^ Academy of Managemenl Review. 1989, Vol. 14, No. 4, 486-489.

Nothing Is Quite So Practicalas a Good Theory

ANDREW H. VAN DE VENUniversity of Minnesota

A central mission of scholars and educators inprofessional schools of management, health,education, and social work is to conduct re-search that contributes knowledge to a scientificdiscipline, on the one hand, and to apply thatknowledge to the practice of management as aprofession, on the other (Simon, 1967). To do thiswell, we need to design our research so that itprovides an intimate understanding of the prac-tical problems facing the profession. Equally im-portant, we need to appreciate and strengthenour skills in developing good theory so that re-search conducted about these problems will ad-vance the knowledge that is relevant to both thediscipline and the profession. Lewin's (1945)statement that "nothing is so practical as a goodtheory" captures a theme that is as importanttoday as it was in Lewin's time. Good theory ispractical precisely because it advances knowl-edge in a scientific discipline, guides researchtoward crucial questions, and enlightens theprofession of management. This special forumfocuses on criteria and methods for buildinggood theory. Its purpose is to suggest ways tostrengthen our theory-development capabilities,and thereby link better theory with the disci-plines and professions represented in the Acad-emy.

Overview

This forum consists of six papers and an edi-tor's commentary that address three related as-pects of our theme: (a) the characteristics of goodtheory, (b) the process of building good theories.

and (c) the methods to improve our current the-orizing.

1. What are the characteristics of good theory?For example, what is required to have a goodtheory of organizational learning, selection, orchange? Although most of us can readily pointto an example of a good theory, we are hard-pressed to systematically articulate how andwhy a theory is good or better than an alterna-tive theory. The advancement of theory in theAcademy requires that far more attention begiven to communicating and illustrating the cri-teria or characteristics of good theories on a va-riety of organization and management topics.

In his capacity as Editor of AMR, David Whet-ten struggles every day with this challenge ofcommunicating theory-building principles insimple and practical ways both with authorsand reviewers. I am delighted that he acceptedmy invitation to write a reflective editorial onwhat constitutes a publishable theoretical con-tribution. Using the basic questions and practi-cal style of a journalist, David Whetten suggeststhat the essential ingredients of a value-addedtheoretical contribution are explicit treatmentsof: Who? What? Where? When? Why? andHow?—and the greatest of these is Why.

In a similar but in-depth vein, SamuelBacharach sets forth the basic vocabulary andground rules for defining and evaluating orga-nizational theories. By discussing the falsifiabil-ity and utility criteria often used to evaluate con-structs, variables, and relationships of theories,his paper sensitizes readers, one more time, tothe ground assumptions that most social scien-

486

tists have taken for granted. Unfortunately,these ground assumptions are too often ignoredor dismissed without adequate justification, re-sulting in theories built on "shifting sand."

2. How might one build a good theory? Fewanswers to this question have been offered thatextend beyond those provided in standardmethodology textbooks (e.g., Dubin, 1969; Kap-lan, 1964; Kerlinger, 1973; Stinchcombe, 1968).Yet, many scholars report either that these stan-dard theory-building guidelines do not apply tomany topical areas in management, or that theguidelines are too standardized and formalizedto accurately reflect their theory-building expe-riences. Advancements in the process of theorybuilding are needed that not only address thegap between espoused and used methods butalso provide valid and practical ways to buildgood theories.

This forum contains three papers that makesuch advancements. First, Karl Weick providesa refreshing alternative to orthodox theory con-struction, which he indicates many times resultsin trivial theories because of its emphasis on val-idation rather than usefulness to judge the plau-sibility of a theory. He proposes that theorizingcan be substantially improved if we adopt prin-ciples of disciplined imagination and view it asan evolutionary process of artificial selection. Inso doing, Weick enlarges the relatively narrowbaseline criteria reviewed by Bacharach thatgenerally are used to evaluate theory.

Whereas Weick emphasizes vicarious experi-ences and independent thought trials as majorsources for theory-building ideas, KathleenEisenhardt provides a roadmap for building the-ory from case study research. By discussing con-crete steps in conducting and analyzing casestudies, Eisenhardt provides a useful inductivestrategy for building theories that are novel, test-able, and especially appropriate for new topicareas. She also suggests some guidelines forhow to evaluate such efforts and how to linkthem into existing literature.

However, a nagging question remains: To

what extent can knowledge acquired throughcase studies and other intensive idiographicmethods be regarded as valid? Haridimos Tsou-kas argues that idiographic studies are veryuseful for producing valid knowledge when theyare concerned with the generative mechanismsand the contingent factors that are responsiblefor observed patterns. Adopting a realist per-spective, Tsoukas distinguishes between (a) theunderlying generative mechanisms or laws thathave the power to cause events to happen in thereal world, (b) the particular circumstances orcontingencies when these causal mechanismsoperate, and (c) the empirical events that peopleexperience and researchers observe. Validknowledge is produced by inferring and ex-plaining what causal mechanisms operate—mparticular circumstances—to explain the empir-ical events that were observed to occur. Tsoukasimportantly cautions that theoretical explana-tions are inadequate when they focus on the em-pirical domain only by examining associationsbetween observed organizational characteris-tics (as often produced by correlational studies),or by surface-level "detective work" of conjunc-tions among an observed sequence of events.Good theory goes beyond establishing empiri-cally observed patterns, that is, it tries to explainwhat caused them.

3. How can we improve our current theories?Although the papers introduced thus far provideuseful criteria and methods for developing newtheories, the fact remains that most of us are"stuck" with the concepts and theories in whichwe have been trained and socialized. Given thepluralistic nature of the Academy, we now havemany theories competing with each other to ex-plain a given phenomenon. Proponents for eachtheory engage in activities to make their theorybetter by increasing its internal consistency, of-ten at the expense of limiting its scope. As aresult, and as Pogge stated, a way of seeing is away of not seeing. From an overall Academyperspective, such impeccable micro logic is cre-ating macro nonsense!

487

How should we deal with the tensions, de-bates, and forced choices between overly nar-row competing theories to address an issue? Forexample, in the a rea of organizat ional-environment relations, how can we reconcilethe competing assumptions, analyses, and con-clusions between population ecology and stra-tegic choice theories? These questions call forscholars and practitioners to pay more attentionto methods for diagnosing situations, to selectand improve relevant theories, and to becomefacile in constructively using the tensions thatexist between alternative theories to addressthose situations.

In particular, Scott Poole and I argue that thetensions, inconsistencies, and contradictions be-tween theories provide important opportunitiesto develop better and more encompassing theo-ries. Thus, instead of suppressing or dismissingthese apparent paradoxes, either within or be-tween theories, we propose four ways to con-sciously and tenaciously pursue them to im-prove our theories: (a) accept the paradox andlearn to live with it constructively; (b) clarify lev-els of reference (e.g., part-whole, micro-macro,or individual-society) and the connectionsamong them; (c) take time into account in explor-ing when contrary assumptions or processeseach exert a separate influence; and (d) intro-duce new concepts which either correct flaws inlogic or provide a more encompassing perspec-tive that dissolves the paradox. These four meth-ods expand on Weick's recommendation toadopt multiple independent thought trials to im-prove our theorizing.

The four ways to address inconsistencieswithin or between theories can broaden ourtheoretical concepts and extend their range ofapplications. But, as Chimezie Osigweh warns,this reconceptualization process can easily andunwittingly go awry as concepts are malformedand result in misinformation. Osigweh ad-dresses the problem of maximizing the potentialfor concept travelling (fitting precisely a varietyof applications), while simultaneously minimiz-

ing errors of concept stretching (broadening themeaning beyond reason). To deal with thisproblem, Osigweh proposes a negation ap-proach for defining and moving concepts acrosslevels of abstraction (i.e., from being situationalconcepts to being generalizable universals, andvice versa).

Conclusion

The papers in this forum represent the tip of aniceberg of interest and effort in theory buildingwithin the Academy. In addition to countless in-quiries and expressions of interest, over 45 pa-pers were submitted in response to AMR's Callfor Papers. All papers were evaluated using thestandard AMR review process. The subject mat-ter of the papers also was screened, and papersthat deal with topics beyond the scope of thistheory-building forum were evaluated for pub-lication in a regular issue of AMR. Finally, a fewpapers deemed appropriate for this forum couldnot be completed before the deadline, and theywill appear in future issues of AMR. Thus, thisforum is an interim, not final or complete, state-ment on theory building. It represents AMR's on-going commitment to strengthening theory-development activity and its continuing searchfor papers that will contribute to this commit-ment. Only through a never-ending pursuit ofthis commitment will we apprec ia te that"nothing is quite so practical as a good theory."

Einally, I must recognize and applaud thededication and thoroughness of AMR's editorialreview board and the anonymous reviewers ofthese papers. As guest editor, I have had anopportunity to observe directly the inner work-ings of AMR's anonymous editorial review pro-cess. This experience has truly impressed uponme the high quality, penetrating insight, andconstructive suggestions that anonymous re-viewers provide authors of prospective AMR pa-pers. As might be expected, the independentassessments of these heterogeneously selectedreviewers often are in disagreement, and thisresults in the editor's having to make challeng-

488

ing judgment calls. However, such feedback theory building. Thus, although no institution isprovides authors with a rich array of indepen- ideal, I conclude that the anonymous paper re-dent interpretations about their papers that, as view process is alive and well for stimulatingWeick suggests, should substantially improve good theory in AMR.

ReferencesDubin, R. (1969) Theory building. New York: Free Press. Massachusetts Institute of Technology. Sociometry, 8, 126-

135.Kaplan, A. (1964) The conduct of inquiry. San Francisco:

Chandler. Simon, H. A. (1967) The business school: A problem in orga-nizational design, journal of Management Studies, 4, 1 -

Kerlinger, F. N. (1973) Foundations of behavioral research 16.(2nd ed.). New York: Holt, Rinehart & Winston.

Stinchcombe, A. (1968) Consfrucfing social theories. NewLewin, K. (1945) The research center for group dynamics at York: Harcourt, Brace & World.

489