What's Wrong with Sociology?

Stephen Cole

• Publisher: Routledge
• Year: 2018
• Print ISBN: 9780765800398
• eBook ISBN: 9781351326346
• Topic: Social Sciences
• Subtopic: Sociology
• Index: Social Sciences

Since the 1950s sociology has experienced a decline in prestige when compared with the other social sciences. In some highly publicized cases some universities have retrenched their sociology departments, others are contemplating either retrenchment or downsizing of their departments. Although there are some practitioners of the discipline who believe that it has never been in better shape, many sociologists have come to believe that there are very serious problems both in the cognitive and social organization of the discipline. This book contains sixteen essays by sociologists who believe that their discipline faces very serious problems which must be overcome if the discipline is to survive and prosper. The contributors were selected to represent diverse views and thus there is substantial disagreement among them over what the problems are that sociology faces and how they may be remedied. In this highly provocative book readers is likely to find some essays they agree with and others they disagree with; but all the essays present important problems faced by the discipline which must be addressed.Although the authors of the sixteen essays do not agree on what is wrong with the discipline, there are some themes which appear frequently. In his introduction Cole summarizes and comments on these themes. His introduction centers on the question of whether sociology is entirely socially constructed. Is what we believe to be true about society constrained by empirical evidence or is it a result of our ideology, power, authority, and other social processes? One theme which appears in many of the essays is that sociology has become too ideological and as a result has lost credibility among university administrators, politicians and the general public. Many of the essays also stress the view that there are very low levels of consensus in sociology and that it is hard to see evidence of progress. Others criticize the discipline for not dealing with the really important social issues and see much of the work published as being parochial and trivial. Questions are also raised about why the use of causal models has failed to generate solutions to most of the problems the discipline addresses. Some authors believe that the discipline adheres to an overly rational model of human behavior and has failed to keep up with some of the advances introduced by post-modernist theories.This highly readable set of essays should be of interest to all those are concerned about the current state of sociology. They will also be useful in introducing graduate students to some of the most important issues currently being debated in the field.Stephen Cole is currently professor of sociology at the State University of New York at Stony Brook and professor of sociology at the University of Queensland, Australia. He is the author of Making Science: Between Nature and Society, and with Jonathan R. Cole, Social Stratification in Science.

1
Why Sociology Doesn’t Make Progress Like the Natural Sciences

Stephen Cole

In analyzing science it is crucial to make a distinction between two types of knowledge: the core and the research frontier (S. Cole, 1992). The core consists of a small group of theories, methods, and exemplars that are almost universally accepted by the relevant scientific community as being both true and important. Examples of parts of the core in physics would be quantum mechanics, the Weinberg, Salam and Glashow theory of weak interactions, or the Bardeen, Cooper, and Schrieffer theory of superconductivity. An example in molecular biochemistry would be the Watson and Crick model of DNA. And an example in evolutionary biology would be Darwin’s theory of evolution.

The frontier consists of all newly produced knowledge. Most of this knowledge is ignored, a small part is paid attention to and most of that is discarded as being wrong. The core is connected to the frontier through the evaluation process. Only a tiny percentage of new knowledge moves from the frontier to the core, yet it is the core which is the basis of progress in all sciences. The core, similar to Kuhn’s ([1962] 1970) concept of paradigm, is what is taught to new recruits and is the common ground upon which new science is developed.

Prior research which I have conducted suggests that at the research frontier there are not significant differences in the way in which natural sciences and social sciences proceed. The problem with fields like sociology is that they have virtually no core knowledge. Sociology has a booming frontier but none of the activity at that frontier seems to enter the core. There seems to be no sociological work which the great majority of the community will regard as both true and important.

Most sociologists will probably require no evidence to accept the validity of this conclusion. But for the few who may be skeptical I suggest examining introductory textbooks in fields like physics, chemistry, and sociology (S. Cole, 1983). In physics texts, for example, there are a relatively small number of references- about 100. In sociology texts there are usually more than 800 references. In physics texts most of the work cited was produced before the contemporary period; in sociology the overwhelming majority of work cited is relatively recent. The physics texts have very high overlap; the sociology texts have less overlap. The material covered in physics texts used 20 years ago is almost the same as the material covered in physics texts used today. In sociology there have been substantial changes in the material covered in texts used 20 years ago and those used today. In other words, the physics texts are reporting what is in the core and the sociology texts are reporting what is at the ever changing frontier.

In this paper I will argue that there are two variables influencing the development of core knowledge which differentiate the natural and social sciences. These are the extent to which the phenomena we study are mutable, and the extent to which problem choice and decisions made in solving problems are based upon cognitive as opposed to non-cognitive criteria. I will be comparing physics with sociology. Among the natural sciences, physics is the field in which the phenomena studied are least likely to be mutable and in which decisions are least likely to be made on non-cognitive criteria. I am aware that there is considerable variation among the natural sciences on these variables. Biology, for example, is in many ways intermediary between physics and sociology on these variables. Ecologists, for example, study phenomena which change over time. And fields like evolution and genetics, which have relevance for social and political issues, are more likely than physics to have decisions of scientists influenced by non-cognitive criteria. Also in fields like medicine, agricultural science, and fisheries science some of the most interesting problems emerge from the study of phenomena which are not immutable and may be historically and culturally specific, as are most problems in sociology. If my theory about why sociology does not develop core knowledge is correct, then we should expect to see differences in the ease of developing core knowledge among the various natural sciences according to their placement on the causal dimensions I have identified.

Defining Progress

If the purpose of this paper is to understand why sociology doesn’t make progress in the same sense as fields like physics, we must have a clear definition of progress. A field is making progress if it has a core and is developing new knowledge which is being added to the core. This paper is based upon the assumption that although nature constrains what is at any time taken to be scientific “truth,” the extent of this constraint can vary dramatically and it is therefore not useful to employ the concept of “absolute truth.” Since so many theories which were once thought to represent truth, are now seen as incorrect, there is no reason to believe that what we currently see as truth will not later be seen as incorrect (Laudan, 1984). Many contemporary scholars in the social studies of science argue that it is necessary to engage in a whiggish type of history, in which the displaced knowledge is redefined in light of the new knowledge in order to make it appear as if there is a type of linear progress in science (Brannigan, 1981; Elkana, 1970; Kuhn ([1962] 1970). Thus, core knowledge is not knowledge which is “really” true but knowledge which the scientific community currently believes to be true.

It should also be clear that the content of the core of a discipline can change. The history of science reveals hundreds of cases where the core not only changes but sometimes in quite radical ways. Thus Priestely’s notion of phlogiston in chemistry was replaced by Lavoisier’s notion of oxygen. And in geology the notion of fixed continents was very quickly replaced by the concept of continental drift and the paradigm of plate tectonics. Because a contribution to science is ultimately displaced from the core does not mean that it did not represent progress at the time when it was in the core. Progress in science, as in other institutions, must be a time-specific concept.¹ Given the fact that science sometimes changes very rapidly, in some fields at times it would not be unusual to see new knowledge entering the core quickly and at times leaving the core quickly. In my view, though, the fact that knowledge is being produced which enters the core counts as evidence of progress. Quick and consensual displacement of core knowledge may indicate progress rather than stagnation. In fact, the “immediacy effect” or the extent to which a discipline depends upon new knowledge, has been used as a measure of progress in the past. (Cole, 1983; Price, 1970; Zuckerman and Merton, [1971] 1973).

Finally, my definition of progress does not depend at all upon the practical utility of scientific knowledge. Some natural science contributions turn out to have practical utility and others do not; but the latter does not necessarily represent any less intellectual progress. For example, in medical science, work which leads to a successful vaccine for AIDS does not necessarily represent a greater advance in science than fundamental knowledge about the immune system, which might not lead to any practical application for many years, if at all. It is my belief that sociological knowledge has had very little influence on the world in the past and even if it were to make more progress it is unlikely to have any significant influence.

Should Sociologists Act More Like Natural Scientists?

Now that I have defined progress, we can move to the primary purpose of this paper, which is to suggest some reasons why sociology does not develop a core. Some of these reasons are inherent in the subject matter of the discipline and are, therefore, not subject to change. Others are inherent in terms of how we have chosen to practice sociology and are theoretically subject to change— but the chances that they will be changed are very low.

Before going on to discuss these issues, however, there is one important answer proposed to the question of why the social sciences don’t make progress like the natural sciences: the social sciences fail to follow the positivistic methods of the natural sciences. George Homans was a proponent of this view (1967), but the foremost contemporary proponent is the theorist Walter Wallace (1971, 1983, 1988). Wallace argues that sociology is a “natural” science but that it does not develop theory on which there is consensus because sociologists fail to follow procedures which the natural scientists follow. For example, sociologists do not define their concepts clearly so that all who utilize a concept will mean the same thing by it.

Indeed, it is perhaps just this frequency with which we blithely go on using a term so utterly fundamental to our discipline without making a concerted effort to standardize its meaning that comes closest to justifying the charge that, scientifically speaking, we sociologists simply do not know (and may not care) what we are talking about. And never having established a common definition of “social structure” (and many other terms of similar centrality, including “culture,” “status,” “role,” “value,” “norm,” and most important of all the term “social phenomenon” itself) it is no wonder that the catalog of sociological natural laws remains so empty. (Wallace, 1988, p. 24)

Wallace then goes on to give examples of how he would define some of these key terms. For example, he defines a “social phenomenon” as follows:

A social phenomenon may be generically defined as an interorganism behavior regularity - that is, a nonrandom coincidence in time and/or space of two or more organisms’ behaviors. (Wallace, 1988, p. 31).

Implicit in Wallace’s critique is the conclusion that if sociologists were to change their ways of proceeding, they would develop knowledge like that in the natural sciences.

I don’t believe that positivist critiques like those of Wallace provide an adequate answer to our question. Consider only the question of defining concepts. Wallace seems to think that if somehow we could impose uniform definitions of these terms we would have uniform treatment of the concepts they represent. Think, for example, whether accepting Wallace’s definition of a social phenomenon is likely to change anything in sociology. In order to get a definition that is innocuous enough for sociologists not to object, Wallace is forced to ignore substance and be so general that almost anything could be classified as a social phenomenon.

Although I agree with Wallace that it is annoying to have every sociologist define key terms in her/his own way, enforcing a standard wouldn’t do any good unless the consensus on the meaning and significance of the concepts was real and natural. Of course, if all sociologists agreed that norms were rules regulating social behavior, and that these norms determined large areas of our behavior, sociology would be better off. But alas there are many sociologists (ethnomethodologists and some symbolic interactionists) who believe that norms, if they exist at all, have little influence on our behavior. Agreeing on terminology won’t get rid of the underlying lack of consensus.

A problem with Wallace’s positivistic views towards sociology is that they tend to be based upon a distorted view of how the natural sciences are actually conducted.² Wallace (1983, Ch. 14), for example, bases his account of how the natural sciences are conducted primarily on the normative accounts of philosophers such as Popper, Hempel, Nagel, and Braithwaite rather than on the more descriptive accounts of contemporary historians, sociologists, and philosophers. In doing their work the natural scientists simply don’t follow the procedures described by the traditional philosophers. The natural scientists do not do science the way Wallace says that sociologists should. For example, Wallace places a great emphasis on the significance of replication; but work by Harry M. Collins (1985) and others has suggested that replication is actually much rarer in the natural sciences than one would expect from reading the normative accounts of positivist philosophers.

My research on levels of consensus at the research frontier show that there are no significant differences between the natural and the social sciences (S. Cole, 1992, Ch. 5). For example, reviewers of National Science Foundation grants, or of manuscripts submitted for publication, are just as likely to disagree in natural science fields as in social science fields. This suggests that whatever differences there are in procedures followed by natural and social scientists, they do not influence the ability of the community to develop consensus on the frontier of knowledge.

If Wallace’s prescription for the development of sociological knowledge is correct why haven’t sociologists followed it? I believe that sociologists have not adopted these procedures because they don’t work. In fact the natural scientists don’t do science the way the positivist philosophers say they should because these prescriptions do not work in the natural sciences either.

I believe that the differences between fields like physics and sociology have considerably more to do with differences in the phenomena being studied than in the characteristics of the people who study them or the procedures that they employ. One difference in procedure which does make a difference is the ease of conducting experiments. Inability to conduct experiments is one impediment to the cumulation of knowledge; but as Wallace (1983) and others have pointed out, the theory of evolution in biology is not based upon experiment but is widely accepted as true and important.

There are two primary distinctions between the phenomena studied in a field like physics and those studied in sociology. The first is the mutability of the phenomena being studied, and the second is that sociologists— unlike physicists — Study phenomena which they themselves are participants in.

Mutability of Phenomena Being Studied

Physicists study phenomena which as far as we know never change. The structure of an atom today is the same as it was a million years ago and the same as it will be a million years from today. Further, the structure of the atom doesn’t change from one country to another or from one material substance to another. The phenomena of interest in physics are truly general. The phenomena that sociologists attempt to understand, frequently change faster than we are able adequately to describe them.

Wallace (1988) tries to counter this argument by switching from his general positivistic approach to a more relativistic one by claiming that “...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Contents
• Introduction: The Social Construction of Sociology
• 1. Why Sociology Doesn’t Make Progress Like the Natural Sciences
• 2. Why the Social Sciences Won’t Become High-Consensus, Rapid-Discovery Science
• 3. Disintegrated Disciplines and the Future of Sociology
• 4. What’s Wrong With Sociology?
• 5. Sociology in the 1990s
• 6. Progress and Cumulation in the Human Sciences After the Fall
• 7. Dilemmas of Theoretical Progress
• 8. Going Out
• 9. Sociology: A Disinvitation?
• 10. Is Sociology of Gender Stratification Parochial? The Case of Women and Warfare
• 11. Blame Analysis: Accounting for the Behavior of Protected Groups
• 12. The State of American Sociology
• 13. The Transformation of the American Sociological Association
• 14. Institutional Perspectives On Sociology
• 15. Sociology: After the Linguistic and Multicultural Turns
• 16. Writing From Sociology’s Periphery
• Contributors