brockner, 1992)leads decision-makers to support further insuffi-cient investment thereby (re)affirming the correct-ness/usefulness of the earlier decision. As a resultthey start an escalation of commitment whichexcludes more and more reversibility of the oncechosen direction—in our context, a once devel-oped capability.

Similar tendencies stem from other cognitiveeffects such as self-reinforcing ‘selective percep-tion’ (Walsh, 1988) or ‘mind maps’ (Weick andRoberts, 1993), to give just some examples. Thepoint of departure here is the bounded capacityof actors in processing information and the neces-sary building of selection patterns. Due to reinforc-ing tendencies these patterns are likely to becometrapped. Such processes are the more prominent,the more uncertain and ambiguous the situation isperceived to be. Heiner (1983) stresses the effectsof complexity perception: the more difficult it is foran actor to decipher the environmental demands,the more likely the actor will impose familiar pat-terns of response to match the challenge (also

North, 1990: 23). The same is true for perceivedthreats.

Persistence of capabilities in the face of chang-ing environmental demands is also caused bysocialization mechanisms. Managers becomesocialized into the belief system in which thesecapabilities are embedded. Socialization into beliefsystems that take for granted the current capabilitypattern and its internal links is likely to mobilizecognitive and emotional resistance against criticalsignals urging a shift in the familiar patterns of act-ing. They do not reflect on these deep beliefs, theysimply practice them, thereby becoming reluctantto acknowledge the need for changing a once bril-liant problem-solving architecture and its under-lying coordination pattern (Westphal and Bednar,2005).

What is still more intriguing is the fact that evenwhen they are aware of the need to change andwilling to change capabilities, the hidden imprintsof the capability pattern may lead them to lookfor alternatives only in the neighborhood of thecurrent practices (Johnson and Johnson, 2002).Thus, managers reinforce current capabilities (viaproject budgeting and investment policy), therebyunintentionally suppressing new unconventionalproject initiatives (Burgelman, 2002b; Leonard-Barton, 1992).

In order to avoid any misunderstanding, a clar-ifying remark is appropriate here: while broadlyacknowledging the power of inertia drivers, itwould be misleading to conceive of organizationalcapabilities as totally immobile entities. As is truefor all social artifacts, capabilities are subject tosome alteration processes over time. By draw-ing on the life cycle concept, Helfat and Peteraf(2003), for instance, point to the gradual evolutionof capabilities, which automatically take place inthe course of aging (for a similar organic under-standing of capability development see Feldman,2004). This procedural idea of an ongoing processof slight capability development is limited to anorganic development and therefore does not dis-solve the threatening diagnosis of the flipside ofcapabilities and self-reinforcing rigidity dynamics.

Paradox implications for capabilitymanagement

The inherent tendency of capabilities to persistamounts to a strategic threat which cannot beneglected. The management faces a paradoxical



situation: on the one hand, the building of complexand reliable problem-solving architecture consti-tutes strength and allows for developing sustain-able competitive advantages. On the other hand,this advantageous side of capabilities is, however,attained by (unconsciously) suppressing alterna-tives, pluralistic ignorance and reduced flexibility.Any capability therefore contains an inherent risk,i.e., the risk of rigidity and helplessness in the faceof fundamentally changing conditions.

As a consequence, organizations are confrontedwith a dilemma: on the one side, they have todevelop reliable patterns of selecting and linkingresources in order to attain superior performanceand competitive advantages and on the other sidethis endeavor constitutes—at least in volatile mar-kets—a considerable risk of becoming locked intoexactly these capabilities. How can this paradoxbe resolved?

The problem of locked organizational capabil-ities has been addressed in many ways in recentstrategy literature. However, the very first step nec-essary is obviously to identify and to confront theJanus face of capability, its strengths and weak-nesses. Whatever the approach in detail, it hasto address the causes and drivers underlying thecapability paradox: our analysis suggests that theseparadox drivers are primarily path dependency,structural inertia, and psychological commitment(cognitive traps).

The most salient suggestion to overcome thecapability paradox within the capability debate isto develop ‘dynamic capabilities’ (Eisenhardt andMartin, 2000; Kusunoki et al., 1998; Teece et al.,1997; Zollo and Winter, 2002; Winter, 2003). Inthe following section we examine to what extentthe fascinating idea of developing dynamic capa-bilities can actually provide a solution to resolvethe capability–rigidity paradox.

APPROACHES TO DYNAMICCAPABILITIES

At first sight the approaches on dynamic capabili-ties build a fairly homogenous class. A closer look,however, reveals remarkable differences amongthem calling for a differentiated discussion. Fromour point of view, three different theories ofdynamic capabilities can be identified. We havelabeled them (1) the radical dynamizationapproach, (2) the integrative approach, and (3) the

innovation routine approach. The first treatsdynamic capabilities as a functional equivalentto classical capabilities in dynamic environments.The second fosters the idea of amending capabil-ities by adding a dynamic dimension and the lastassigns the task of dynamization to a special typeof routine called innovation routine.

Radical dynamization approach

The core idea of total dynamization is to trans-form the conception of capabilities into full-blownadaptability—at least in high-velocity markets.Based on a differentiation between differentdegrees and patterns of dynamic capabilities, acontingency approach of dynamization dependingon the degree of market dynamic is advocated(Eisenhardt and Martin, 2000). A clear distinctionis drawn between moderately dynamic and high-velocity markets. Accordingly, two broad classesof dynamic capabilities are introduced. ‘Moder-ate dynamic markets’ require dynamic capabilities,which come close to the classical conception ofcapabilities, i.e., the pattern-driven conception ofproblem-solving with some incremental changes.The real challenge, however, is seen in the sec-ond case, namely mastering high-velocity environ-ments with rapidly and discontinuously changingmarket conditions and rules (Bourgeois and Eisen-hardt, 1988). Radical dynamic capabilities are con-ceived to master this volatility. The linking andselection process has to continuously create newcombinations of resources: ‘They are in a continu-ously unstable state’ (Eisenhardt and Martin, 2000:1113). Dynamic capabilities in this sense build dif-ferent types of capabilities, which amount to expe-riential, improvisational, and highly fragile pro-cesses of reconfiguration, integration, and acqui-sition of resources. They make use of real-timeinformation, simultaneously explore multiple alter-natives, rely on quickly created new knowledge,are governed by very few simple rules, do not getstored in the organizational memory, and thus donot produce predictable outcomes. Their strengthno longer flows from architecture but rather fromits ability to continuously produce new constel-lations and solutions. The new basis for buildingcompetitive advantages is seen in the encompass-ing capability to change very quickly and to masterunforeseeable environmental demands (Eisenhardt,2002).



Quite obviously, this conception of dynamiccapabilities comes very close to the functioning ofwhat is known as adhocracy (Mintzberg, 1979) orthe ‘total learning organization’ (Pedler, Burgoyne,and Boydell, 1991; Vaill, 1996). The distinguishingcharacteristic of the learning organization is thatall activities permanently operate in the learningmode, i.e., they are not bound to history/experienceor any rules. The learning organization is alwaysready to revise hitherto cognitions and changeexpectations; they are in flux or, as Weick (1977)puts it, they are ‘chronically unfrozen’.

Discussion

This solution to the capability paradox in high-velocity markets is a radical one. The subject to bedynamized is in danger of getting lost as actuallythe generation and reproduction of capabilities areno longer needed or even become dysfunctional.The only organizational capability left in high-velocity markets is the ability to learn quicklyand to improvise effectively. Problems are solvedwithout relying on previously built expertise andcompetitive advantages can only be gained fromrapid learning and flexible pacing (Eisenhardt andMartin, 2000: 1116).

This conception of dynamic capabilities is with-out doubt an appealing one, which nicely matchesthe prevalent feeling that all parameters of ourlife are continuously changing. From a theoreti-cal point of view, however, the suggested type oforganization and its functioning raises some fun-damental questions.

The major concern addresses the logic of suchflexible organizations. A total learning system, assuggested in this approach, is supposed to reactto any signal from a volatile environment in anew way. Organizations could no longer observeand handle environmental developments on thebasis of proven selection patterns and operatingrules. The handling would have to be created caseby case from scratch without any guidance fromthe past and experiences of successful practices.Actually this approach advocates spontaneous act-ing throughout the system. Any capability struc-ture for guiding the development of these activ-ities would hinder the advanced full flexibility.However, it is hard to see how organizations canbuild resource heterogeneity and sustainable supe-rior performance on this basis. The working ofsuch improvised solutions cannot be anticipated

because they are supposed to be new each time andthus there are no experiences that allow for prop-erly assessing the effects of the new solutions. Thesuccess of mere spontaneous reactions is likely todepend on mere luck and/or intuition only. Obvi-ously, this mode of adaptive acting does not meetin any way the basic dimensions of a capabilityas outlined at the beginning of this article. Simi-larly, Winter (2003) points out that such streamsof newly created activities and spontaneous adap-tations cannot be understood as exercising capa-bilities; rather they represent a completely differ-ent mode of acting and practicing, namely ad hocproblem-solving.

Winter (2003) holds, however, that the mode ofad hoc problem-solving can be considered as afunctional equivalent to building (dynamic) capa-bilities. In his view organizations, in volatile cir-cumstances, are well advised to calculate whetherad hoc problem-solving is—compared to capabil-ity building—the preferable option since it doesnot require longer-term investment in resources. Itis, however, hard to see that this is really a viableoption. How should an organization exist withoutany investments in tangible and intangible assetsand any specific patterns of doing business? It iseven harder to see how this mode of acting couldamount to a sustainable advantage. Rather it raisesthe more fundamental question as to why thistype of coordinating activity should be conductedwithin the realm of an organization. Actually, thelogic of mere ad hoc problem-solving comes soclose to the (unpatterned) mode of market coor-dination that the boundaries blur. Ultimately, themode of total flexibility (ad hoc problem-solving)eliminates the very reason for creating organi-zations/hierarchies instead of market coordination(Williamson, 1975). There is no rationale why adhoc problem-solving in organizations should out-perform market coordination.

Modern systems theory (Luhmann, 1995) andcognitive schools of organizational thought (e.g.,Daft and Weick, 1984; March and Simon, 1958)can provide further insights to substantiate thisargument. They all see pattern building and struc-turing as both preconditional and being the actualmotive for creating and maintaining organizations.Their argument starts with environmental com-plexity and ambiguity and the requirement to pro-vide orientation by developing workable schemesfor enabling action. Volatile environments do notappear on the ‘organizational screen’ in terms of



clear-cut problems; rather actors must actively con-strue models for understanding and decipheringthe complex world in order to survive. Collec-tive actors/organizations are considered to providesuperior schemes, which is ultimately the reasonfor their existence. Organizations have to safeguardand cultivate their knowledgeable and actionableschemes to guarantee effectiveness. At the sametime the safeguarded schemes form the organiza-tional boundary by drawing a distinction betweeninside and outside. If organizations refrained fromdoing so, they would simply merge with the envi-ronment after having solved a specific problem.The distinction, and thus the boundary, would van-ish as there is no such thing as a boundaryless orga-nization. A world of fully dynamized ‘capabilities’would come close to a world without organiza-tions.

A similar argument can be drawn from educa-tional psychology (Carroll, 1993; Piaget, 1970).Research has shown how learning is structurallybound to the existence of cognitive patterns ormental maps. They are simply the preconditionfor perceiving and thinking and subsequently forlearning. The lesson is that there is no uncon-ditioned observation and perception. In the sameway, to be able to act and to learn, organizationsneed their own sense-making patterns to reduceenvironmental complexity to an appropriate level.

As shown in the first section the notion of orga-nizational capability essentially builds on patternsand maps. Furthermore, it is the very function ofa capability to enable an organization to skillfullyget along with these complex challenges from avolatile environment and possibly provide a plat-form to master these challenges in a better waythan competitors are able to. The most surprisingconclusion from this discussion therefore is that afull-blown dynamization of capabilities means inthe final analysis not only eliminating the operat-ing basis of an organization but also to drop theidea of capability building. If there are no pat-terns, no organizational memories, and no assetsthen there is no basis to grow for any capability atall—irrespective of whether we address classicalor dynamic capabilities (Helfat and Peteraf, 2003).

Eisenhardt and Martin (2000) are quite aware ofthis radical consequence. Relying on complexitytheory they seek to find a way out of this dilemmaby stressing the necessity for some minimal struc-tures and a few simple rules. A critical number of

routines and rules are assumed to prevent organi-zations sliding into chaos or dissolution. But thissuggestion raises subsequent questions. By pursu-ing the claim for (minimal) structures and (simple)routines systematically, it becomes very hard todraw the line as to where the dynamic conceptionends and the classical conception of capabilitiesbegins. To make this suggestion operational, thesestructures and rules would have to be properlydefined. At a minimum they must work in a reli-able and repeatable manner. In other words, takingthe suggestion of minimal structures and routinesseriously brings us back to the familiar evolve-ment of pattern-driven problem-solving. The dif-ference between the classical capability conceptionand the radical dynamic capabilities (as well as adhoc problem-solving) would become a matter of(uncertain) degree only, and no longer a radicaldeparture.

Integrative approach

The most prominent approach towards a theory ofdynamic capabilities has been provided by Teeceet al. (1997). It is the most salient one among var-ious attempts of in-built flexibility. The authors donot start with the paradox introduced above; rather,they refer to empirical evidence finding com-petitive advantages to be primarily accomplishedthrough responsiveness and flexible product inno-vation. Successful organizations proved flexible interms of ‘dynamic capabilities’. Dynamic capa-bilities are conceived to be the mechanisms ofadapting, integrating, and reconfiguring integratedclusters of resources and capabilities to matchthe requirements of a changing environment: ‘Theterm “dynamic” refers to the capacity to renewcompetencies’ (Teece et al., 1997: 515).

More precisely, dynamic capabilities are con-ceptualized by three dimensions: (1) positions,(2) paths, and (3) processes:

1. ‘Positions’ refers to both internal and exter-nal positions. The internal position relates tothe specific set of resources available in afirm (financial, technological, reputational, andstructural). The external side refers to the spe-cific market position/assets of the focal firm.The current position of a firm determines to acertain extent the future decisions a firm canreach and realize.



2. ‘Paths’ represents the history of an organiza-tion; i.e., the current position of a firm is basi-cally shaped by the patterns evolved from thepast. And also, where a firm can go in the futuredepends on its current paths and their shapingforce.

3. The dimension ‘processes’ is at the heart ofthis capability conception and is twofold. Onthe one hand, processes are devoted to coordi-nating and integrating available resources. Thisis understood as being the static component.On the other hand, processes refer to orga-nizational learning and the reconfiguration ofresources. The latter two sub-dimensions rep-resent the dynamic component, which is sup-posed to guarantee permanent adaptation andchange of the organization. The dynamic sub-dimension ‘learning’ covers both processes ofincremental improvements (amendments of thecurrent positions) and processes of identifyingnew opportunities. The second dynamic sub-dimension ‘reconfiguration’ addresses the trans-formation of a firm’s asset structure accom-plished through alert surveillance of the envi-ronment for discontinuities and subsequent rad-ical changes.

In successful firms, the interaction of these staticand dynamic components is assumed to convergeto a full-blown ‘dynamic capability’. In order toavoid any misunderstanding, it seems importantto stress that here the term ‘dynamic capability’explicitly comprises both dynamic and static ele-ments, which is the very reason we call it the‘integrative approach.’

Discussion

Without doubt, the idea of integrating dynamicmechanisms into the concept of capabilities offersa very plausible reaction to the problem of thecapability paradox. In our view, however, it is atleast questionable as to what extent this idea ofintegrated dynamics can actually provide a wayout of the paradox and its underlying conceptualdilemma. The core idea of the suggested method ofdynamizing capabilities is to expand the scope ofthe construct by including learning and transfor-mation processes. The new dynamic features areadded to the two more static dimensions of theproblem-solving architecture in order to inject themissing flexibility. The guiding logic seems to be

the following: the classical patterns of organiza-tional capabilities—complex problem-solving andappreciated practicing which is reliable—buildthe core. The additional dynamic dimension isdesigned to overcome the inherent risk of becom-ing rigid and trapped.

The authors suggest ‘reprogramming’ organiza-tional capability as a two-dimensional notion con-sisting of both a stable and a dynamic dimension.In other words, the integrative approach tries tobe two things at once: it focuses on exploiting thepositive effects of patterned capabilities and simul-taneously overcoming its inherent problematic sideby adding a dynamic dimension.

In our view, this approach tends to underratethe inertial dynamics of organizational capabili-ties. It builds on two contradictory notions of logicat the same time: reliable replication and contin-uous change—two dimensions that hardly mix.Dynamizing in the defined sense attempts to trans-form reliable and routinized action patterns intoflowing operations. Stable patterns are thought toprovide their reliable service while at the sametime continuously changing. Reliable and replica-ble patterns cannot, however, evolve without sta-bilization; patterns cannot be conceived in termsof continuous change. Making them subject to thecontinuous adaptation inevitably means dissolvingtheir replicable essence. The very core of a capa-bility—as defined above, as well as by Teece et al.(1997: 519): ‘distinct ways of coordinating andcombining’—would simply get lost and becomesubstituted by a chain of subsequent singular andpatternless acts.

Obviously, the idea of adding the learning func-tion to the capabilities concept, which at firstglance seemed very plausible, runs into a serioustheoretical problem. In terms of a consistent the-ory it is not possible to simply add the missingdynamic feature as the two dimensions are con-tradictory in nature. To stress the one dimensionnecessarily means negating the other—at least toa significant degree.

It should be noted, however, that Teece, Pisano,and Shuen are sensitive to this conceptual implo-sion. They try to attenuate the dissolving effectof adding a dynamic dimension by qualifying thepenetrative power of the learning function. Theystress that learning has a tendency to be localand close to previous activities (Teece et al., 1997:522). Following this line of thought, the scope ofinnovative action would be constrained by the past



and by historical paths. Furthermore, the authorspoint to economic reasons for limiting the numberof changes: ‘Change is costly’ (Teece et al., 1997:521). These arguments, however, easily lead tofostering small-scale, incremental changes. Path-driven learning and small-scale transformations areonly likely to modify the established method ofpatterned problem-solving and to focus on fine-tuning, similar to the evolutionary stage in theframework of Tushman, Newman, and Romanelli(1986). As a consequence, the basic assumptionsand values underlying the predominant capabilitypattern are not called into question and a funda-mental change or renewal of the existing capabil-ity cannot occur. But precisely this fundamentaltransformation is regularly claimed to be the mostimportant function of dynamic capabilities.

In conclusion, the idea of building an integratedconception of dynamic capabilities by simultane-ously relying on both patterned replication and thelearning function is likely to conceptually implode.The attempt to merge continuous learning into thecapability conception inevitably leads us back tothe paradox; the suggested integration logic obvi-ously overstretches the capacity of the concep-tion of capability. Thus, the question arises as towhether the term ‘dynamic capability’ does notinevitably lead to a contradiction in terms.

Innovation routines

The third approach to dynamizing capabilitiesbasically suggests supplying the missing dynamicdimension by installing separate innovation rou-tines that allow a firm to overcome the rigiditytrap (paradox) of organizational capabilities (Nel-son and Winter, 1982; Zollo and Winter, 2002).Innovations are defined as the creation of any sortof novelty which, focusing on capabilities, meansthe creation of novel problem-solving patterns.Innovation routines are characterized as organi-zational procedures directed towards change: ‘Adynamic capability is a learned and stable patternof collective activity through which the organi-zation systematically generates and modifies itsoperating routines in pursuit of improved effective-ness’ (Zollo and Winter, 2002: 340). In the case ofvolatile environments—then and only then—theorganization is well advised to build (or accept theevolution of) higher-order search routines in orderto bring about regular modifications of establishedoperating (lower-order) routines. Such innovation

routines, somewhat similar to the Japanese systemof continuous improvement, are expected to pro-duce revisions, even radical changes, in a system-atic and predictable fashion (Nelson and Winter,1982: 17). This approach also includes the expla-nation of the genesis of routines. Both operatingand innovation routines (‘dynamic capabilities’)arise from learning and are conceived as con-densed results of former trial-and-error behavior interms of collective learning acts. Zollo and Win-ter therefore suggest establishing separate ‘learningmechanisms’ to develop a firm’s routines. Theselearning mechanisms are specified as stages ina recursive cycle of ‘experience accumulation,’‘knowledge articulation,’ and ‘knowledge codifica-tion’—a specification which comes close to Non-aka’s knowledge spiral (Nonaka, 1994).

This interesting contribution has considerablyadvanced the discussion on the nature and meritsof ‘dynamic capabilities.’ The major distinguish-ing feature is its emphasis on the notion of inno-vation routine. A discussion of this contributiontherefore has to focus on this core element. Themajor question is whether or not separated innova-tion routines actually provide a promising avenuefor overcoming the capability trap and conceivingorganizational dynamics.

Discussion

Routines are doubtless at the heart of the classicalcapability conception although, as shown above, acapability is supposed to consist of much morethan interlinked routines. Replicable clusters ofproblem-solving activities constitute an organiza-tional capability and inevitably concurrently pro-duce the inherent risk of entrapment. The idea hereis to overcome the risk of getting trapped in chang-ing environments by installing another routine orset of routines directed to modify (operating) rou-tines. Whatever the specific subject and the levelof routines, the very logic of their functioningis always the same: they replicate and stabilizerecursively a pattern of activities. Construing thedynamization of capabilities within this conceptualframework means to duplicate the same structureor architecture on a different (higher) level. Asthe capability paradox has been shown to resultfrom inherent structural dynamics (simplification,selectivity, etc.), such duplication is prone to evokeon a higher level the same set of problems which



dynamic capabilities set out to overcome. Inno-vation routines (and the ‘learning mechanisms’too) are conceived as ‘stable patterns’ of collec-tive activity, which means they inevitably shapeattention, learning, and action within a predefinedframework, thereby replicating a programmed typeof thought and problem framing. Their under-lying logic simply does not allow transcendingthe inscribed program—otherwise they would nolonger be routines. In sum, it is hard to attain inno-vation through routines.

Innovation routines are thus likely to bringabout the same form of inside world, whichhas been identified as an element of the capa-bility trap. The empirical findings by Leonard-Barton (1992) confirm this conclusion. Her studyfocused on the evolvement and working of capa-bilities in R&D—product development in particu-lar—showing that it was precisely innovation rou-tines that brought about the risk of a dysfunctionalcapability flip and the fatal restrictions in explor-ing the scope of alternatives. The trouble comesfrom the logic of functioning: the effectiveness of(innovation) routines is bound to the recurrenceof a specific situation or problem structure. Thetasks and challenges (innovation requirements) arestudied again and again within the preprogrammedframing. This is promising as long as new signalsdo not ask for radical frame-breaking changes; i.e.,it works for familiar triggers but not for surprisesand discontinuities. In high-velocity situations notonly is the pace high, but rather the type of envi-ronmental signals is likely to be new and unpre-dictable in nature as well, with routines by theirvery character being not suited to meeting theseextraordinary requirements.

Zollo and Winter are quite aware of this prob-lem and try to escape this rigidity trap in unpre-dictable environments by requiring dynamic capa-bilities and learning mechanisms ‘to be updatedrepeatedly’ (Zollo and Winter, 2002: 341). Such‘updates,’ however, can definitively not be accom-plished again by routines, but only by frame-breaking approaches which are beyond the rou-tines’ program. In dynamic and complex environ-ments the suggested ‘learning mechanisms’ can-not fulfill this function to the necessary extentbecause they are again based on routines. Thespecified set of learning routines for accumulat-ing experience, articulating hidden knowledge, andcodifying knowledge has the same grammar ofprogramming. This obviously brings us into aloop with a frame-breaking routine, which amountsto a contradiction in terms. Whatever level weapproach the essential logic always remains thesame, namely that routines are likely to concur-rently turn valuable capabilities into rigidities.

Summary

In summation, the discussion of dynamic capabil-ities reveals the following picture: all approacheshighlight a dynamic feature of organizational capa-bilities and the need for changing capabilitiesin order to address dynamic markets. The wayin which the dynamization is conceptualized dif-fers primarily in terms of the favored learningdirection, the basic mechanism of dynamization,and the specific emphasis on organizational rou-tines. Table 1 summarizes the logic of the threeapproaches from a comparative view.

Table 1. Comparison of approaches on dynamic capabilities

Radicalized dynamizationapproach

Integrated dynamizationapproach

Routinized dynamizationapproach

Dominantlearningdirection

Experiential (ahistorical)learning

Historical learning within theframe of capabilities

Historical learning within theframe of innovation routines

Mechanisms ofdynamization

Regimes of ad-hocproblem-solving processes

Processes internal to capability Processes external to capability

Importance ofroutines

Low : Avoid routines as far aspossible

Medium: Build routines andpaths

High: Build multiple level ofroutines

Capabilities inthe context ofRBV

Paradigm change: Capabilityas strategic resource obsolete

Paradigm modification:Capability as dynamic strategicresource

Paradigm variation: Capabilityas strategic resourceMeta-capabilities as mediumfor dynamization



The discussion of the approaches revealed someirritating contradictions, raising doubts whether thenotion of dynamic capabilities actually providesthe most promising avenue for overcoming thecapability paradox.

Three conclusions are salient. A first result isthat capabilities cannot be thought of as beingfully flexible; this would mean overstretching thescope of the conception. Whatever the perspec-tive, a capability refers to patterned and replicableactivities oriented toward specific tasks (see alsoWinter, 2003: 992).

A second result is that overcoming the capabilityparadox calls for frame-breaking changes and notonly incremental steps. Due to the fact that orga-nizational capabilities are likely to stick to theirunderlying logic, a real dynamization in the faceof changed market conditions has to bring abouthigher-level changes.

A third result is that alternatively conceptual-izing dynamic capabilities in terms of innovationroutines does not yield a fully convincing solutionto the paradox either. This approach bounds thedynamization to the execution of routines and istherefore likely to restrict the scope of change tothe logic of familiar programs.

How to overcome the blind alleyAs, on the one hand, the solution of dynamiccapabilities seems to lead to contradictions and,on the other hand, dynamization of capability-based management is imperative, it might be advis-able to look for structurally different solutionsbeyond overstretching the conception of capabil-ity and underrating the necessity of frame-breakingchanges. In the next section we aim at developing asolution with a different set of logic. It conceives oforganizational capabilities and dynamization of theproblem-solving architecture as two separate func-tions that a successful organization has to addresssimultaneously. At the core is a dual model ofcapability management which is based upon coun-tervailing processes. Our conception draws on thethree-step model of strategic control developed bySchreyogg and Steinmann (1987).

ORGANIZATIONAL CAPABILITIESAND DYNAMIZATION: A DUALPROCESS MODEL

In a nutshell, the basic idea we are suggestingis counterbalancing patterned selection (capability)

and risk compensation. The guiding idea is toexploit on the one hand the power of patternedproblem-solving and on the other hand to compen-sate for its inherent risk of dysfunctional flip byinstalling alert environmental surveillance designedto give early indication of any unexpected changenecessities. Instead of dynamizing the capabil-ity conception, capability evolvement and systemdynamization are conceived as two separate coun-tervailing processes, which are performed simulta-neously.

The logic of dual processes

The point of departure is the conclusion that,as opposed to the idea of a dynamic capabil-ity, the strengths of patterned problem-solving anddynamization cannot be merged into one concep-tion. Our proposed approach therefore favors apolicy of keeping the two strategic functions sep-arate and treating them as countervailing forces.Such endeavor requires first of all shifting theperspective from the capability level to the sys-tems level. This is because we need an encom-passing perspective, which allows for conceivingtwo separate processes that complement each otherin order to secure the system’s survival and itsenduring success. From a systems theory pointof view (Luhmann, 1995), capability building anddynamization cannot only be treated as two sep-arate system functions but can also be conceivedof as countervailing processes designed to man-age the contradictory requirements of exploitationand exploration (fundamental change). Since oneand the same process cannot comprise concur-rently stabilizing and destabilizing forces, the pro-cesses of organizational capability evolvement anddevelopment have to be temporarily and locallydeskewed (Antonacopoulou and Tsoukas, 2002).By conceptually deskewing practicing from reflect-ing on practicing, the practice of an organizationalcapability can be maintained within its logic offunctioning since a reflection process separated intime and space deals with the adaptation of the pat-terned practice as part of a permanent interactionbetween organization and environment.

Risk compensation through monitoring

Organizational capabilities provide a problem-solving architecture, which enables theorganization to make sense out of an ambiguous



unpredictable environment and to master complextasks in an effective and replicable way. To makeuse of these advantages implies that only a lim-ited set of procedures can be employed repeat-edly, while at the same time potentially availablealternatives are (consciously and unconsciously)excluded (Winter, 2003). Thus capabilities areselective by their very nature.

This inherent selectivity of capabilities (includ-ing dynamic capabilities in terms of innovationroutines!) brings forth a structural risk, namelythe application of traditional patterns to new tasks.Furthermore, the closing dynamics of recursivereplication of once successful problem-solvingarchitectures are likely to render the system igno-rant of fundamental change requirements and waysof mastering the new challenges. Discontinuities,misapplications and ignored events are likely tothreaten the system’s survival. Thus this inherentthreat has to be skillfully observed and kept undercontrol. In other words, a system cannot focus onlyon developing and exploiting capabilities—it mustsimultaneously find ways to handle the inherentrisk of relying on selective and recursive prac-tices. Risk compensation therefore amounts to aseparate high-ranking system function, which takescare of change requirements and adaptation needsevolving from unpredictable environments. To putit differently, dynamization is accomplished by risk

compensation. In conclusion, at the heart of suc-cessful capability management stands the balancedduality of patterned selection and dynamizationthrough compensation.

Dynamization through compensation meansfirstly monitoring the system’s capabilities, itsevolvement, its usage, its effects inside and out-side the organization, as well as critical issuesand discontinuities in the system’s environment.By continuously observing (scanning) the capa-bility landscape, its practices, recursion, blinders,potential failures and maladjustments can be iden-tified—preferably at a very early stage. By becom-ing aware of these critical signals the issue ofpotential change requirements is put on the agendaof organizational decision making. This systempresses itself to make a decision as to whetheror not the approved problem-solving architectureshould be abandoned.

Obviously the suggested monitoring changes theinternal status of capabilities. Instead of automaticreplication they become an explicit issue in away which is alienated from the taken-for-grantedworld. From an organizational point of view,installing monitoring requires self-observation asthe organization has to observe in a separate pro-cess its own practices and the effects of its capa-bilities in light of a discontinuous environment.

As depicted in Figure 1 the monitoring pro-cess looks at the practice of capabilities from a

t0 t1 t2 t3 t4 tn…

Operational level

Observational level

Capability Practices

internal environment

external environment

…

Capability Monitoring

lock-ins? inertia? cognitive traps?

Figure 1. A dual-process model of capability dynamization



non-practicing point of view; i.e., it is primar-ily designed as a modus of reflection (and notdirect acting). The reflecting activity we have inmind is a ‘second-order observation’ in terms ofsecond-order cybernetics (von Foerster, 1982), i.e.,an observation of first-order observations locatedat the operational level.

Following Maturana’s (1970) insight: ‘anythingsaid is said by an observer,’ we can understand anypractice as an observation—more precisely as afirst-order observation. The monitoring suggestedhere is conceived of as a second-order observa-tion, i.e., an observation of first-order observers(practitioners). In organizational capabilities, dis-tinctions are drawn and replicated without beingaware of them. These distinctions thus amount tothe blind spot of any first-order observer. Thisblind spot can be seen by second-order obser-vation processes only. By asking how the dis-tinctions are drawn, second-order observation putsthe framing of first-order practice into perspective(Antonacopoulou and Tsoukas, 2002; Luhmann,1993; Giddens, 1984). In order to compensate therisk of a dysfunctional flip or to reside in a capabil-ity trap, the modus of a second-order observationis a necessary precondition. This observation pro-cess can potentially provide the organization with‘enriched’ irritations. By observing the processesof linking resources (organizational capabilities)and the underlying distinctions, the focal organi-zation gains insights, which—due to the differentframe of reference—differ from the observationson the operational level.

Second-order observation is not part of thisfirst-order acting-practice. For a moment, throughsecond-order observation, the organization takestime out in order to reflect on the reasons for aspecific way of doing something.

Quite obviously, second-order observationcomes close to ‘double-loop learning’ or, as it iscalled by March, ‘explorative learning’ (Argyris,1976; March, 1991), focusing both on the basicassumptions and the blind spots of organizationalactivity. Double-loop learning calls into ques-tion the ‘theories-in-use’ which guide performancethrough selection and linking processes (capabil-ity) where the ‘certain way of doing things’ ismade the subject of extraordinary reflection andthereby opened up to changes responding to dis-continuous events (Argyris and Schon, 1978).

To put it differently, through this type ofreflection capabilities lose their taken-for-grated

status and become reframed as potentially revis-able (Putnam and Majia, 1992; Tversky and Kah-neman, 1981). Monitoring brands replication prac-tices with an index of uncertainty. As a conse-quence, the validity and temporality of capabilitysteadily becomes a critical issue during the strate-gic discourse, thereby keeping the system alert torequired change.

Non-routine dynamization

From a managerial point of view the monitoringof organizational capabilities is a separate functionto be executed using its own logic. By continu-ously checking whether and why the capability-driven activities still apply or registered signals arelikely to threaten the validity of the capability inthe future, the firm gains flexibility and adaptabil-ity.

At first glance, it might appear that the advo-cated second process of monitoring is similar tothe idea of establishing innovation/search routines.If the monitoring process is to succeed as partof the dual model outlined above, it has to bedesigned extraordinarily and the basic requirementis to keep the process open. Opposed to the ideaof innovation routines, we deem that monitoringshould aim at refraining from any routinization.As has been shown, routinization means program-ming, and programming means selectivity and nar-rowing the scope. Routinization thus hampers theprocess and counteracts the logic of compensat-ing for the risk of stable competence patterns. Asargued above, the logic of capability monitoringhas to set a counterpoint to the logic of routiniza-tion by opening the scope, which has been nar-rowed by the problem-solving architecture. Onlythen is there a good chance of detecting extraor-dinary, unforeseen signals which may call intoquestion the ongoing validity of the current com-petence patterns. As nobody can know in advancewhich kind of signals or events occur and wherethey show up, any prestructuring in terms of gen-eral rules is likely to blind the observers. Dueto the incidental and potentially unfamiliar char-acter of the threatening situation, the deliberateinstallation of routines is likely to mislead the orga-nization in new and unfamiliar situations (as, forinstance, Weick, 1993, demonstrated in his well-known analysis of the Man Gulch disaster). Rou-tines reflect familiar problem situations and theirsolutions and not the handling of unknown events.



To guarantee a firm’s responsiveness and flexibil-ity the scanning process therefore has to be influx.

It should, however, be pointed out that regis-tered discrepancies or rigidities are not supposedto automatically lead to actual change activities.Rather, the organization always has the optionto change the method of selecting and linkingresources, or to stay with the established pat-terns. It should be pointed out, however, that thisdichotomy represents only an ideal and simpli-fied type of organizational behavior. In reality avariety of change options exist, differing in extentand intensity. They reside in the area between thetwo extremes: total change at the one end andstaying with established problem-solving architec-ture at the other end. The dual-process model thusreplaces the unrealistic idea of permanent transfor-mation with the idea of a combination of learningto change and learning to stay (Schreyogg andNoss, 2000). In many cases, there are good rea-sons to stay with the established capability pat-terns, notably that the threatening signals maynot be strong enough, potential negative effectstoo vague, change costs too high, or the firmperceives good chances of changing the threat-ening context in such a way that the establishedcapability remains valid (e.g., a change in thecompetitive rules or the acquisition of competi-tors).

Quite obviously, the risk of capability building,as such, cannot be eliminated by the suggestedmonitoring, as patterning always bears a risk. But itcan provide a measure to contain it at an acceptablelevel. The risk level varies to a certain degree withthe intensity of monitoring efforts. As monitoringefforts are costly, there is no general rule sayingthat the greater the amount of monitoring, thebetter is the organization’s performance. In somecases it might be advisable for cost reasons toaccept a higher level of risk.

Furthermore, it should be noted that the advo-cated idea of monitoring implies that evolved pat-terns of organizational capability are reversibleand not completely congealed. While being deeplyembedded in organizational practices and rootedin paths, capabilities as linking practices provedto be subject to change interventions—at least inmost cases (Argyris, 1990; Schein, 1985). Other-wise capability monitoring would not make muchsense as the reframing of organizational capabili-ties would be impossible.

The design of capability monitoring

When executing the model of capability monitor-ing, it should first of all be taken into accountthat any monitoring needs a frame of reference inorder to produce information. As capabilities usu-ally evolve over time in the context of complexand partly implicit experiences, organizations oftenlack a well-articulated understanding of their owncapabilities. As shown above, organizational capa-bility is successfully practiced but rarely reflected.As a consequence, as a first step in setting upthe monitoring, the capability in use and theircritical indicators have to be identified in termsof (1) complexity, (2) practice, and (3) reliability,and brought into an observable format. Techniquesthat can help here are, for example, ‘cognitivemapping’, ‘conversational analyses’ or ‘patternrecognition’ (Ambrosini and Bowman, 2002; Hen-derson and Cockburn, 1994; Johnson and Johnson,2002).

The actual observation activity focuses on boththe internal and external environment (Aguilar,1967). The field to be observed is endless andthus not ultimately identifiable; the environment,as opposed to organizations, is boundaryless (seeLuhmann, 1995). The surveillance is supposed toaddress first of all the possible paradox-driversidentified above: (1) path dependency,(2) structural inertia, and (3) commitment. Thisthreefold focus can be used for scanning the group,the unit, the division, and the corporate level, andpossibly the cross impacts among these. Figure 2summarizes the features.

Since the scanning can never be all encompass-ing it becomes inadvertently selective in its ownpart too. The resulting danger of overseeing sig-nals and missing incidents is, however, limited bythe fact that missed critical signals from the inter-nal and external environment do not stand still,but evolve as they finally manifest themselves as acrisis. In other words, missed signals bring abouta focus by themselves when they literally becomefelt in terms of a crisis.

As crises are strong signals at a very late stage,the range of options may be dramatically reduced.The ‘best’ options may no longer be availableat that late stage. Therefore, monitoring shouldaim at detecting crisis signals as early as possi-ble. From research on detecting crisis signals, it isknown that crises are regularly preceded by weaksignals. Organizations should therefore become



Unit

Levels of capability practices

Par

ado

x d

rive

rs

Unit

Path dependency

Structural inertia

Commitment

Group CorporationDivisionUnit

Figure 2. The focus of capability monitoring

skilled in recognizing weak signals and at properlyinterpreting them (Ansoff, 1976; Hensgen, DeS-ouza, and Kraft, 2003; Sheaffer, Richardson, andRosenblatt, 1998). The observation should be keptas open as possible and the interpretation of theincoming signals and triggers should be run as adhoc problem-solving. The organization has to nur-ture improvisational skills to deal with surprisingsignals and events (Mendonca et al., 2004).

While establishing search routines generallycounteracts the compensating function of themonitoring process, in interpreting and handlingthreatening signals some rules might be helpfulnevertheless. For the interpretation of signals asystem of checks and balances proved to be useful,as well as the use of dialectical principles (Pearsonand Mitroff, 1993; Preble, 1997). For getting ahandle on pressing signals the installation of alarmsystems proves to be useful in many fields. Thesesystems are designed to ensure a quick responseby the organization. Well-known systems are ‘warrooms,’ ‘distribution lists,’ or a ‘red telephone’(Weick and Sutcliffe, 2001).

The foregoing discussion on organizing andmanaging the task of monitoring also implies thatcapability monitoring cannot be assigned to a sin-gle specialized position or department. It needsrather to be managed as a broadly scattered activ-ity across the entire organization. This requirementnot only echoes the open character of the com-pensating function but also results from the factthat it is unknown where and in which contextthe relevant information enters the firm (Schreyoggand Steinmann, 1987). In principle, any (sub)unitand member of an organization may come into

the situation where they are confronted with amost important signal of the threatened validityof the current capability. The scope of possibili-ties is extremely broad, ranging from internationalconferences, customer claims, negotiations withsuppliers, and accreditation auditors to reading anewspaper or having a conversation with a partnerin a golf club. It would therefore be misleadingto prestructure the activity with operating rules asmanaging the process rather means to encourageand motivate all units, subunits, and members ofthe organization to actively participate in the capa-bility monitoring process.

The suggestions provided so far may evoke theimpression that implementing a capability moni-toring system is primarily a question of technicaldesign. We know, however, from many similarcontrol systems that such activities are likely tobecome subject to various organizational dynam-ics such as political processes, cultural bias, andemotional resistances. There is evidence that orga-nizations may fail to assimilate critical signalson current capabilities and to develop responsesto them because of deep emotional involvement(Gilbert, 2005). Another cause for missing signalsfrom surveillance activities is the power struc-ture and maneuvers to conserve it. Critical signalson the validity of current practices more oftenthan not are experienced as a profound critique ofthose authorities who have developed and advo-cated the method of doing business and gain-ing competitive advantages in question. They notonly feel humbled by critical surveillance activitiesbut also threatened in their position by anticipat-ing the erosion of former competence attributions,



loss of resources, and reputation, etc. Members ofthe power structure therefore often launch indi-rect threats and intimidating messages to keep thestream of critical signals under control. Also, thosewho do not obey and communicate actively criti-cal signals may have to experience punishments inthe form of social distancing (Westphal and Bed-nar, 2005), diminished chances for promotion, orbeing messed around.

All of these tendencies may hamper the will-ingness of organizational members to participatein the badly needed alert surveillance activities.Any effective design of a monitoring system hasto take such tendencies into account and counter-act such adverse effects. There are many measuresavailable. The most important is to institutionallyreflect on those activities and to establish self-critical forums to search for such tendencies, e.g.,by inviting outsiders to challenge familiar waysof transferring and interpreting critical signals,the application of six-eyes principles, encourag-ing devil’s advocacy, and authentic dissent (Muss-weiler, Strack, and Pfeiffer, 2000; Nemeth, Brown,and Rogers, 2001). A great deal can be learnedhere from the way certified auditing agencies dealwith their own ‘rigidities’ experienced in the past(Herrbach, 2001). A very important provision is todevelop a supportive context, which facilitates thedual-model logic. For example, an organizationalculture and structure that accepts no-sayers andmavericks (Nystrom and Starbuck, 1984; O’Reillyand Tushman, 2004) and stresses the importanceof a high customer and competitor orientation(Atuahene-Gima, 2005). These measures are asimportant for effectively managing capability mon-itoring as installing surveillance systems and build-ing up scanning procedures.

CONCLUSION

The suggested dual-process model is designedto manage the paradoxical side of organizationalcapabilities which is likely to turn effectivenessinto rigidity. The model advocates developingtwo countervailing processes, namely the recursivepracticing of distinctive organizational capabilitiesaiming at excellence and efficiency, and simulta-neously the reflexive monitoring of these capabil-ities in order to check their ongoing workabilityin the light of a potentially changing unpredictableenvironment. This observation activity is expected

to initiate a capability change if necessary. Theformer process fosters the regular exploitation ofan outstanding capability and the other destabilizesthis practice by continuously calling into questionits ongoing environmental fit. The dual-processmodel is designed to capture both the bright andthe dark side of capabilities. The organization isexpected to skillfully handle the two countervailingsystem functions primarily by separating them inspace and time. The balancing of two sets of logichas been suggested as a response to attempts aim-ing at integrating dynamics into capabilities, whichin our view eliminates or overstretches the concep-tion of capability. Practiced skillfully, the monitor-ing process provides the missing dynamization ofthe system. Practical tests are needed to explorethe workability of this dual-level conception andto specify in further detail the conditions of itssuccessful functioning.

REFERENCES

Aguilar F. 1967. Scanning the Business Environment .Macmillan: New York.

Ambrosini V, Bowman C. 2002. Mapping successfulorganizational routines. In Mapping Strategic Knowl-edge, Huff AS, Jenkins M (eds). Sage: London;19–45.

Amit R, Schoemaker PJH. 1993. Strategic assets andorganizational rent. Strategic Management Journal14(1): 33–46.

Ansoff HI. 1976. Managing strategic surprise by responseto weak signals. California Management Review20(2): 21–33.

Antonacopoulou E, Tsoukas H. 2002. Time and reflexiv-ity in organization studies. Organization Studies 23(6):857–862.

Argyris C. 1976. Single-loop and double-loop models inresearch on decision making. Administrative ScienceQuarterly 21: 363–375.

Argyris C. 1990. Overcoming Organizational Defenses .Prentice-Hall: Englewood Cliffs, NJ.

Argyris C, Schon DA. 1978. Organizational Learning:A Theory of Action Perspective. Addison Wesley:Reading, MA.

Arthur WB. 1983. Competing technologies and lock-in by historical events: the dynamics of allocationunder increasing returns. Analysis Paper WP-83-92, International Institute for Applied Systems:Laxenburg, Austria.

Arthur WB. 1989. Competing technologies, increasingreturns, and lock-in by historical events. EconomicJournal 99: 116–131.

Ashby WR. 1965. An Introduction to Cybernetics .Methuen: London.



Atuahene-Gima K. 2005. Resolving the capability–rigidity paradox in new product innovation. Journalof Marketing 69: (October): 61–83.

Barney JB. 1991. Firm resources and sustained com-petitive advantage. Journal of Management 17(1):99–120.

Barney JB. 1997. Looking inside for competitive advan-tage. In Core Competency-Based Strategy , Camp-bell A, Sommers LK (eds). International ThomsonBusiness Press: London; 13–29.

Beatty RW, Ulrich DO. 1991. Re-energizing the matureorganization. Organizational Dynamics 20: 16–30.

Benner MJ, Tushman ML. 2003. Exploitation, explo-ration, and process management: the productivitydilemma revisited. Academy of Management Review28(2): 238–256.

Bercovitz JE, de Figueiredo JM, Teece DJ. 1996. Firmcapabilities and managerial decision-making: a theoryof innovation biases. In Innovation: Oversights andForesights , Garud R, Nayyar P, Shapira Z (eds).Cambridge University Press: Cambridge, U.K.;233–259.

Bourgeois LJ, Eisenhardt KM. 1988. Strategic decisionprocesses in high-velocity environments: four cases inthe microcomputer industry. Management Science 34:816–835.

Brockner J. 1992. The escalation of commitment to afailing course of action: toward theoretical progress.Academy of Management Review 17(1): 39–61.

Brockner J, Rubin JZ. 1981. Face-saving and entrapment.Journal of Experimental Social Psychology 17: 68–79.

Burgelman RA. 2002a. Strategy as vector and the inertiaof coevolutionary lock-in. Administrative ScienceQuarterly 47(2): 325–357.

Burgelman RA. 2002b. Strategy is Destiny. HowStrategy-Making Shapes a Company’s Future. FreePress: New York.

Carroll JB. 1993. Human Cognitive Abilities: A survey ofFactor-Analytic Studies . Cambridge University Press:New York.

Cohen WM, Levinthal DA. 1990. Absorptive capacity:a new perspective on learning and innovation.Administrative Sciences Quarterly 35(1): 128–152.

Collis DJ. 1994. Research note: how valuable areorganizational capabilities. Strategic ManagementJournal 15(8): 143–152.

Cook SN, Brown JS. 1999. Bridging epistemologies: thegenerative dance between organizational knowledgeand organizational knowing. Organization Science10(4): 382–400.

Cowan R, Gunby P. 1996. Sprayed to death: path-dependence, lock-in and pest control strategies.Economic Journal 106: 521–542.

Cyert RM, March JG. 1963. A Behavioral Theory of theFirm . Prentice-Hall: Englewood Cliffs, NJ.

Daft RL, Weick KE. 1984. Toward a model oforganizations as interpretation systems. Academy ofManagement Review 9: 284–295.

Danneels E. 2002. The dynamics of product innovationand firm competencies. Strategic Management Journal23(12): 1095–1121.

David P. 1985. Clio and the economics of QUERTY.American Economic Review 75: (May): 332–337.

Dierickx I, Cool K. 1989. Asset stock accumulation andsustainability of competitive advantage. ManagementScience 35: 1504–1510.

Dosi G, Hobday M, Marengo L. 2003. Problem-solvingbehavior, organizational forms, and the complexityof tasks. In The SMS Blackwell Handbook of Orga-nizational Capabilities , CE Helfat (ed). Blackwell:Malden, MA; 167–192.

Dosi G, Nelson RR, Winter SG. 2000. Introduction: thenature and dynamics of organizational capabilities.In The Nature and Dynamics of OrganizationalCapabilities , Dosi G, Nelson RR, Winter SG (eds).Oxford University Press: New York; 1–22.

Duncan RB. 1972. Characteristics of organizational envi-ronments and perceived environmental uncertainty.Administrative Science Quarterly 17: 313–328.

Eisenhardt KM. 2002. Has strategy changed? SloanManagement Review 43: (Winter): 88–91.

Eisenhardt KM, Martin JA. 2000. Dynamic capabilities:what are they? Strategic Management Journal , SpecialIssue 21(10–11): 1105–1121.

Esser JK. 1998. Alive and well after 25 years: a reviewof groupthink research. Organizational Behavior andHuman Decision Processes 73(2–3): 116–141.

Feldman MS. 2004. Resources in emerging structuresand processes of change. Organization Science 15(3):295–309.

Festinger L. 1957. A Theory of Cognitive Dissonance.Peterson: Evanston, IL.

Ghemawat P. 1991. Commitment: The Dynamic ofStrategy . Free Press: New York.

Ghemawat P, Del Sol P. 1998. Commitment versusflexibility. California Management Review 40(4):26–42.

Gherardi S, Nicolini D. 2002. Learning in a constellationof interconnected practices: canon or dissonance?Journal of Management Studies 39(4): 419–436.

Giddens A. 1984. The Constitution of Society: Outline ofthe Theory of Structuration . University of CaliforniaPress: Berkeley, CA.

Gilbert C. 2005. Unbundling the structure of inertia:resource versus routine rigidity. Academy of Manage-ment Journal 48(5): 741–763.

Hannan MT, Freeman J. 1977. The population ecologyof organizations. American Journal of Sociology 82:929–964.

Hannan MT, Freeman J. 1984. Structural inertia andorganizational change. American Sociological Review49: 149–164.

Heiner RH. 1983. The origin of predictable behavior.American Economic Review 73(4): 560–595.

Helfat CE. 1994. Evolutionary trajectories in petroleumfirm R&D. Management Science 40(12): 1720–1747.

Helfat CE, Peteraf MA. 2003. The dynamic resource-based view: capability lifecycles. Strategic Manage-ment Journal 24(10): 997–1010.

Henderson RM. 1993. Underinvestment and incompe-tence as responses to radical innovation: evidencefrom photolithographic alignment equipment industry.Rand Journal of Economics 24(2): 248–270.



Henderson RM, Clark KB. 1990. Architectural innova-tion: the reconfiguration of existing product technolo-gies and the failure of established firms. AdministrativeScience Quarterly 35(1): 9–30.

Henderson RM, Cockburn IM. 1994. Measuring com-petence? Exploring firm effects in pharmaceuti-cal research. Strategic Management Journal 15(8):63–84.

Hensgen T, Desouza KC, Kraft GD. 2003. Games, signaldetection, and processing in the context of crisismanagement. Journal of Contingencies and CrisisManagement 11(2): 67–77.

Herrbach O. 2001. Audit quality, auditor behavior and thepsychological contract. European Accounting Review10(4): 787–802.

Hofer CW, Schendel D. 1978. Strategy Formulation:Analytical Concepts . West: St. Paul, MN.

Janis IL. 1982. Groupthink: Psychological Studies ofPolicy Decisions and Fiascoes (2nd edn). HoughtonMifflin: Boston, MA.

Johnson P, Johnson G. 2002. Facilitating group cognitivemapping of core competencies. In Mapping StrategicKnowledge, Huff AS, Jenkins M (eds). Sage: London;220–236.

Kogut B, Kulatilaka N. 2001. Capabilities as real options.Organization Science 12(6): 744–758.

Kogut B, Zander U. 1992. Knowledge of the firm,combinative capabilities, and the replication oftechnology. Organization Science 3: 383–397.

Kusunoki K, Nonaka I, Nagata A. 1998. Organizationalcapabilities in product development of Japanesefirms: a conceptual framework and empirical findings.Organization Science 9(6): 699–718.

Lei D, Hitt MA, Bettis R. 1996. Dynamic core compe-tencies through meta-learning and strategic context.Journal of Management 22(4): 549–569.

Leonard-Barton D. 1992. Core capabilities und corerigidity: a paradox in managing new productdevelopment. Strategic Management Journal 13(5):111–126.

Levinthal D. 2000. Organizational capabilities in complexworlds. In The Nature and Dynamics of OrganizationalCapabilities , Dosi G, Nelson RR, Winter SG (eds).Oxford University Press: Oxford; 363–379.

Levinthal D, March JG. 1993. The myopia of learning.Strategic Management Journal 14(8): 95–112.

Levitt B, March JG. 1988. Organizational learning.Annual Review of Sociology 14: 319–340.

Luhmann N. 1993. Deconstruction as second-orderobserving. New Literary History 24: 763–782.

Luhmann N. 1995. Social Systems . Stanford UniversityPress: Stanford, CA.

March JG. 1991. Exploration and exploitation inorganizational learning. Organization Science 2(1):71–87.

March JG, Olsen JP. 1976. Ambiguity and Choice inOrganization . Universitetsforlaget: Bergen.

March JG, Simon HA. 1958. Organizations . Wiley: NewYork.

Markides CC. 1998. Strategic innovation in establishedcompanies. Sloan Management Review 39: (Spring):31–42.

Maturana HR. 1970. Biology of cognition. BiologicalComputer Laboratory Research Report bcl 9.0,University of Illinois: Urbana, IL.

Mendonca S, Cunha MP, Kaivo-oja J, Ruff F. 2004.Wild cards, weak signals and organisationaleimprovisation. Futures 36(2): 201–218.

Miller D. 1993. The architecture of simplicity. Academyof Management Review 18: 116–138.

Miller D. 1994. What happens after success: the perilsof excellence. Journal of Management Studies 31:325–358.

Miller DT, Nelson LD. 2002. Seeing approach motiva-tion in the avoidance behavior of others: implicationsfor an understanding of pluralistic ignorance. Journalof Personality and Social Psychology 83: 1066–1075.

Mintzberg H. 1979. The Structuring of Organizations .Prentice-Hall: Englewood Cliffs, NJ.

Mussweiler T, Strack F, Pfeiffer T. 2000. Overcomingthe inevitable anchoring effect: considering theopposite compensates for selective accessibility.Personality and Social Psychology Bulletin 26:1142–1150.

Nelson RR, Winter SG. 1982. An Evolutionary Theoryof Economic Change. Belknap Press of HarvardUniversity Press: Cambridge, MA.

Nemeth C, Brown K, Rogers J. 2001. Devil’s advocateversus authentic dissent: stimulating quantity andquality. European Journal of Social Psychology 31(6):707–720.

Nonaka I. 1994. A dynamic theory of organizationalknowledge creation. Organization Science 5(1):14–37.

North DC. 1990. Institutions, Institutional Change andEconomic Performance. Cambridge University Press:Cambridge, U.K.

Nystrom PC, Starbuck W. 1984. To avoid organizationalcrisis—unlearn. Organizational Development 13(1):53–65.

O’Reilly CAI, Tushman M. 2004. The ambidextrousorganization. Harvard Business Review 82(4): 74–81.

Pearson CM, Mitroff II. 1993. From crisis prone tocrisis prepared: a framework for crisis management.Academy of Management Executive 7: 48–59.

Pedler M, Burgoyne J, Boydell T. 1991. The LearningCompany: A Strategy for Sustainable Development .McGraw-Hill: London.

Peteraf MA. 1993. The cornerstones of competitiveadvantage: a resource-based view. Strategic Manage-ment Journal 14(3): 179–191.

Peteraf MA, Barney JB. 2003. Unraveling the resource-based tangle. Managerial and Decision Economics 24:309–323.

Piaget J. 1970. Carmichael’s Manual of Child Psychol-ogy . Wiley: New York.

Polanyi M. 1958. Personal Knowledge, University ofChicago Press: Chicago.

Polanyi M. 1966. The Tacit Dimension . Routledge &Kegan Paul: London.

Preble JF. 1997. Integrating the crisis managementinto the strategic management process. Journal ofManagement Studies 34(5): 769–791.



Putnam LL, Majia H. 1992. Framing, reframing, andissue development. In Communication and Negotia-tion , Putnam LL, Roloff ME (eds). Sage: London;128–155.

Repenning NP, Sterman JD. 2002. Capability traps andself-confirming attribution errors in the dynamicsof process improvement. Administrative ScienceQuarterly 47(2): 265–295.

Sanchez R, Mahoney JT. 1996. Modularity, flexibility,and knowledge management in product and organi-zation design. Strategic Management Journal , WinterSpecial Issue 17: 63–76.

Schein EH. 1985. Organizational Culture and Leader-ship: A Dynamic View . Jossey-Bass: San Francisco,CA.

Schreyogg G, Noss C. 2000. Reframing change inorganizations: the equilibrium logic and beyond. InAcademy of Management Conference Proceedings,Toronto; B1–B6.

Schreyogg G, Steinmann H. 1987. Strategic control: anew perspective. Academy of Management Review12(1): 91–103.

Sheaffer Z, Richardson B, Rosenblatt Z. 1998. Early-warning-signals management: a lesson from theBarings crisis. Journal of Contingencies and CrisisManagement 6(1): 1–22.

Staw BM. 1976. Knee-deep in the big muddy: a study ofescalating commitment to a chosen course of action.Organizational Behavior and Human Performance 16:27–44.

Staw BM, Ross J. 1978. Commitment to a policy deci-sion: a multi-theoretical perspective. AdministrativeScience Quarterly 23: 40–64.

Teece DJ, Pisano G, Shuen A. 1997. Dynamic capabili-ties and strategic management. Strategic ManagementJournal 18(7): 509–533.

Tripsas M, Gavetti G. 2000. Capabilities, cognition, andinertia: evidence from digital imaging. StrategicManagement Journal , Special Issue 21(10–11):1147–1161.

Tushman ML, Newman WH, Romanelli E. 1986. Con-vergence and upheaval: managing the unsteady paceof organizational evolution. California ManagementReview 24(1): 29–44.

Tversky A, Kahneman D. 1981. The framing of decisionsand the psychology of choice. Science 211: 453–458.

Vaill PB. 1996. Learning as a Way of Being: Strategiesfor Survival in a World of Permanent White Water .Jossey-Bass: San Francisco, CA.

von Foerster H. 1982. Observing Systems . Intersystems:Seaside, CA.

Walsh JP. 1988. Selectivity and selective perception: aninvestigation of managers’ belief structures. Academyof Management Journal 31: 973–896.

Weick KE. 1977. Organization design: organizations asself-designing systems. Organizational Dynamics 6:31–46.

Weick KE. 1993. The collapse of sensemaking inorganizations: the Man Gulch disaster. HarvardBusiness Review 38(4): 628–652.

Weick KE, Roberts KH. 1993. Collective mind inorganizations: heedful interrelating on flight decks.Administrative Science Quarterly 38(3): 357–381.

Weick KE, Sutcliffe KM (eds). 2001. Managing theUnexpected: Assuring High-Performance in an Age ofComplexity . Jossey-Bass: San Francisco, CA.

Weinert FE. 2001. Concept of competence: a conceptualclarification. In Defining and Selecting Key Compe-tencies , Rychen DS, Salganik LH (eds). Hogrefe &Huber: Seattle, WA; 45–65.

Wernerfelt B. 1984. A resource-based view of the firm.Strategic Management Journal 5(2): 171–180.

Westphal JP, Bednar MK. 2005. Pluralistic ignorancein corporate boards and firm’s strategic persistenceto low firm performance. Administrative ScienceQuarterly 50: 262–298.

Williamson OE. 1975. Markets and Hierarchies: Analysisand Antitrust Implications . Free Press: New York.

Winter SG. 2000. The satisficing principle in capabilitylearning. Strategic Management Journal , Special Issue21(10–11): 981–996.

Winter SG. 2003. Understanding dynamic capabilities.Strategic Management Journal 24(10): 991–995.

Zahra SA, George G. 2002. Absorptive capacity: areview, reconceptualization, and extension. Academyof Management Review 27: 185–203.

Zollo M, Winter SG. 2002. Deliberate learning andthe evolution of dynamic capabilities. OrganizationScience 13(3): 339–351.


HOW DYNAMIC CAN ORGANIZATIONAL … · Strategic Management Journal Strat. Mgmt. ... How Dynamic Can Organizational Capabilities Be? 915 ... tional capability is also a historical

Documents