See Bottom of the Home page for links to additional documents
Beyond Sociology's Tower of Babel:
Reconstructing the Scientific Method
Bernard Phillips
In memory of C. Wright Mills
who gave us a vision for sociology's future
(New York: Aldine, 2001)
PREFACE
PART ONE THE SCIENTIFIC METHOD: BUREAUCRATIC AND INTERACTIVE PARADIGMS
Chapter 1 SOCIOLOGY AND THE SCIENTIFIC METHOD
Problems with Bureaucratic Science
Orientation to Problems
Moving up Language’s Ladder of Abstraction
Moving Down the Ladder of Abstraction
Integrating Knowledge
Worldviews
A Scientific Method for Sociology
Definition of the Problem
High Level of Abstraction
Low Level of Abstraction
Integration of Knowledge
Reflexive Analysis and Interactive Worldview
Chapter 2 CULTURAL AND SOCIOLOGICAL PARADIGMS
Cultural Paradigms and Worldviews
John Dewey
Thomas Kuhn
Harold Kincaid
Sociology’s Paradigm
David A. Snow
Thomas J. Scheff
PART TWO ILLUSTRATING THE WEB APPROACH TO THE SCIENTIFIC METHOD
Chapter 3 THE INVISIBLE CRISIS OF MODERN SOCIETY
Anomie
Alienation
Social Stratification
Relative Deprivation: A Missing Link
Chapter 4 ADDRESSING THE INVISIBLE CRISIS
Interaction and the Crisis: General
The Idea of Interaction
Interaction among a Web of Sociological Concepts
Interaction and Feedback Loops
Interaction and the Crisis: Specific
Revolutions
Satyagraha
The Individual and Awareness of Emotions
PART THREE SOME IMPLICATIONS FOR SOCIOLOGY
Chapter 5 REFLEXIVITY
Toward a Reflexive Sociology
Gouldner’s Vision
Carrying Forward Gouldner’s Vision
Social Technology for a Reflexive Sociology
Basic Research
The Reconstructed Method as a Tool for All Social Scientists
Some Programmatic Ideas for Using a Web Culture Lag Theory
The Iron Law of Oligarchy
The Pygmalion Effect
The Process of Secularization
The Unanticipated Consequences of Purposive Action
Increasing Divorce Rate
The Life Cycle of the Church
The Self-Fulfilling Prophecy
The Change in Emphasis from
Production to Consumption
The New Social Movements of the Twentieth Approach: Social Change Illustrated
Century
Applied Research
Teaching
A Negative Example: Consultants Using System Dynamics
Dewey, Freire, Illich, Pecotche and Gandhi
John Dewey
Paulo Freire
Ivan Illich
Carlos Bernardo Gonzalez Pecotche
Mohandas Gandhi
Toward Reflexive Teaching
Chapter 6 LANGUAGE AND EMOTIONS
Strengthening Linguistic Tools
Abstract Concepts
Causal-Loop Diagrams
Allegories for Understanding Language
Nineteen Eighty-Four
The Languages of Pao
Back to the Future
Part One The Scientific Method: Bureaucratic and Interactive Paradigms
Part Two Illustrating the Web Approach to the Scientific Method
Part Three Some Implications for Sociology
Some Concluding Remarks
PREFACE
C. Wright Mills, who turned me as a student at Columbia College in 1950 from medicine to sociology, embodied Auguste Comte’s Enlightenment dream as to the promise of sociology. If the biophysical sciences had created the basis for problem-solving technologies now known as the industrial revolution, surely the discipline of sociology could manage to do the same for the enormous social problems facing society. If the scientific method had been the chief instrument for transforming societies from the seventeenth century onward, why couldn’t that same instrument help us to direct that transformation toward ends that human beings desire? The attention given to Mills by the discipline, despite the fact that he was a loner who wrote mainly for a popular audience, is illustrated by the rating given to The Sociological Imagination in 1998 by the members of the International Sociological Association. It achieved the second highest rating among books published in the twentieth century considered to be the most influential for sociologists, preceded by Weber’s Economy and Society and followed by Merton’s Social Theory and Social Structure, Weber’s The Protestant Ethic and the Spirit of Capitalism, Berger’s and Luckman’s The Social Construction of Reality, Bourdieu’s Distinction, Elias’ The Civilizing Process, Habermas’ The Theory of Communicative Action, Parsons’ The Structure of Social Action and Goffman’s The Presentation of Self in Everyday Life.
Despite that rating along with the widespread lip service given to the idea of “the sociological imagination,” Mills’ methodological achievements in that book and elsewhere have been barely noticed relative to his contributions to political sociology. He gave us a vision of a sociology that would dare to define the most fundamental problems facing the human race as research problems. I recall a day in the Spring of 1958 when we were both on a plane to Champaign-Urbana where he was to deliver a lecture based on his book, The Causes of World War III. In addition to his defining basic research problems, the breadth of the approach he advised for all of us sociologists, and others as well, is suggested by this familiar quote:
. . .that imagination is the capacity to shift from one perspective to another-- from the political to the psychological; from examination of a single family to comparative assessments of the national budgets of the world. . . [1959: 7].
And there is also his reflexive orientation, illustrated in his Appendix--”On Intellectual Craftsmanship”-- with these words: “the most admirable thinkers within the scholarly community you have chosen to join do not split their work from their lives.” This was an orientation strengthened in 1970 by Gouldner’s call for a “reflexive sociology” in The Coming Crisis of Western Sociology.
What is the state of modern sociology with respect to Mills’ orientation to basic problems, to intellectual breadth and to a reflexive orientation? In the early 1940s while still a graduate student at the University of Wisconsin, Mills analyzed some fifty textbooks on social problems in order to learn about “the style of reflection and the social-historical basis of American sociology”:
The level of abstraction which characterizes these texts is so low that often they seem to be empirically confused for lack of abstraction to knit them together. They display bodies of meagerly connected facts, ranging from rape in rural districts to public housing, and intellectually sanction this low level of abstraction. . .Collecting and dealing in a fragmentary way with scattered problems and facts of milieux, these books are not focused on larger stratifications or upon structured wholes [1943: 166].
Mills’ criticism could just as easily apply to contemporary sociology. We now have some forty Sections of the American Sociological Association with their “meagerly connected facts.” And a basis for this state of affairs is still a “low level of abstraction,” given our orientation to Merton’s “theories of the middle range.” The problems we tend to define for study are no more fundamental than “rape in rural districts” and “public housing.” And despite Mills and Gouldner, our research is characterized by little reflexivity.
It appears that at this time in history we have experienced a century of sociology’s failures to achieve the rapid cumulative development sought by Comte, Mills, Gouldner and the rest of us, and this has been coupled with a century of awesome and apparently escalating problems in modern society. Given this state of affairs, it is no wonder that almost all of us have turned to “the falsification of memory” and “the technique of particularization,” procedures described by Vidich and Bensman in their 1960 study of Springdalers. We have largely succeeded in burying our Enlightenment dreams for the promise of sociology by immersing ourselves within highly specialized areas of study. We behave much like the Springdaler who “instead of entertaining the youthful dream of a 500-acre farm, entertains the plan to buy a home freezer by the fall.” And the result for us appears to be much like that for the Springdalers:
Because they do not recognize their defeat, they are not defeated. The compromises, the self-deception and the self-avoidance are mechanisms which work; for, in operating on the basis of contradictory, illogical and conflicting assumptions, they are able to cope in their day-to-day lives with their immediate problems in a way that permits some degree of satisfaction, recognition and achievement. . .[1960: 320].
Yet as we examine our situation in the 21st century, we contemporary sociologists find ourselves in a position to open up to those falsified memories and to challenge our techniques of particularization by building on Mills’ ideas, such as his emphasis on basic problems. For example, the very fact that modern problems appear to be increasing--such as the escalation of technologies for delivering weapons of mass destruction--makes it ever more difficult for us to continue to bury our heads in the sand. It is becoming more and more obvious to us sociologists that we have failed to give society the platform of knowledge it needs as a basis for constructing social technologies in all institutions which can confront our complex social problems. If we look outside the discipline we find little credibility given to us as scientists, And we are learning to see ever more clearly the contradiction between our scientific ideals and what we have in fact achieved within the discipline. Instead of sociological knowledge based on the full range of our findings, what we have are separate pieces of knowledge located within the diverse areas of the discipline. Instead of knowledge that is rapidly cumulating, we have fads and fashions in the ideas and terms we use with relatively little cumulative development. And we even have questions raised by some postmodernists as well as others as to the possibility of any “scientific method” that can be applied to human behavior.
In addition to this increasing attention to fundamental problems outside and inside of the discipline, we are now in a much better position to follow Mills’ lead of shuttling up and down language’s ladder of abstraction, giving us increasing ability to integrate our “bodies of meagerly connected facts, ranging from rape in rural districts to public housing.” For example, the very fragmentation of sociology into forty Sections has yielded greater understanding of the enormous complexity of human behavior, suggesting that pessimism and cynicism about the discipline’s achievements and potential is premature. Further, a new sociological understanding of the limitations of positivism’s quantitative one-sidedness opens up a much broader approach to the scientific method. Abstract concepts and theory have come to the fore as essential ingredients for developing a science of sociology, just as they have proved to be essential for the biophysical sciences. And the contemporary philosophy of science and social science points up alternatives to positivistic assumptions, such as the inevitability of being carried to truth by the scientific method, the importance of isolated hypotheses and the centrality of exact prediction. Instead, we emerge with a web orientation to the scientific method, where no proposition is seen in isolation from all others and where--following Mills--we shuttle up and down language’s ladder of abstraction.
It is also possible to go beyond Mills’ and Gouldner’s ideas about reflexivity. To illustrate, we might begin with Kuhn’s view of scientific revolutions as stemming from awareness of contradictions within an existing scientific paradigm that are resolved within an alternative one. By extending his idea of scientific paradigms to cultural paradigms, his thesis implies that cultural revolutions also require awareness of existing contradictions that are resolved within an alternative cultural paradigm. Reflexivity points us exactly in this direction, examining our work and life with an eye toward uncovering contradictions and coming up with an alternative cultural paradigm where they could be resolved. Such reflexivity need not be limited to a vague metaphor. Instead, the concept of “cultural paradigm” can become a highly systematic one when linked to other abstract concepts within the reconstructed scientific method suggested by Mills’ notion of shuttling up and down the ladder of abstraction. From this perspective, we can proceed systematically to question--reflexively--our taken-for-granted assumptions, following the direction suggested metaphorically by Gouldner:
The historical mission of a Reflexive Sociology. . .would be to transform the sociologist, to penetrate deeply into his daily life and work. . .and to raise the sociologist’s self-awareness to a new historical level [1970: 489].
The difficulties involved in reconstructing our present approach to the scientific method appear to be extraordinary, for that approach is nested within our cultural paradigm. Just as we have relatively isolated Sections within the American Sociological Association, so do we have institutions within society along with organizations within the economy and occupations within organizations which have little contact with one another. Granting that a reflexive approach would raise up this nested relationship to full view, we would then be presented with an enormous contradiction between scientific ideals and practices that are supported by our fundamental way of life. And just as in the case of the Springdalers, we would be presented with youthful ideals which we have failed to reach. Under these circumstances, what would prevent us from reverting to our present cultural paradigm which gives us procedures for the falsification of memory through techniques of particularization? By so doing, we would at least be able to obtain “some degree of recognition, satisfaction and achievement.” This is exactly the difficulty I am facing as author of this book. I am asking readers to raise to the surface taken-for-granted assumptions about the scientific method and to examine their departure from scientific ideals, a task that also will show the departure of our cultural paradigm from our basic cultural values.
Yet if we follow Kuhn’s argument further--an argument strengthened by much that sociologists have learned about society--we find that awareness of such contradictions can yield changes in a scientific paradigm. If a new paradigm is developed within which those contradictions are resolved, then we can expect a shift to that paradigm, and this is exactly the approach I take in the following chapters. I outline new scientific and cultural paradigms and sketch their potential for resolving contradictions within the old ones. I do this not because I believe that I have succeeded in demolishing our present scientific and cultural paradigms. Rather, I believe it essential for a reader to be presented with these alternatives in order to even consider the possibility of such drastic and all-encompassing changes. Indeed, in my own view I do no more than open a door to the possibility of such shifts in our approach to the scientific method as well as modern culture. Fortunately from my own perspective, I am not alone in this approach, for it is the basis for the work of a group made up largely of sociologists, the sociological imagination group, introduced in a web site at <www.uab.edu/ethicscenter/SI.html>. We ask other sociologists and social scientists to join us in our present efforts to test the utility of this approach to the scientific method. We feel that it is worthwhile to abandon our present degree of “recognition, satisfaction and achievement,” given the urgent problems in sociology and society.
In Part One, “The Scientific Method: Bureaucratic and Interactive Paradigms,” I present in Chapter 1 a direction for reconstructing the scientific method. It builds on C. Wright Mills’ work and is oriented to shuttling far up and down language’s ladder of abstraction, by contrast with “grand theory” or “abstracted empiricism” which emphasize the top or the bottom, respectively, of that ladder. His analysis includes his vision of the centrality of developing a sociological imagination, and here he points beyond sociology to society as a whole, just as his books were written for a popular as well as an academic audience. Chapter 2 uses Kuhn’s analysis as a basis for taking up our cultural paradigm as well as our sociological paradigm, with the latter nested within the former just as any given epistemology is nested within a metaphysics. This chapter carries further Chapter 1’s analysis of the contrast between bureaucratic and interactive cultural and scientific paradigms. We look to the ideas of John Dewey, Thomas Kuhn and Harold Kincaid for philosophical and historical insights. David Snow and Thomas Scheff --devoting particular attention to sociology’s methods and theory--give us further understanding of sociology’s present situation as well as its future possibilities. Overall, Part One presents an outline of a reconstructed scientific method, taking into account some of its implications for modern society.
Part Two, “Illustrating the Web Approach to the Scientific Method,” aims to exemplify both the substantive fruitfulness and the applied implications of that approach. Yet the focus of Part Two remains on the web approach to the scientific method, and not on the validity of the few substantive and applied illustrations presented. Chapter 3 centers on anomie, alienation, social stratification and relative deprivation, all aspects of one fundamental social problem which might be called “the invisible crisis of modern society.” The web approach to the scientific method may prove to yield understanding of very broad and basic problems, as well as those of limited scope, that encompass the full range of substantive questions within the discipline. Whereas Chapter 3 centers on the utility of the web approach for understanding problems, Chapter 4 focuses on its utility for moving toward solutions. There we begin with a general examination of what it would take to change both sociology’s research paradigm and the cultural paradigm within modern society. We follow this with specific analyses of three illustrations: revolutions in general, the Gandhian technique of satyagraha, and change in a two-person social relationship. Both parts of Chapter 4 are oriented to the problem of how social structures can be altered so as to yield a higher degree social interaction.
Part Three, “Some Implications,” looks to several implications of the foregoing chapters for the discipline of sociology. In Chapter 5 we center on Gouldner’s idea of a reflexive sociology and point toward procedures which would carry forward Gouldner’s vision. In particular, we take up a number of approaches to social change along with educational procedures. In both cases we examine the impact of reflexivity. Thus, for example, we look to what a reflexive approach would yield for culture lag theory. And we also carry forward the reflexive implications of some of the educational ideas of Dewey, Freire, Illich, Pecotche and Gandhi. Chapter 6 centers on language as a key thread tying together earlier chapters. In an initial section, “Strengthening Linguistic Tools,” we introduce additional concepts to help us integrate earlier material, and we carry further our understanding of causal-loop diagrams. Also, we summarize two allegories--Nineteen Eighty-Four and The Languages of Pao--for penetrating the nature of language. Then, in a second section, we apply these linguistic tools to earlier chapters, centering on the problem of emotional repression and the possibilities for emotional expression.
If the arguments in this book for the problematic nature of our present interpretation of the scientific method prove to be credible, and if the alternative interpretation sketched here proves to be fruitful, then the implications of those arguments and that interpretation extend far beyond the few that are examined in Part Three. For example, the conclusions drawn for every single social science study that has ever been conducted would be open to reinterpretation. This would result from past failures to take into account systematically the enormous complexity involved within any given instance of human behavior. Our present piecemeal and specialized approach assumes implicitly that the pieces of the human jig-saw puzzle can at some point be put together so as to yield a coherent picture of human behavior. Yet if each piece is itself deficient, then no coherent picture emerges when we attempt to put the pieces together. Shifting from this metaphor to research procedures presently in use, one example has to do with the impact of the investigator on the investigation at every stage of the research process. These “investigator effects” are not taken into account by our non-reflexive approach to the scientific method, and we continue to publish studies which almost invariably include no information about such effects despite occasional questions raised as to the unscientific nature of such practices. This is much like a trial where neither the prosecutor nor the defense attorney is allowed to cross-examine witnesses yet where we expect an accurate verdict.
Carrying this implication one step further, if all of our conclusions from the social sciences become suspect, then so does the worth of all of the actions based on those conclusions which we have performed as individuals and societies. This includes past and present decisions made within every one of our institutions, and this is not limited to the relatively few decisions based on proposals by social scientists. They extend to the subtle influences of the social sciences as a result of their location within our formal and informal educational systems at all levels. All of this is also implied by the limitations within the cultural paradigm of modern society, a paradigm that encompasses the scientific paradigm governing our research procedures. In one sense this far-reaching critique of modern society is not a new idea, since much of postmodernist literature suggests the importance of deconstructing present assumptions within all aspects of society. Yet what is new is the acceptance of such deconstruction coupled with an alternative approach to the scientific method that is optimistic in its assessment of human possibilities and that promises to resolve fundamental contradictions within our scientific and cultural paradigms.
That alternative approach also gives the social scientist a special role within contemporary society, as illustrated by Gouldner’s vision of the future of the social sciences:
. . .At decisive points the ordinary language and conventional understandings fail and must be transcended. It is essentially the task of the social sciences, more generally, to create new and “extraordinary” languages, to help men learn to speak them, and to mediate between the deficient understandings of ordinary language and the different and liberating perspectives of the extraordinary languages of social theory. . . .To say social theorists are concept-creators means that they are not merely in the knowledge-creating business, but also in the language-reform and language-creating business. In other words, they are from the beginning involved in creating a new culture [Gouldner, 1972: 16].
Following Gouldner’s vision, perhaps the twenty-first century will not come to be seen as the century which witnessed an acceleration of the catastrophes of the twentieth century. Instead, perhaps it will come to be seen as the age of the social sciences, where the Enlightenment visions of Comte, Mills, Gouldner and the rest of us for a society able to confront its fundamental problems became a reality for the first time in human history.
In a book where I attempt to build bridges linking sociological knowledge, everything I have ever read inside and outside of the discipline has influenced me. But in particular I am indebted to every individual cited in the following chapters for making this book possible. Earlier ideas from Marx, Simmel, Durkheim, Weber, Nietzsche and Korzybski have been particularly influential, as is the case for more recent ideas from Mills, Gouldner, Kuhn and Williams. I owe a great deal to Harold Kincaid, whose philosophical ideas helped to start me on this journey, to Tom Scheff, whose part/whole methodology and publications helped to broaden my orientation, to Richard Koffler for his faith in my ideas and to Louie Johnston for his unflagging enthusiasm. I want to thank those who read parts of this manuscript at one stage or another or who encouraged me to proceed, including Dave Asavanond, Steve Baran, Stu Bennett, Larry Busch, Lee Cass, David Christner, Hank Everett, Joe Feagin, Sandy Klein, Marty Kozloff, Jack Levin, Felice Levine, Tony Levy, John and Joanne Livingstone, Clem Malin, Marvin Nadel, David Phillips, Seymour and Phyllis Pustilnik, John Rice, Dave Stearns and Emek Tanay. And I have learned much from all those who are a part of the sociological imagination network, with special thanks to Dave Britt, Tom Conroy, Dick Edgar, John Hall, Matt Hoover, Joe Hopper, Chanoch Jacobsen, Jim Kimberly, Richard Lachmann, Lauren Langman, Donald Levine, Bronwen Lichtenstein, Dave Maines, John Malarkey, Stjepan Mestrovic, Alfonso Morales, Gil Musolf, Joe Perry, Gary Reed, Jay Weinstein, Doris Wilkinson and Andy Ziner.
PART ONE THE SCIENTIFIC METHOD:
BUREAUCRATIC AND INTERACTIVE PARADIGMS
It is in Genesis that God punishes the human race for attempting to build “a town and a tower with its top reaching heaven”:
Now Yahweh came down to see the town and the tower that the sons of man had built. “So they are all a single people with a single language!” said Yahweh. “This is but the start of their undertakings! There will be nothing too hard for them to do. Come, let us go down and confuse their language on the spot so that they can no longer understand one another.” Yahweh scattered them thence over the whole face of the earth, and they stopped building the town. It was named Babel therefore, because there Yahweh confused the language of the whole earth. It was from there that Yahweh scattered them over the whole face of the earth [Genesis 11: 5-9; The Jerusalem Bible, 1966: 14].
From the perspective of the Old Testament, the tower of Babel becomes a metaphor for the division of the human race into groups unable to communicate with one another. Applying that metaphor to contemporary sociology, we appear to have achieved a more subtle procedure than speaking languages from different cultures. We have learned to speak the languages of different subcultures within our discipline. Unless an individual learns the language of a given field by becoming familiar with its literature, he or she will remain unable to communicate with others in that field.
Part One is about an approach to the scientific method which aims at building bridges across our subcultures or fields of sociology, changing our tower of Babel into a discipline where we can all gain from learning to follow the scientific ideal of communicating with one another. God’s fears as to the result might in fact actually be fulfilled: “This is but the start of their undertakings! There will be nothing too hard for them to do.” Following the work of Thomas Kuhn, we should not underestimate the difficulties confronting any major shift in our discipline. This is particularly true when a scientific paradigm is itself nested within a cultural paradigm. Yet also following Kuhn, scientific revolutions can indeed occur when a discipline becomes aware not only of its fundamental contradictions but also of a direction for resolving them. In Chapter 1 we sketch a contrast between our present interpretation of the scientific method and an alternative interpretation, with the former labeled “bureaucratic” and the latter “interactive.” Granting the achievements we have made with the aid of our past interpretation, the method appears to be unable to cope with the enormous complexity of human behavior. Our interactive interpretation or “web approach” aims to take fully into account that complexity. In Chapter 2 we pursue the paradigmatic basis for making a fundamental change to that web approach. For example, we examine the nature of our cultural paradigm as well as a cultural alternative. And we also look to an approach to sociological theory which can enable us to follow that cultural alternative.
CHAPTER 1
SOCIOLOGY AND THE SCIENTIFIC METHOD
The work of C. Wright Mills, someone who I was fortunate enough to know personally and who envisioned both what our discipline lacks and what it might proceed to achieve, plays an important role in this chapter. Mills is known largely for his contributions to political sociology as well as his metaphor of “the sociological imagination,” yet it is time that we begin to take seriously what he contributed to our understanding of the scientific method in sociology. In the first section, “Problems with Bureaucratic Science,” I sketch--within the context of other materials--his profound critique not just of the sociology of his own times but also of how we are presently going about our business. To complete the picture which he drew, I also bring in aspects of what we have learned since his time from the history and philosophy of science as well as from sociology. This critique of bureaucratic science constitutes the basis, in the second and final part of this chapter, for a more systematic presentation of the approach to the scientific method which I believe we desperately need for substantive progress and which society urgently requires. Mills somehow succeeded in giving voice to the aspirations for sociology which continue to lie buried under layers of cynicism and pessimism within the rest of us sociologists, waiting to take wing. Yet we must go beyond Mills’ insights and metaphors and build on his work in a highly systematic way if we are to chart a direction for contemporary sociologists and social scientists in general. That chart is begun in this chapter and this book, yet it remains for the discipline as a whole to carry it further.
PROBLEMS WITH BUREAUCRATIC SCIENCE
Many of us are familiar with Mills’ concept of the sociological imagination:
. . .that imagination is the capacity to shift from one perspective to another--from the political to the psychological; from examination of a single family to comparative assessments of the national budgets of the world; from the theological school to the military establishment; from considerations of an oil industry to studies of contemporary poetry. . .[1959: 7].
Mills reveals the same breadth of vision which characterized the classical sociologists and is so central to the present ideals of the discipline [see for example Horowitz, 1983; Chasin, 1990]. He saw that kind of breadth as essential for fulfilling the Enlightenment “promise of sociology.” Although he never developed a systematic direction for just how sociologists should proceed to employ the scientific method, the body of his work suggests five components: (1) We should not shirk from addressing absolutely fundamental problems within society. (2) We should move far up language’s ladder of abstraction so as to utilize very abstract concepts. (3) We should come far down that ladder so as to examine the concrete evidence that bears on our ideas. (4) We should work to integrate our knowledge so that our approach is broad enough to enable us--as indicated in the above quote--to shift from one perspective to another. (5) We should develop ourselves as individuals with the ability to think in this broad way, developing a “sociological imagination” that suggests a new vision of society. We take up these five components in the following subsections, emphasizing critiques of our present approach to the scientific method.
Orientation to Problems
Mills’ orientation to problems is illustrated by a body of work that included an examination of the power of elites in subverting democratic ideals [1948; 1956], interest in the alienation of “the new middle class” [1951], concerns about the coming of World War III [1958], and an analysis of the personal troubles of the individual in modern society [1959]. We can begin to understand his zest for conflict--expressed in his relationships to sociological colleagues--from a story he told a class about an encounter with Eisenhower, then President of Columbia University. When Eisenhower walked into his classroom one day unannounced and quietly took a seat in the back row, Mills unhesitatingly altered his lecture, presenting a systematic plan for a violent revolution to overthrow the U. S. government. The class would operate as a key cell in directing the course of the revolution, which he claimed would be opposed by the ruling class. Eisenhower sat stonily silent as Mills proceeded. Finally, Eisenhower stood up and quickly walked out, with Mills never seeing him again. Mills told this story with a broad smile on his face, apparently having no particular ax to grind against Eisenhower but simply as a joke he managed to play on the powers that be within the university.
We can understand more fully Mills’ interest in avoiding trivial problems from a study he completed during World War II while still a graduate student at the University of Wisconsin. He analyzed some fifty textbooks on social problems in order to learn about “the style of reflection and the social-historical basis of American sociology,” a rather tall order even for an eminent professor. These texts generally centered on how the individual might “adjust” to society in order to solve his or her problems rather than on how society might change in response to those problems:
Use of “adjustment” accepts the goals and the means of smaller community milieux. At the most, writers using these terms suggest techniques or means believed to be less disruptive than others to attain the goals that are given. They do not typically consider whether or not certain groups or individuals caught in economically underprivileged situations can possibly obtain the current goals without drastic shifts in the basic institutions which channel and promote them. The idea of adjustment seems to be most directly applicable to a social scene in which, on the one hand, there is a society and, on the other, an individual immigrant. The immigrant then “adjusts” to the new environment. . .[1943: 179-180].
For Mills, the idea of adjustment works to foreclose the aspirations of the immigrant. An alternative would be for the immigrant to retain those high aspirations and for society to undergo “drastic shifts in the basic institutions which channel and promote them.”
Mills also cites the emphasis of these texts on the idea of “cultural lag,” based on the work of Ogburn [1957] with his idea that “adaptive” or “nonmaterial culture” lags behind “material culture”:
. . .The model in which institutions lag behind technology and science involves a positive evaluation of natural science and of orderly progressive change. Loosely, it derives from a liberal continuation of the enlightenment with its full rationalism, its messianic and now politically naive admiration of physical science as a kind of thinking and activity, and with its concept of time as progress. . .[1943: 177].
This use of the idea of cultural lag is analogous to an emphasis on the adjustment of the individual. Not only must the individual adapt to existing norms of society but the institutions of society must adapt to the physical and biological technologies built on the continuing development of the biophysical sciences, for continuing “progress” supposedly depends on such adaptation. Yet we can construct alternatives to this approach, just as Mills saw alternatives to accepting the goal of the adjustment of the immigrant as one that sociologists should adopt. Society need not bow down to physical and biological technologies involving “material culture.” Rather, it is possible for us to create the kind of society in which those technologies move in directions that strengthen “nonmaterial culture.”
Mills’ approach to the scientific method within sociology preceded Thomas Kuhn’s analysis of scientific revolutions within the physical and biological sciences [1962, 1977, 1992] by several years, yet the two orientations are similar in their view of the forces blocking us from confronting basic problems. Kuhn’s argument was based on an intuitive application of the sociology of knowledge. He saw a community of scientists as being swayed not just by evidence alone but also by such factors as tradition, social hierarchy and the personality of the scientist. For Kuhn, the problem of achieving a scientific revolution is a massive one, for the very subculture of a science or its “scientific paradigm” must be challenged and not just particular studies. Mills is also concerned with the subculture of a science. For example, he suggests that “academic departmentalization may well have been instrumental in atomizing the problems which they [the authors] have addressed” [1943: 166], thus yielding the failure to confront the question of changing our institutions. Mills also suggests that those authors came from similar backgrounds, shared common perspectives, and thus tended to conform to relatively conservative norms when it came to any question of fundamental changes in society. As for hierarchy, Mills points to the relationship between teacher and student or author of a text and students who read the text. The result is an emphasis on the systematization of existing ideas rather than on questioning those ideas or attempting to discover new ones.
If we take Kuhn’s analysis one step further, then we can discover problems which are even more fundamental than subcultural ones, namely, cultural ones. Extrapolating Kuhn, research can succeed in confronting contradictions within the researcher’s cultural paradigm. For an example we turn to Friedrich Nietzsche’s The Gay Science:
The greatest recent event--that “God is dead,” that the belief in the Christian god has become unbelievable--is already beginning to cast its first shadows over Europe. . . .how much must collapse now that this faith has been undermined. . .for example, the whole of our European morality. . . .we philosophers and “free spirits” feel, when we hear the news that “the old god is dead,” as if a new dawn shone on us. . .at long last our ships may venture out again, venture out to face any danger; all the daring of the lover of knowledge is permitted again. . .[(1887) 1974].
Mills no less than Nietzsche was most interested in investigating major transformations of society. He believed that “The existence of mass estrangement among workers, anxiety among professionals, and anomie among middle sectors invalidated the ‘modern’ period.” And he suggested that the new epoch that was dawning might be labeled the “postmodern era,” most likely the first usage of that term. And just as Kuhn wrote of contradictions which come to light within the old paradigm, Mills wrote of the end of an epoch: “When what is happening in the social world as well as what is widely felt and widely thought can no longer be satisfactorily explained by the received principles, then an epoch is ending and a new one needs to be defined [1960; quoted in Horowitz, 1983: 323, 327].
Moving Up Language’s Ladder of Abstraction
Yet just how are major problems, such as our present failure to understand “what is happening in the social world as well as what is widely felt and widely thought,” to be addressed? Or using Nietzsche’s example, just how is the “death of God” to be understood and confronted? Mills’ analysis suggests an answer when he employs the concept of “anomie” and when his usage of “estrangement among workers” and “anxiety among professionals” implies the concept of “alienation.” It is, then, sociology’s abstract concepts which come to function as a basic part of the scientific method within the discipline. Alvin Gouldner, in a reply to a review of his The Coming Crisis of Western Sociology [1970]--quoted in the preface--explains more clearly what Mills was implying:
At decisive points the ordinary language and conventional understandings fail and must be transcended. It is essentially the task of the social sciences, more generally, to create new and “extraordinary” languages, to help men learn to speak them, and to mediate between the deficient understandings of ordinary language and the different and liberating perspectives of the extraordinary languages of social theory. . . .To say social theorists are concept-creators means that they are not merely in the knowledge-creating business, but also in the language-reform and language-creating business. In other words, they are from the beginning involved in creating a new culture [Gouldner, 1972: 16].
Concepts like anomie and alienation are, then, the sociologist’s most powerful tool for understanding the fundamental problems of society.
From this perspective we can draw an analogy between ordinary language and the language of sociology, looking to both similarities and differences. On the one hand ordinary and sociological concepts are tools for understanding our world and addressing problems, and both can be employed within sentences or propositions that state the nature of the world and how we might proceed to solve problems. Also, both are abstract to at least some degree, being at least one stage removed from our nonverbal experiences with phenomena: they are linguistic and reflect on experience, as distinct from the tools used by other forms of life. On the other hand the linguistic tools of the sociologist carry along with them the weight of sociological knowledge, and such knowledge can be most useful in understanding the world and addressing our problems. In addition, the concepts of sociology--like anomie and alienation--generally are more abstract than our everyday concepts, which tend to stay closer to whatever we experience concretely. As a result, sociological concepts tend to cover far more ground. When we speak of anomie, for example, we can refer to contradictions within any culture, past or present. Yet if the sociologist remains unaware of the importance of these differences then it will be all too easy to rely far too much on ordinary usages, using sociological language only in passing. Also, the sociologist will fail to define sociological concepts in a sufficiently abstract way, thus losing generality.
Mills wrote about such deficiencies within sociological usage in his analysis of texts on social problems:
The level of abstraction which characterizes these texts is so low that often they seem to be empirically confused for lack of abstraction to knit them together. They display bodies of meagerly connected facts, ranging from rape in rural districts to public housing, and intellectually sanction this low level of abstraction. . . .Collecting and dealing in a fragmentary way with scattered problems and facts of milieux, these books are not focused on larger stratifications or upon structured wholes [1943: 166].
Mills suggests--in this quote repeated in the preface--that we require abstract concepts like social stratification, concepts general enough to apply to a very wide range of situations, if we are to avoid the “bodies of meagerly connected facts” to be found in those texts. Yet usage of abstract concepts has immediate implications for our approach to problems. If patterns of social stratification are a partial cause of social problems, then solutions will require fundamental changes in society. By avoiding such usages, the sociologist is able to maintain a more conservative stance on addressing social problems. What Mills succeeds in doing here is to confront sociologists with contradictions between their scientific ideals and their actual research procedures. In Kuhn’s terms, he alerts us to paradigmatic contradictions within the subculture of sociology as well as within modern culture.
Mills is taking on not just sociologists who center on social problems but rather an emphasis within the discipline as a whole, as illustrated by what Merton has called “theories of the middle range”:
Every effort should be made to avoid dwelling upon illustrations drawn from the “more mature” sciences--such as physics and chemistry. . .because their very maturity permits these disciplines to deal fruitfully with abstractions of high order to a degree which, it is submitted, is not yet the case with sociology [1968: 139-140].
Merton’s approach appears to be a classic illustration of what he himself called a “self-fulfilling prophecy.” By defining sociology as immature and unable to employ “abstractions of high order,” we create that very situation of immaturity. Some three decades ago Willer and Webster launched a profound critique of Merton’s approach to the scientific method, basing their argument largely on the philosophy of science [1970; see also Peirce, 1955; Hempel, 1965; Willer, 1967]. They maintained that the more developed sciences, versus sociology, construct abstract concepts. For example, there are assertions about “mass” and “specific gravity” in physics, about “bonds” and “valences of molecules” in chemistry, and about “heredity,” “natural selection” and “genes” in biology. They argued that sociology’s “immaturity” derives in large measure from its failure to use abstract concepts and theory. This approach taken by Willer and Webster has been updated by a variety of analyses pointing in the same direction [see for example Phillips, 1972, 1979, 1980, 1985, 1988, 1990; Lauderdale, 1990; Wallerstein, 1980, 1991, 1998].
Sociology’s failure to emphasize concepts at a very high level of abstraction also derives from the relationship between the disciplines of sociology and philosophy coupled with the pragmatic stance of much of American sociology. As a discipline, sociology has in large measure distinguished itself from philosophy by emphasizing its empirical stance. That stance has been seen as enabling it to confront the practical problems of everyday life. Mills saw this situation as follows:
The ideal of practicality, of not being “utopian,” operated . . .as a polemic against the “philosophy of history” brought into American sociology by men trained in Germany; this polemic implemented the drive to lower levels of abstraction. A view of isolated and immediate problems as the “real” problems may well be characteristic of a society rapidly growing and expanding, as America was in the nineteenth century and, ideologically, in the early twentieth century. . .the practice of the detailed and complete empiricism of the survey is justified by an epistemology of gross description . . .[1943: 168].
Mills reveals the contradiction between ideals for practical action and scientific procedures which tend to atomize problems and prevent the sociologist from understanding them. The very tools required for such understanding come to be seen negatively as “philosophical” and are as a result avoided. The result is the kind of fragmentation of problems which Mills noted in his analysis of textbooks on social problems.
Moving Down the Ladder of Abstraction
Let us bear in mind as we proceed to examine movement down language’s ladder of abstraction how intimately it is linked to both commitment to a problem and the use of abstract concepts. It is useful to return to Nietzsche at this point, for he conveys the spirit of science’s empirical achievements by contrast with the situation of humanity prior to the rise of science:
It is a profound and fundamental good fortune that scientific discoveries stand up under examination and furnish the basis, again and again, for further discoveries. . .To lose firm ground for once! To float! To err! To be mad! That was part of the paradise and the debauchery of bygone ages, while our bliss is like that of a man who has suffered shipwreck, climbed ashore, and now stands with both feet on the firm old earth--amazed that it does not waver [(1887) 1974: 111].
Nietzsche here captures science’s ability to discover firm knowledge--not necessarily absolute truth--which far transcends pre-scientific opinions. It is this ability which is fundamental to the motivation of sociologists in their efforts to move down language’s ladder of abstraction to obtain facts. Yet when that motivation becomes isolated from commitments to solving major problems and using abstract concepts in the process, then it yields what Mills criticized as trivial research despite its helping us to plant “both feet on the firm old earth.”
Sociologists often move down the ladder of abstraction by using procedures which oversimplify the complexity of human behavior. Implicitly, this appears to be an effort to imitate both the simplicity and predictive power of much of the physical sciences, as illustrated by the simple yet powerful formula, F = ma, or force = mass times acceleration. Herbert Blumer criticized this approach many years ago:
The objective of variable research is initially to isolate a simple and fixed relation between two variables. . . .This is accomplished by separating the variable from its connection with other variables through their exclusion or neutralization.
A difficulty of this scheme is that the empirical reference of a true sociological variable is not unitary or distinct. When caught in its actual social character, it turns out to be an intricate and inner-moving complex. To illustrate, let me take what seems ostensibly to be a fairly clear-cut variable relation, namely between a birth control program and the birth rate of a given people. . . .For the program of birth control one may choose its time period, or select some reasonable measure such as the number of people visiting birth control clinics. For the birth rate, one merely takes it as it is. . . .
Yet, a scrutiny of what the two variables stand for in the life of the group gives us a different picture. Thus, viewing the program of birth control in terms of how it enters into the lives of people, we need to note many things such as the literacy of the people, the clarity of the printed information, the manner and extent of its distribution, the social position of the directors of the program and of the personnel, how the personnel act, the character of their instructional talks, the way in which people define attendance at birth control clinics, the expressed views of influential personages with reference to the program, how such persons are regarded, and the nature of the discussions among people with regard to the clinics . . .[ Blumer, 1956: 688].
Blumer is here criticizing the simplifying assumptions, largely invisible, which lie behind what he calls “analysis of the variable” within quantitative sociology. Granting the importance of moving down language’s ladder of abstraction to concrete measurements of particular factors, we sociologists should have learned enough from our research to realize that we cannot learn much by centering on only two variables within a complex context of factors and ignoring the rest with some phrase like “other things being equal” or ceteris paribus. And we can even add many other factors to Blumer’s example that emphasizes situational factors close to his own symbolic-interactionist perspective. For example, there is the matter of cultural values and norms as well as patterns of social organization in society as a whole like social stratification and bureaucracy. Most of quantitative methodology pushes aside such considerations, for they would interfere with the tools of measurement which we presently have and which yield definite findings. Those tools build on certain aspects of what quantitative sociologists take to be the basis for the successes of the physical sciences. If we return to the simple formula, F = ma, only a very few variables have sufficed to yield extremely accurate predictions. Such formulae can help us sociologists to understand the importance of using concepts defined at very high levels of abstraction, of linking concepts with one another systematically--although not necessarily mathematically--and of testing theoretical ideas. Unfortunately, we have learned instead to isolate phenomena from their complex contexts so as to yield the kinds of measurements that prepare the way for using mathematics to help us make predictions.
None of the above arguments should be taken to imply that measurement procedures, including those based on a quantitative approach, are necessarily deficient. Problems arise, however, when such procedures are not combined with important theory, and thus the researcher remains at a low level of abstraction. This is illustrated by efforts which follow the direction of employing as much mathematics as possible--remaining unconcerned with theoretical questions--and thus conform to the interests of early philosophers of science who were steeped in mathematics. We have paid little attention to the pragmatist doctrine put forward by Abraham Kaplan:
It is one of the themes of this book that the various sciences, taken together, are not colonies subject to the governance of logic, methodology, philosophy of science, or any other discipline whatever, but are, and of right ought to be, free and independent [Kaplan, 1964, quoted in Diesing, 1991: 82].
Instead of such independence, which would look to assess the actual achievements resulting from methodology, we have quantitative measurement procedures geared to the movement from nominal to ordinal to interval to ratio scales, supporting a focus on mathematical prediction. Further, the focus often is on increasing the reliability and precision of measurements while ignoring the range of contextual factors involved. For example, there is interest in specifying the “operational definition” of a given concept, moving down language’s ladder of abstraction, without moving up that ladder as well..
Yet another aspect of efforts to move down that ladder has to do with procedures for obtaining probability samples, typically for surveys of a given population. Such procedures employ mathematical assumptions which enable the researcher to conclude, say, that a given sample of individuals represents a much larger population of individuals to within a specified degree of sampling error. And this in turn becomes the basis for quantitative analyses of the resulting data. However, the focus of such sampling procedures is generally quite narrow, employing concepts at a low level of abstraction. Once again this serves to simplify enormously complex situations and enables the researcher to move down the ladder abstraction and draw simplistic conclusions about the relationship between two or several variables. For example, it is a rare study that takes into account populations in the past as well as the present; the focus is on the present. Even the rare “panel study” takes place over a limited number of years, avoiding very long-term change like that from preindustrial to modern society or oral to literate society. If culture is indeed an important concept, then how are probability samples to tell us anything of significance about basic cultural change, granting the very rare sampling of written materials over long periods of time? Sampling procedures were originally developed not by sociologists but within other disciplines. For example, agricultural economists were interested in improving crop yields, geneticists wanted to learn about heredity, and the U. S. Armed Forces wanted to plan bombing runs during World War II [Schutte, 1977: Appendix I]. None of this speaks to our own problem of attempting to understand cultural change.
Integrating Knowledge
We turn now from sociology’s movement down language’s ladder of abstraction to its movement across the discipline’s specialized fields, as illustrated by the forty Sections within the American Sociological Association. Overall, what becomes obvious as we proceed with this analysis is the way in which all aspects of the sociologist’s usage of the scientific method are intimately tied together within the same scientific paradigm. For example, tied closely to the above analysis of our orientation to problems and our movements up and down language’s ladder of abstraction are our procedures for drawing statistical inferences about the relationship between two variables. Here once again we make substantial use of mathematics and attempt to move toward prediction. This generally results in statements with very limited utility either for understanding phenomena or for solving practical problems. For example, we may be able to reject the null hypothesis that there is no relationship whatsoever between, say, the number of people visiting a birth control center and the birth rate of a given people. In other words, we find that the birth control program “works” at least to a very limited extent, and this has at least some utility in efforts to evaluate its effectiveness. But what have we learned as a result? Theoretically, we have learned little more, for example, about the impact of cultural values and norms, about patterns of bureaucracy and social stratification, about anomie and alienation, of patterns of conformity and deviance, or about relative deprivation and reinforcement. And there appears to be little practical impact for our conclusion, since we surely did not really previously believe the null hypothesis that the birth control program had absolutely no effect on the birth rate. And what have we learned about comparing this program with many others as well as how to improve any of these programs?
In an effort to answer at least some of these questions, quantitative analysts have utilized procedures for correlation and regression, both of which encourage precise measurement procedures. Efforts at correlation quantify the degree to which variation of a given variable is accounted for by variation in another variable, getting beyond the simple statement that there is at least some relationship between the two which could not easily arise as a result of sampling error or chance. And regression procedures specify mathematical formulae which we can use to make predictions from what we know about one variable to what we don’t know about another. In our birth control example, correlation would tell us just how much the birth rate would be affected by a birth control program, and regression would enable us to make a prediction as to the change in birth rate from our knowledge of the existence of birth control programs. One problem, however, is that use of such more sophisticated quantitative procedures--which also require more assumptions that may not hold true--does not necessarily yield high correlation’s or accurate predictions. Arguably, sociological efforts to correlate and predict have generally yielded low correlation’s and inaccurate predictions. This is quite understandable once we take into account Blumer’s critique of the analysis of variables as well as the general failure of such quantitative research to conceptualize variables at very high levels of abstraction. Rather than yield sophisticated knowledge, sophisticated tools generally have shown up our enormous ignorance.
One response on the part of quantitative sociologists to these problems is to move further in a quantitative direction. If high correlations and accurate predictions cannot be made from knowledge of one variable, how about many variables? For example, they use such procedures as factor analysis, cluster analysis, partial and multiple correlation, multiple regression, path analysis and discriminant analysis. There are definitely occasional instances in which such procedures, when coupled with abstract theoretical concepts, have advanced our understanding, and one illustration will be presented in Chapter 4. And there are also instances when such analyses have yielded better bases for evaluation research, where higher correlations or more accurate predictions aided in the overall assessment of certain projects or procedures over others. Yet it is arguably the case that such quantitative approaches have, in general, further diverted attention from abstract conceptualization and theory and taken us still further away from developing a sociology which is integrated, credible and cumulates rapidly. Such a sociology requires us to face up to our present divorce between methods and theory. Quantitative procedures, by contrast, generally place mathematics once again in the saddle, riding the horse of a long-dead philosophy of science.
Efforts to integrate knowledge have not been limited to such relatively complex mathematical procedures. For example, we might have reference here to the cross-tabulational tradition within sociology, primarily based on procedures for the analysis of surveys. Historically, many of such procedures grew out of research sponsored by the U. S. Armed Forces during World War II, followed by great interest in survey research and accompanied by electrical-mechanical inventions for the analysis of data with the aid of punched cards. “Cross tabulation” is simply a way of obtaining the distribution of one variable within the categories of another. In a simple example, we might determine whether a greater percentage of those attending birth-control clinics had a lower birth rate than those not attending them. Sociologists have emphasized cross-tabulations, even to this day, in their search for cause-and-effect relationships between variables. Perhaps the greatest influence on procedures for examining such relationships was “the elaboration model,” developed by Paul Lazarsfeld and his co-workers at Columbia University in order to interpret data obtained on the American soldier [Stouffer, 1949] during World War II [Lazarsfeld and Rosenberg, 1955; see also Hyman, 1972; Phillips, 1985, 430-442]. Such procedures often focus on secondary analyses, and they generally involve the introduction of a third variable and a detailed cross-tabulation of three variables. Many such analyses have yielded considerable insight, but generally they have contributed to the imbalance between theory and methods.
Worldviews
The fact that all of these problems associated with our present approach to the scientific method are intertwined with one another suggests not only the existence of a scientific paradigm or subculture which yields them but more generally a cultural paradigm within which that scientific paradigm is located. As for the nature of that cultural paradigm we can turn to several analyses, taking into account what sociology as a whole reveals about this matter. Let us begin with a different quote from Nietzsche, who appears to have captured much of the nature of our worldview:
On the doctrine of poisons--So many things have to come together for scientific thinking to originate; and all these necessary strengths had to be invented, practiced and cultivated separately. As long as they were still separate, however, they frequently had an altogether different effect than they do now that they are integrated into scientific thin