Davis Marquis 05 OS

7/27/2019 Davis Marquis 05 OS

1/12

Organization ScienceVol. 00, No. 0, Xxxxx 2005, pp. 112

issn1047-7039 eissn1526-5455 05 0000 0001

informs

doi 10.1287/orsc.1050.0137

2005 INFORMS

Prospects for Organization Theory in theEarly Twenty-First Century:

Institutional Fields and Mechanisms

Gerald F. DavisUniversity of Michigan Business School, 701 Tappan Street, Ann Arbor, Michigan 48109-1234, [email protected]

Christopher MarquisHarvard Business School, Boston, Massachusetts 02161, [email protected]

This paper argues that the research in organization theory has seen a shift in orientation from paradigm-driven work toproblem-driven work since the late 1980s. A number of paradigms for the study of organizations were elaborated duringthe mid-1970s, including transaction cost economics, resource dependence theory, organizational ecology, new institutionaltheory, and agency theory in financial economics. These approaches reflected the dominant trends of the large corporations

of their time: increasing concentration, diversification, and bureaucratization. However, subsequent shifts in organizationalboundaries, the increased use of alliances and network forms, and the expanding role of financial markets in shapingorganizational decision making all make normal science driven by the internally derived questions from these paradigmsless fruitful. Instead, we argue that problem-driven work that uses mechanism-based theorizing and research that takes thefield rather than the organization as the unit of analysis are the most appropriate styles of organizational research underconditions of major economic changesuch as our own era. This sort of work is best exemplified by various studies underthe rubric of institutional theory in the past 15 years, which are reviewed here.

Key words: organization theory; social mechanisms; organizational fields; paradigms

Organization theory has found itself at an interest-ing crossroads at the turn of the century. On the onehand, we are constantly reminded that we live in a

world in which large organizations have absorbed soci-ety and vacuumed up most of social reality (Perrow1991). Multinational corporations (MNCs) have tran-scended political boundaries in their sales and theirproduction processes, and dozens of them have annualrevenues that surpass the GDP of all but a few nations.Their employees are like citizens with rights, benefits,and legalistic grievance procedures (Dobbin and Sutton1998). Just as nation-states in centuries past came to bethe dominant locus of power through their monopoly onthe legitimate use of physical force, MNCs dominate theworld economy (and thus society) through their concen-trated control of capital. States are largely stuck with

agreed-upon land borders, but MNCs and their mobileinvestments get to choose their jurisdictions in the mar-ketplace of laws. It hardly seems a fair fight, as largeorganizations continue their drive to vacuum up what-ever is left of social life. This situation locates organiza-tion theory as the queen of the social sciences, uniquelyable to explain the dominant social structures of ourtimes.

Yet close inspection by our theoretical confreresin law and economics reveals most organizations tobe mere legal fictions with no inside or outside

analogous to bordersthey are simply dense spotsin networks of contracts among sovereign individuals(who may themselves be mere fictionsJensen and

Meckling 1976). With corporations, there is no theretherethey are simply legal devices with useful prop-erties for raising finance. While counting new incorpo-rations may give the impression that we are living ina Cambrian Explosion of organizations (Aldrich 1999),counting may not be that informative. It is triviallyeasy to incorporate in the United States, with or with-out a recognizable organization. Enron had upwards of3,500 subsidiaries and affiliates, often organized as cor-porations or limited liability companiesentities thatwere often both legal and accounting fictions. Orig-inal equipment manufacturers (OEMs) routinely mar-ket products designed, manufactured, and distributed by

other firms. Firms have no more long-term attachmentto their employees than consumers do to their grocers(Alchian and Demsetz 1972). Coase (1937, p. 388, quot-ing Robertson) noted that in a market economy we findislands of conscious power in this ocean of unconsciouscooperation like lumps of butter coagulating in a pail ofbuttermilk, but it is becoming increasingly difficult todistinguish the lumps, much less count them. By someaccounts, the imposing objects of organization theoryhave evaporated. No longer queen, organization theorymay be more like the phrenology of the social sciences.

1


2/12

Davis and Marquis: Prospects for Organization Theory in the Early Twenty-First Century2 Organization Science 00(0), pp. 112, 2005 INFORMS

This paper argues that an appropriate aspiration fororganization theory in the early twenty-first century isproviding a natural history of the changing institutionsof contemporary capitalism. By natural history we meanexplaining patterned variation over time: how do spe-cific ecosystems of institutions change? Institutions arethe sources of social and economic orderwhether orga-

nizational or otherwise. Contemporary capitalism high-lights the prominent economic trends of our times:globalization in finance and trade flows, postindustri-alism and the declining importance of manufacturingrelative to service in advanced economies, and a pre-dominant place for financial markets in generating bothorder and disorder. North (1990, p. 6) argues that thecentral puzzle of human history is to account for thewidely divergent paths of historical change. How havesocieties diverged? What accounts for their widely dis-parate performance characteristics? Organization the-ory has a distinctive toolkit for addressing this puzzleas it plays out today, including a well-elaborated set oftheoretical mechanisms that can illuminate how macro-level changes (e.g., governmental policies to eliminatediscrimination, the growth of the environmental move-ment, a bursting asset price bubble) have their impacton the ground. This is a pivot point in the academicdivision of labor, a crossroads between sociology, eco-nomics, psychology, and political science. Put simply,organization theorists are best placed to address someof the critical questions of our time because organiza-tional processes are often the drive train by which socialand economic change are effected. As Lounsbury andVentresca (2002, p. 6) point out, this would in effect

be a return to the sociological approach to organizationsof the 1950s, when organizations were seen as sitesfor understanding the constitution and consequences ofmodern forms of power rather than objects of theoryin their own right. A review of recent work suggeststhat the field has largely moved in this direction with-out particular conscious guidance (see Davis 2004 for afull account). Moreover, we argue that new institutionaltheory has the best chance of accomplishing these under-standings because it focuses on fields, mechanisms, andchange, particularly market incursions into tradition-ally stable fields, of the sort most common today.

The paper is organized as follows. We first argue that

empirical work in organization theory has shifted overthe past two decades from paradigm-driven research, inwhich topics to be studied flow directly out of problemsof theory, toward problem-driven work oriented towardevents in the world. We argue that this shift followsfrom three factors: changes in the composition of orga-nizational fields in recent times, changes in the empiri-cal relations among core constructs in the theories, anda greater sensitization to cross-national differences andwhat they imply for general theories of organization. Wedescribe social mechanisms and their place in theories

about organizations. We argue that problem-driven workdrawing on (organizational) mechanisms is particularlyapt during a time of significant social and economic tran-sitions, when the explanatory power of old theories hasbroken down. We then review studies in the traditionof new institutional theory that exemplify the kind ofmechanism-based theorizing that we advocate. We con-

clude with some implications for what organization the-ory can and should look like going forward.

Organization Theory: FromParadigm-Driven to Problem-Driven1

For the first decade after the foundational statementof March and Simon (1958), organization theory hadthe prospect of following a single paradigm. As in thebureaucracies described by March and Simon, the fieldcould decompose the problem of understanding orga-nizations into subproblems amenable to discrete piecesof research that might aggregate back up into a grandtheory of organizations. One might study the induce-ment/contribution theory, or problemistic search, or theoperation of departmental information filters, with theexpectation that the findings would add up to an under-standing of organizations. Thompsons great 1967 syn-thesis Organizations in Action summarized the resultsof the intervening years of research and, along withLawrence and Lorsch (1967), pointed the way to a morecontingent model of organization. In subsequent years,the field brachiated into a set of partially overlapping,partially competing theories of organization. Transac-tion cost economics focused on where organizations

placed their boundaries and how relations among sepa-rable parts were governed (Williamson 1975). Resourcedependence argued against the efficiency orientation ofeconomic approaches and posited that much of an orga-nizations structure and action repertoire flowed from itsposition in networks of exchange and the power imbal-ances these created (Pfeffer and Salancik 1978). Newinstitutional theorists focused on the symbolic natureof much organizational action and structure and theirdecoupling from internal operations, oriented as theywere toward the requirements of external evaluators(Meyer and Rowan 1977). And organizational ecologistsargued that bureaucracies were prone to such rigidities

that the most sensible research approach was not tostudy internal dynamics and adaptation, but the birthsand deaths overextended periods of different forms thatwere more or less fit for their selection environment(Hannan and Freeman 1977).

Theoretical debates among divergent paradigms canbe a positive sign of health for a discipline, and thewaning hegemony of a single approach may signal mat-uration. Baum and Rowley argue that a diversity ofperspectives need not lead to fragmentation and a lackof consensus Indeed, multiple views are vital to


3/12

Davis and Marquis: Prospects for Organization Theory in the Early Twenty-First CenturyOrganization Science 00(0), pp. 112, 2005 INFORMS 3

scientific advancement (2002, p. 23). The elaborationof divergent perspectives allows for critical tests thatadvance scientific understandingand may the best the-ory win (Stinchcombe 1968). Indeed, the burst of found-ings of new theories in the mid-1970s is somethingthat students of organizations might have expected fromStinchcombes (1965) discussion of social structure and

organizations. Presumably, a winnowing process wouldfollow to select out the weak theories and allow thestrong to survive. Research following the theory pro-liferation tested core ideas in each of these divergentapproaches, and scholars occasionally found contexts toconduct critical tests (e.g., Singh et al. 1986). One mighthave hoped to see another grand synthesis at the endof the 1980s that selected judiciously from the compet-ing paradigms, drawing on meta-analyses of the cumula-tive bodies of findings in each approach (Aldrich 1999).Based on this synthesis, new work could proceed tocumulate knowledge in ever-greater territories of orga-nizational life.

Yet nothing of the sort has happened. Instead,with the notable exception of population ecologists,macro-organizational scholars since 1990 have largelyabandoned the idea of cumulative work within a par-ticular paradigm in favor of problem-driven work thatis theoretically agnostic. To document this, we clas-sified 89 articles published in Administrative ScienceQuarterly from 1991 through 2001 that could be cat-egorized as contributions to organization theory. Wefound that ten articles (11%) followed a theory-testingmodel in which the research question flowed directlyfrom the logic of a particular theory. In each case

the ultimate dependent variables were organizationalbirth and/or death rates, while the independent vari-ables included organizational density and competition,change, and governmental regulation. Outside of ecol-ogy, however, the problems that occupied researchersattention were, in general, broadly topical, and theapproaches theoretically eclectic. With a few exceptions,empirical studies examined events from the past fewdecades: corporate acquisitions (who did them, whatadvisors they used, how much they paid, where the tar-gets were located), takeovers (who was threatened, howthey responded), boardroom dynamics, alliances, diver-sification and strategic change, and so on. Moreover,

the broader questions were often core sociological prob-lems for which organizations mediate between economicand social forces and individual or collective outcomes.How have Japanese firms labor practices changed inresponse to the extended economic downturn? What isthe organizational texture of Chinas transition to capi-talism? How do market pressures change the way healthcare is delivered? How do corporate elites maintain theirpower in the faces of challenges arising out of share-holder capitalism? The most engaging recent work hassought to make sense of the intersection of biography

and history in social structures, as Mills (1959) urgedlong ago. It happens that the social structures that mat-ter are often organizational structures (Lounsbury andVentresca 2002).

We want to be clear that by problem-driven work wedo not imply that the research was intended to providesolutions to the problems faced by business managers

(although it may have been). Rather, problem-drivenwork is distinguished by its orientation toward explain-ing events in the worldstarting with the questionwhy is it that ? Paradigm-driven work, in contrast,begins with hypotheses deduced from theory intended tobe general. Events in the world are primarily contextsfor testing those hypotheses in paradigm-driven work.Stokes (1997) argues in the case of the natural sciencesthat the distinction between basic and applied isa false dichotomy: Research can be driven by a questfor fundamental understanding with no prospect for use(Bohrs quadrant), for individual, group, or societal usewithout regard for fundamental knowledge (Edisonsquadrant), or fundamental knowledge inspired by use(Pasteurs quadrant). (Research may also be done withneither use nor fundamental knowledge in mind, e.g.,lab training.) In the post-war era, he argues, Pasteursquadrantknowledge for usehas been the source ofthe most significant scientific advances. By the sametoken, some of the great advances in social scientificthinking have taken the form of problem-driven work.Arguably, both the 18th Brumaire of Louis Bonaparte(written during the time that Napoleons nephew and hisfollowers were dissolving the assembly and establishinga dictatorship) and The Protestant Ethic and the Spirit

of Capitalism are problem-driven works in the sense wehave described.Why the shift from paradigm-driven to problem-

driven work in organization theory? There are severallikely causes. First, the product of the Cambrian Explo-sion of organization theories in the mid-1970s appearsin retrospect to be a set of divergent perspectives onAmerican organizations, especially large firms, from thatpeculiar era. American firms might have vertically inte-grated their suppliers with high asset specificity, diver-sified out of industries that were overly constraining,adopted a set of decoupled structures to ward off outsideevaluators, and become structurally inert in the process.

However, like Nixon, disco, and polyester clothing, thesewere not necessarily permanent features of the sociallandscape. Manufacturing conglomerates largely disap-peared in the United States during the 1980s, alliancesand other networks displaced vertical integration, thefinancial revolution drastically changed the characterof outside evaluators for public corporations and othertraded entities, and constant change became the clichordering business. Statistical relationships among vari-ables turned out to be highly unstable over time; forinstance, patterns of merger, acquisition, and executive


4/12


succession looked wildly different in the 1960s and the1980s. In the earlier decade, conglomerates run by socialoutsiders did the acquiring and single-industry firmswere the targets, while in the 1980s it was conglom-erates (now run by insiders) that were taken over andbusted up while focused firms were largely safe. (Onepossibility, of course, is that such instability is a sign of

low internal validitysee Cook et al. 1990, p. 497.)In hindsight, it seems nave to have imagined that

things could have been otherwise. Harvey (1990, p. 343)notes that producing novelty, in products and orga-nizations, is one of the essential features of capital-ism: capitalism creates its own distinctive historicalgeography. Its developmental trajectory is not in anyordinary sense predictable, precisely because it hasalways been based on speculationon new products,new technologies, new spaces and locations, new laborprocesses and the like There are laws of pro-cess at work under capitalism capable of generating aseemingly infinite range of outcomes out of the slight-est variation in initial conditions or of human activityand imagination. By the 1990s we saw a floweringof organizational forms, from a revival of preindustrialputting-out systems (Lazerson 1995) to forms that wouldhave been unimaginable a few years prior. Web-basedretailer would not be a comprehensible phrase in 1993;in 1999 there were hundreds of them; and in late 2003one of them (Amazon.com) had a market capitalizationthat approximated that of General Motors, the worldslargest manufacturer. As Sabel and Zeitlin (1996) put it,It is as though the prehistoric and imaginary creaturesin the industrial bestiary had suddenly come to life.

Cross-national research highlighted the idiosyncrasyof the United States and its large organizations. Forexample, while industrial districts were regarded as anextinct form in the United States (when they were con-sidered at all), they provided longstanding and robustalternatives to vertical integration elsewhere in the world(Piore and Sabel 1984) and evidently a good model fororganizing high technology industries (Saxenian 1994).American managers were protected from their share-holders by the dispersed ownership that accompaniedlarge corporate sizethe famous separation of owner-ship and control. However, outside the United Statesand the United Kingdom, the vast majority of large cor-

porations had dominant shareholders, either foundingfamilies or governments, who were not removed fromactive management (Davis and Useem 2002). Businessgroups, relatively rare in the United States, turned outto be pervasive in other economies, making the sepa-rable, countable, autonomous organization of Americanorganization theory the rare exception rather than therule (Granovetter 1994). Some even argued that West-ern theoretical views of atomistic firms reflected ageneral ethnocentrism that revealed itself in pervasivemisunderstandings of East Asian enterprise (Biggart and

Hamilton 1992). Indeed, legal scholars find that thequestion How do we know when an enterprise exists?is often in principle unanswerable in China, given incon-sistent approaches to recognizing firms among diversegovernmental agencies (Clarke 2003).

Finally, there is an increased sophistication about whatsocial science can achieve with respect to organizations.

The notion of a theory of organizations now seems likenave scientism, like a theory of diesel trucks, or a theoryof hitchhiking. Organizations simply are not the kind ofthing amenable to general theory. Applying a templateto organizational phenomena (such as an open systemsschema of inputs, transformation processes, and outputs)often obscures more than it illuminates in the case ofthe postindustrial organizations that surround us today.A brief case example follows. At the start of the 1990sWestinghouse was a 100-year old industrial conglomer-ate employing upwards of 100,000 people. During thenext decade, it sold of a number of industrial busi-nesses, acquired broadcaster CBS, liquidated its remain-

ing industrial operations (shedding half its employees),changed its name to CBS, moved its headquarters fromPittsburgh to New York, and was acquired by Viacom,where it is now part of a stable of media properties.This shiftin product, industry, employees, identity, andgeographydoes not accord well with any of the the-ories we have mentioned, and while extreme in degree,it is far from unique. If not predictable, we can at leasthope to make such changes explicable, and a single gen-eral contingency theory of organizations will not be upfor the task.

There are dangers to a discipline that lacks a domi-nant paradigm to guide the accumulation of knowledge

(Pfeffer 1993). Lab psychologists can do meta-analysesand come to some conclusions that might generalize(Aldrich 1999), but problem-driven work risks beingsimply business journalism with regressions. The valueof cataloguing novelty may not be self-evident. Andtheoretical eclecticism can easily degenerate into laun-dry lists of variables to include in statistical analyses.Sorensen (1998, pp. 238239) laments the fascination,if not an obsession, with statistical models and concerns,and a neglect of the need to develop sociological mod-els mirroring conceptions of mechanisms of social pro-cesses leading to statistical models with a conceptuallymeaningless list of variables preventing any kind of sub-

stantive conclusion (p. 243). Our own assessment is thatthe next step for organization theory is not to find andenforce the singular overarching paradigm (as some readPfeffer 1993 to say), but to move toward mechanism-based theorizing about change in the world of the sortthat recent work in new institutional theory exemplifies.

Mechanism-Based Theorizing

How are we to proceed without Theory? What Systemof Thought have these Reformers to present to this mad


5/12


swirling planetary disorganization, to the Inevident Wel-ter of fact, event, phenomenon, calamity? Show methe words that will reorder the world, or else keep silent.If the snake sheds his skin before a new skin is ready,naked he will be in the world, prey to the forces ofchaos. Without his skin he will be dismantled, lose coher-ence, and die (speech by Aleksii Antedilluvianovich

Prelapsarianov, worlds oldest living Bolshevik, in theKremlins Hall of Deputies, from Angels in America,Part 2: Perestroika).

What do we mean by mechanism-based theoriz-ing? Social mechanisms are sometimes-true theories(Coleman 1964, p. 516) that provide an intermediarylevel of analysis in-between pure description and story-telling, on the one hand, and universal social laws, onthe other (Hedstrom and Swedberg 1998). If a regres-sion tells us about a relation between two variables. Forinstance, if you wind a watch it will keep runningmechanisms pry the back off the watch and show how.Mechanisms describe a set of interacting partsan

assembly of elements producing an effect not inher-ent in any one of them. A mechanism is not so muchabout nuts and bolts as about cogs and wheels thewheelwork or agency by which an effect is produced(Hernes 1998, p. 74). Hedstrom and Swedberg (1998)collect several examples of mechanism-based theoriesin sociology: the self-fulfilling prophecy in which aninitially false definition of a situation evokes behav-ior that makes it become true (e.g., a run on a bank),diffusion through networks in which actors adopt aninnovation because their network contacts have previ-ously adopted, and threshold-based behavior in which an

actors propensity to join in a collective action dependson his or her observation of a number of others whohave already joined in. In each case, collective outcomes(bank runs, widespread adoption of an innovation, col-lective action) result from individual actors observingwhat others have done, which shapes their own propen-sity to take that action, which in turn makes them evi-dence for the next actor.

A general typology of micro/macro linkages includessituational mechanisms (from macro to micro), action-

formation mechanisms (from micro to micro), andtransformational mechanisms (from micro to macro)(Hedstrom and Swedberg 1998, pp. 2223). McAdam

et al. (2001, pp. 2526) provide an alternative typol-ogy oriented toward political outcomes, such as in socialmovements: environmental mechanism (mean exter-nally generated influences on conditions affecting sociallife), cognitive mechanisms (operate through alter-ations of individual and collective perception), andrelational mechanisms (alter connections among peo-ple, groups, and interpersonal networks). These mech-anisms are especially apt for making sense of socialchange processes and are broadly applicable to organi-zational phenomena in times of economic transition.

Campbell (2004) further catalogs a number of mecha-nisms that have been used in the study of organizationsand social movements: framing (the use of metaphorsand symbols that organize perceptions of issues and cuecourses of action by linking problems and actions toprevailing cultural conceptions), diffusion (the spreadof ideas, structures, and practices, often via networks),

translation (how ideas that diffuse are modified andimplemented to work in specific local contexts), brico-lage (recombining elements, often borrowed from othercontexts, to create a new configuration of social activ-ity), network cultivation and strategic leadership (orinstitutional entrepreneurship). Stinchcombe (2002) addsthree more: commensuration (the institutional processby which entities become comparable and thus competi-tors, as firms in an industry), evangelism (the reverse ofdiffusion, where adopters or their agents eagerly spreadorganizational practices), and the bases of truth telling.Of course there are many more mechanisms than thosein this brief set, but this gives a sense of the shape ofthe domain.

Importantly, mechanism-based theorizing can aspire toexplain but not predict. Elster quotes George Vaillant: Perhaps for every child who becomes alcoholic inresponse to an alcoholic environment, another eschewsalcohol in response to the same environment Bothreactions embody mechanisms: doing what your parentsdo and doing the opposite of what they do. We can-not tell ahead of time what will become of the childof an alcoholic, but if he or she either turns out to beeither a teetotaler or an alcoholic, we may suspect weknow why (Elster 1998, p. 45). The quest for novelty

in capitalist economies suggests that we will often be inthis situation: confident that actors will respond to par-ticular pressures, but uncertain in what direction. Thus,we may be able to explain afterward but not predictprospectively.

In some sense, mechanism-based work is a return toorganization theorys roots. March and Simons expla-nation of hierarchy in terms of bounded rationality is asuperb example. How are cognitively simple individualsable to accomplish ends far greater than themselves?They may accomplish this through the differentiationof subunits into mental bite-sized chunks and the art-ful reaggregation of those chunks via a hierarchy. Their

1958 book is a masterpiece of mechanism-based the-orizing about what organizations do. Subsequent workadded new mechanisms to our portfolio. We havemany instances of selection processes (through differ-ential births and deaths, through takeovers, throughlegal changes). We have countless studies of diffu-sion, through networks and otherwise (see Strang andSoule 1998 for a review). We know how industry con-centration can prompt the births of specialists throughresource partitioning (Carroll 1985). Network struc-tures and their consequences are now reasonably well


6/12


documented (Burt 1992). The formats of transactioncost-minimizing contracts are diffused through the lit-erature. The legacy of several decades research is, inshort, a virtual compendium or toolkit of organizationalmechanisms, albeit one that has not been well cataloguedas such.

Importantly, in contemporary work organizations are

often not the object of explanation but its subject. Thatis, researchers are prone not to ask why are thereso many kinds of organizations but how do organiza-tions act as the wheelwork producing a social outcome?(such as stratification, or state policy, or labor mobility).Doing organization theory does not signal that one isseeking to explain, say, why is there hierarchy? (paceMarch and Simon 1958). Rather, organization theory isbest seen as a commitment to a level of analysis, anorganizational mode of explanation. The point is not toaccumulate findings about what is generally true aboutorganizationsthis proves to be a fruitless endeavorbut to use organizational mechanisms to explain socialphenomena, to locate the intersection Mills (1959)wrote about.

Examples from New Institutional TheoryInstitutional theory, broadly construed, has the bestchance of accomplishing the aspiration for organizationtheory that we have described. Most obviously, insti-tutions are the core construct of the approach. Scott(1995, p. 33) defines them as follows: Institutionsconsist of cognitive, normative, and regulative struc-tures and activities that provide stability and mean-

ing to social behavior. Institutions are transported byvarious carriersculture, structures, and routinesandthey operate at multiple levels of jurisdiction. Sec-ond, this approach is distinguished by taking the fieldas the relevant unit of analysis and remaining agnosticabout whether it is composed of organizations, individ-uals, or other combinations of actors: The concept offield identifies an arenaa system of actors, actions,and relationswhose participants take one another intoaccount as they carry out interrelated activities. Ratherthan focusing on a single organization or movement, oreven a single type of organization or movement (pop-ulation), it allows us to view these actors in context

(McAdam and Scott 2004). By field we do not mean toinvoke field theory as Martin (2003) defines it. Althoughhe includes DiMaggio and Powells (1983) version ofnew institutional theory under this rubric, we see theirproposal to take the field as the unit of analysis as com-patible with mechanisms, an explicable chain of ele-ments banging into one another (Martin 2003, p. 16).

The players that populate fields and the nature of theirplay can change over time. For instance, medical carein the United States traditionally had a relatively simpleorganization. In 1950, physicians that were certified, and

were dues-paying members of the American MedicalAssociation saw patients and when appropriate referredthem to local nonprofit community hospitals. Physicians,hospitals, and the AMA were the most important actorsdefining health care, and the AMA acted as a centralgatekeeper preventing chiropractors, psychologists, faithhealers, and cosmetologists from joining their field. By

1990, health care included free-standing clinics for dial-ysis, plastic surgery, and drug abuse treatment, for-profitHMOs, and dozens of new medical specialties organizedinto a range of new forms and represented by a wildlybrachiating set of professional organizations beyond theAMA, which no longer exercised the same exclusivecontrol on entry into the field (Scott et al. 2000). Count-ing organizations within a particular industry would notin itself be an especially informative way to make senseof health care in the second half of the twentieth cen-tury; one must take a field-level approach, as indicatedby institutional theorists (DiMaggio and Powell 1983).

A field-level approach is especially appropriate duringunsettled times such as today, when new industry seg-ments proliferate and when the boundaries around exist-ing industries can shift from permeable to nonexistent. Inthe United States the sharp legal divisions among com-mercial banking, investment banking, and insurance, andthe geographical restrictions on the operations of finan-cial institutions, were largely eliminated, prompting theassembly of conglomerates providing every conceivablebusiness and consumer financial service under one roof.Media and communications companies have similarlyeffaced the distinctions among channels (providingbroadcasts, telephone service, cable television, broad-

band, and Internet access) and content (the stuff dis-tributed over the channels). The inevitable term mediaconglomerate ends up providing very little informationabout what a company actually does. In such contexts,a field-level approach provides a useful framework forcharacterizing how settlements come about.

Consider Silicon Alley, the Web industrial district inthe approximate vicinity of the Flatiron Building in NewYork that flourished during the late 1990s. At its for-mative stage, it was highly uncertain what Web busi-nesses would actually do. Possibilities included retailing,auctions, data provision, media content, advertising,community services, and many others, as well as design-

ing the Websites for people and businesses. The indus-try included individual freelancers, small start-ups, andunits of established companies in other industries. Newentrants drew founders, employees, and business mod-els from constituent industries that included software,

journalism, advertising, retail, publishing, and broad-cast media, prompting rampant institutional bricolage.Robbins (2002) studied the evolution of job titles inSilicon Alley, a fascinating cut on the problem of newindustry formation. There was little agreement amongWeb practitioners about what tasks should be grouped


7/12


together as a job, how jobs should be labeled, or howthey should be filled (a failure of Stinchcombes 2002commensuration mechanism). What should firms claimto be looking for when recruiting employeescontentdesigners, Web artists, or what? And what should poten-tial employees or contractors claim to be? Clearly thenomenclature of employment is important for matching

talent to tasks, yet Robbins found persistent idiosyn-crasy in the titling of positions that had not reacheda settlement by the time the financial bubble that heldSilicon Alley aloft popped. Moreover, this case exem-plifies a problem common across instances of industryconvergencewhose language and assumptions aboutorganizing wins when there is no hegemon to enforce therules? The answer is not obvious in advance, but mech-anisms can help provide an explanation for the processof settlement.

While industries typically morph out of prior indus-tries, drawing on elements such as what to call

jobs and how to organize work, the recycling indus-try could effectively date its founding to Earth Day1970, when pressures from the environmental move-ment spontaneously created demand for a solution tothe solid waste problem. Lounsbury (2001) examinedhow colleges and universities responded: Some sim-ply added new responsibilities for recycling to theirexisting waste management departments, while otherscreated new recycling departments, often managed byyoung activists. The latter schools were frequently dis-tinguished by the fact that their students were connectedto a national activist organization that disseminated tac-tics for activism and recycling implementation. Once in

place, the activist-recyclers evolved into a self-organizedprofession, creating a nationally qualified organizationthat helped members deal with the various pressuresthey faced. Moreover, once they were defined as a pro-fession, they could generate best practices, spread stan-dardized tools and techniques, and advocate for newsolutions within their respective organizationsthuscreating another overlay connecting members of the fieldof higher education.

Even in instances where an industry would seemto be reasonably well defined, a field-level approachproves essential for a full understanding of institutionalchange. Hoffman (1999, 2001a) analyzes how the U.S.

chemical industry responded to the advent of environ-mentalism from the early 1960s (when the publicationof Rachel Carsons Silent Spring focused public atten-tion on the environmental consequences of DDT) tothe mid-1990s (by which point the Chemical Manu-facturers Association had implemented a responsiblecare program of environmental principles binding ontheir members). Environmentalism had gone from heresy(perceived by industry members as the unscientific opin-ion of antiprogress fringe elements) to dogma, sub-scribed to by all responsible manufacturers (Hoffman

2001a). Hoffman defined the field not based on a com-mon product or technology, but according to interestin and impact on a central issue: for chemical indus-try environmentalism, Field membership was definedby who participated in the legal process and thereforehad a voice in determining institutional norms regardingenvironmentalism (Hoffman 1999, p. 364). One way

to understand this issue is to examine lawsuits amongorganizations as a way to trace changes in field member-ship. Thus, over time members expanded from chemicalmanufacturers and their trade associations to govern-mental organizations (particularly following the creationof the Environmental Protection Agency in 1970) andnongovernmental organizations (such as environmentalgroups that filed lawsuits against alleged polluters), ulti-mately including insurance companies who were poten-tially liable for waste cleanup for their clients. Theseaggregate-level processes in turn shaped the activities ofdiscrete firms within the industry by offering a delimitedmenu of feasible responses to pressures around envi-ronmental practice (Hoffman 2001b). Hoffmans workhighlights processes of framing, diffusion, translation,and bricolageall mechanisms identified by Campbell(2004).

While work on fields occasionally documents howactors outside the market, such as social movements,influence organizational fields, a recurring theme inmuch institutional research is how fields respond to mar-ket encroachment. That is, how do traditional practiceschange due to increases in market pressures (broadlydefined)? Ahmadjian and Robinson (2001) examined thechanging nature of the employment relation in Japanese

firms during the 1990s. During the extended period ofeconomic growth that Japan experienced from the 1950sto the late 1980s, core firms were able to make (andkeep) promises of long-term employment for their work-ers, recruiting college graduates who intended to staywith the organization for their entire careers. The sus-tainability of this practice was of course contingent oncontinuous growth. After the economic bubble burstaround the turn of the decade, Japanese firms facedstrong pressures to downsize their workforces, a prac-tice quite common in the United States, yet an anathemain Japan. Firms that tried to shrink their labor forces,particularly large and visible firms that were not able

to point to dramatic losses, found themselves subjectto television exposs and protest demonstrations. Even-tually downsizing occurred on a large scale, as firmsfound that if they shrunk their employment roles as partof a herd, they were less likely to be singled out foropprobrium. Again, this represents a field-level approachthat explains large-scale historical events by linking theactions of individuals (corporate managers) to collectives(herds of companies) through social mechanisms.

Field-level change in college publishing firms isdemonstrated by the change in executive succession


8/12


practices as publishers moved from an editorial logic(grounded in family ownership and strong ties betweeneditors and authors) to a market logic (where publish-ing firms were often units of conglomerates evaluated bybottom-line profitability). Thornton and Ocasio (1999)document a shift in the allocation of power at the top ofthe publishing hierarchy as the field moved from the edi-

torial to the market model, but more tellingly they arguethat the rules of the game had changed: Why executiveswould leave depended on whether the industry was seenas a gentlemans game or just another commodity busi-ness. Such changes have implications for what kinds ofcontent make it into print and thereby into the handsof college students. The field-level conception indicatesthat what counts as good performance for a publisher,and thus which executives are regarded as successes andfailures, changes over time as new rules replace old.

Another exemplary study examined how rural hos-pitals responded to competitive changes in their envi-ronments. The changes in the composition of the fieldof health care described by Scott et al. (2000) gener-ated a fair amount of competitive pressure for hospitals.While one might imagine that these organizations werestuck between a rock and a hard place, their admin-istrators were often quite resourceful in constructingsolutions to their problems. Hospitals transformed intodiverse organizations within the broad field of healthcare, including nursing homes, drug treatment facili-ties, and outpatient clinics (DAunno et al. 2000). Thisstudy shows the micro-structure of field-level change bydemonstrating how actors of one type (hospitals) cantransform into new types, changing the aggregate struc-

ture of the field (and moreover the availability of certaintypes of services for patients).In the United States, the question of how human

resource practices have responded to legal changes hasreceived a fair amount of attention from Dobbin (1998),Sutton (1996), Edelman (1992), and their collaborators,and they have generated a model to account for this pro-cess. While one might perceive governmental regulationto require firms to do certain well-specified things and toavoid others, law is often much more ambiguous partic-ularly when it comes to corporate structures. (Americancorporate law is described as being generally enablingrather than mandatory.) The various Civil Rights acts of

the mid-1960s are a case in point: Legal mandates toavoid discrimination do not exactly provide a roadmapto compliance. Firms experimented with various formsaimed at demonstrating complianceequal employmentpolicies, affirmative action offices, and othersuntilcourt testing showed one to be sufficient, after whichthe sanctioned approach diffused widely among firms(Edelman 1992). Moreover, whether the creation of newhuman resource offices was intended as a cynical faadedetached from actual operations or a sincere effort torectify past discrimination, the outcome was the creation

of a constituency within firms for innovations in person-nel policies and a professional network linking compa-nies into a field with respect to their human resources(Sutton and Dobbin 1996). Human Resource profession-als end up portraying their profession not as the haplessimplementer of unwanted and ambiguous state policies,but as the originator of modern and progressive employ-

ment practices (Dobbin and Sutton 1998). Once again,field-level organizational mechanisms are the devicesstanding between governmental policy on discriminationand practices on the ground such as maternity leave andaffirmative action that shape individual life chances.

Even in the core domain of shareholder capitalism, afield-level, mechanism-oriented approach provides dis-tinctive insights into the large American corporation.The structures, practices, power relations, and generalorientation of U.S. firms changed dramatically duringthe 1980s and 1990s as the ideology of shareholder cap-italism (and its quasi-theoretical underpinning in agencytheory) came to dominate the corporate economy. Hos-tile takeovers aimed at unlocking shareholder valuewere the most visible device for changing corporatepractice, but (as with legislation against discrimination)corporate managers and boards were resourceful in howthey responded. It is a truism of social psychology that,in the face of ambiguity, decision makers often taketheir cue from what other actors have done. For man-agers and boards, this often takes the form of followingthe lead of the firms with which they share directorsor which are located in the same city. A compari-son of the spread of poison pills (which make hostiletakeovers much more difficult) and golden parachutes

(which richly compensate those who lose their jobs aftera hostile takeover) shows how mechanisms at the fieldlevel shaped aggregate corporate responses to takeovers(Davis and Greve 1997). Whereas pills spread rapidlythrough a board-to-board contagion process in whichdirectors of firms that adopted them encouraged theother boards on which they served to adopt, parachutesspread slowly through geographic proximitythat is,firms adopted parachutes to the extent that others head-quartered in the same city had done so. The divergentspreads of these devices reflects their legitimacy in theeyes of the relevant constituenciescorporate managersand directors. In contrast to their evaluations by share-

holders, corporate directors typically considered poisonpills to be unobjectionable but golden parachutes to behighly questionable. Given this background condition,the two innovations had rather different diffusion pro-cesses, with pills diffusing rapidly across the countryand parachutes gaining prevalence much more slowly.Norms also varied by region. For example, most firms inDallas had parachutes early, most in northern Californianever did.

Other studies that examine the geography of legiti-macy include the analysis by Marquis (2003) of how


9/12


traditions in corporate governance practice vary acrosslocal U.S. business communities. Communities that wereestablished before the advent of air travel (i.e., thatgrew up prior to the 1920s) such as St. Louis main-tained more densely connected corporate networks intothe twenty-first century than younger communities ofcomparable size such as Phoenix. This network imprint-

ing effect reflects local norms passed on by prominentlocal companies to newcomers. Thus, a new companyin an established business community is more likely toappoint directors to its board from among local corpo-rate peers than is a comparable new firm in a youngerbusiness community, which in turn socializes the new-comer into local practice.

As these studies show, a good working knowl-edge of organization theory provides a large toolkitof mechanismssometimes-true theoriesfor explain-ing the evolution of economic institutions, and there issense in remaining agnostic ex ante rather than beingwedded to, say, birth and death, or diffusion, or adapta-tion, as the true-or-false explanation. For instance, oneof the central problems in the world economy todayis convergence in corporate governance practices. Thereare many equally plausible explanations for whether andwhen firms or national systems of governance wouldmove toward a common type. In a recent paper (Davisand Marquis 2004) we studied the most likely case forconvergence: foreign firms listed on the New York StockExchange and Nasdaq. Prior arguments suggested thatlisting in the United States was a short path toward con-vergence on the American model. We contrasted severalaccounts that might have turned out to be true: For-

eign firms might list in the United States only whenthey have already turned American. Firms might turnAmerican over time after listing in the United States.Firms might turn American to the extent that they aretied to American firms (e.g., through shared directors).Or old firms might remain inert, while new firms areborn American. Our results were consistent with thelast explanation: While French firms privatized by thestate are listed on the New York Stock Exchange andrecruit American directors, they retain their distinctivepractices over time. In contrast, Israeli biotech and start-up software firms are in many respects indistinguishablefrom their Silicon Valley counterparts. One process of

interest in the contemporary economy (convergence incorporate governance) was susceptible to four organiza-tional explanations, but during the time we studied (thelate 1990s) one turned out to be the most applicablesometimes-true theory, while we could tentatively ruleout the others.

These examples do not describe the basis of a gen-eral theory of organizations. They are part of a naturalhistory of a particular set of economically consequen-tial institutions. Most of the findings we describe arehighly context specific, describing how American firms

in a federalized legal system responded to the CivilRights Acts of the 1960s, or how Japanese corpora-tions changed their employment practices in response tothe end of the 1980s bubble economy. Thus, it makessense for these accounts to be assembled from bits ofsometimes-true theoriesthat is, from mechanismsrather than seeking the observed results as deductions

from general covering laws. Note that these studies wereproblem driven not in the sense that they yield use-ful insights for managers, but because they are orientedtoward explaining events in the world rather than chosenpurely as contexts for testing hypotheses derived fromtheory.

ImplicationsWe have argued that an appropriate goal for organiza-tion theory in the early twenty-first century is a betterexplanation of the economic institutions of our time.The studies we reviewed in the previous section provide

guidance for how to do this: They exemplify problem-driven research drawing on organizational mechanismsto make sense of more or less singular historical occur-rences in institutional fields. The aim of this work is notto generate a general theory of what organizations areor do, but to create an understanding of how historicalshifts in economy and society have their impact on theground (cf. Lounsbury and Ventresca 2002).

We close with what we hope will be useful thoughtsfor the future. First, how can we evaluate scientificprogress in a discipline that is theoretically eclectic,rooted in mechanisms rather than paradigms? One signof progress is that weak theories are selected out,

as researchers favor progressive theories capable ofaccounting for observed regularities while making novelpredictions (e.g., Baum and Rowley 2002). We are will-ing to assert that organization theory as a discipline hasno history of such selection and little prospect for it inthe future: The theories of the 1970s, for instance, con-tinue to hang on independent of empirical confirmation,and efforts at disconfirmation are both rare and relativelyineffective. To our knowledge, no organizational the-ory has ever been rejected (as opposed to falsified).However this should not be surprising: Organizations aresimply not the kinds of things susceptible to a theory,and researchers in practice have largely abandoned the

notion of a theory of organizations. An alternative formof progress is in the discovery (or creation) and refine-ment of mechanisms. The most productive theoreticalwork going forward will be in cataloging and develop-ing organizational mechanisms. Mechanisms, unlike the-ories, are not falsified; rather, they are employed as toolsfor explanations (which can themselves be rejected, ofcourse).

One might hold out hope that a general theory offields will eventually fill the space held by organiza-tion theory. That is, rather than abandoning the idea of


10/12


a unifying approach for the discipline, we might adoptan alternative that avoids the difficulties of taking organi-zations as units of analysis. Some have already declaredvictory in this regard (see Martin 2003 on new insti-tutionalism). However this is premature: An appropri-ate lesson from the analysis of Stokes (1997) Pasteursquadrant is that problem-driven work is far more likely

to produce basic understandings than armchair theoriz-ing. This will remain the case at least until a stock offield-level studies with close attention to mechanismshas accumulated.

Three areas of research deserve the greatest atten-tion in light of our argument for problem-driven workon contemporary economic institutions. First, relativelyfew studies (other than those we reviewed above) havetaken seriously the injunction to study fields as units ofanalysis. As a result, our understanding of the empiri-cal properties of fields is somewhat limited. The birth ofnew industries (e.g., the Internet industry) and blendingof old industries (e.g., information and communicationtechnologies, financial services, media) provide particu-larly rich contexts for studying fields and their genesis,as these are situations where the admissions standardsand rules of play are revealed and contested. Second,market incursions into previously protected fields arean ongoing feature of contemporary life, as profes-sional discretion gives way to financial rationalization inpublishing, higher education, health care, the press, andelsewhere. Understanding the organizational impact ofmarketization as it occurs in new contexts is a pressingneed for future research. Finally, while market incursionsinto sacred domains is a venerable part of modernization

(see, for instance, the Communist Manifesto), the incur-sion of financial markets and their peculiar ontology isdistinctly postmodern. One of the evident pressures isfor corporations and those that run them to prove theirmerit relative to other financial assets (such as mortgage-backed securities or credit card debt bonds). The mech-anisms of commensuration (Stinchcombe 2002) amongfinancial assets, including corporations, are particularlyapt subjects for future research.

We note in conclusion that our argument does notrequire a substantial change in current research practice.Indeed, a survey of the past decade and a half of researchin the macro end of organization science shows that the

field has already swung in the direction we describe,toward problem-driven rather than theory-driven work(Davis 2004). Beyond simple ratification of a changethat has already occurred, however, we encourage afocus on fields and mechanisms: fields because substan-tial economic change does not stay contained withinorganizational or industry boundaries, and mechanismsbecause the quality of explanation is enhanced by anexplicit focus on the cogs and wheels behind the regres-sion coefficients. Such a turn would greatly enhance theprospects for organization theory in the early twenty-first

century to provide useful understandings of the contem-porary era.

AcknowledgmentsThe authors thank Ron Burt, Kim Elsbach, Scott Sonenshein,Kathie Sutcliffe, Marc Ventresca, Jim Walsh, participants inthe Frontiers of Organization Science Conference, and two

anonymous reviewers for helpful comments on an earlier draft.

Endnote1The argument in this section draws on Davis (2004).

ReferencesAhmadjian, C. L., P. Robinson. 2001. Safety in numbers: Downsiz-

ing and the deinstitutionalization of permanent employment inJapan. Admin. Sci. Quart. 46 622654.

Alchian, A., H. Demsetz. 1972. Production, information costs, andeconomic organization. Amer. Econom. Rev. 62 777795.

Aldrich, H. E. 1999. Organizations Evolving. Sage, Thousand Oaks,CA.

Baum, J. A. C., T. J. Rowley. 2002. Introduction. J. A. C. Baum, ed.Companion to Organizations. Sage, London, 134.

Biggart, N. W., G. C. Hamilton. 1992. On the limits of a firm-basedtheory to explain business networks: The Western bias of neo-classical economics. N. Nohria, R. G. Eccles, eds. Networks andOrganizations: Structure, Form, and Action. Harvard BusinessSchool Press, Boston, 471490.

Burt, R. S. 1992. Structural Holes. Harvard University Press,Cambridge, MA.

Campbell, J. L. 2004. Where do we stand? Common mechanismsin organizations and social movements research. G. F. Davis,D. McAdam, W. R. Scott, M. N. Zald, eds. Social Movementsand Organization Theory. Cambridge University Press, NewYork. Forthcoming.

Carroll, G. R. 1985. Concentration and specialization: Dynamics ofniche width in populations of organizations. Amer. J. Sociology90 12621283.

Clarke, D. C. 2003. How do we know when an enterprise exists?Unanswerable questions and legal polycentricity in China.Working paper, University of Washington School of Law,Seattle, WA.

Coase, R. H. 1937. The nature of the firm. Economica 4(16) 386405.

Coleman, J. S. 1964. Introduction to Mathematical Sociology. FreePress, New York.

Cook, T. D., D. T. Campbell, L. Peracchio. 1990. Quasi experimen-tation. M. D. Dunnette, L. M. Hough, eds. Handbook of Indus-trial & Organizational Psychology, 2nd ed., Vol. 1. ConsultingPsychologists Press, Palo Alto, CA, 491576.

DAunno, T., M. Succi, J. A. Alexander. 2000. The role of insti-tutional and market forces in divergent organizational change.Admin. Sci. Quart. 45 679703.

Davis, G. F. 2004. Firms and environments. N. Smelser, R. Swedberg,eds. Handbook of Economic Sociology, 2nd ed. Russell Sage/Princeton University Press, Princeton, NJ. Forthcoming.

Davis, G. F., H. R. Greve. 1997. Corporate elite networks and gover-nance changes in the 1980s. Amer. J. Sociology 103 137.

Davis, G. F., C. Marquis. 2004. The globalization of stock marketsand convergence in corporate governance. R. Swedberg, V. Nee,eds. The Economic Sociology of Capitalism. Princeton Univer-sity Press, Princeton, NJ. Forthcoming.


11/12


Davis, G. F., M. Useem. 2002. Top management, company directors,and corporate control. A. Pettigrew, H. Thomas, R. Whittington,eds. Handbook of Strategy and Management. Sage, London, UK,233259.

DiMaggio, P. J., W. W. Powell. 1983. The iron cage revisited: Insti-tutional isomorphism and collective rationality in organizationalfields. Amer. Sociological Rev. 48 147160.

Dobbin, F., J. R. Sutton. 1998. The strength of a weak state: Therights revolution and the rise of human resources managementdivisions. Amer. J. Sociology 104 441476.

Edelman, L. B. 1992. Legal ambiguity and symbolic structures: Orga-nizational mediation of civil rights law. Amer. J. Sociology 97

15311576.

Elster, J. 1998. A plea for mechanisms. P. Hedstrom, R. Swedberg,eds. Social Mechanisms: An Analytical Approach to SocialTheory. Cambridge University Press, New York, 4573.

Granovetter, M. 1994. Business groups. N. J. Smelser, R. Swedberg,eds. Handbook of Economic Sociology. Princeton UniversityPress, Princeton, NJ, 453475.

Hannan, M. T., J. Freeman. 1977. The population ecology of organi-zations. Amer. J. Sociology 82 929964.

Harvey, D. 1990. The Condition of Postmodernity: An Enquiry intothe Origins of Cultural Change. Blackwell, Cambridge, MA.

Hedstrom, P., R. Swedberg. 1998. Social Mechanisms: An Analyt-ical Approach to Social Theory. Cambridge University Press,New York.

Hernes, G. 1998. Real virtuality. P. Hedstrom, R. Swedberg, eds.Social Mechanisms: An Analytical Approach to Social Theory.Cambridge University Press, New York, 74101.

Hoffman, A. J. 1999. Institutional evolution and change: Environmen-talism and the US chemical industry. Acad. Management J. 42351371.

Hoffman, A. J. 2001a. From Heresy to Dogma: An InstitutionalHistory of Corporate Environmentalism (expanded edition).

Stanford University Press, Stanford, CA.

Hoffman, A. J. 2001b. Linking organizational and field-level analy-ses: The diffusion of corporate environmental practice. Organ.Environment 14 133156.

Jensen, M. C., W. H. Meckling. 1976. Theory of the firm: Manage-rial behavior, agency cost, and ownership structure. J. FinancialEconom. 3 305360.

Lawrence, P. R., J. W. Lorsch. 1967. Differentiation and integrationin complex organizations. Admin. Sci. Quart. 12 147.

Lazerson, M. 1995. A new phoenix? Modern putting-out in theModena knitwear industry. Admin. Sci. Quart. 40 3459.

Lounsbury, M. 2001. Institutional sources of practice variation:Staffing college and university recycling programs. Admin. Sci.

Quart. 46 2956.Lounsbury, M., M. J. Ventresca. 2002. Social structure and organi-

zations revisited. M. Lounsbury, M. J. Ventresca, eds. SocialStructure and Organizations Revisited: Research in the Sociol-

ogy of Organizations 19 336.

March, J. G., H. A. Simon. 1958. Organizations. Wiley, New York.

Marquis, C. 2003. The pressure of the past: Network imprinting inintercorporate communities. Admin. Sci. Quart. 48 655689.

Martin, J. L. 2003. What is field theory? Amer. J. Sociology 109 149.

McAdam, D., S. Tarrow, C. Tilly. 2001. Dynamics of Contention.Cambridge University Press, New York.

McAdam, D., W. R. Scott. 2004. Organizations and movements.G. F. Davis, D. McAdam, W. R. Scott, M. N. Zald, eds. SocialMovements and Organization Theory. Cambridge UniversityPress, New York. Forthcoming.

Meyer, J. W., B. Rowan. 1977. Institutionalized organizations: Formalstructure as myth and ceremony. Amer. J. Sociology 83 4162.

Mills, C. W. 1959. The Sociological Imagination. Oxford, New York.

North, D. C. 1990. Institutions, Institutional Change and EconomicPerformance. Cambridge University Press, Cambridge, UK.

Perrow, C. 1991. A society of organizations. Theory Soc. 20 725762.

Pfeffer, J. 1993. Barriers to the advance of organizational science:Paradigm development as a dependent variable. Acad. Manage-ment Rev. 18 599620.

Pfeffer, J., G. R. Salancik. 1978. The External Control of Organi-zations: A Resource Dependence Perspective. Harper & Row,New York.

Piore, M. J., C. F. Sabel. 1984. The Second Industrial Divide. Basic,New York.

Robbins, G. E. 2002. Titles and Tasks: New Jobs for New Mediain Silicon Alley? Unpublished doctoral dissertation, Columbia

University, New York.

Sabel, C. F., J. Zeitlin. 1996. Stories, strategies, structures: Rethink-ing historical alternatives to mass production. C. F. Sabel,J. Zeitlin, eds. Worlds of Possibility: Flexibility and MassProduction in Western Industrialization. Cambridge UniversityPress, Cambridge, UK.

Saxenian, A. 1994. Regional Advantage: Culture and Competitionin Silicon Valley and Route 128. Harvard University Press,Cambridge, MA.

Scott, W. R. 1995. Institutions and Organizations. Sage, ThousandOaks, CA.

Scott, W. R., M. Ruef, P. J. Mendel, C. A. Caronna. 2000. InstitutionalChange and Healthcare Organizations. University of Chicago

Press, Chicago, IL.Singh, J. V., R. J. House, D. J. Tucker. 1986. Organizational change

and organizational mortality. Admin. Sci. Quart. 31 587611.

Sorenson, A. B. 1998. Theoretical mechanisms and the empiricalstudy of social processes. P. Hedstrom, R. Swedberg, eds.Social Mechanisms: An Analytical Approach to Social Theory.Cambridge University Press, New York, 238266.

Stinchcombe, A. L. 1965. Social structure and organizations.J. G. March, ed. Handbook of Organizations. Rand McNally,Chicago, IL, 142193.

Stinchcombe, A. L. 1968. Constructing Social Theories. Universityof Chicago Press, Chicago, IL.

Stinchcombe, A. L. 2002. New sociological microfoundations fororganizational theory: A postscript. M. Lounsbury, M. J.Ventresca, eds. Social Structure and Organizations Revisited:Research in the Sociology of Organizations 19 415433.

Stokes, D. E. 1997. Pasteurs Quadrant: Basic Science and Tech-nological Innovation. Brookings Institution Press, Washington,D.C.

Strang, D., S. A. Soule. 1998. Diffusion in organizations and socialmovements: From hybrid corn to poison pills. Annual Rev. Soci-ology 24 265290.

Sutton, J. R., F. Dobbin. 1996. The two faces of governance:Responses to legal uncertainty in U.S. firms, 1955 to 1985.Amer. Sociological Rev. 61 794811.


12/12


Thompson, J. D. 1967. Organizations in Action. McGraw Hill,New York.

Thornton, P. H., W. Ocasio. 1999. Institutional logics and the histori-cal contingency of power in organizations: Executive succession

in the higher education publishing industry, 19581990. Amer.J. Sociology 105 801843.

Williamson, O. E. 1975. Markets and Hierarchies: Analysis andAntitrust Implications. Free Press, New York.

Documents

Davis Marquis 05 OS