Article Psycho-medical discourse in South African research on teenage pregnancy Catriona Macleod and Kevin Durrheim There is currently a fairly large body of research on teenage pregnancy in South Africa. This research can roughly be divided into two categories, viz research that (1) investigates the consequences of early child-bearing, and (2) explores factors associated with conception in the teenage years (Macleod 1999a, 1999b). There are various points of tension in these literatures, which are by no means seamless bodies of knowledge. The main tension runs between what has been called 'mainstream' and 'revisionist' approaches in the American literature (eg, see the exchange between Furstenberg (1991, 1992) and Geronimus( 1991)). 'Mainstream' writers see teenage pregnancy as a social problem that leads to the disruption of schooling, poor obstetric outcomes, inadequate mothering, poor child outcomes, relationship difficulties with family, partner and peers, and to demographic concerns around an increasing population (see Macleod 1999a for review of this literature in South Africa). Postulated contributory factors include reproductive ignorance, risk-taking behaviour, precocious pubertal development, single-parent, female-headed households, family dysfunction, poverty, poor self-esteem and moral development, poor health services, peer influence, coercive sexual relations, the breakdown of tradition and, conversely, the cultural value placed on fertility (see Macleod 1999b for review of this literature in South Africa). Preston-Whyte and colleagues (Preston-Whyte 1991, Preston-Whyte and Allen 1992, Preston-Whyte and Zondi 1989,1991,1992) present a revisionist argument. They postulate that early reproduction represents a rational reaction to a number ofpersonal and structural constraints experienced by teenagers in the African community. Indeed, early reproduction may be encouraged by the number of successful, single parent women in the African community and the fact that motherhood is viewed as an avenue to achieving adulthood status in the event of a delay TRANSFORMATION 51 (2003) ISSN 0258-7696 74 Psycho-medical discourse in South African research on teenage pregnancy of marriage through the necessity of amassing bridewealth. Therefore, 'there is little incentive to strive for the achievement ofnorms and values which are, after all, largely middle-class, and in the South African context, white' (Preston-Whyte and Allen 1992:215). Historical studies show that 'teenage pregnancy' (as opposed to unwed motherhood) is a topic of rather recent scientific interest. It emerged as a social problem within scientific debates in the late 1960s and early 1970s in the United States (Arney andBergen 1984, Vinovskis 1988) and somewhat later - late 1970s - in South Africa (Macleod 1999a). An explanation advanced for this sudden and recent concern is that the age of menarche has decreased owing to better nutrition, thereby putting adolescents more 'at risk' of conceiving at an earlier age. This is questioned by Vinovskis (1988), who argues that while scientific interest in the area has increased, rates of early conception and reproduction have been decreasing in the United Sates fromapeakin 1957. Arney and Bergen (1984), in their paper entitled 'Power and visibility: the invention of teenage pregnancy' (our emphasis), outline a Foucauldian history of scientific discourse on the subject of teenage pregnancy. They trace how the morally loaded concepts of 'unwed mother' and 'illegitimate child' dissolved into a single new scientifically neutralised concept of 'teenage pregnancy' in the late 1960s. This shift did not take place, they state, because our understanding of the phenomenon was becoming more accurate, or because the problem demanded more humane treatment, but rather because it allowed for the deployment of the scientific discourse. Pregnant teenagers became technical problems requiring endless scrutiny and measurement, and an in-depth knowledge of their structure. This shift in power meant that women who conceive in their teenage years needno longer be objects of moral exclusion, but may rather be disciplined by scientific inclusion. In this paper, we dc not join the debate as to whether teenage pregnancy is indeed a problem or not in South Africa. We are not concerned with the dominant question of the consequences or causes of early reproduction. Rather, we turn an inspecting eye on scientific discourse itself, analysing the ways in which this discourse medicalises and psychologises the sexual and reproductive adolescent. In other words, we turn to the question of how this scientifically neutralised concept - teenage pregnancy - is deployed in scientific discourse in South Africa and how these constructions may re- produce practices that act as modes of regulation. To do this we employ a Foucauldian-based discursive analysis of the 75 Catriona Macleod and Kevin Durrheim knowledge (savoir) of teenage pregnancy produced in South African research and literature. This savoir uses the language of truth to construct the sexual and reproductive teenager as a subject, and adolescent and reproductive health as an obj ect, of government. As such, it has implications in terms of professional practices and interventions in the everyday lives of teenagers (see discussion in the final section of the paper). In particular we draw on Butchart (1995, 1998) and Rose (1990, 1992), to analyse the deployment of: (1) the medicalised strategies of sanitary science (which constructs the anatomical identity of the pregnant teenager), social medicine (which extends medical surveillance to the adolescent-in-context as well as the population) and community health (in which the community is invoked as a repository of culture, tradition and health); and (2) psychologisation (which renders the self, emotion, mind and psyche of the pregnant teenager visible and cognisable) in the South Africanresearch and academic literature. Data collection and analysis The text we analyse includes published and unpublished research reports and other academic literature on teenage pregnancy in South Africa from 1970 to 1997. This material was collected for a larger study on teenage pregnancy (Macleod 1999c) through (1) conducting searches on international and national bibliographic data archives; and (2) sending letters to heads of departments of relevant social science, education, and medical departments of all South African universities, as well as relevant non-governmental organisations, requesting bibliographic details of all research on teenage pregnancy conducted in their organisations. The result was a collection of 77 research reports, theses, articles and chapters. Forty-one of these are published in local professional journals such as Nursing RSA, Salus and Outlook, local scientific journals such as South African Medical Journal, South African Journal of Psychology, South African Journal of Sociology, and international journals such as Social Science and Medicine, International Journal of Adolescence and Youth and The Journal of Social Psychology, with South African Medical Journal attracting more articles (13) than other journals. Eleven of the unpublished manuscripts originated from historically white Afrikaans- speaking universities, 11 from historically white English-speaking universities, seven from historically black universities, two from a bilingual (English and Afrikaans) university, and three from research councils. Table one indicates the time periods in which the documents appeared. Much of 76 Psycho-medical discourse in South African research on teenage pregnancy the research analysed here was conducted prior to the 1994 elections, a time when racial (as well as gender and class) power relations were permeated by apartheid ideology and practice. As expected, race appears as a robust signifier in the literature (readers are referred to Macleod and Durrheim 2002). Table One: Time period in which the documents analysed appeared Time period No of documents 1975-1980 2 1981 -1985 6 1986- 1990 29 1991 - 1995 27 1996/1997 13 The data (ie, all the documents collected) were analysed using discourse analysis. What is analysed is the discursive 'event' (Fairclough 1992) which is simultaneously a piece of text, an instance of discursive practice and an instance of social practice. The discursive events in this instance are simultaneously the texts written on teenage sexuality and pregnancy, the discursive construction contained in that text, and the research practices engaged in to collect information and produce the text. The aim of the analysis was to investigate scientific discourse in terms of its codifying effects concerning the known (the effects of veridiction) (Foucault 1991b). Concretely, this translated into (1) reading and re-reading the texts; (2) chunking the material according to themes (the psychological subject, biological themes, medical and psychological interventions); (3) applying Parker's (1992) seven basic criteria for identifying discourses; and (4) infusing the analysis with theoretical insights which draw on Foucault's analytics ofpower/knowledge and governmentality. In the analysis the data were treated as one body (ie. with no distinction between documents) so as to allow for similarities across documents to emerge. The 21 extracts featured in the paper were chosen for their salience in illustrating the issues emerging in the analysis, and not on the basis of the document from which they were drawn. The medicalisation of teenage pregnancy In the History of Sexuality: Vol /, Foucault talks of the emergence of a discourse on sex within the medical establishment during the eighteenth 77 Catriona Macleod and Kevin Durrheim century: 'one had to speak of [sex] as of a thing to be not simply condemned or tolerated but managed, inserted into systems of utility, regulated for the greater good of all, made to function according to an optimum' (1978:24). He traces how the rituals of the confession, which first appeared in European Christianity in the thirteenth century, came to function within the norms of scientific regularity. The speech of the patient was codified in clinical terms through, inter alia, the examination, the questionnaire, the deployment of a set of signs and symptoms; sex became endowed with 'polymorphous causal power' (1978:65), but was, simultaneously, latent, or elusive by nature, thus requiring confession; the one receiving the confession no longer merely forgave or judged, but deciphered and interpreted: 'he (sic) was the master of truth' (1978:67); the domain of sex was 'medicaiised', or 'placed under the rule of the normal and the pathological' (1978:67, our emphasis). In this section we analyse how teenage (hetero)sexuality and reproduction is managed, normalised and pathologised within a medicaiised discourse. Of pivotal importance in the process of medicalisation are the expert gaze and surveillance. In The Birth of the Clinic: an archaeology of medical perception, Foucault (1976) describes how the modern understanding of disease broke with that of classical medicine. He ascribes this to the development of what he calls the 'clinic', which was simultaneously a new institution (the clinical hospital) and a new style of medical thought and practice. Central to this was the clinical gaze, which Foucault outlines in historical narrative: At the beginning of the nineteenth century, doctors described what for centuries had remained below the threshold of the visible and the expressible, but this did not mean that, after over-indulging in speculation, they had begun to perceive once again, or that they listened to reason rather than to imagination; it meant that the relation between the visible and invisible ... changed its structure, revealing through gaze and language what had previously been below and beyond their domain. (1976:xii) Thus, the gaze shifted from a perception of signs of illness on the surface of the body to one in which pathology was signified by the secrets and essences internal to the body - that is, an 'invisible visibility' (Foucault 1976:149). This visibility was simultaneously an audibility as language was used to describe what was 'seen' below the surface. The medical gaze is not, as Butchart (1995, 1998) points out, only a cognitive or perceptual skill 78 Psycho-medical discourse in South African research on teenage pregnancy developed by the doctor through training, but is the technology through which the body of the patient is simultaneously constructed and restrained. It allows and limits what the doctor can see, how s/he describes the functions of the body, and how s/he practices or intervenes in the bodily area of concern. The expert gaze and visibility achieve effects ofpower through hierarchical observation and normalising judgement (Foucault 1977). For example, the clinical gaze, which makes the calibration of obstetric risk possible (eg the calculation of the possibility of cephalopelvic disproportion), produces a normalising and hierarchical knowledge of the reproductive subject (eg normal pregnancy and childbearing practices, as well as which personal characteristics and behaviours are most likely to lead to obstetric problems). A system of surveillance is established in which the subject is observed, measured and monitored with the subject eventually conducting her own conduct on the basis of normalised judgements. Medical and psychological discourse function as interlinked apparatuses of surveillance which are structured around a knowledge of the normal and pathological, and a series of risks to be avoided, which each individual under the weight of the inspecting gaze 'will end b> intcriorising to the point that he is his own overseer'(Foucault 1980:155). lathe following, Butchart's (1995,1998) Foucauldian-basedgenealogy of the socio-medical discourses producing the African body in South Africa is used to show how three different moments of medical discourse in South Africa, viz sanitary science, social medicine, and community health are deployed in the construction of young women and their sexual and reproductive behaviour. These domains of medical practice function to render the 'problem' visible, audible and calculable from the vantage of science, thus establishing a system of surveillance. Each simultaneously individualises and totalises, normalises and pathologises, the sexual and reproductive adolescent, but in different ways. Sanitary science Sanitary science, which emerged in the mid- to late-1800s, operates as a medical gaze that brings into view the anatomical identity of the body and the exchange of energy and matter between human bodies and external space (Butchart 1995, 1998). Substances belonging to neither the body nor the environment - but simultaneously to both (eg urine, saliva, faeces, and in our case, foetuses) - emerged as the dominant objects of sanitary science. 79 Catriona Macleod and Kevin Durrheim Sanitary science has as its aim the control of the exchange of matter between the inside and the outside of the body through external coercion. In this process, internal and external bodily diseases, disorders and lesions are described, observed and diagnosed. Hierarchical observation and normalising judgement are instituted on a biological level, with certain anatomical states taking on the definition of health. In the scientific literature on teenage pregnancy, sanitary science inserts the medical gaze into the body of the adolescent, either normalising or, mostly, pathologising her reproductive ability: Extract 1 The younger the teenager, the more dangers are believed to attend the young mother and her infant during pregnancy and birth. Among these obstetrical problems are pregnancy induced hypertension, anaemia, premature labour, occipito-posterior presentation, cephalopelvic disproportion, vaginal lacerations as a consequence ofthe tight nulliparous perineum and low birth mass infants. (Boult and Cunningham 1992:163/ 164) Extract 2 Chronological age perse has no bearing on the obstetric performance of adolescents when the latter are compared with young adults of similar parity and socio-economic status. (Ncayiyana and Ter Haar 1989:231) Swanson (1977) and Butchart (1995,1998) have illustrated how, during the bubonic plague at the turn of the century, sanitary science advanced the cause of racialised urban segregation through the creation of locations that contained the threat created by Africans' failure to manage the exchange of matter between their bodily boundaries. In much the same way, the biological discourse of sanitary science intersects with a developmental framework to segregate young women from their older counterparts, rendering them reproductively inadequate on an anatomical level. The pregnant teenager is depicted as 'dangerously' pathological (which threatens both herself and her infant) by definition of her physical immaturity, which leads to internal and external lesions such as anaemia and vaginal lacerations, as well as imbalances in the reproductive processes (eg cephalopelvic disproportion). In other words, the adolescent's ability anatomically to manage the body/ environment boundary in pregnancy and birth is seen as impaired. Extract 2, on the other hand, counters this developmental reading, normalising adolescent reproduction within the confines of the anatomical identity of reproductive women in general. Normal and pathological reproduction in 80 Psycho-medical discourse in South African research on teenage pregnancy this reading becomes attached to the anatomical identity of parity and the social identity of socio-economic status (see social medicine discussed later). The clinical examination and ante-natal care are the intervention practices associated with sanitary science in pre-natal reproductive health. In these, the health service provider inspects the interior of the body by monitoring foetal growth and maternal weight and manages the exchange of matter between the interior and the exterior of the body (eg by prescribing vitamin tablets and advising against the use of alcohol). Extract 3 The tendency of pregnant teenagers to present late at antenatal clinics, or not attend at all has been consistently referred to ... Several authors ... make an association between paucity of antenatal care and LBW [low birth weight]. This is because nutritional and iron and foliate supplements are not able to be administered, no maternal weight gain monitored, leading to foetal growth deprivation. (Boult and Cunningham 1993:47) Although some authors contest the allegation that adolescents attend ante-natal care more irregularly than do adults (cf Ncayiyana and Ter Haar 1989), the dominant construction of adolescents as poor attenders contributes to their anatomical pathologisation. The missing of opportunities for reproductive inspection, surveillance and repair places the pregnant teenager outside the gaze of the expert, implying reproductive health cost. The clinical examination and monitoring which occurs in ante-natal care is conditioned on totalising and individualising technologies. Hierarchical observation, which defines and understands individuals in terms of their relative positioning in a population, and normalising judgement, provides the link between individualisation and totalisation. Extract 4 Management obviously depends on the assessment of the patient and many factors have to be considered. (Rockey 1986:17) Extract 5 Each case is assessed on its own merits. (Nash 1990:310) The health care provider assesses the individual teenager as a unique medical subject, and then prescribes interventions appropriate to her particular reproductive condition. This individualising practice, however, only has meaning in the context of the (totalising) framework of normal (adult) reproductive health within which she is located as normal or pathological. 81 Catriona Macleod and Kevin Durrheim Social Medicine Social medicine emerged later than sanitary science and criticised the latter for dividing the indivisible person through a de-contextualised and biological focus on anatomical units rather than the person as a whole. Social medicine understands the individual's health and illness in the context of his/her social and environmental location. Extract 6 H aving said that maternal age alonedoesnothave adverse effects on the outcome of pregnancy, it is still true that adolescent childbearing is associated with several problems, most of which are of a social nature. (Mukasa 1992:423) Extract 7 Any preventive effort therefore should be part of a multi-disciplinary, holistically integrated approach, addressing the social and welfare needs of adolescents. (Schoeman 1990:unpaginated) Extract 6 counters the pathologisation of teenagers' anatomical capacity to reproduce. Instead, the 'problem' is formulated in terms of its 'social nature'. What this highlights is the fact that while teenage pregnancy may be biologically or physically unproblematic, social difficulties arise, for example, in the disruption of schooling and family and peer relations, and the creation of unstable partnerships outside the context marriage. Thus, as Extract 7 shows, a multidisciplinary and holistic approach is required to ensure that inspection and intervention is based on a proper understanding of all aspects of the teenager's life. A second associated feature of social medicine is its understanding of health as a social phenomenon, distributed across a population of individuals in social and environmental contexts. Thus the focus of intervention shifts out of the clinic to the broader population, where the public health gaze medicalises everyday life by bringing into view health and illness patterns of individuals in collectives. The power ofthis discourse is that it incorporates not only the pregnant teenager but also non-pregnant teenagers and partners into its net of surveillance and intervention. Extracts Underpinning the provision of quality health care to women ... is the need to strengthen preventive programmes that promote adolescent well-being. (Parekh and De laRey 1997:223) 82 Psycho-medical discourse in South African research on teenage pregnancy Extract 9 These figures could possibly be lowered ifteenagers could be counselled before they became sexually active. Sex counselling programmes should not only be directed at females, but should include the male adolescent, [sic] it is important to highlight all the possible consequences of sexual activity, because he must understand that the decision to become sexually active can change his life as well as that of his sex partner dramatically. (Keogh 1988:30) Extract 10 They [adolescent multiparas] need education and information to enable them to make reasoned choices about both the spacing oftheir pregnancies and the prevention of any more unplanned and unwanted pregnancies. (Boult 1992:17) Through the construction of adolescent sexual and reproductive health as a personal and public concern, teenagers and their partners are incited in social medicine to modify their habits in the direction of greater attention to their own well-being. The emphasis shifts from the clinical gaze focussed on the individual anatomy of the pregnant teenager to the social management of teenage (hetero)sexuality and reproduction. The divide between the unwell (pregnant) and the well (non-pregnant) thus becomes blurred; the promotion of 'well-being' allows for the extension of surveillance to all adolescents. This is effected through 'preventive' programmes (Extract 8) which allow for the deployment of psychological (Extract 9) and pedagogical (Extract 10) interventions such as sexuality education and counselling, which have as their aims behaviour change (eg abstinence or reproductive planning). In social medicine, unlike sanitary science, the aim of knowledge of the population is not only to render the individual case meaningful (ie, against a normative framework). Instead, population becomes an object in its own right as it is the population as a social entity rather than the individual that is the point of intervention. Thus totalising technologies are enhanced. At the same time, individualising tendencies are re-deployed in new ways, as the agent of intervention is no longer the expert prescribing anatomical interventions but rather the individual monitoring and managing his/her own behaviour under expert guidance (through sexuality education or counselling). For example, the task of the sex counselling programmes referred to in Extract 9 is to inform adolescents about the possible consequences of sexual activity (ie, expert knowledge). In addition the expert incites self-management on the part of the adolescent as they are 83 Catriona Macleod and Kevin Durrheim encouraged to understand risk and make proper decisions in the light of potential dramatic life changes. This is an intensely individualising technology. Although the adolescent ostensibly has freedom of choice, totalising strategies are apparent in that the content of counselling and education programmes are circumscribed by underlying normative frameworks. Remnants of the concept of pathology remain, as correct and incorrect decisions and conduct are implicit in all these programmes. For example, should the adolescent choose to engage in early sexual activity, having been assisted by the expert in 'understanding' the 'consequences', s/he is positioned either as irresponsible, unreasonable or deficient (Extract 9). Community Health Community health takes the globalising tendencies of social medicine and makes them culturally and locally sensitive. It emerged as a counter to social medicine's uniform and regimented approach to the health of a population as whole. It articulates a discourse of community as a repository of culture and a dynamic agent of regulation and change. Within community health, health is understood to be located, not only in individuals and populations, but especially in tradition and culture. Thus, we see in the literature an explosion of knowledge around cultural specificities. Extract 11 Additional sociocultural and political factors regarding the avoidance of contraceptive use is evident in African countries. High cultural and social value is placed on fertility, which makes pregnancy a valued state. ... In oppressive African regimes political reasons for the non-use of contraceptives exist. They revolve around a claim for autonomy from state interference in family affairs, as in Algeria, and state imposition of contraceptive use, as in Namibia. (Carolissen 1993:13) Cultural and political diversity are cited here as factors in the perpetuation of teenage pregnancy. Instead of a uniform notion of wellness and health, the community health discourse introduces peculiarity and difference. The peculiarity and difference referred to, however, tends to centre ar °undracialised and economised boundaries. The signifiers 'race', 'culture', ethnicity', 'poverty', and 'lower socio-economic status' are utilised in scientific discourse on teenage pregnancy in South Africa to highlight differences'in adolescent sexual and reproductive behaviour, with'tradition' culture' and 'poverty' sanitising or disguising an underlying racialising 84 Psycho-medical discourse in South African research on teenage pregnancy project. 'Blacks' and 'the poor' (who are generally co-extensive) are exoticised and rendered strange and thus open to scrutiny, monitoring and intervention (see Macleod and Durrheim (2002), and Macleod (2002) for more in-depth discussion of this aspect of the scientific discourse on teenage pregnancy in South Africa). Many of the practices of community health are the same as for social medicine, ie, the deployment of sexuality education and counselling. However, the content andrationale is tailored to particular communities, and community health is practised in a spirit of tolerance, flexibility and understanding. Health service providers are encouraged to adapt their interventions to local beliefs and practices. Extract 12 For sexuality education to be acceptable for black youths, it should fit into their sub-culture. (Schoeman 1990:17) Extract 13 There is a need for the acceptance by medical personnel of cultural beliefs such as witchcraft. (Fouche 1992:77, translated from Afrikaans) Extract 14 Cultural values such as that placed on fertility in many parts ofthe black community need to be recognised. (Preston-Whyte 1991:46) Community health co-opts the community and peer group into the production of health. Extract 15 The eventual dream still remains that the total community must take responsibility—ministers, health authorities, politicians and business. ... With the total involvement of community leaders in co-operation with parents and teachers ... (De Villiers 1991:232, translated from Afrikaans) Extract 16 A useful community based approach involving the 'community health activists' could possibly be developed in preventative education. This could involve community health workers training peer groups in general sexuality education and life skills.... The trained peers could then form zonal groups and establish 'teenage clubs' which focus not only on sexuality education but also have other activities which are of interest to teenagers. The important difference between this approach and established ones is the fact that teaching methods will be less didactic, encouraging participation. (Carolissen 1993:42) 85 Catriona Maclaod and Kevin Durrheim Preventive action is now not merely the domain of the service provider or expert, but is rather in the hands of the 'community'. 'Total involvement' and commitment to the well-being of the community is incited. The health worker's task shifts from one of overseeing the individual teenager, to monitoring groups of adolescents. S/he apparently has been divested of power, as s/he no longer instructs or directs (s/he is not didactic). Instead, s/he mobilises the community and elicits their participation. This 'participatory' rhetoric masks the operation ofpower. The community is now attributed with a free will and decision-making power, much the same as with the humanistic understanding of the individual. The community becomes the overseer of its members' health, managing the circumstances surrounding potential or actual pregnancy. The individualising and totalising tendencies of community health operate in a similar manner to those of sanitary science and social medicine; however, the added dimension of the community as a separable, culturally specified unit is introduced. The individual is transformed into a 'microcosm of his culture as a whole' (Butchart 1995:273), and thus is normalised with respect to the observed and documented peculiarities of his/her community. The community, on the other hand, is homogenised as a unit (for example, with the peer group being spoken of in the singular). This unit is contrasted with other communities in terms of its sexual and reproductive health. Thus, the individual is simultaneously individualised and totalised with respect to the normative framework of his/her community, while the community is simultaneously homogenised as unique and totalised with respect to the normative judgements regarding healthy communal living. While sanitary science, social medicine, community health take different objects as their points of focus, they do not represent separable discursive events in the management of adolescent sexual and reproductive health. Rather, they act as multiplicative layers of discourse and practice (along with psychologisation discussed below). For example, social medicine, while concentrating on the teenager-in-context and the population as a social entity, does not abandon the anatomical identity of the pregnant adolescent. Community health utilises the technologies of normalising judgement deployed in sanitary science and social medicine, but tailors them to a community/cultural discourse. 86 Psycho-medical discourse in South African research on teenage pregnancy The psychologisation of teenage pregnancy Psychologisation has strong resonances with medicalisation, in particular with social medicine and community health (which emphasise, inter alia, the importance of counselling). It refers to those processes that render the body, the self, emotions, the mind, the psyche and the relationship visible as psychological objects (Rose 1990). In psychologisation, the expert gaze extends beyond the biological (while still incorporating it) to psychological essences of the being. As these psychological objects become visible, so they become calculable and manageable through various processes (assessment, evaluation, diagnosis, therapy, remediation), or what Rose (1992) calls the psy-complex. Thepsy-complex allows for the apparently 'public' issue of rationalities of government to be linked to the 'private' question of how one should behave, how one 'conducts' one's own conduct. Rose (1992) outlines three principle forms of connection between psychological expertise and modern government: rationality (truthful knowledge and efficacious technique), private space (outside the formal scope of public powers), and autonomy (the construction of a regulated autonomy). The 'Janus face of [this] expertise'(Rose 1992:367) is that it operates as arelay between government and privacy- 'their claims to truth and efficacy appealing, on the one hand, to government... and, on the other hand, to those ... attempting to manage their own private affairs efficaciously". Within the private realm, the person is constructed as autonomous, free to choose or decide on the basis of individual motives, needs and aspirations. However, the private becomes intensely public and political through the plethora of psychologised texts andmethods concerning the private conduct of life. Psychological expertise tabulates and regulates a population of modern subjects who anticipate and understand themselves in psychological terms. This is not to say that the psy-complex determines forms of subjectivity, but rather that it is successful as a form of government to the extent that individuals come to experience themselves according to the psychological types, behaviours, characteristics and qualities attributed to them in psychologised discourse. The wide-ranging focus of psychological discourse is evidenced in the South African teenage pregnancy literature. The psychologised gaze spans, inter alia: individual emotions (eg Pond 1987) indicates that teenage mothers feel inferior and insecure); unconscious dynamics (eg Pond 1987) postulates that pregnant teenagers experience covert anger and hostility towards males); cognitive notions (eg Fouche 1992) argues that teenage mothers 87 Catriona MacJeod and Kevin Durrheim have negative irrational thoughts concerning their babies); moral development (eg Oosthuizen 1990) believes that teenagers displaying Kohlberg' s stages three and four moral reasoning are more at risk for pregnancy); identity issues (eg Brits 1989) shows that pregnant teenagers have a poorly defined sense of identity and self-concept); familial patterns (eg Blom 1989) implicates parenting styles and communication patterns in the occurrence of teenage pregnancy); 'cultural' arrangements (eg Preston-Whyte 1991) emphasises the value placed on fertility in 'African' communities). This breadth of scope allows for multiple layers of normalisation and pathologisation of the pregnant teenager. On one level, she is pathologised as an individual, with pregnancy indicating deviant behaviour, emotions, developmental processes or psychological status. At a second level, her individual response is normalised within pathological family relations and formation, social relations, and the vagaries of adolescence. The revisionists introduce a third level, in which the individual and family are normalised, but social arrangements areproblematised. The following extracts are illustrative of each of these levels. Extract 17 Unmarried mothershave also been found to have feelings of inferiority. Each of the subjects in this study confirmed this as they are plagued by feelings ofinsecurity, inadequacy and lack of achievement. Thus all the subjects display a tendency towards depression, yet all hide their feelings. Adele, Delia and Elaine's depression was manifest, however, by evasiveness, passivity and withdrawal. (Pond 1987:164) Extract 18 Many unmarried mothers report a negative family influence and unsatisfactory parent-child relationships. The most common form of disturbance is parental domination or rejection. In most cases where a parent is domineering, the other is passive and unassertive in the home. Parental discord and the subsequent loss of a parent (usually the father) through separation or divorce is also prevalent. The absence of the father seems to have resulted in object relation and role growth problems, as well as a large proportion of disturbed mother-daughter relationships characterized mostly by much ambivalence. (Pond 1987:51) Extract 19 [T]he number of teenage births will not be reduced without radical changes in the socio-political dispensation to which teenagers are the heirs. (Preston-Whyte 1991:46) 88 Psycho-medical discourse in South African research on teenage pregnancy In Extract 17 the pregnant teenager is pathologised on an individual level - she is defined as lacking in terms of the normalised traits of security, adequacy, achievement, activity, engagement, forthrightness, and mental well-being. The gendered and political assumptions contained in these norms remain implicit (see, for example, Sampson 1990) who deconstructs the notion of normality as largely coterminous with the characteristics ascribed to white males living in a liberal democratic society). At the second level, as evidenced in Extracts 18, the pregnancy is rendered 'normal' within a pathological familial context. This pathologisation relies on the assumption of the 'normal' family consisting of two heterosexual parents with children between whom there is a lack of discord. This level of normalisation and pathologisation allows for an extension of psychologised surveillance to the family- it is no longer just the individual teenager who needs to be reformed, but also familial relations and interactions. This level interconnects with social medicine's pre-occupation with the person-in-context while the third level resonates with the community health discourse. For example, in Extract 19 the individual and the family are exonerated from blame. Instead, they are the innocent heirs of a pathologised socio-political dispensation. Their reaction is normalised, and the work that has to done is to 'radically change' the peculiarities of a particular society. We have seen how sanitary science renders the anatomical identity of the body visible, social medicine, the whole person and the population, and community health, the peculiarities of culture and tradition. Psychology, through the deployment of clinical instruments, scientific method and clinical observation, brings the inner being, the essence of self and relationships into the plane of sight. Extract 20 The incomplete sentence test emphasised that the pregnant teenagers have a low self image. (Brits 1989:196, translated from Afrikaans) Extract 21 The findings of this study confirmed that the transition to motherhood is accompanied by a number of psychological consequences that place the teenage mothers at risk in terms of later life adjustment. (Parekh and Dele Rey 1997:227) Examples of clinical instruments of measurement used in the literature analysed include psychodynamic tests such as the Thematic Apperception Test and the Rorschach (Pond 1987), personality tests (Brits 1989), and tests of relationships such as the Arizona Social Support Interview Schedule 89 Catriona Macleod and Kevin Durrheim (Tanga andUys 1996) and the Quality ofFamily Function during Adolescence (Blom 1990). The scientific methods used ranged from anthropological vignettes to quantitative statistical analyses, with data being collected through individual and focus group discussions, standardised and non- standardised questionnaires, essay-writing, psychometric tests, andhospital records. These instruments and methods are used to reveal, in a multi- layered fashion, the emotional, psychological and social/relational status of the individual, making visible her degree of psychological normality or pathology. In Extract 20, for example, we see how completing sentences specified by the expert as eliciting meaningful responses allows for the detection of pregnant teenagers' low self image. In Extract 21 the qualitative analysis of focus group discussions reveals 'psychological consequences', and allows for the specification of future scenarios in which psychological pathology and poor adjustment are to be expected. Clinical instruments and scientific method are supplemented by clinical observation. For example, in Extract 17 the expert is able to uncover an underlying emotional state (depression) through the observation of particular behaviours (evasiveness, passivity and withdrawal). One of the chief disciplinary mechanisms utilised in psychologisation is the confession (cast as counselling or therapy). This practice, in which the most private and intimate details of the person's inner being are 'amplified to audibility' (Butchart 1995:336), forms the interface between the private conduct of the person's life and the public issue of psychological normality and pathology. The teenager renders herself truthfully to the confessor (the health service provider, social worker, psychologist, counsellor) who hears, evaluates and encourages appropriate self-monitoring and self-regulation. Extract 22 The nurse... has [the] opportunity... to discover how the patient views her pregnancy - what she really feels and not what she has been told to feel by her family. By just listening the nurse can help the patient correlate her real thoughts from the conflicting ideas clogging her mind. (Rockey 1986:17) Power and knowledge are linked in the confessional through the will to truth. The nurse in Extract 22 extracts 'real thoughts' and what the teenager 'really feels'. This will to truth is interpenetrated with normalised pronouncements regarding adolescence and teenage pregnancy. These pronouncements, legitimated through scientific method, form an intricate part of the discursive space between nurse and teenager. By 'just listening', 90 Psycho-medical discourse in South African research on teenage pregnancy the nurse requires the teenager to become audible, binding her to discourse and, finally, to self-regulatory practices. Implications in terms of practices and transformation The questions guiding this section of the paper are: (1) what are the links between scientific knowledge and professional practice; and (2), what are the implications of a Foucauldian perspective for the transformation of practices (whether scientific or professional)? In answering these we invoke the Foucault's (1991a) thoughts on governmentality, and arguments concerning the political utility of a Foucauldian discourse. The psycho-medical discourses which produce knowledge (savoir) of teenage pregnancy in the language of truth construct the teenager as a subject, and adolescent sexual and reproductive health as an object, of government. Foucault used the terms government and governmentality in inter-related ways. He defined government as the 'conduct of conduct' (Gordon 1991:2), that is, as an activity that aims to shape, guide or affect the behaviour, actions, or comportment of people. Thus, for example, clinical examinations, ante-natal classes, sexuality education, counselling, community programmes, etc, all act as modes of government in shaping, guiding or affecting the health and illness-related behaviour of individuals, families, and groups of people. By governmentality, Foucault meant an ensemble of 'institutions, procedures, analyses, reflections and tactics' (1991 a: 102) through which power operates. He described governmentality as the art or rationality of government, as the way or system of knowing about the nature of the practice of government (the 'conduct of conduct'). Government requires ways of knowing the population, of thinking about it and rendering it the object of dissection and analysis. It necessitates the institution of an enterprise of enquiry into the population's state and condition: 'The constitution of a savoir of government is absolutely inseparable from that of a knowledge of the processes related to population in its larger sense' (Foucault 1991 a: 100). Thus, for example, in producing knowledge of the object, 'teenage pregnancy', scientific expertise and rationality construct discourses concerning a subject, the pregnant and/or sexual teenager, and re-produce practices which act as modes of regulation that 'conduct the conduct' ofyoung women and their sexual and reproductive behaviour. These regulatory practices are founded on a field of visibility and calculability provided by savoir. The mundane operation of scientific documentation 'makes the individual stable through constructing a perceptual 91 Catriona Macleod and Kevin Durrheim system, a way ofrendering the mobile and confusing manifold ofthe sensible into a cognizable [visible, audible, and calculable] field' (Rose 1989:124). Thus practices of intervention, the actions carried out upon the lives and conduct of teenagers and their families by those vested with the authority to do so, are intricately connected to the truths generated by intellectual technologies regarding the nature of pregnant teenagers. This is not to say that the knowledge generated by research 'influences' the behaviour of service providers in a directly linear fashion, but rather that the practices of health production and those of knowledge production are intimately tied up with one another, each simultaneously producing and constraining the other. Present-day service provider practices represent, in part, the technical means by which the discursive constructions and governmental processes circulating in the scientific literature, lecture halls, text-books, etc, take effect in the lives of teenagers and their families. As Rose (1993:291) puts it 'Knowledge... is an apparatus for the production, circulation, accumulation, authorization and realization of truth. And truth is a technical matter — it is the "know how" that promises to make government possible'. Government, inFoucauldian terms, operates, in large part, by transforming the object, which it has rendered visible, audible and cognisable, into a subject. Experts act on the lives of teenagers and their families, who, in turn, act on their own lives in accordance with the tenets of the knowledge gleaned by the experts. Thus, surveillance turns to self-surveillance, regulation to self-regulation, monitoring to self-monitoring. The teenager is not only an object of scientific investigation and professional practice, but also a subj ect, an individual who labours to understand herself, to become a certain type of person, to behave in certain ways, to control her sexual urges, to patrol access to her body, to make herself attractive to males, to attend to her own health and welfare, and to achieve particular educational, career and social goals. This process is referred to by Dean (1994) as governmental self- formation. Post-colonial authors (eg Stoler 1995, Vaughan 1991) have pointed to the promises and limitations of a Foucauldian analysis within African settings. Of pertinence to the above discussion is Vaughan's (1991) questioning of the extent to which colonial and post-colonial power operates through individual subjectivities. She argues that a great deal of what she calls 'unitization' went on in colonial states, which is not the same as the creation of the speaking subject. While we would argue that 'unitization' has strong linkages with Foucault's 'totalization', which he sees as interweaved with 92 Psycho-medical discourse in South African research on teenage pregnancy individualisation, this type of critique alerts us to the possible limits of disciplinary technology in the lives of certain reproductive adolescent women. This in no ways undermines the argument developed in this section, however, as the epistemology and research activities that have driven academia in South Africa in general, and the research featured in this article, have, to a large extent, been Eurocentric in nature (Nkomo 1991). This epistemology, in turn, informs the pedagogy and practice ofservice providers, as well as the systems within which they operate. What does all this mean in terms of the transformation of the practices of scientific research and health service provision with regards to young sexually active or reproductive women? Many feminists (eg Deveaux 1994, Di Leonardo 1991, Harding 1992,Hartsock 1990) have found Foucauldian theory unsatisfactory as it seems to provide no morally evaluative or politically committed stance by which to judge either one regime of truth as superior to another, or societies or relations as better or worse than each other. These authors argue that since power, in Foucauldian terms, is everywhere, it is ultimately nowhere, making social improvement or transformation impossible, as successful resistance means simply changing one discursive identity for another, thus creating new oppressions. Defenders of Foucauldian discourse respond to the above criticisms in a number of ways. It is true that Foucault was reluctant to delineate a clear- cut political agenda because of his sense of the dangers of programmes based on grand theory. However, his writings were clearly political, and he didmake certain value judgements (Grimshaw 1993;Hoy 1988). For example, he suggests that we would be better off without the modern obsession of classifying people according to their sexual preferences and using distinctions such as that between normal and deviant. In the same way, it is possible to say that we would be better off without the specified category of teenage pregnancy and without the need to produce volumes of data on the normal or pathological emotional, educational, social and maternal functioning of the sexually active or reproductive adolescents (this does not mean that we cease to engage in any research activities but rather that our endeavours have a different focus). The claim to be 'better off need not entail, however, an appeal to a sexual Utopia (Hoy 1988). Liberation is not seen as transcendence or global transformation, but rather as a freeing from the assumption that prevailing ways of understanding ourselves and others are necessary and self-evident. In Foucauldian discourse, determining the liberatory status of any discourse is a matter of historical inquiry, not 93 Catriona Macleod and Kevin Durrheim theoretical orpoliticalpronouncement(Sawicki 1988). For example, Foucault observes thatpsychoanalysis played a liberating role in relation to psychiatry, but as a global theory it has contributed to forms of social control and normalisation. In the same way obstetric practice may have played a liberating role in many women's lives, while at the same time providing the tools for regulating others'. For Foucault liberation is intimately associated with resistance. For example, in 'The Subject andPower', he proposes a 'new economy ofpower relations' in which the starting point is the forms of resistance to various forms of power (Foucault 1982). These forms of resistance have features in common, three of which are: (1) they question the status of the individual whether by asserting the right to be different or by criticising the separation of the individual from the community; (2) they oppose the effects of power which are linked to knowledge; and (3) they are a refusal of abstractions which determine who one is. In terms of our practices regarding young women and their sexual and reproductive behaviour, these may translate into: (1) a questioning of the taken-for-granted assumptions concerning sexually active and pregnant teenagers as well as the associated notions invoked in discussions on teenage pregnancy, such as mothering, the family, marital status, cultural and racial issues (see Macleod 1999c, 2001); (2) deconstructing current scientific knowledge about teenage pregnancy, problematising its claim to objective neutrality andrevealing its, for example, gendering effects; and (3) refusing abstractions such as the term 'teenage pregnancy', or expert status. Foucault advocates this type of resistance as an antidote to our modern forms of power: Maybe the target nowadays is not to discover what we are but to refuse what we are. We have to imagine and to build up what we could be to getrid of this kind of political 'double bind', which is the simultaneous individualisation and totalisation of modern power structures. ... We have to promote new forms of subjectivity through the refusal of this kind of individuality which has been imposed on us for several centuries (Foucault 1982:785). This does not imply abandoning scientific endeavours or the provision of health services. Indeed much of what we do may, on the surface, look the same (eg interviewing young women for research purposes or providing clinical treatment for young pregnant women in need of care). However, these practices would be infused by the 'new economy of power relations' referred to above. 94 Psycho-medical discourse in South African research on teenage pregnancy Conclusion The purpose of this paper has been to inspect scientific discourse on teenage pregnancy in South Africa, on the basis that these discursive events are not innocent ofpolitical effects. The proliferation of savoir regarding the pregnant (and potentially pregnant) teenager has made available the technologies to render the reproductive adolescent a subject of governmentality. Through the mutually reinforcing processes of medicalisationandpsychologisationthe 'pregnant teenager' and the 'sexual teenager' are constructed, produced, normalised and pathologised. As a subject of government, the teenager's hidden bodily, psychological, social and cultural processes are brought into the plane of visibility and simultaneously rendered calculable and manageable though scientific and clinical method. Construction of the object - the pregnant (or potentially pregnant) teenager - occurs through normalising judgement and simultaneous individualisation and totalisation. Each of the discursive events discussed (sanitary science, social medicine, community health and psychologisation) deploy these technologies in specific but interconnected ways. Normalising judgement depends on hierarchical observation effected through the clinical gaze and surveillance which eventually turns to self- monitoring. The pregnant teenager, her anatomy, behaviour, psyche and relationships are evaluated according to (mostly unstated) norms — norms concerning, inter alia, the optimal reproductive body, the timing of reproduction, the perfect mother, the developing adolescent, the self- actualised, fulfilled, economically active individual, gender and conjugal relations, and the culturalised individual. But, as Rose points out, 'Normality is not an observation but a valuation' (1989:123), with these valuations having political implications - judgements about what is desirable (eg reproduction within certain age ranges and within a conjugal bond) are made which translates into an injunction as to the goal to be achieved. This is not to say that a state of valuelessness is desirable (or even possible), but rather that the political implications of these values need careful inspection. There are two effects of the multiplication of the processes of governmentality surrounding adolescent sexuality and reproduction reflected inscientific discourse. The first is thatnew aspects of adolescents' lives (not only their bodies but also their behaviour, emotions, social and cultural interactions) are brought into the plane of visibility. The practices of intervention (the clinical examination, ante-natal care, counselling, sexuality 95 Catriona Macleod and Kevin Durrtieim education) discussed in the literature are implemented not only in the lives of those falling outside the range of the normal, but also those deemed normal as they are incited in the labours required to maintain normality. The surface of the object of government (the pregnant and non-pregnant adolescent) is mapped in finer lines, allowing for more points of gradation, and a more intricate and in-depth knowledge of her nature. The second effect is that power relations between the expert and teenager have shifted from overt to covert regulation. The health service provider no longer prescribes programmes (as in sanitary science), but rather invokes the participation of the teenager, her family and the community in the production of adolescent sexual and reproductive health and welfare. S/he does not impose, but relies on action at a distance, whereby teenagers, their families and communities are incited to monitor their own behaviours against the backdrop of the normative framework provided by the expert. The expert, teenager, family, community, and policy-makers and public servants all become the overseer and the overseen. The teenager is incited to be true to her nature, the family to fulfil normal family functions, the expert to be knowledgeable concerning the nature of teenage pregnancy and proficient in efficacious methods, the community to take responsibility for localised issues, and the policy-maker and public servant to institute population-based interventions. Acknowledgements We thank the colleagues who responded to our requests for information regarding research conducted on teenage pregnancy in their departments or organisations, and the Johan Jacobs Foundation for financial assistance. References Arney, WR and BJ Bergen (1984) 'Power and visibility: the invention of teenage pregnancy', Social Science and Medicine 18(1). Blom, SM (1990) 'An investigation into family life of unmarried pregnant adolescents'. Unpublished Masters thesis, University of the Free State. Boult, BE (1992) 'Children having children: black adolescent multiparas', Salus 14(3). Boult, BE and PW Cunningham (1992) 'Black teenage pregnancy: asocio-medical approach', Medicine and Law 11. (1993) 'Some aspects of obstetrics in Black teenage pregnancies: a comparative study of three age groups'. Research paper C28, University of Port Elizabeth. 96 Psycho-medical discourse in South African research on teenage pregnancy Brits, WJ(1989) 'Die ontoereiken.de verwerkliking van sosio-pedagogiese essensies as aanleiding tot tienerswangerskap'. Unpublished Masters thesis, University of Pretoria. Burman, E, A Kottler, A Levett, and I Parker (1997) 'Power and discourse: culture end change in South Africa', in A Levett, A Kottler, E Burman and I Parker (eds) Culture, Power and Difference: discourse analysis in South Africa. Cape Town: University of Cape Town Press. Butchart, A (1995) 'On the anatomy of power: bodies of knowledge in South African socio-medical discourse'. Unpublished Doctoral thesis, University of South Africa. (1998) The Anatomy ojPower: European constructions oj'the African body. Pretoria: Unisa Press. Carolissen, RL (1993) 'The social context of adolescent pregnancy: the case of Mamre'. Unpublished Masters thesis, University of Cape Town. Dean, M (1994) '"A social structure of many souls": moral regulation, government, and self-formation', Canadian Journal of Sociology 19(2). De Villiers, VP (1991) 'Seksonderrig onder tieners in die Paarl', South African MedicalJournal 80(2). Deveaux, M (1994) 'Feminism and empowerment: a critical reading of Foucault', Feminist Studies 20(2). Fairclough, N (1992) Discourse and Social Change. Cambridge: Polity Press. Foucault, M (1976) TheBirth of the Clinic: an archaeology of medical perception. London: Tavistock. (1977) Discipline and Punish: the birth of the prison. London: Penguin. (1978) The History of Sexuality: voll. London: Penguin. (1980) 'The eye of power', in C Gordon (ed) Power/Knowledge: selected interviews and other writings 1972 - 1977. New York: Pantheon. (1982) 'The subject and power', Critical Inquiry 8 (summer). (1991a)' Governmentality', in G Burchell, C Gordon and P Miller (eds) The Foucault Effect: studies in governmentality. Chicago: University of Chicago Press. (1991b) 'Questions of method', inG Burchell, CGordonandPMiller(eds). The Foucault Effect: studies in governmentality. Chicago: University of Chicago Press. Fouch6, M (1992) 'Die sielkundige behoeftes van swanger adolessentete Pelonomi Hospitaal'. Unpublished Masters thesis, University of the Free State. Furstenberg, FF (1991) 'As the pendulum swings: teenage childbearing and social concern', Family Relations 40(2). 97 Catriona Maclood and Kevin Durrheim (1992) 'Teenage childbearing and cultural rationality: a thesis in search of evidence', Family Relations 41 (2). Geronimus, AT (1991) Teenage childbearing and social and reproductive disadvantage: the evolution of complex questions and the demise of simple answers', Family Relations 40(4). Gordon, C (1991) 'Governmental rationality: an introduction', in G Burchell, C Gordon and P Miller (eds). The Foucault Effect: studies in governmentality. Chicago: University of Chicago Press. Grimshaw, J (1993) 'Practices of freedom', in C Ramazanoglu (ed) Up Against Foucault: explorations of some tensions between Foucault and feminism. London: Routledge. Hartsock, N (1990) 'Foucault on power: a theory for women?', in LJ Nicholson (ed) Feminism/Postmodernism. New York & London: Routledge. Hoy, DC (1988) 'Foucault: modern or postmodern?' in J Arac (ed) After Foucault: humanistic knowledge, postmodern challenges. New Brunswick: Rutgers University Press. Keogh, J (1988) 'Sex counselling of the male adolescent', Nursing RSA 3(9). Macleod, C (1999a) 'Teenage pregnancy and its "negative" consequences: review of South African research- Part V, South African Journal of Psychology 29(1). (1999b) 'The "causes" of teenage pregnancy: review of South African research - Part 2', South African Journal of Psychology 29( 1). (1999c) "The governmentality of teenage pregnancy: scientific literature and professional practice in South Africa'. Unpublished Doctoral thesis, University of Natal, Durban. (2001) 'Teenage motherhood and the regulation ofmothering in the scientific literature: the South African example', Feminism and Psychology 11(4). (2002) 'Economised security and the scientific literature on teenage pregnancy in South Africa', Gender and Society 16(5). Macleod, C and K Durrheim (2002) 'Racialization in the scientific literature on teenage pregnancy: the South African example', Ethnic and Racial Studies 25(5). Mukasa, FM (1992) 'Comparison of pregnancy and labour in teenagers and primigravidas aged 21 -25 in Transkei', South African Medical Journal 81(8). Nash, E (1990) 'Teenage pregnancy - need a child bear a child?' South African Outlook 120(1432). Ncayiyana, DJ and G Ter Haar (1989) 'Pregnant adolescents in rural Transkei', South African Medical Journal 75(5). Nkomo, M (1991) 'Epistemological and disciplinary transformation in post- Apartheid South Africa' in J.Jansen(ed) Knowledge and Power in South Africa: 98 Psycho-medical discourse in South African research on teenage pregnancy critical perspectives across the disciplines. Johannesburg: Skotaville. Oosthuizen, V (1990) 'The social problem of teenage pregnancy - a review', Southern African Journal of Child and Adolescent Psychiatry 2(2). Parekh, A and C De la Rey (1997) ' Intragroup accounts of teenage motherhood: a community-based psychological perspective', South African Journal of PsychologyTKA). Parker, I (1992) Discourse Dynamics: critical analysis for social and individual psychology. London: Routledge. Pond, TS (1987) 'A psychodynamic approach to the adolescent unmarried mother'. Unpublished Masters thesis, University of Pretoria. Preston-Whyte, E (1991) 'Teenage pregnancy in selected coloured and black communities'. Unpublished report, Human Sciences Research Council, Pretoria. Preston-Whyte, E and J Allen (1992) 'Teenage pregnancy in the Coloured community', in S Burman and E Preston-Whyte (eds) Questionable Issue: illegitimacy in South Africa. Oxford: Oxford University Press. Preston-Whyte, E and M Zondi (1989) 'To control their own reproduction: the agenda of black teenage mothers in Durban', Agenda 4. (1991) 'Adolescent sexuality and its implications for teenage pregnancy and AIDS', South African Journal of Continuing Medical Education 9( 11). (1992) 'African teenage pregnancy: whose problem?' in S Burman and E Preston-Whyte (eds) Questionable Issue: illegitimacy in South Africa. Oxford: Oxford University Press. Rockey, T (1986) 'Teenage pregnancy', Nursing RSA 1(11). Rose, N (1989) Governing the Soul: the shaping of the private self. London: Routledge. — — (1990) 'Psychology as a "social" science', in I Parker and J Shotter (eds) Deconstructing Social Psychology. London: Routledge. (1992) 'Engineering the human soul: analyzing psychological expertise', Science in Context 5(3). (1993) 'Government, authority and expertise in advanced liberalism', Economy & Society 22(2). Sampson, EE (1990)' Social psychology and social control', in I Parker and J Shotter, (eds) Deconstructing Social Psychology. London: Routledge Sawicki, J (1988) 'Feminism and the power of Foucauldian discourse', in J Arac (ed) After Foucault: humanistic knowledge, postmodern challenges. ~Nev/Brunswic Rutgers University Press. Schoeman, MN (1990) 'Sexuality education among black South African teenagers: what can reasonably be expected?', Curationis 13(3). 99 Catriona Maclsod and Kevin Durrheim Stoler, A (1995). Race and the Education of Desire: Foucault's History ofSexuality and the colonial order of things. Durham: Duke University Press. Swanson, M (1977) 'Sanitation syndrome: bubonic plague and urban native policy in the Cape colony, 1900 - 1909', Journal of African History 18(3). Tanga, TT, and LR Uys (1996) 'The social support systems of unmarried pregnant adolescents with particular reference to parents', Curationis 19(1). Vaughan, M (1991) Curing their Ills: colonialpower and African illness. Cambridge: Polity Press. Vinovskis, MA(1988)^n 'Epidemic'ofAdolescent Pregnancy: some historical and policy considerations. New York: Oxford University Press. 100