Reflections on the Relation Between Sociology and Social Policy

Critical Rationalism

In the 1940s, Karl Popper (1966: ii, 231) developed a perspective that he named critical rationalism. In the social sciences, this philosophy sees the growth of scientific knowledge as starting from attempts to solve intellectual problems. Its focus is on the ways by which problems are explained. The lessons learned are added to existing knowledge in a process of cumulation that in turn generates new problems. Popper did not explore the ways in which improved explanation may depend upon the elaboration of a specialised language. Nor have critical rationalist studies paid sufficient attention to the sources of the hypotheses that are selected for testing. Any appreciation of how knowledge grows has to start with problem selection.

Max Weber’s methodological work can be seen as pioneering the same perspective. He maintained that while ‘it is the investigator and the prevailing ideas of the time that determine what becomes the object of investigation’ (2004: 383), once that decision has been taken, the investigator’s priority must be the ‘need for conceptually ordering empirical reality in such a manner as to lay claim to validity as experiential truth’ (2004: 365). This need is for a form of proof that would be acceptable to a Chinese. This was Weber’s way of declaring that sociological conclusions must be formulated in a manner free from cultural assumptions, whether of geographical or historical origin. The individual sociologist was responsible for eliminating such forms of bias.

Teaching in the social sciences sometimes reflects the first part of Weber’s doctrine, in that a teacher may discuss the research findings that he or she considers most relevant to the political priorities of the day. With respect to the second part, a teacher may concentrate upon the findings of the research that, in his or her view, contributes most to the student’s education as a social scientist. Some students learn lessons about research in their undergraduate years that they carry with them into subsequent careers as researchers and teachers. Their view of their discipline influences their selection of problems to investigate.

The Two Disciplines

Sociology and social policy address different kinds of problem and they seek different kinds of knowledge. Sociology aims to account for observations about social practices and relations; because it seeks new knowledge in the form of better explanations, it often has to find improved or new concepts. ¹ Work in social policy seeks practical knowledge useful to the improvement of existing practice. The distinction between the two fields is not invalidated by the recognition that some researchers have worked on both sides of the dividing line. ²

In the study of social policy, research reports for the most part employ ordinary language, whereas sociology, if it is to account for observations of high generality, has to discover and employ culture-free concepts. The distinction between the two disciplines is therefore in significant part a conceptual distinction, not one that can justify claims to intellectual territory. As is generally recognised, research reports may develop sociological theory in the course of explaining observations that have practical significance. For example, when Durkheim set out to explain variations in the suicide rate, he gave his book the subtitle A Study in Sociology (1962) because he was presenting a case for sociology as a distinct discipline; he had to introduce some new concepts (like anomie) to account for the regularities he had uncovered. At the same time, his book demonstrated that the new theory had valuable applications.

Studies in social policy address circumstances in particular countries. They focus on state institutions and the possibilities of improving their operation. If they are comparative, the comparison is employed to draw conclusions that could be applied in particular states. Studies in social policy seek to build practical knowledge and aim to present it in terms that will be understood by those who might learn from them.

It is relatively easy to find a research problem in the field of public policy. The mass media highlight such problems every day. Necessarily, they describe them in ordinary language. Words such as anti-Semitism, ethnicity, homophobia, Islamophobia, multiculturalism and racism are currently essential to political discussion in the English language. In ordinary language, the meanings of words are decided by their daily use in many different kinds of situation and in changing circumstances. The words used may therefore have many different shades of meaning. To discover which usage is considered correct or appropriate to the circumstances, the inquirer consults a dictionary.

Teaching and research in social policy has to rely on ordinary language conceptualisations because its product is addressed to those who might vote in favour of its recommendations or have to implement them. The case studies in the book The Blunders of our Governments, by two political scientists (King and Crewe, 2013), provide striking examples of the deficiencies in the knowledge of politicians and administrators about the problems that were created as a result of the policy changes they introduced, and the obstacles to better utilisation of public knowledge.

Problem-finding in Sociology

The hold that ordinary language conceptions have upon thought is one of the reasons why it is usually more difficult to find a good research problem in sociology. The formulation of such a problem may demand fundamental criticism of ordinary language words and their replacement by words that have places in a technical vocabulary. In a technical language, the meanings of words are also decided by their use, but that use is strictly controlled. In experimental research, the attempt to check someone else’s findings depends upon replicating a procedure previously used, and upon employing standard definitions. The best definition of a concept is the one that proves to have the greatest explanatory power. Language in the world of theory, whether experimental or not, strives to be context-free, to be addressed to whom it may concern.

Advances in knowledge about such matters often depend upon new uses of existing words, and sometimes upon the invention of new words. Innovations of this kind have met resistance. Thus, the second edition of Modern English Usage (Gowers, 1965: 569–570) introduced a new entry on Sociologese. This contended that:

Sociology is a new science concerning itself not with esoteric matters outside the comprehension of the layman, as the older sciences do, but with the ordinary affairs of ordinary people. This seems to engender in those who write about it a feeling that the lack of any abstruseness demands a compensatory abstruseness in their language.

It quoted passages from sociological writing about industrial relations, class differences in speech, and family life. The editor of the revised third edition considers that Gowers’ examples are ‘worth preserving’. He might have noted that much of what earlier was described as sociologese has since passed into ordinary language usage. It has been remarked that ‘a considerable part of Sociology consists of cleaning up the language in which common people talk of social and moral problems’ (Hughes and Hughes, 1952: 131), but it goes further than that. New words may be needed to introduce new knowledge, and those words may not remain novel. Words and expressions such as electoral swing, inflation, moral panic, presentation of self, role model, socio-economic status, etc., are now part of ordinary language in the English-speaking world. ³

Halsey’s History of Sociology in Britain focused on institutional expansion. From the perspective of critical rationalism, this approach needs to be complemented by study of the ways in which institutions cultivate the growth of knowledge and how successful they are in doing so. Academic life is organised to reward researchers for original contributions, such that, after the researcher’s findings have been made known, the stock of knowledge is greater; in some perhaps tiny but maybe important measure, it is greater than it was before. Universities also reward students for demonstrating an understanding of how knowledge can be won. Sociologists may remember the observation of one of their founding figures, that ‘If there is such a science as sociology, it can only be the study of a world hitherto unknown’ (Durkheim, 1962: 310), but they also know that the focus on publication responds to career structures as well as to academic vocations, and that new knowledge is not always appreciated.

Future authors could well examine particular fields within sociology. They could inquire into what was known at an earlier time, such as at the first meeting of the Sociological Society in 1904, or at the foundation of the BSA in 1951. What was then known about, for example, the organisation of industry, deviant behaviour, the social significance of the biological differences that were then identified as those of race and sex? The list continues: social stratification, the practice of religion, of education, mass communication, sport, and many more specialisms. Studies of their intellectual history could identify ways in which good problems have been found. Only rarely are these properly presented in textbooks.

The criticism of sociologese started with a complaint about use of the expression ‘relatively unstructured conversational interaction’ in the study of industrial relations. A review of the growth of knowledge in this field might well highlight the advance that was embodied in Roethlisberger and Dickson’s Management and the Worker of 1939. It showed how a distinction between formal and informal organisation helped explain a lot about conversational interaction in industrial settings. It did not consider how the formal organisation was affected by technological innovation and the resulting change in the market for the products in question.

The next step depended on the exercise of sociological imagination. Tom Burns has described his puzzlement when, interviewing managers in the new electronics industry, he found a discrepancy between their conceptions of their jobs and the evidence about how they actually spent their time. Generalising about this, he concluded that ‘sociology seems at any one time to be pursuing not so much the right kind of knowledge as the right kinds of question … questions do not suggest themselves … They arise from doubt’ (Burns, 1995: 167–170). In this case, they arose from his doubt about whether what the managers regarded as an abnormal departure from the pattern of activities as they should be, was not in fact a predictable response to new conditions.

Burns’ recognition of a discrepancy between the prevailing model of formal organisation and the results from his examination of actual practice reminded him of Durkheim’s distinction between the mechanical and the organic division of labour. Reporting on The Management of Innovation, he and his colleague described mechanistic and organic styles of management as associated with two kinds of production, an old style adapted to a stable market and a new one that had to respond to continuous technological change. His experience illustrated the way that analogy with existing explanations can stimulate the discovery of new explanations.

Sociological inquiry need not be tied to traditional institutions. One field within social policy took its orientation from the criminal justice system. Its focus was displayed in the title of the British Journal of Delinquency, published from 1950. In 1960, it became the British Journal of Criminology. When, later, a new generation took up the study of deviance, comprehending all forms of deviance and not just those defined as criminal, they created a new body of sociological knowledge. ⁴

In its day, the best of sociology textbooks was Kingsley Davis’ Human Society. In it, he criticised the assumption that the differences in the behaviour ascribed to men and women sprang from biological differences (Davis, 1948: 99). He went no further. Sociological knowledge about the nature of these differences was transformed once they were represented as differences of gender. Another new body of distinctively sociological knowledge was established.

The study of social stratification came alive in a new way when Davis, together with Moore, maintained that it was an unconsciously evolved device by which societies insure that the most important positions are conscientiously filled by the most qualified persons. The political implications of such a thesis brought it intense critical attention. An intellectual history of the subsequent growth in empirical knowledge about stratification in industrial societies might document how differences of socio-economic status and social mobility came to be identified, measured, related to other social institutions, publicised, and later passed into public knowledge. Similarly, in some of the other specialised fields mentioned, historical studies might throw new light upon the ways in which sociologists have discovered theoretically rewarding research problems.

Practical and Theoretical Vocabulary

Knowledge can cumulate more effectively when it can draw upon a specialised vocabulary. Some authors have discussed the difference between the two kinds of language, or vocabulary, as a contrast between folk and analytical concepts. A simpler distinction is that drawn by American anthropologists between emic and etic constructs. An everyday example of the difference is that when a patient goes to a doctor for treatment, he or she reports his or her symptoms in ordinary language using emic constructs. The doctor makes a diagnosis, drawing upon technical knowledge expressed in etic constructs. In one formulation, emic constructs are accounts expressed in categories meaningful to members of the community under study, whereas etic constructs are accounts expressed in categories meaningful to the community of scientific observers (Lett, 1996). The constructs in the technical language may be called concepts, recognising the distinction in logic between denotation and connotation. Ordinary language constructs denote things listed in the dictionary; in addition, many of them have further associations for individuals, varying from one person to another. This makes them multi-vocal. It can be useful to distinguish concepts from ordinary language constructs because concepts have connotations for specialists that are lacking in ordinary language usage. They provide foundations for theories.

The difference between constructs and concepts arises because they serve different ends. Ordinary language facilitates the business of everyday life and the accumulation of practical knowledge expressed in emic constructs. Technical language makes possible the accumulation of theoretical knowledge expressed in propositions using etic constructs, or concepts. This parallels the distinction between applied science and pure science. It is a reformulation of Popper’s (1957: 26–34) distinction between essentialism and nominalism. The constructs of ordinary language have to be multi-vocal, but improved explanation demands use of univocal constructs, corresponding to methodological nominalism. The multi-vocal character of ordinary language words means that any definition is likely to search for something essential to all the meanings. In everyday usage this does not matter, but any attempt to formulate a theoretical proposition that does not isolate a singular meaning will generate confusion. ⁵

Explanation

The proposed distinction between sociology and social policy is the more useful because it recognises that, in seeking to cumulate knowledge, sociology has continuously to restrict its vision, concentrating on improved explanation. The distinction therefore embodies an intellectual criterion for assessing work and trends within sociology as a discipline. Knowledge about social policy has to be more diverse and cumulation is not so high a priority.

From the standpoint of critical rationalism, if sociological teaching starts from the discussion of concepts it is back-to-front. Knowledge grows more rapidly if the sociologist first identifies a puzzling observation and then seeks the best explanation. This entails a differentiation of the explanandum, the things to be explained, and the explanans, that which explains the explanandum. To start with a discussion of the concept is to stay within ordinary language and its explanans, instead of starting from the explanandum and assessing the value of alternative explanations.

If one explanation is alternative to another, it should be possible to express each alternative as series of propositions, so as to bring out the assumptions on which each rests and the precise nature of the differences between them. One merit of a legal education is that it obliges students to learn the arguments for and against particular propositions so that they can speak for either side in a dispute. This encourages a focus on reasoning and on alternative modes of utilising evidence. Students of social policy may be taught along these lines because they may, one day, have to propose a particular measure and will need to be aware of the contrary arguments with which they will have to contend. Students of sociology seem not to be taught in this way. Instead, much attention is paid to the thought of particular theorists.

If one explanation is confronted with another, it may be possible to show that one of them is wrong. This can be instructive. Indeed, Popper (1969: ix) contended that ‘all our knowledge grows only through the correcting of our mistakes’. The formal expression of an argument (which in some fields is improved by expression in mathematical terms) promotes clarity in the evaluation of argument.

Confrontation of alternative explanations can take the form of a debate. In this event, there may be a sequence starting from a motion, or proposition, advanced by its proponent, the challenge of a contrary argument, and a general discussion ending in a vote. What passes as ‘debate’ in current sociology is often the rehearsal of competing opinions between persons who do not engage with one another’s reasoning or use of evidence. ⁶

Conclusion

Discussion of the differences, in research and teaching, between sociology and social policy as distinct disciplines, draws attention to the distinction between the search for new knowledge and the better utilisation of existing knowledge. Studies in intellectual history can uncover the processes by which new knowledge has been won.

Arguments about social policy have to be capable of expression in ordinary language. If sociology is seen as a search for new knowledge, this accounts for the centrality of the search for concepts of greater explanatory power. The acceptance of such a conclusion offers a basis for the criticism of sociological practice.