Sociological Forum, VoL 11, No. 2, 1996
The Prominence of Formal Theory in Sociology David Wilier 1,2
In sociology, work in formal theory is more extensive and intensive today than at any time in the past, but remains invisible to some, including S. Zhao [(1996) "The Beginning of the End or the End of the Beginning: The Theory Construction Movement Revisited," Sociological Forum, Vol. 11, No. 2, pp. 305-318]. I trace the growth of formal theory in sociology focusing on theoretical research programs and show that, whereas formal theory came after the books on theory construction, it was not based on them. Elementary Theory's research program provides an example of that discontinuity. Some misunderstandings concerning formal theory can be traced to the Maryland Theory Conference of 1990 and a critique of selected ideas of that Conference is offered. KEY WORDS: formal theory; theory growth; theory programs; elementary theory.
INTRODUCTION "Rick, what in heaven's name brought you to Casablanca?" "My health. I came to Casablanca for the waters." "Waters? What waters? We are in the desert." "I was misinformed." Claude Rains and Humphrey Bogart in Casablanca
The leading formal theories of sociology are today broader in scope, better constructed, and more fully tested than at any previous time (Berger and Zelditch, 1993a). Those theories and their tests are published in leadResearch leading to developments of Elementary Theory mentioned in this paper were supported by grants from The National Science Foundation. I thank Bernard Cohen and Murray Webster for comments on an earlier version of this manuscript. lDepartment of Sociology, University of South Carolina, Columbia, South Carolina 29208. 2To whom correspondence should be addressed. 319 0884-8971/96/0600-0319509.50/09 1996PlenumPublishingCorporation
320
Wilier
ing sociology journals (cf. Lovaglia et al., 1995; Fisek et al., 1995; Markovsky et al., 1993; Skvoretz and Wilier, 1993; Berger et al., 1992). And the scientists who build and test them routinely give presentations at the meetings of the American Sociological Association, their own meetings, and other prominent venues. Yet Zhao asserts that "the formal mode of theory construction is dead now" (Zhao, 1996:305). My purpose here is not to give evidence that there is a substantial and growing body of formal theory work: proof of that assertion is found in leading journals as referenced above and elsewhere in this paper. Instead, I seek to understand how Zhao has been misinformed. Since Zhao is not alone in making this error, the issues I address extend beyond his paper to a larger literature focusing on the Maryland Theory Conference of 1990 sponsored by the National Science Foundation. This paper has three main sections. In the first, I show that the series of books on theory construction that spans the 1960s and into the 1970s does not constitute a "theory construction movement." Then I turn to the claim that formal theory is dead and show that what Zhao interprets as a termination of the drive toward formal theory is, instead, a discontinuity. The theory construction books came before the development of formal theory, but formal theory did not rest on their foundations. I trace this discontinuity to three problems that the books on theory construction were not able to overcome. The second section shows why Zhao's measure for the life and death of formal theory necessarily misinforms him. To do so I note the steps through which theory develops beginning with methods of theory construction and extending through cycles of testing and theory extension that are my daily work and the daily work of others who are engaged in cumulative theory development. This work is the kind that Lakatos called "theoretical research programmes" (1970 and 1978). Though Lakatos bases his ideas of how theory develops exclusively on physical, chemical, and biological sciences, the processes in the development of sociological theory that I trace are the same as those he finds for other sciences. The third section critiques misinformation from Maryland Theory Conference of 1990 that Zhao highlights in his introduction. At the conference only one of the presenters, Bernard P. Cohen, and I, as a discussant, knew from firsthand experience the then current state of formal theory. Even though Cohen traced important recent developments in formal theory, the other papers, astonshingly, offered reasons for formal theory's death. Hoping to inform the other presenters of their errors prior to publication, both Cohen and I cited extensively from the literature of formal theory. Whereas in science it is expected that evidence will be carefully weighed, when pub-
Prominence of Formal Theory
321
lished the conference papers were effectively unchanged (Hage, 1994a). Errors in Zhao reflect errors of the Maryland Conference and I am pleased to have this opportunity to confront them. In fact, some of this section is based on my discussion of the papers with which the conference was closed, but I also take this opportunity to correct Zhao's misinterpretation of my earlier work.
THE DISCONTINUITY Thirty years ago there were no successful theoretical research programs in sociology, but writings on how to construct theory were beginning to appear. These writings were scattered and, contrary to Zhao, there was no theory construction movement in sociology, at least as the term "movement" is normally used. Acts qualify as a "movement" only if there is some minimal amount of cooperation linking them. Mere temporal contiguity is not enough. That the authors of books on theory construction mentioned by Zhao, Zetterberg (1954), Wilier (1967), Stinchcombe (1968), Blalock (1969), Dubin (1969), Mullins (1971), Reynolds (1971), Wallace (1971), Gibbs (1972), and Hage (1972) worked independently from each other is immediately obvious to anyone who would examine the indexes of their books. 3 For example, Stinchcombe, Dubin, Reynolds, and Wallace do not reference my 1967 book. Dubin does not mention Stinchcombe. Reynolds refers to Stinchcombe once in a footnote but not at all to Dubin. Wallace mentions Stinchcombe and Dubin but only in scattered footnotes. More generally, no later work examines any earlier work in detail, and none bases a substantial part of its plan for theory building on the ideas of others. Today formal theorists hold regular meetings, but authors of theory construction books held no meetings.4 Many of the authors never met. 5 There was a collection of works on theory construction, but the collection was no movement. Furthermore, there is a discontinuity between the methods proposed in theory construction books and methods actually used when formal theo3Zhao lists 12 books, but Merton ([1949/1968) wrote much earlier and Glaser and Strauss 1.(1967) did not have formal theory as a goal. oday meetings for Theory and Research on Group Processes are held annually just before (or after) the American Sociological Association Meetings and are open to all interested in formal theory. Formal theorists also meet annually at the Sunbelt Social Networks Conference and other venues. 5I base this assertion, in part, on my own experience. Though I meet regularly with others working on formal theories today, previously I encountered only 2 of the other 9 authors of theory construction books and then only in passing. By contrast Bernard Cohen met all ten, but only he and I exchanged ideas. (Personal communication).
322
Wilier
des were later constructed. Evidence for this discontinuity is in Berger and Zelditch (1993a) where the histories and developments of leading theoretical research programmes in sociology are given by those centrally engaged. 6 None of the theoretical research programs mention or use ideas from any of the 10 theory construction books, with one exception. Blalock (1969) is referenced and then only once for a brief critical comment (Wilier and Markovsky, 1993). The discontinuity can be traced to three problems that the theory construction books failed to overcome. First, all of the books shared with mainstream sociology a fundamental confusion between statistical procedures of empirical generalization on the one hand and the methods of theoretical science on the other. In Systematic Empiricism: Critique of a PseudoScience (Wilier and Wilier, 1973), the methods of sociology were traced from the British empiricist J. S. Mill (1843/1967) through Pearson (1892/1957) of the product-moment correlation and Fisher (1935) of the t-test based experiment to their introduction into sociology. To explain in brief: Mill, Pearson, and Fisher conflated empirical generalizations, which are always specific to time and place (such as the correlation between amount of fertilization and size of yield), with scientific laws (as with F = MA), which are universal. Pearson was highly critical of physicists for using equals signs in laws, which he felt was an idealization. He believed all scientific laws to be probabilistic and called for physicists to change their methods to the ones he developed. Fisher asserted that experimentation in his area, agricultural extension work, was more advanced than any other science. Unlike Galileo, Newton, and Maxwell, no theory testing occurred in Fisher's experiments. In fact, Mill, Fisher, and Pearson evidenced no understanding of the then older theoretic methods of the sciences and, as a consequence, invented a new methodology of systematic empiricism. When introduced by Stouffer (1950) to become the mainstream methods of sociology, systematic empiricism was generally believed to be science. That belief persists. It was the failure of systematic empiricism to produce theoretic science that called the theory construction books into being (Wilier and Wilier, 1973). 7 And it is because the methods proposed in those books failed to transcend the limitations of systematic empiricism that they did not become the basis for formal theory when it did develop. 6Here and throughout, a theoretical research programme is a theory and the research supporting it. 7Zhao (1996:307) asserts that it was the failure of the "Parsonsian paradigm" that originated the theory construction movement. He is correct that Merton's call for "middle range theory" as coupled to a critique of "grand theory" was influential.
Prominence of Formal Theory
323
Second, the authors of the theory construction books failed to present their readers with the methods of science. Instead, a philosophy of science was presented as if it were science's methodology. The reason for this conflation is that, unlike sociology, which has an abundance of methods books, there are no books on the methods of theoretical sciences. The sciences neither have nor need methods books because their method is theory. This error might have been self-correcting had philosophy of science presented an accurate picture of how science works, but as shown in Systematic Empiricism: Critique of a PseudoScience (Wilier and Wilier, 1973), the picture was distorted. The then dominant school of philosophy of science was logical positivism, the central concern of which was to draw a line of demarcation between science and religion. Their central proposition, which Hempel (1965) later called the "narrower thesis of empiricism," was the assertion the concepts and propositions of science could be reduced to observational statements. While that thesis placed theology on the outside, Carnap discovered that it also ruled theory out of science (1936/1937). Only after publication of the last theory construction book was it recognized that the portrayal of science by logical positivism, including the early Popper ([1935] 1959), was inadequate (Suppe, 1974). When introduced into sociology in the theory construction books, that particular philosophy of science offered an empiricist interpretation of science that was not suited to overcoming the limitations of systematic empiricism, s Third, authors of the theory construction books tried to develop a content free methodology for building theory, but the form and functioning of theory proved too difficult to separate from its content. Today it is easy to explain how formal social theory is developed and applied. One simply selects a theory such as Expectations States or Elementary Theory, traces its basic ideas and explains how it has been applied. In the 1960s, either such theories did not yet exist or they were very new--embryonic, and their practitioners were just mastering their uses. So the authors of the theory construction books had no specifically sociological examples to use which were not pretheoretic and thus misleading. Furthermore, since none of the authors actually had seen a social theory of the kind they wished to promote and none had the opportunity to work with one, none of the authors was actually qualified to write about the development of formal social theory. 9 8philosophy of science may help in theory construction and its exposition, but should never be mistaken as a methodology of science. My experience suggests that the history of science offers more to the formal theorist in sociology than does the philosophy of science. 9I mean none of the authors were qualified, including myself. By contrast, only a decade later ideas proven in theory formulation and testing were central in Cohen (1980).
324
Wilier
So the 10 books on theory construction proved to be a dead end. The result was the discontinuity between the theory construction books and the later development of formal theory, a discontinuity in which there was an almost complete turnover of the people working on the problem of building formal theory. In fact, it proved easier to develop formal theories in sociology than to set down rules of how to do so prior to their development. But I want to emphasize that dead ends are not failures. Science advances by trying many paths and, frequently, the right path is found only by elimination.
GROWTH OF A PROGRAM
According to Zhao, after a first "surge of interest" in theory construction it "disappeared completely": Zhao quotes Hage (1994b) that there have been no "new articles or books on the topic" (1996:306). Presuming that the absence of books and articles on "theory construction" meant the death of "formal theory," Zhao offers a periodization of the "movement" to show "how it was modified and abandoned by the authors" (1996: 306). Since theory construction was not a movement and since formal theory work has not been abandoned, I will not review Zhao's periods. Instead, I briefly trace what actually happened in the case of one program of theory development. After publication of my 1967 book, I employed its theory construction methods in research and found that they did not result in scientific theory. My previous experience in physics meant that I would recognize scientific theory when encountered. The problem of how to build such theory remained, however. The failure of the 1967 ideas resulted in a reevaluation of the foundations of science in Systematic Empiricism, some of the conclusions of which were just mentioned. But Systematic Empiricism only criticized method: it did not propose a direction to take in building social theory. I learned how to build and use theory only by actually building, testing, and extending a particular theory called Elementary Theory. The basic structure of Elementary Theory was influenced by my earlier experience in physics. At the core of the theory is an interpreted geometry, a "modeling procedure," which is based on graph and network theory. The modeling procedure is used to represent social relations of actors connected by sanction flows. Connecting relations at actors' positions gives abstract representation for structures of social relations that can be compared to ongoing social structures or used to design experiments. The theory contains a rationality principle and a law for its interpretation and a resistance principle and a law interpreting resistance. The principles and laws are ap-
Prominence of Formal Theory
325
plied to models, and resulting dynamics are compared to empirical cases for prediction and explanation. These basic formulations are given in Wilier and Anderson (1981) together with institutional, experimental, and historical applications. The basic formulations are also given in Wilier (1987) together with a series of experiments on exchange and coercive structures supporting the theory. According to Lakatos referenced above, as theories develop they come into contact with other theories, and contradictory predictions lead to critical tests. By 1986, the concepts for structure in Elementary Theory had come in conflict with those of power-dependence theory (Cook et al., 1983). Power-dependence theory had developed a method for locating power positions called "Vulnerability." As shown in Wilier (1986), Vulnerability produces contradictory predictions for some structures. Subsequently, with the development of the Graph-theoretic Power Index, a critical test was published (Markovsky et al., 1988). Previously, research had focused only on exclusion as a structural power condition. A second condition, "inclusion," was presented and resistance-based predictions for it tested (Patton and Wilier, 1990). By 1992, weak power had been discovered: I edited a special edition of Social Networks in which three theories competing with Elementary Theory were published. The next year marked the first tests of Elementary Theory's predictions for weak power (Markovsky et al., 1993), tests among the four theories (Skvoretz and Wilier, 1993), and the first crossnational experimental test of any formal theory (Wilier and Szmatka, 1993). When Elementary Theory is used as a method, it is employed to plan research including measurement procedures and the design of experiments. It organizes investigations indicating what is best studied first, what second, and what to follow, and information from each investigation becomes cumulative knowledge by incorporation into the theory. The newly expanded theory is then used to find the next directions for study. Through cumulation, the scope of application of Elementary Theory has expanded such that an ever broader range of cases can be explained and predicted. Theory is the method of physics, chemistry, and biology, but today, theoretical research programs in sociology, such as Elementary Theory, evidence the qualities found in those sciences (Wilier, 1987; Berger and Zelditch, 1993b). The problem of constructing and using formal theory in sociology has been solved. Because it has been solved, theory construction books are no longer needed. Here it matters not that the problem was not solved by books proposing context-free methods, but, instead, by building specific theories such as Elementary Theory and Status Characteristics Theory. With the solution of the problem, the publication of theory construction books ended. Thus, to take the ongoing publication of new books in theory
Wilier
326
construction as the measure for the existence of formal theory, as did Zhao and Hage before him, is a major error.
MISINFORMATION OF THE MARYLAND THEORY
CONFERENCE Undoubtedly, Zhao would not have been mistaken about the development of formal theory in sociology had the papers published by the Maryland Theory Conference not also been in error. Some of the errors made there are difficult to understand, particularly those of Hage, the organizer of the conference. According to Hage, "[T]he failure to develop for-
mal sociological theory has emanated from the lack of collective effort to write sociology" (1994b:53, italics original). Here, two errors are conjoined: (1) that formal theory did not develop, and (2) that failure was due to the lack of cooperative efforts. Twenty-eight years before the Maryland Conference, Berger, Zelditch, and Anderson edited Sociological Theory in Progress (Vol. 1, 1966), and eight years after that Berger, Conner, and Fisek edited Expectations State Theory: A Theoretical Research Program (1974). That the two contained papers on Status Characteristics and Expectations States Theory by an array of authors clearly indicates extensive collective efforts. Further collective efforts are seen by the multiple authorship of many papers on that theory in leading sociology journals up to the time of the Maryland conference and beyond. I already have noted steps in the development of Elementary Theory, and references there also clearly indicate collective efforts. Work elsewhere may be individualistic and single authored, but that is not the rule in formal theory. 1~ Of the other errors made at the conference and subsequently published, two are particularly egregious: Blalock's claim that formal theory is impeded because our subject matter is too complex and Turner's assertion that scientific development is furthered by bureaucratic centralism. Since formal theory did develop in sociology and science based on it is ongoing, both claims are wrong. Yet comment is called for because both are frequently heard and both contain dangerous falsehoods. According to Blalock (1994), "social reality is highly complex" (121), and "since reality is indeed complex, so must be both our theories and our research" (122). Simple theories are only useful for polemics because they
l~ et al. (1993), while brief, has five authors. Upon its publication, Skvoretz, who is one of the authors, noted with some irony that, "Five authors? It must be science" (Personal Communication).
Prominence of Formal Theory
327
explain only a few percent of the variance. "[M]any of our problems stem from a tendency to strive toward theoretical parsimony" (128). Parsimony might be attained in other sciences, but cannot reasonably be expected in sociology. "In other sciences, parsimonious explanations may remain the ideal, but are not really expected. In sociology . . . . parsimony will probably have to be sacrificed" (129). Blalock is quite wrong. Parsimonious explanations are common to all theoretic sciences. In fact, any reality, social or otherwise, will necessarily appear complex when we do not have a theory with which to comprehend it. Put somewhat differently, as Friz Heider, the creator of balance theory, explained to me, "The purpose of science is to make the world simple" (personal communication). In science, theory simplifies. In fields such as sociology, where systematic empiricism is still a dominant method, empirical research never simplifies, for its aim is not to apply theory but to find findings, and findings are no more than lists of information particular to time and place. So, research in science simplifies, whereas research using systematic empiricism makes the world more complex. According to Turner (1994), "Mature and formal theorizing are the result of levels of organizational control that most sociologists would never be willing to accept or tolerate" (48). Science can only develop when the lay public and competing organizations are excluded from influencing knowledge and when all resources from prestige through research funding are centrally controlled (48-49). The resulting bureaucratic centrism will practice thought control to "[s]tandardize cognitive orientations and criteria for defining intellectual problems" (49). But bureaucratic centrism has not developed because its "conditions would be offensive to sociologists who, for all of their collectivist biases at the ideological level, are . . . 'rugged individualists'" (48). Thus, sociology is "the impossible science," and there is only "one social science that is organized to produce formal theory-economics." (51). Either economics is a science and bureaucratically centralized or Turner is wrong. This semester I am the visiting scholar in an economics department, so I asked about the system of "organizational control" that makes the science of economics possible, but no economist had encountered it. No one could identify the authority, which must exist, according to Turner, to concentrate resources and impose thought control. In fact, economists have a much wider array of sources of research support than do sociologists, and, having few "collectivist biases," they are quite individualistic. How is it possible that they have a science? Contrary to Turner, since science was first established during the enlightenment and throughout the conflicts with 20th-century totalitarian dictatorships, it is well known that scientists have stood for the free exchange
Wilier
328
of ideas. The opposition of scientists to bureaucratic centrism and the control of ideas is frequently principled but is not entirely altruistic. In fact, the opposition to the control of ideas rests on interests growing from the enterprise of science itself. Without the free interchange of ideas, science cannot be self-correcting, and without self-correction, it cannot be progressive. Science and thought control through bureaucratic centrism are antithetical.
CONCLUDING REMARKS While Zhao bears no responsibility whatsoever for the faults of the Maryland Conference as published in Hage (1994a), he is responsible for uncritically taking its errors at face value while ignoring the excellent points concerning the growth of theory made by Cohen (1994) published in the same work. Cohen does not hide his disagreement with the others and, because his paper was included, there is a bibliography of works on developing and developed theories that Zhao does not appear to have consulted. Zhao is also responsible for misinterpretations of my work. Though there were many errors in my Scientific Sociology: Theory and Method, that scientific laws apply under conditions of nonlimited scope is not one of them. Though it should have been clear that nonlimited scope is a condition of laws in all sciences (Walker and Cohen, 1985), Zhao (1996) took it as a quality specific to sociology and portrayed its introduction as a "retreat from the pursuit of general laws" (311). He adds that, "by redefining the concept of universality, Wilier shifted the focus of interest in theory construction from generality to conditionality" (311) and concluded that "obtaining pure universality in sociological theorizing is highly unlikely" (314). Here, generality and universality are being conflated. All theories must be universal because they claim to apply to all cases within their scope without specific exceptions. This bold claim is necessary, for without it theory has no utility, being limited, like findings, to particular times and places. In fact, theories can be tested for universality and at least one such test has been carried out (cf. Willer and Szmatka, 1993). By contrast, the generality of theory refers to its scope of application--the variety of cases and conditions to which the theory applies. Interestingly, the scope of some well-known theories is (technically) zero. For example, there are no "perfect gasses," so the scope of the gas laws is nil. There are no markets with an unlimited number of wholly unrelated and completely rational actors, so the scope of neoclassical microeconomics is nil. Yet both theories are robust and give excellent predictions for a significant range of cases.
Prominence of Formal Theory
329
In fact, both theories are quite universal however constricted their scope. Certainly, the ideas of universality and generality are altogether distinct, and it is time that they were understood in sociology. Upon reading this paper, Bernard Cohen wrote that the current state of formal theory "is more varied, more complex, and more promising than even you suggest" (personal communication). In fact, structural social psychology is only one of the important streams of current formal theory work. Today, formal theory is found in areas as diverse as social networks and rational choice theory, while The Journal of Mathematical Sociology is now in its 20th year of publication. When I began work on my 1967 book I wrote that "There are no scientific laws of society" (vii). What was true in 1967 is not true now. Then, no one had constructed and successfully tested a formal sociological theory and only a handful were trying. Today, sociologists can deploy a number of powerful theories, and hundreds are working to extend them. What was unknown 30 years ago is the most dynamic part of sociology today: tomorrow formal theory may be its core.
REFERENCES Berger, Joseph, Morris Zelditch, Jr., and Bo Anderson 1966 Sociological Theories in Progress, Vol. 1. New York: Houghton Mifflin. Berger, Joseph, Thomas J. Conner, and M. Hamit Fisek 1974 Expectation States Theory: A Theoretical Research Program. Cambridge, MA: Winthrop Publishers. Berger, Joseph, Robert Z. Norman, James Balkwell, and Roy E Smith 1992 "Status inconsistency in task situations: A test of four status processing principles." American Sociological Review 57:843-855. Berger, Joseph, and Morris Zelditch, Jr. 1993a Theoretical Research Programs: Studies in Theory Growth. Stanford CA: Stanford University Press. 1993b " O r i e n t i n g strategies and theory growth." In J. Berger and M. Zelditch, Jr. (eds.), Theoretical Research Programs: Studies in Theory Growth: 319. Stanford CA: Stanford University Press. Blalock, Hubert M. 1969 Theory Construction: From Verbal to Mathematical Formulation. Englewood Cliffs, NJ: Prentice-Hall.
1994 "Why have we failed to systematize reality's complexities?" In Jerald Hage (ed.), Formal Theory in Sociology: 121-136. Albany: State University of New York Press. Carnap, Rudolf 1936/1937 "Testability and meaning." Philosophy of Science 3:419-71 and 4:140. Cohen, Bernard P. 1980 Developing Sociological Knowledge: Theory and Method. E n g l e w o o d Cliffs, NJ: Prentice-Hall. 1994 "Sociological theory: The half-full cup." In Jerald Hage (ed.), Formal Theory in Sociology: 66-89. Albany: State University of New York Press. Cook, Korea S., Richard M. Emerson, Mary R. Gillmore, and Toshio Yamagishi 1983 "The distribution of power in exchange networks: Theory and experimental results." American Journal of Sociology 89:275-305. Dubin, Robert 1969 Theory Building. New York: The Free Press.
330 Fisek, M. H., Joseph Berger, and Robert Norman 1995 "Evaluations, and the formation of expectations." American Journal of Sociology 101:721-746. Fisher, Ronald A. 1935 The Design of Experiments Edinburgh: Oliver and Boyd. Gibbs, Jack 1972 Sociological Theory Construction. Hinsdale, IL: Dryden Press. Glaser, Barney and Anselm Strauss 1967 The Discovery of Grounded Theory. New York: Aldine. Hage, Jerald 1972 Techniques and Problems of Theory Construction in Sociology. New York: Wiley. 1994a Formal Theory in Sociology. Albany: State University of New York Press. 1994b "Sociological theory: Complex, fragmented and politicized" In Jerald Hage (ed.), Formal Theory in Sociology: 52-65. Albany: State University of New York Press. Hempel, Carl G. 1965 Fundamentals of Concept Formation in Empirical Science. Chicago: University of Chicago Press. Lakatos, Imre 1970 "Falsification and the methodology of scientific research programmes." In Imre Lakatos and Alan Musgrave (eds.) Criticism and the Growth of Knowledge." 91-196. Cambridge: Cambridge University Press. 1978 The Methodology of Scientific Research Programme~ New York: Cambridge University Press. Lovaglia, Michael, John Skvoretz, David Wilier, and Barry Markovsky 1995 "Negotiated outcomes in social exchange networks." Social Forces 74:123-155. Markovsky, Barry, David Wilier, and Travis Patton 1988 "Power relations in exchange networks." American Sociological Review 53:220-236. Markovsky, Barry, John Skvoretz, David Wilier, Michael Lovaglia and Jeffrey Erger 1993 "The seeds of weak power: An extension of network exchange theory." *Original publication date.
Wilier American Sociological Review 58:197209. Melton, Robert K. [1949] 1968 "On sociological theories of the middle range." In Robert K. Merton. (ed.), Social Theory and Social Structure: 39-72. New York: Free Press. Mill, John S. 1967 A System of Logic. (1843") London: Longmans, Green and Co. Mnllins, Nicholas 1971 The Art of Theory Construction and Use. New York: Harper & Row. Patton, Travis, and David Wilier 1990 "Connection and power in centralized e x c h a n g e n e t w o r k s . " J o u r n a l of Mathematical Sociology 16:31-49. Pearson, Karl 1957 The Grammar of Science. (1892") New York: Meridian Books. Popper, Karl R. 1959 The Logic of Scientific Discovery. (1935") New York: Basic Books. Reynolds, Paul D. 1971 A Primer in Theory Construction. Indianapolis, IN: Bobbs-Merrill. Skvoretz, John and David Wilier 1993 "Exclusion and power: A test of four theories of power in exchange networks." American Sociological Review 58:801-818. Stinchcombe, Arthur 1968 Constructing Social Theory. New York: Harcourt Brace Jovanovich. Stouffer, Samuel 1950 "Some observations on study design." American Journal of Sociology 55: 355-361. Suppe, Frederick 1974 The Structure of Scientific Theories. Urbana: University of lllinois Press. Turner, Jonathan 1994 "The failure of sociology to institutionalize cumulative theorizing." In Jerald Hage (ed.), Formal Theory in Sociology: 41-51. Albany: State University of New York Press. Walker, Henry A., and Bernard P. Cohen 1985 "Scope statements: Imperatives for evaluating theory." American Sociological Review 50:288-301.
Prominence of Formal Theory Wallace, Walter 1971 The Logic of Science in Sociology. New York: Aldine. Wilier, David 1967 Scientific Sociology Theory and Method. Englewood Cliffs, NJ: Prentice-Hall. 1986 "Vulnerability and the location of power positions." American Journal of Sociology" 92:441-444. 1987 Theory and the Experimental Investigation of Social Structures. New York: Gordon & Breach. Wilier, David, and Bo Anderson, eds. 1981 Networks, Exchange and Coerciom New York: Elsevier/Greenwood. Wilier, David, and Barry Markovsky 1993 "Elementary theory: Its development and research program." In Joseph Berger and Morris Zelditch, Jr. (eds.), Theoretical Research Programs: Studies in Theory Growth." 323-363. Stanford CA: Stanford University Press.
331 Wilier, David, and Jacek Szmatka 1993 "Cross-national experimental investigations of elementary theory: Implications for the generality of the theory and the autonomy of social structure." In E. J. Lawler, B. Markovsky, K. Heimer, and J. O'Brien (eds.), Advances in Group Processes, Vol. 10:37-81. Greenwich CT: JAI Press. Wilier, David, and Judith Wilier 1973 Systematic Empiricism: Critique of a Pseudoscience. Englewood Cliffs N J: Prentice-Hall. Zetterberg, Hans L. 1954 On Theory and Verification in Sociology. Totowa, NJ: Bedminster Press. Zhao, Shanyang 1996 "The beginning of the end or the end of the beginning: The theory construction movement revisited." Sociological Forum, 11:305-318.