Although there is a variety of ways in which sociology may be applied, as I have endeavoured to show in the previous chapter, it is plain that many sociologists, and most of those who are engaged in practical social work, think of applied sociology pre-eminently in terms of its capacity to provide (or at least to suggest) remedies for particular social evils. The attempt has rarely been made, however, to show exactly how sociological principles, concepts or data have been used, or might be used, in a direct way to solve practical problems.1 The actual state of affairs may be illustrated from the experience of the International Sociological Association in undertaking a systematic review of the application of sociological knowledge in a number of different fields.2 The authors of the principal papers found so few instances of the direct and successful application of sociology to the different problems with which they dealt, that they had to confine themselves for the most part to reviewing current research or reflecting upon methodological questions.
But even if it is impossible to point, as yet, to any problems solved with conspicuous success by means of sociological reasoning or research, it is still useful to consider what has been accomplished on a modest scale, and what may be expected and hoped for if sociological studies are intelligently developed, Let us begin by examining what constitutes a social problem. Raab and Selznick say that this is ‘a problem in human relationships which seriously
1It is worthy of note that despite much talk of an applied science there are very few books or journals which deal at all thoroughly with applied sociology. A journal actually entitled Applied Sociology enjoyed a brief existence in the USA from 1921–27. At the present time the most useful and interesting journal in this field is Social Problems (published at Indiana University by the Society for the Study of Social Problems). Among the few books which even attempt to discuss social problems in a systematic relation with sociological theory and research the work by E.Raab and G.J. Selznick, Major Social Problems (New York 1959) is outstanding despite its text book presentation. In the present chapter I shall draw heavily upon these two sources.
2See Transactions of the Fourth World Congress of Sociology (London 1959), Vol. II.
threatens society itself or impedes the important aspirations of many people’.1 They go on to say, about the first aspect: ‘A social problem exists when organized society’s ability to order relationships among people seems to be failing; when its institutions are faltering, its laws are being flouted, the transmission of its values from one generation to the next is breaking down, the framework of expectations is being shaken. The widespread contemporary concern with juvenile delinquency, for example, is only partly that delinquency is the doorway to crime or is a threat to personal safety and property. It is also a fear that society is failing to transmit positive social values to its youth… It is seen, in other words, as a breakdown in society itself.’2 On the other hand, Barbara Wootton, in her Social Science and Social Pathology,3 defines more narrowly what she terms ‘social pathology’: it includes ‘all those actions on the prevention of which public money is spent, or the doers of which are punished or otherwise dealt with at the public expense’. Her intention is to define the field of study as sharply as possible, and to avoid the difficulties which arise from divergent subjective ideas of what are to be regarded as ‘social problems’. But this definition restricts the field unduly, in that it refers only to actions, and not to situations, and includes only such actions as attract the attention of the state at a particular time. Thus it excludes from consideration many important situations and kinds of behaviour which are very widely regarded as constituting social problems; e.g. poverty, some types or degrees of industrial conflict, and in modern times, war. I shall, therefore, adopt here the broader definition proposed by Raab and Selznick.
Among the great array of problems thus defined there are two— crime and delinquency, and industrial relations—which have been investigated with exceptional thoroughness; and the research carried out in these fields reveals very clearly both the difficulties and the potentialities of sociology as an applied science. Crime and delinquency have perhaps attracted more public attention than any other problem in the industrial countries in recent years, partly because of the continued increase in their incidence, partly because other problems (such as poverty) have declined in importance. An examination of the achievements of sociological research in this sphere is greatly helped by the work of Barbara Wootton, already mentioned, which provides a very comprehensive review and
analysis of recent work. Wootton selects twenty one major investigations for consideration,1 and finds that they refer to twelve different factors as being possibly associated with criminality or delinquency: (i) the size of the delinquent’s family, (ii) the presence of other criminals in the family, (iii) club membership, (iv) church attendance, (v) employment record, (vi) social status, (vii) poverty, (viii) mother’s employment outside the home, (ix) school truancy, (x) broken home, (xi) health, (xii) educational attainment.2 She shows that none of these factors can be regarded as causes, in any strict sense, and arrives at the general conclusion that ‘this collection of studies, although chosen for its comparative methodological merit, produces only the most meagre, and dubiously supported, generalizations’.3 Similar views have been expressed by other investigators; thus John Mack observes that ‘the total of possible factors which may be specially connected with delinquency is limited only by the patience of the investigator and by the number of methods extant and professionally favoured at the time of the investigation, It is now generally accepted that all that these comparisons can establish is the fact of correlation, the fact that delinquency is frequently accompanied by defective home discipline and by temperamental instability, and by intellectual disabilities such as backwardness and dullness, and so on’.4
It is now widely held that the main positive contribution of research in this field has been to show that many popular explanations of crime and delinquency are untenable. As Barbara Wootton observes:
‘Up till now the chief effect of precise investigations into questions of social pathology has been to undermine the credibility of virtually all the current myths. Solid evidence that irreligion, or lack of interest in boys’ clubs, or life in the squalor of a problem family or a mother’s absence at work have the corrupting effects that they are said to have, or that the younger one embarks on a career of crime, the longer one is likely to stick to it, or that the delinquencies of the young are “all the fault of the parents”, or that problem families repeat themselves generation after generation—solid evidence for
1On grounds of methodological soundness.
2Op cit. Chapter III, pp. 81–135.
3Op. cit. p. 134.
4John Mack, ‘Juvenile Delinquency Research: A Criticism’, Sociological Review 3(47) July 1955, quoted by Gordon Rose, ‘Trends in the Development of Criminology in Britain’, British Journal of Sociology IX(1) March 1958.
any of this is conspicuously lacking; and any evidence that can be found has a way of falling to pieces on closer inspection.’1
Similarly, Hermann Mannheim writes, in the course of a short survey of American criminology: ‘It was not only laborious, it was also courageous and unpopular to prove that it was in fact not the immigrant but the American-born white who was largely responsible for the high crime rate; that American society rather than the Negro had to bear the blame for much of Negro crime… And, lastly, it may have required more than just ordinary courage to show, as the late Edwin H.Sutherland did, that some of the most powerful American business enterprises were in fact “habitual criminals”…’2
Such findings are useful, both in restraining public authorities from inappropriate remedial action and over a longer period in influencing public opinion, but they can hardly be said to provide a firm basis for practical measures to deal directly with the problems. Among workers in the field who are trying to enhance the practical value of their studies, two principal lines of thought may be distinguished at the present time. There are those who argue that students of crime and delinquency have so far only mapped out in a very general way the main areas of social life which produce delinquency, and that the next stage will be to distinguish different types of delinquency and to seek the specific causes of each type.3 This view implicitly re-states Durkheim’s rule that every social fact has a single cause, and that where a phenomenon appears to have several causes this is an indication that we are dealing, not with a single phenomenon, but with several distinct phenomena, each one of which has its own specific cause.4 But the rule is only re-stated, without considering its difficulties and without showing how it might actually be applied in the study of crime and delinquency. The difficulties have long been recognized in connection with Durkheim’s
1Op. cit. p. 326.
2Hermann Mannheim, ‘American Criminology: Impressions of a European Criminologist’, British Journal of Sociology V(4) December 1954.
3See Gordon Rose, ‘Trends in the Development of Criminology in Britain’, British Journal of Sociology IX(1) March, 1958, p. 62.
4See The Rules of Sociological Method, Ch. VI. Durkheim in fact gives as examples of phenomena which appear to have several causes, suicide and crime; and he argues ‘if suicide depends upon more than one cause, this is because there are in reality several types of suicide. And it is the same with crime’. The application of the rule ‘one effect—one cause’ is shown in great detail in Durkheim’s study of suicide.
own study of suicide; first, that there is something arbitrary in his distinction between different types of suicide,1 and secondly, that the causal connections which he establishes are by no means as solid or convincing as might be wished. Most sociologists today would be more inclined to accept J.S.Mill’s contention that social phenomena depend upon a ‘complication of causes’ than Durkheim’s criticism of it; and the investigator of a social problem may feel that he would be helped more by some kind of vector theory than by a theory which involves a one-to-one relationship between a phenomenon and its cause. Unfortunately, although much social analysis is couched in terms of ‘factors’, ‘forces’ and ‘pressures’, which are alleged to produce a particular happening or condition,2 there has been little progress in measuring the various acting forces with precision, in order to make prediction possible. Moreover, the forces or factors involved may be so numerous and heterogeneous that their interconnections can never be properly disentangled.
The second line of thought, recognizing these difficulties of causal explanation, proposes a different approach. It is well formulated by Barbara Wootton, in a passage which deserves to be quoted at length:
‘Much of the research reviewed in these pages is inspired by the hope that the connections between various manifestations of social pathology and other specific phenomena may prove to be a cause-and-effect relationship; and that the elimination of causes will result in a cure. It cannot, however, be said that this type of enquiry has been very successful. The generalizations that have been reached are shaky; few of them are supported by the work of any considerable number of investigators; and even in those that do recur, the quantitative variations are apt to be very large. Many of these generalizations, moreover, are quite untenable as causes… By contrast, the record of experiments in prediction is much more encouraging, and is also rapidly improving. Yet, as has already been observed, the predictive factors used by some of the most successful enquiries are obviously far removed from ‘causes’ as we have defined them. The demonstrably reliable findings of the Mannheim and Wilkins enquiry,3 which show that previous experience of an
1That is to say, the distinction seems to be determined not by observable differences in the phenomena under examination, but by some pre-conception of the possible causes.
2For example, the notion of ‘cross-pressures’ influencing the individual’s political allegiance.
3H.Mannheim and L.T.Wilkins, Prediction Methods in Relation to Borstal Training (London 1955).
approved school or of probation, together with frequent changes of job, augur badly for an offender’s post-Borstal record, can hardly be interpreted to imply that the abolition of approved schools and of the probation system, or the enactment of a law forbidding employees to discharge themselves would improve the outlook. And, although it is true that the factors used by the Gluecks1 in their predictive tables (such as inconsistent discipline in the home) could more credibly be regarded as causes of delinquency, the validation of these is still awaited.
‘The moral seems to be that it is in their role as the handmaidens of practical decision that the social sciences can shine most brightly. Prediction may be a less ambitious goal than causation, but it is certainly more often within the reach of our present capacities and techniques. Though still unable to say much about the why or wherefore of any given social events, we may yet be in a position to indicate which of a limited range of decisions is most likely to produce desired results; and the reason for this is just the fact that the range of possible alternatives is so closely limited. The factors, for instance, which throw light upon the relative success of alternative methods of dealing with particular types of offender are much more manageable than are the multifarious observations necessary to establish why people commit the crimes that they do.’2
Other social scientists have come to similar conclusions. Sargant Florence makes a distinction between ‘applied science’ and ‘operational research’. He writes, ‘When a science…is called applied it seems to assume that a body of general theory, principles or doctrine has been built up by ‘pure’ science…which is fairly certain. It is then only a question of applying the general theory by deduction to a particular case… Operational research, on the other hand, does not necessarily assume any theory or principle to be certain or reliable and deduce from it. The official definition of operational research as ‘the use of the scientific method to provide executives with an analytical and objective basis for decisions’ (First Report of the Committee on Industrial Productivity, 1949, p. 17, Cmd. 7665) clearly stresses method rather than doctrine.’3 And L. T.Wilkins, who co-operated with H.Mannheim in the prediction
1S.Glueck and E.Glueck, 500 Criminal Careers (New York 1930).
2Op. cit. pp. 323–25.
3P.Sargant Florence, ‘Patterns in Recent Social Research’, British Journal of Sociology I(3) September 1950.
study on Borstal training, has advocated an extension of prediction methods in operational research.1
It is unnecessary to choose between these two approaches in the sense of excluding one of them entirely; for research can well be pursued along both lines. But in considering their relative importance in the whole strategy of research, two further points need to be made. In the first place, operational research is not applied sociology. It is a procedure which appears to depend upon commonsense or practical wisdom, rather than upon any specialized sociological knowledge or sociological mode of thought. Its connection with sociology consists chiefly in the use of techniques of enquiry derived from sociology, and of statistical methods which have been largely employed in sociological research.2 By all appearances it could well proceed in the absence of any sociological (or psychological) theories or descriptions. But this is unsatisfactory, both theoretically and practically. For in this case it is not, as Barbara Wootton claims, the social sciences which ‘shine most brightly’ as handmaidens of practical decision; it is only certain techniques of investigation which do so, while the social sciences (as bodies of theory and description) play little part. Even in the matter of practical guidance the contribution of operational research, at least in the form of prediction studies, is not quite so promising as has been claimed. For the ability to predict success or failure in a course of action does not always imply the ability to solve the social problem to which the action relates. The study of Borstal training, which Wootton particularly cites, is a case in point; for the social problem is constituted by the ‘failures’, and the successful prediction of failure provides no clue to its causes or, consequently, to the remedies.
Secondly, in the attempts to formulate causal explanations of crime and delinquency, extremely heterogeneous factors have been juxtaposed or associated. Thus, explanations may be offered which refer to a range of individual psychological characteristics (temperamental instability, mental backwardness, etc.), and to a variety of social circumstances (divorce, parental discipline, criminal environment, etc.). In this particular field of study the general problem of the relation between psychological and sociological explanation is very acutely posed; but there are few works in which the problem
1‘Some Developments in Prediction Methodology in Applied Social Research’, British Journal of Sociology VI(4) December 1955.
2It will be shown later that ‘applied sociology’ is often conceived in terms of a method of investigation.
is even clearly formulated, and none, to my knowledge, in which a plausible theoretical model encompassing both sets of data has been constructed.
The second field of applied research which I propose to consider is that concerned with industrial relations. The principal aim of research has been to discover the causes of industrial conflict, in the individual enterprise and on a national scale, and of other impediments to high productivity such as high rates of absenteeism, illness, labour turnover, etc. The same difficulties of causal explanation as in the study of crime and delinquency have been encountered, but there are additional difficulties which arise from the ramifications of industrial problems.
Before considering these questions I should like to take up the point made earlier, that the principal contribution of sociology is generally seen as consisting in skilled investigation. This is very apparent in industrial sociology. The authors of a recent survey of sociological research in industry1 observe that ‘it is often the case no doubt, that to the firm which calls upon his services, the sociologist appears less as the possessor of a particular kind of knowledge which is capable of application, than as a person trained in the use of certain tools (interview, etc.), or indeed as one who has the advantage of appearing neutral to everyone concerned’.2 This is very close to the view advanced by Barbara Wootton according to which the contribution of the social sciences to practical life is found in a scientific attitude and in the use of certain techniques of enquiry. This conception is very prevalent. In the sphere of medical research, for example, it has been claimed that ‘Many physicians approach sociologists, if they approach them at all, as I did, thinking of them as people with technical skill in interviewing and questionnaire construction’.3 But a contribution of this kind can only be regarded as a very minor part of applied sociology, more especially since these techniques of investigation are common to all the social sciences. Clémens and Evrard recognize this, and while admitting that there are few instances of the successful application of sociological principles to solve specific problems, they make a case for applied industrial sociology in respect of its contributions to improved organization. The contributions are of two kinds: first, studies of
1René Clémens et Pol Evrard, ‘La connaissance sociologique et son application à la vie industrielle’, Transactions of the Fourth World Congress of Sociology, II pp. 1–12.
2Op. cit. p. 3.
3Mary E.W.Goss and George G.Reader, ‘Collaboration Between Sociologist and Physician’, Social Problems 4(1) July 1956.
the social structure of business enterprises which make possible improvements in the authority system, the system of communication, the constitution of working groups, etc.; and secondly, training courses for managers and supervisors. Clémens and Evrard regard the latter as the more successful development: ‘It is in the sphere of training for those who hold positions of authority in industrial life, that the most serious attempts to apply sociology in a systematic way are to be found. The ever more numerous training courses for managers, technicians and supervisors, or for trade union officials, constitute an important factor of social change in modern industry…’1 But this favourable assessment would not receive universal assent. As William F.Whyte observes: ‘To be sure, important changes in human relations have taken place in industry, but there is little reason to believe that they have come to any considerable degree through research and through training that has been based upon such research. There have been hundreds of thousands of human relations training programmes in industry. To my knowledge, only two of them have been subjected to the sort of solid research evaluation which would measure their effectiveness. In one case, an International Harvester programme, workers reactions to the foreman who had been trained were slightly more negative after the programme than they had been before. In the other case, Detroit Edison, there was a loss registered in one division which was more than counterbalanced by a gain in another’.2
The inadequacy of applied research in industry is due, Whyte suggests, to the failure to take account of the wider social context of industrial problems. Similar criticisms have been expressed more fully by Georges Friedmann, who has made an outstanding contribution to the study of problems of industrial work in relation to different types of economic and political system.3 Some of the principal difficulties of applied research in this field arise, then, from the fact that in investigating the causes which have produced a practical problem, we encounter a network of interrelated factors which extends far beyond the limited situation in which the problem itself occurs. The study of obstacles to greater productivity, of resistances to technological change, of ways to increase the efficiency of management, or of the prevention, mitigation or peaceful settlement
1Op. cit. p. 3.
2William F.Whyte, ‘Problems of Industrial Sociology’, Social Problems 4 (2) October 1956, pp. 148–60. The article as a whole sets out admirably the problems of applied research and makes some excellent suggestions for more effective studies.
3See especially his Industrial Society (Glencoe 1956).
of industrial conflicts, leads inescapably to the consideration of much broader questions concerning the property system, social stratification and occupational mobility, family structure, ideologies and cultural traditions.1 Furthermore, when this intricate web of circumstances is explored it may disclose not only the ‘complication of causes’ in the production of a specific social problem but also interconnections among social problems such that the solution of one problem may require the prior solution of others, or that the solution of one may impede the solution of another or even create a new problem.
In seeking to apply his knowledge, therefore, the sociologist has to contend with two kinds of difficulty: first, that of establishing precise causal connections, and secondly that of delimiting a problem without doing violence to its relations with important general features of the social structure and culture. The consideration of these difficulties leads, I believe, to a recognition that there are different types of social problem, varying both in their importance and in the extent to which they can be resolved. Some social evils are, in the strict sense, ineradicable. Durkheim argued that crime is a ‘normal phenomenon’ in human societies which could only be eliminated by social controls and sanctions so repressive as to destroy many other cherished values; and further, that each type of society has its own ‘normal’ rate of crime. The high rates of crime and delinquency in modern industrial societies may therefore be the counterpart of the relative laxity of social control in these societies. Similarly, high divorce rates may be explained, in part, as the outcome of the pursuit by individuals of other ends—personal freedom, an ideal romantic love—which are in themselves good. In the under-developed countries, many social problems have to be seen at the present time as concomitants of the disturbing processes of industrialization and urbanization. It should not be concluded that in these cases social research is futile. Sociological studies may encourage a more realistic approach to such problems, and in particular restrain those immoderate moral denunciations which often exacerbate the difficulties. They may also, at the least, suggest means of diminishing the evils without harm to other social values, and more effective ways of dealing with their consequences. Thus, modern criminology is
1Although such ramifications are particularly evident in the case of industrial problems, they can also be traced in other areas. The rates of crime and delinquency in a society may be influenced by a dominant way of life which emphasizes very strongly economic competition and individual achievement judged in terms of wealth, or by inadequate provision for public education.
concerned to discover ways of reducing crime and delinquency, and more successful treatment methods for offenders. Studies of marriage and divorce aim to provide knowledge of the factors predisposing to marital conflict and divorce, which can be used, in marriage counselling and in other ways, to reduce the incidence of such problems or to find solutions to them without breaking up the family.
There are other social problems which can be solved, or which constitute such a grave danger to human society that a radical solution has to be sought. In the first category comes the problem of poverty in economically developed societies. Here, the solution requires from the sociologist primarily an investigation of the facts, in the course of which one or a small number of factors may be revealed as the cause of the problem. In other cases, the causation may be much more difficult to discover, but sociological research may at least eliminate some popular but erroneous beliefs about the causes, and provide a sufficient description of the problem situation to be a useful guide in dealing with it. Experiments with different remedial measures and careful evaluation of the results, as well as operational research, may make it possible to resolve or control the problem, even though we remain largely ignorant of its causes. In the second category, of supremely dangerous problems, the pre-eminent example, in this age of nuclear weapons, is war. No one is likely to suppose that sociologists alone, or sociologists and psychologists together, will provide a universally acclaimed solution. Equally however, no-one will contest that a war fought with nuclear weapons would be an immense and probably final disaster for mankind, that sociological and psychological research can at least contribute significantly to understanding how crucial situations of tension and conflict develop and so help responsible leaders to avoid them, and that sociologists ought consequently to make an exceptional effort to investigate the problems of war and peace, and to disseminate their findings as widely as possible. It is sad that so few have undertaken the task,1 even though it is so formidable, so exposed to the incursion of political strife, and so little assimilable to the neat research designs of much current enquiry.
This last example brings out very clearly a feature which I have tried to bring into relief at various places in this and the preceding chapter. The sociologist can rarely solve a problem directly or
1Two important contributions are those of Erich Fromm, May Man Prevail? An Enquiry into the Facts and Fictions of Foreign Policy (New York 1961); and Raymond Aron, Paix et Guerre entre les Nations (Paris 1962).
propose a policy which is exactly appropriate, even when he is sure what is wrong; for every solution of a problem and act of policy is a political decision. It expresses the resolve of a social group to maintain or to change a particular way of life, and to act in accordance with certain social ideals. The sociologist may supply information, elucidate the context of problems, point to causes or conditions, indicate the advantages and costs of alternative courses of action; in the longer term, his studies may, and I believe do, influence social ideals themselves. But in the last resort political decisions rest upon judgment, or political wisdom, and upon interests. To suppose otherwise, and to assign sociologists the role of philosopher kings in modern dress, is to revert to the delusions of Comte’s positive politics.
It is perhaps well, at the close of this book, to dispel any possible misconception that the value of sociology culminates in its various applications to practical life which we have just been considering. To the question, “What is the use of sociology?” I would answer rather that it widens our sympathies and imagination, and increases our understanding of other human beings outside the narrow circle of our own time, locality, and social situation, than that it simply provides means to discover the remedies for present ills. But these purposes are not radically antithetical and they are perhaps of equal practical significance in the long run. Most sociologists would feel that in all spheres of their work they were making some contribution to the improvement of social life.
As I have indicated in the text, the literature of applied sociology is not abundant. The reader can usefully consult the following books and journals:
E.Raab and G.J.Selznick, Major Social Problems (New York, Row, Petersen & Co., 1959).
Transactions of the Fourth World Congress of Sociology, Vol. II (London, International Sociological Association, 1959).
R.M.Titmuss, Essays on ‘The Welfare State’ (London, Allen & Unwin, 1958).
Social Problems (published by the Society for the Study of Social Problems, 1953 onwards).
The best survey of applied social research, mainly in the field of crime and delinquency, is Barbara Wootton, Social Science and Social Pathology (London, Allen & Unwin, 1959).
There has been little writing on the sociological aspects of planning. The symposium on social planning in Transactions of the Fourth World Congress of Sociology, Vol. II, provides a useful introduction. Different aspects of planning and social policy in India are discussed in the following books:
S.C. Dube, India’s Changing Villages (London, Routledge & Kegan Paul, 1958).
Report of a Seminar on Some Aspects of Social Planning (Agra, Institute of Social Sciences, 1960).
R.N.Saksena (ed.) Sociology, Social Research and Social Problems in India (Bombay, Asia Publishing House, 1961).