The deviance of women: a critique and an enquiry

Abstract

The purpose of this paper is quite simple: it is to focus on an obscure and largely ignored area of human behaviour, namely deviance in women. It is suggested. that the problems associated with this neglect may, in themselves, be interesting from a sociological point of view. that the topic has both intrinsic interest for the sociologist today and that it has wider relevance to, for example, aspects of social structure. that, in so far as it is an example of possibly curious counter-tendencies to certain widespread trends in deviance, it may be of especial interest. The deviance of women is one of the areas of human behaviour most notably ignored in sociological literature. This apparent lack of interest is remarkable for a number of reasons: deviance in general (i.e. in practice either male and female deviance together or simply male deviance alone) has long aroused considerable interest. Since Durkheim (1938: ch. 3) argued that ‘deviant’ or ‘pathological’ behaviour has social as well as individual aspects, the sociology of deviance has been an important and developing field of sociological theorizing and enquiry. Therefore the general unconcern with the potential deviance of approximately half the members of any human society is surprising. Interest in the changing position of women as it has affected marriage, the family and the division of roles within it has been considerable and has been productive of a vast range of studies, from, moreover, almost every known ‘type’ of sociologist: from American functionalist theoreticians to the more empirically-minded British.1 As will be indicated below, differences between the patterns and manifestations of male and of female deviance were long ago observed (see Quetelet 1838; Durkheim 1897: bk. 3, ch. 2); they were observed, moreover, to differ with that kind of regularity and uniformity which normally attract the attention of the social scientist (Radzinowicz 1937). Nevertheless the focus of research has been very much away from this particular area, so that a wide-ranging selection of readings on the sociology of crime and delinquency, published in 1962, did not contain one extract on female criminality (Wolfgang, Savitz and Johnston 1962). Despite this gap, this has remained an often crucial, although unacknowledged dimension of deviance. Of course, one can advance simple, straightforward reasons for this gap in sociological literature. Women appear to have low rates of participation in deviant activities as measured by such indices as official criminal statistics (Radzinowicz 1937; Mannheim 1965: vol. 11, ch. 26), the range and number of their socially disapproved sexual outlets (Kinsey et al. 1953; Schofield 1965), suicide rates (Durkheim 1897), involvement in activities labelled as vagrant (see, e.g. O'Connor 1961), and, although with certain provisos (but see Little 1965), when defined as mentally ill. (Perhaps it should be made clear at this point that the activities here selected are not put forward in any sense as integrated components of a comprehensive definition of deviant behaviour; they simply represent a range of categories, commonly thought of as deviant.2 They thus correspond to the concept defined by Howard S. Becker (Becker 1963: ch. 1): deviance is not a quality of the act the person commits, but rather a consequence of the application by others of rules and sanctions to an ‘offender’. The deviant is one to whom that label has successfully been applied; deviant behaviour is behaviour that people so label.3 These modes of behaviour also correspond to Lemert's secondary deviance in that they largely receive institutional reinforcement as deviating behaviour and may become deviant careers (Lemert 1951, 1967: ch. III) for individuals resorting to them. Not only do woman appear to be remarkably conformist,4 they seem to have consistently lower rates of deviance than men, with a fairly constant ratio which remains despite fluctuations in rates over time. (There is some suggestion of a diminution of this ratio in recent years, but data for the U.K. do not, on the whole, support this view.) It is perhaps therefore inevitable that male delinquents and deviants should be studied in preference to females: there are more of them and they thus have greater social visibility, are more accessible because more widespread, and are much more likely to appear – and be defined as – a pressing social problem. The last mentioned factor will ultimately affect the research funds available for a particular type of project.5 These reasons to some extent explain the concentration on male deviance, but not the almost total exclusion of studies of females from serious literature. One might argue that there is a strong element of pragmatism in this (as in so many other) areas of sociological research. Current focal concerns of society and especially of its rulers and policy-makers have a way of becoming the consuming interests of sociologists.6 (Note the number of sociologists in the U.K. listed in a recent survey as working on the sociology of education and compare this with involvement of successive governments in educational research and reform.) Certainly students of the sociology of deviance have been often closely (often inevitably, for purposes of data collection) involved with the operators of the social control system. At the very least most workers in this field tend to approach their subject with the dual purposes of investigating then eradicating what they term social ‘disorganization’ or ‘pathology’ (see Soc. Rev.: vol. 14 no. 3). Given this orientation amongst social pathologists, it is interesting to observe, as Professor Mays (Wright Mills 1942) did recently in a critical review, that concern over a subject – and notably that of juvenile delinquency – is in almost directly inverse relationship to the efficacy of policy. It may well be that the very formulation of a mode of behaviour as a ‘problem area’ for study may draw public and press attention to it and hence, through greater likelihood of reporting, stepped up police vigilance etc., immensely increase its dimensions; it has been suggested (Cohen and Rock 1970) that in the case of the ‘Teddy boys’ of the nineteen-fifties the ‘menace’ existed in the public mind long before any activities of the teenagers concerned warranted it.7 One need not go as far as to argue that high rates of sociological interest in a problem area produce appropriately high ‘problem situation’ responses; but one might well be forgiven for wondering whether the deviance of women is a non-problem both to the social scientist and to society in general, because so little effort has been devoted to studying it. On one of the very few occasions in recent years that the deviant behaviour of women became a matter for public concern, the reaction of the committee set up to investigate it was to suggest that the social visibility of the activities should be so reduced that the public concern could largely cease (The Departmental Committee on Homosexual Offences and Prostitution 1956), although the deviant behaviour would continue.8 As was suggested earlier, when certain factors are taken into account, the picture of deviance amongst women, certainly in the U.K. since the early years of this century, has been an increasingly calm and conforming one. Of course, this statement must be heavily qualified, not least by the view that deviant women may have benefited far more than their male counterparts by the extension of the concept of the sick role and its more widespread application, in that women who might once have been adjudged ‘delinquent’ may now be defined as ‘sick’ and are hence excluded from the population of ‘deviants’. Nevertheless, whatever the caveats it is broadly true to say that, for example, the number of women convicted for offences of drunkenness has declined very markedly since 1900, and much more rapidly than has the equivalent rate for men.9 Similarly, it seems to be widely agreed that the numbers and proportion of women involved in prostitution has declined. Undoubtedly the legislation introduced after the Wolfenden Committee reported was aimed at minimizing the social visibility (and hence society's awareness) of prostitution, but this decline seems to be a common international trend.10 Where indictable offences are concerned, the picture is more complex, but it appears that the numbers of women convicted for, for example, violent offences against the person tended to decline between the wars and began to rise again only after 1945. Rates of convictions of women for indictable offences have increased since the war, but for most categories of offences the proportionate rise has kept pace with the parallel rise in rates of males convicted although rarely has it been as high. This suggests several interpretations, among them that increased willingness to report offences, greater police efficiency in certain directions and a decrease in tolerance of crime amongst the public have affected female offenders only as much but no more than males; again what may either be a general increase in the opportunities for certain types of crime11 or simply an overall rise in criminal activity has involved women to no greater extent than men. The most curious feature of this analysis is that one would expect a more than proportionate rise in the conviction rates for women over this period; as Barbara Wootton has pointed out (Baroness Wootton of Abinger 1963), given that the participation of women in social, business and industrial life has been increasing steadily, one might expect a commensurate development in their criminal activities. That this does not appear to be the case reflects significantly upon the deviant behaviour of women and, perhaps more importantly, upon the societal reaction to it. In this context of societal reaction to female deviance the legislation based upon the recommendations of the Wolfenden Committee (Street Offences Act 1959) was very curious. Basically it rationalized and strengthened statutes against open soliciting, by abolishing the requirement of proven ‘annoyance’ as a basis for prosecution; yet it retained the term ‘common prostitute’, thereby preserving a process of damning the accused before conviction, let alone sentence which is, to say the least, unusual in English Law.12 Thus, contrary to most modern trends in the growth of devices for labelling deviants and the increase in the number of categories available,13 although the old style term was retained, its potential use was much reduced. For there seems to be widespread agreement that an effect of the 1959 Act was by ‘driving’ prostitution off the streets (and indeed, any public place) to make it more difficult to observe and check trends and also to ‘catch’ prostitutes and reform them. This was anticipated in the Report, as it was that middlemen, taxi-drivers, hotel porters, café proprietors, would enter the ‘game’ and make matters more complex and difficult to control.14 That a society is prepared, while keeping its formal, deviant-defining structure to reduce the toll of those so defined and, indirectly, to lose some possibility of controlling them, is surely remarkable. Many more notable features in the trends of delinquency reported amongst women can be observed – for instance, the sharp rises during the two world wars. Other forms of deviance, while more difficult to study from this point of view, seem to display also certain interesting and often unexpected characteristics: thus it has been suggested that the rising rates of illegitimacy are especially amongst middle class girls (see e.g. Vincent 1961; Roberts 1966) and this may in part represent a deviant phenomenon of some dimensions – in terms, for example, of its rejection both of the accepted sexual mores of the group and its ‘rational’ ethic in regard to birth control. But the most remarkable common feature of all the varied data on female deviance is the way in which it largely lacks consideration in the appropriate literature. Yet by any standards there would appear to be sufficient grounds for such study, either the pragmatic ones of concern for social problems, or the simple natural-historical one of wishing to observe, measure, categorize and discuss. The answer to this mystery seems to lie not so much in the reasons listed above but in the fact that most theories of crime or deviance within a sociological framework have to make passing reference to female behaviour. Most studies at least begin with the framework imposed by the legal system, which imposes exact definitions of categories of behaviour, and these are often treated as discrete and meaningful groupings.15 Except for certain specific instances, such as rape, most legal provisions are considered as applying with equal force to both sexes.16 Behaviourial categories not dealt with by law usually follow parallel forms in referring both to male and female deviance from them as deviant, (although perhaps not always treating the misbehaviour of the two sexes as equally heinous). Any hypotheses put forward, any theories offered as explanation or for understanding deviant forms of behaviour are thus bound to take account of both male and female behaviour. Given the ubiquitous nature of the sex differential in crime and in deviance in general, any generalizing theory must point to factors which operate with very different effects upon men and upon women or, alternatively, that their like deviance is dealt with in very different manners by facets of the social system.17 Indeed whether a particular theory explains feminine delinquency as well as masculine and offers within its framework an explanation for the apparent difference in rates is often a useful touchstone of its validity. Most sociologists working in this field are aware, with perhaps some discomfort, of the sex based variation in modes of, and participation in, deviance and do try to allow for it, however awkwardly. The very stability and ubiquity of the sex differential in crime rates and other forms of deviance already instanced have meant that sociologists involved in the study of deviance have had to put forward explanatory theories based on the male and on masculine forms of acting out behaviour in a social frame – whether the behaviour was genetically determined, culturally framed or an adaptation based on responses to dysfunctional features of industrial/capitalist society. The female side, since it loomed much less significantly, was fitted in as an aspect of such explanations. Early theorists of crime and typologists of criminal behaviour, such as Lombroso, tended to suggest modes of behaviour not formally recognized as crime, notably prostitution, as available to women, but not men, hence their lower participation rates in ‘normal’ crime. More recent approaches have, like Lombroso, seen deviant behaviour as a single meaningful category of action, but have insisted that in practice the sexes choose different responses or ‘outlets’ for deviance because of socially prescribed focal concerns and structural characteristics of the social framework. The following brief outline of a few of these approaches will indicate some of their weaker points and suggest that their inadequacies stem from a fundamentally awkward and misconceived approach to the subject. One major grouping (although it includes Pollak, a more recent writer) may be labelled the ‘female iceberg’ theorists, since they deny our current perception of the deviant situation and suggest that women are only too deviant but in a different, or a more successfully concealed, manner (Lombroso and Ferrero 1893). Lombroso and Ferrero argued that women criminals are essentially occasional rather than habitual, although they, like their masculine counterparts, had certain allegedly atavistic features, notably unfeminine features and build and dark masculine hair (sic). Lombroso also argued (Lombroso 1911) that women would appear to be as criminal as men were their prostitutive activities included in the criminal statistics. In other words, prostitution is the female functional equivalent of crime. This view may be challenged on many counts: first and simply that prostitution appears to be in secular decline, while ‘male’ crime manifestly is not. Secondly, if female behaviour as in prostitution is not characterized as crime (although being, as it were, an ethical equivalent) why is this so? Surely a topic of considerable interest in itself. Thirdly female delinquency, as defined by Lombroso, consists of acts by definition involving men, in fact many more men than women.18 Fourthly, as Davis (op. cit.) shows, prostitution flourishes when family ties are strong and the status of women, especially wives, is low. Hence the decline of prostitution in modern society – with the emancipation of women and rise in status – to be replaced by ‘free, mutually pleasurable intercourse’. Can this also be defined as deviant? Finally, of course, there is the question of how the facts of female deviance, particularly the predominance of property offences, notably larceny, fit into this view. Otto Pollak (1950) saw the problem of ‘explaining’ female deviance in terms of feminine deviousness. Like Lombroso, he argues that much female deviance goes unrecorded, but unlike Lombroso he claims that women commit the same types of offences as men – and indeed have certain special opportunities for theft and murder – but that they are both equipped biologically to dissemble and socialized into so doing. They thus deviate as much as males but take advantage of their talents of dissembling and concealment and their positions in home and family to avoid discovery, and hence social awareness, of the deviant behaviour. This interpretation, although ingenious, begs far too many questions; to select a few of the more obvious: is there any evidence that hidden female crime is much greater than hidden male crime (such as embezzlement, white collar crime, ‘knocking-off’ in factories, docks, etc., motoring offences). These are likely to be as least as frequent, or, alternatively, quite as immeasurable. Pollak does not explain why some female deviance does surface and is ‘processed’. Nor does he, any more than Lombroso, answer the query as to what functional needs of society are fulfilled by tacitly permitting women to engage in deviance such as murdering their families, seducing children and stealing from employers and clients. Modern neofunctionalist theory (as e.g., in Parsons 1954; Cohen 1955; Cloward and Ohlin 1960) has made serious attempts to include female deviance in its scheme. The functionalist approach, stressing as it does the impact of social structural features on individuals and groups and also the structuring of situations which lead to opportunities for deviance, has had enormous influence on modern studies of deviance, notably on studies of juvenile delinquency, but also on aspects of mental health and illness, and, rather less directly, on studies of institutions. The concern of this approach has been especially with the effect of certain societal goals (and the structural means, or lack of them, for attaining them) and their impact on specific social groups found, or forced, to be inadequate in these terms, such as lower socio-economic groups, ethnic minorities of low status. The reaction, it is argued,19 will be into forms of deviance, criminal or retreatist perhaps, depending on the availability of such subcultural opportunities. In these discussions it is assumed that growing adolescent males accept as the focal point of their lives the occupational/financial syndrome and that their actions can be interpreted in this light. It is argued (e.g. by Parsons and by Cohen) of girls, whose ‘focal concerns’ are said to be sexual and marital, that female delinquency follows different lines. Women are not required in our society to perform the same instrumental functions in the labour market or in support and protection of their families; their role, on the contrary, is to act through men, through whom they acquire status and are related to the economic system. As a result, sex, and their ability to bargain with it becomes their major interest. As Kingsley Davis puts it: ‘Women must depend on sex for their social position much more than men do.’ Hence women are more likely to be involved in violating sexual mores than in delinquent activities. The fullest statement of this view appears in G. H. Grosser, Juvenile Delinquency and Contemporary American Sex Roles (1951), where male and female delinquents are equally fully considered. It is apparent that a number of subsequent writers have delved into Grosser's theoretical formulations and relied on his empirical data – Cohen for one acknowledges such a debt. Yet another curious footnote to this whole discussion might be to wonder why this seminal work has remained unpublished. These theoretical approaches which have been briefly stated and it is hoped, with a minimum of distortion, pose several initial and straightforward problems. First they ignore, or cannot subsume, substantive knowledge about deviance: their assumptions that young girls find ‘role expression’ through sexual delinquency rather than other forms of deviance are not borne out by the data, for example, in the U.K. where property offences remain the single largest category of offences – and where, for example, over the age of seventeen, female shoplifters exceed males. The reasons for the lower observed rates of deviant participation are nowhere fully articulated. Are they the result perhaps of sexual success being comparatively easier to obtain than occupational achievement – success criteria in the former context being rapidly fulfilled by marriage, which once achieved is a fairly plateau-like position, there being no promotion available? If this is so, then ‘wayward’ behaviour in girls may be seen as parallel to delinquency in boys; except that it can hardly be regarded in Grosser's terms as the moral equivalent of stealing, since theft accepts legitimate ends – the acquisition of goods, enhancement of life style, etc. – but rejects the legitimate means of attaining them. Promiscuous behaviour is then much more akin to the negativistic, anti-utilitarian theft-and-vandalism syndrome described by Cohen (1955) which rejects both legitimate means and ends. Now this exegesis of one aspect of female deviance is surely overcomplex. Difficulties arise because the masculine and feminine poles of deviance have to be forced on to a single axis. Yet it is quite clear that this leads to the denial of certain characteristics of female deviance and of the nature and meaning of the modes of activity which comprise it. Thus Grosser, who, with his concept of sex-role-expressive behaviour, perhaps comes nearest to success in explanatory hypotheses concerning female deviance, still adds to a central confusion about sexuality and the female role. He talks, for example, of a continuum of female sex offences from prostitutive to impulsive; but these points represent surely essentially different types of behaviour in terms of its meaning to the actor. A single shot attack on received values (both bourgeois and respectable working-class) with regard to sexual behaviour – chastity, monogamy, sex should be linked to (responsible) procreation, etc. – can hardly be equated, either in its subjectively perceived aspects, or its societal effects, with any form of prostitution, whether ‘casual’ or fully professional, where an economic gain is one of the major concerns. This failure to see that prostitution is a deviant career structure, with its own economic organization and even apprenticeships and fee-splitting,20 leads to it being equated with hysterical gestures of sexual deviance and actual sexual inversion. Yet basically, all they have in common in terms of situational definitions, is that women are, in both instances, being deviant in what Grosser calls a role-expressive way. This is merely tautological and has little analytical value as regards the problems of female deviance. Possibly the most fruitful source of interpretation and analysis of female deviance in recent thinking could have been that of what Jack P. Gibbs has called the ‘new perspective’ of the study of societal reactions to deviance (Gibbs 1966). Examples of this approach suggest (in summary) that in contrast to what Gibbs calls the ‘biological’ approach and ‘analytic’ approach of previous studies attention should focus on the process of definitions of deviance and the labelling of deviants by means of a complex of feedbacks and reinforcements through societal reaction. Gibbs chooses for analysis and critique statements from Kitsuse, Erikson and Becker, but the present writer would suggest that much of the work done by, for instance, Lemert21 could also be included, while the contribution of Wilkins (1964) is important too and the recent book by Scheff probably represents this approach carried to its ultimate conclusion (Scheff 1966). Such a view would seem to have considerable potential in examining some of the problems posed by female deviance: clearly the fact that this approach focuses on the societal reaction to deviance should make it illuminate an area apparently much influenced by social definitions and attitudes. It is rather remarkable, then, that very little work in this field appears to have been done. One of the few exceptions to this is Lemert's chapter on prostitution in his Social Pathology. Although his stated aim in this study is to examine the way in which primary deviance becomes reinforced and redefined into secondary deviance through societal reactions and personal definitions, in fact, when dealing with prostitution he concentrates on structural aspects of prostitution: the socio-economic background of prostitutes, occupations associated with prostitution, ecological factors and red light districts. While he indicates the type of women likely to drift into prostitution (e.g. unskilled migrants to urban areas) and suggests that there may be a process of self-definition involved, his actual analysis does not differ markedly from the structural functionalist form taken by Davis (1963). In a sense, Lemert exemplifies, through his own implicit attitudes which emerge in this article, the importance of societal reaction to, and attitudes towards, deviance, and the process of defining deviance and the deviant. For instance, he begins by regarding prostitution as an example of sexual deviation.22 Now, while the activities of a prostitute clearly deviate from the mores laid down for sexual relations – that is monogamy and fidelity, affection and desire, rather than cash, as the basis for intercourse – nevertheless from her own point of view, they are a ‘normal’ (within the subculture) instrumental means of earning a living. Bryan (1965), for instance, has shown how a rudimentary apprenticeship system exists in prostitution as a recognized means for girls to learn the ways of the ‘game’. Autobiographical accounts of the careers of prostitutes also lay stress on the professional aspects of prostitution: the narrator of Streetwalker describes her annoyance at being expected to respond to a leer ‘out of business hours’ and her account of her discussions and gossip with another young prostitute and her landlady (who has retired from the streets) bear all the marks of occupational ‘intalk’. Wayland Young (1964) has instanced the learning and association patterns connected with prostitution, the passing on of skills, tricks and techniques. Again, Jackman et al. (1963), in their small-scale study of the prostitute's self-image demonstrate the existence of various ‘strategies’ for managing a career in prostitution despite the status denigration of the occupation; these appear to consist of taking an essentially instrumental view of the prostitute's role and its relation to her existence.23 The foregoing somewhat limited account of a number of approaches to female deviance should serve to demonstrate their common inadequacy and inappropriateness for the topic. This is not, of course, to deny that considerable illumination has emerged through many of these works. The point at issue is rather that much more meaningful insight and understanding might be obtained by other approaches, and that distortion of the inherent characteristics of the material is often a consequence of trying to force it into an inappropriate schematic or conceptual framework. Major common flaws appear to be grounded in the laudable scholarly aim of intellectual economy. Thus the sex differential in reported deviance (and the cultural perception of women as generally more conformist in social behaviour) was so ubiquitously observed that all theories of deviance have had to take some cognizance of it; however, they have tended to do so by constructing single, homogeneous hypotheses which account for both male and female deviance and their apparent differences. Thus Lombroso saw atavistic hysterical females taking prostitution as an alternative to ‘normal’ crime; Pollak, concentrating on opportunity structures, argues that women commit at least as much crime as men but conceal it better; the neofunctionalists (Grosser, Cohen, Cloward and Ohlin) claim that the focal concerns of the male in modern society are occupational and financial, for the female they are primarily sexual, hence the difference in deviance and patterns of deviance. All these views acknowledge differences in the social-structural components of sex roles even if only implicitly; they tend to ignore on the one hand the empirical realities of the deviance of women – e.g. the recorded figures on crime which demonstrate that the most (statistically) typical offence for females is theft as it is for males and that so-called sex offences by females tend to be a result of concern over protection of adolescent girls on the one hand or behaviour by prostitutes on the other, of which the latter can just conceivably be so defined, but is probably more helpfully perceived as occupational deviance. Thus a much more meaningful approach would take female deviance as an aspect of the female sex role and its relationship with social structure, rather than trying to make it conform to patterns apparently observed in the male role and its particular articulation with social structure. It would analyse components of the role, alternative role sets, opportunities for role playing in society, supportive agencies available for aid in role playing, and would view the deviance of women as related to and within this perspective.24 Lest this should seem as fruitless a task as some of the attempts mentioned above, it is perhaps useful here to point to two very important contributions to the study of deviant women which have appeared in the last two years and both of which consider social structure in institutions for women (Ward and Kassebaum 1966; Giallombardo 1966). Both these illuminating studies take as their starting point the roles of women in society and the consequent adaptation necessary within a total institution. Their findings not only throw light on institutional life for women – in showing, for instance, that feminine adaptation and structuring of social relations within prison tends to be in terms of homosexual roles (Ward and Kassebaum) or in familial/homosexual clusters (Giallombardo) rather than in occupational or criminal/political terms as in men's prisons25– they also throw considerable doubt on many assumptions and assertions about the impact of total institutions and the structuring of inmate society (see e.g. Sykes 1958). Any survey of the literature indicates that at present we barely possess the basic components for an initial analysis of the deviance of women. These are lonely, uncharted seas of human behaviour. We lack, for this area, the rich resources of past documentation and research that one can call upon in studying male juvenile delinquency or alcoholism, we have no equivalent of The Jackroller, The Professional Thief, still less is there a Delinquent Girls. We start very much from scratch, although a recent book by a journalistic writer using a penetrating case-study technique (Parker 1955) illustrates the kind of initial data required and which, apart from a few studies of prostitutes, by themselves and others, scarcely exist. Despite the considerable groundwork which has to be put in, there can be little doubt that this will be rewarded by an interesting and valuable outcome. As was suggested in the introduction to this paper, this area qualifies for study both in its own right as being characterized by certain unusual features such as lower levels of deviance participation, differential perception and definition of this deviance by the community, and countertrends to contemporary patterns of deviance. But further as we have seen, neglect of this area and misapprehension about it make it peculiarly interesting to the sociologist of sociology and could perhaps make one wonder about the ‘ideology’ of social pathologists on this point (cf. Wright Mills 1942). Greater knowledge of the sociology of female deviance would enhance our knowledge of feminine behaviour, of sex roles – their characteristics, norms and socializing processes associated with them; it would broaden our view of deviance in general and especially would it make more meaningful the study of male deviance by freeing it from an embarrassing need to include problems of female deviance.26 To summarize, what seems to be needed in the study of female deviance is a crash programme of research which telescopes decades of comparable studies of males. First of all we do not even begin to have a ‘natural history’ of female deviance, our knowledge of its parameters,27 its structure and substructures, the types and nature of activities which make it up, are all exceedingly limited. Even those few sources which exist, such as the autobiographical accounts of prostitutes, have nowhere been systematically analysed and studied. Then, as Mannheim has admirably put it, we need to (Mannheim 1965: vol. 1) ‘avoid the frequent mistake of studying the subject solely under . . . comparative aspects . . . An objective and scientific approach should try to treat female crime as a topic in its own right. Nor, should [we] try to understand female criminality exclusively from the sexual angle.’ But it seems at the same time desirable and likely to prove fruitful to continue some aspects of modern approaches, to take, for example, descriptions of the structures of modern societies and analyse how these provide a framework within which certain roles are acted out, such as those of the adolescent girl or the adult woman and how deviance occurs, is reinforced (or not) and deviant career patterns and roles for women emerge. Eventually, no doubt, it will be possible to reintegrate the study of male and female deviance, not simply for reasons of intellectual economy, but because a sociology of deviance can only be fully developed if it takes as its field all deviant phenomena. But first, of course, these phenomena must have been properly studied within their own context. Where the deviance of women is concerned, there may be a syndrome of ‘modification’ of female deviance within the social system, rather than the ‘amplification’ of deviance amongst adolescent males (as discussed by Cohen 1967; Cohen and Rock 1970), due perhaps to certain factors of the female role in society and social perceptions of its importance. At present such a suggestion must await further developments in the study of the deviance of women before we have sufficient material to begin to formulate it into fully fledged hypotheses. I hope that in this brief critical enquiry I have indicated some of the features that this research should not have, a few which it could possess and above all, the grounds for deeming it both a proper, necessary and potentially most interesting field for sociological enquiry.