Sex and Youth: Contextual Factors Affecting Risk for HIV/AIDS

Sex and youth: contextual factors affecting risk for HIV/AIDS

Joint United Nations Programme on HIV/AIDS

UNICEF • UNDP • UNFPA • UNDCP UNESCO • WHO • WORLD BANK UNAIDS/99.26E (English original, May 1999)

© Joint United Nations Programme on HIV/AIDS (UNAIDS) The designations employed and the presentation of the 1999. All rights reserved. This document, which is not a for- material in this work do not imply the expression of any mal publication of UNAIDS may be freely reviewed, quoted, opinion whatsoever on the part of UNAIDS concerning the reproduced or translated, in part or in full, provided the legal status of any country, territory, city or area or of its source is acknowledged. The document may not be sold or authorities, or concerning the delimitation of its frontiers used in conjunction with commercial purposes without and boundaries. prior written approval from UNAIDS (Contact: UNAIDS Information Centre).

UNAIDS – 20 avenue Appia – 1211 Geneva 27 – Switzerland Tel.: (+41 22) 791 46 51 – Fax : (+41 22) 791 41 65 e-mail: [email protected] – http://www.unaids.org Sex and youth: contextual factors affecting risk for HIV/AIDS A comparative analysis of multi-site studies in developing countries

UNAIDS BEST PRACTICE COLLECTION

Young people and risk-taking in sexual relations Community responses to AIDS Use of the female condom: genders relations and sexual negotiation

Young people and risk-taking in sexual relations This set of studies presents a comparative analysis of data collected in interviews and discussions with nearly 3000 young people in 7 countries in Africa, Asia and the Americas. Strikingly similar themes and issues emerge, including concepts of youth, the challenge to traditional cultures, modernization and urbanization revealing a complex and heterogeneous situation for young people and no one HIV/STD prevention strategy. Future prevention efforts must take into account the impact of dominant sexuality frameworks, the onset of sexual activity, the importance of the body for young people, mass media, risk assessment and safer sex.

Community responses to AIDS A comparative analysis of the resulting data from 5 countries identified local beliefs about HIV/AIDS, the community and household responses and the inter-relations between the two. Key factors influencing the responses include the existing economic situation, prevailing relations between men and women in the communities and households, local beliefs in health and health care and local levels of stigmatization. Recommendations are made for policy and programme development.

Use of the female condom: genders relations and sexual negotiation This third set of studies first collected data in 4 countries on gender relations, sexual communication and negotiation followed by an intervention to strengthen womens capacity in these latter areas. The comparative analysis clearly identified economic dependence on men and gender stereotypes as the two major factors constraining women in their sexual behaviour. The report finishes with specific recommendations.

This bare summary of these 3 pioneering sets of studies investigations into the determinants of HIV-related vulnerability cannot convey the extraordinary wealth of data and the richness of experiences and feelings reported by the participants with striking frankness. The volume will be read and re-read by national authorities, programme designers and managers, researchers and intervention specialists. In addition it will be of great interest and value to all those who are interested in the issues surrounding young people and HIV/AIDS, sexual behaviour, communication and negotiation, the improvement and strengthening of responses for the benefit of people living with HIV/AIDS, their carers and their communities, gender roles and the options for women who want to protect themselves against HIV and other STDs as well as pregnancy.

UNAIDS, 20 Avenue Appia, CH-1211 Geneva 27, Switzerland. Tel.: (+41 22) 791 46 51 - Fax: (+41 22) 791 41 65 e-mail: [email protected] Sex and youth: contextual factors affecting risk for HIV/AIDS

A comparative analysis of multi-site studies in developing countries

Joint United Nations Programme on HIV/AIDS

UNICEF • UNDP • UNFPA • UNDCP UNESCO • WHO • WORLD BANK

UNAIDS Geneva, Switzerland 1999 Sex and youth: contextual factors affecting risk for HIV/AIDS

A comparative analysis of multi-site studies in developing countries

Part 1 Young people and risk-taking in sexual relations Gary Dowsett1 and Peter Aggleton2

Part 2 Community responses to AIDS Peter Aggleton and Ian Warwick2 (from an original analysis by Ian Warwick)

Part 3 Use of the female condom: gender relations and sexual negotiation Peter Aggleton, Kim Rivers2 and Sue Scott3 (from an original analysis by Sue Scott)

1 Centre for the Study of Sexually Transmissible Diseases, La Trobe University, Melbourne, Australia 2 Thomas Coram Research Unit, Institute of Education, University of London, UK 3 Department of Applied Social Science, University of Stirling, UK ACKNOWLEDGEMENTS We would also like to thank Roger Ingham who provided valuable advice in relation to protocol design, and Anne Malcolm and Annie George who pro- Part 1 vided technical assistance during the conduct of the studies. In Geneva, Eva Young people and risk-taking in sex- Rodrigues, Penny Ratcliffe, Sylvie ual relations Schaller, Paula Soper and Julie Dew were among those who provided secre- We thank the principal investigators tarial and administrative support. In and research teams in each study site, London, Helen Thomas, Maria Harrison without whom these investigations and Paula Hassett did the same. could never have been conducted: Séverin-Cécile Abega at l’Université de Last but not least we would like to thank Yaoundé, Cameroun; Carol Jenkins at all those who participated in the studies the Papua New Guinea Institute of themselves. Without their involvement Medical Research, Goroka, Papua New and willingness to speak openly about Guinea; Teresa Marshall at the their experiences, this work could never Corporación de Salud y Políticas have been completed. Sociales, Santiago, Chile; Agnes Runganga at the University of Zimbabwe, Harare, Zimbabwe; Jacobo Gary Dowsett and Peter Aggleton Schifter at the Instituto Latinamericano Sydney and London de Prevención y Educación en Salud, San José, Costa Rica; Michael Tan at the Health Action Information Network, Part 2 Quezon City, Philippines; and Chou Community responses to AIDS Meng Tarr at the Cambodian AIDS Social Research Project, Phnom Penh, We thank the principal investigators Cambodia. We would also like to thank and research teams without whom Purnima Mane for her detailed com- these multi-site investigations could ments on an earlier draft of this never have been conducted: Shalini Comparative Analysis. Bharat of the Tata Institute of Social Sciences, Mumbai, India; Melkizedeck The multi-site studies were designed Leshabari of Muhimbili University and implemented according to a College of Health Sciences, Dar es General Research Protocol developed Salaam, the United Republic of by Gary Dowsett, a member of the for- Tanzania; Roberto Castro of the Centro mer Social and Behavioural Studies and Regional de Investigaciones Support Unit of the World Health Multidisciplinarias, Cuernavaca, Organization’s Global Programme on Mexico;4 Rafael Garcia of Universidad AIDS. He is now Deputy Director of the Autónoma de Santo Domingo, Centre for the Study of Sexually República Dominicana; and Anchalee Transmissible Diseases at La Trobe Singhanetra-Renard of Chiang Mai University, Melbourne, Australia. University, Thailand. We would also like

4 Previously of the Instituto Nacional de Salud Pública, Cuernavaca, Mexico.

1 to thank Purnima Mane, Michel Caraël, Enseñanza en Nutrición y Salud Mark Connolly and Binod Mahanty of (INCIENSA) and Gisela Herrera of the UNAIDS for their helpful comments on Departmento Control del SIDA in Costa an earlier draft of this report. Rica. In Mexico, they were coordinated by Griselda Hernandez of the Consejo The protocol for the studies described Nacional de Prevención y Control del here was designed by Jane Verrall SIDA (CONASIDA). In Senegal, they working within the Social Studies and were coordinated by Cheikh Ibrahim Support Unit of the World Health Niang at Université Cheikh Anta Diop Organization’s former Global de Dakar. In Indonesia they were coor- Programme on AIDS. She is now dinated by Bernadette Setiadi, of the Director of the International Union for Lembaga Psikologi Terapan, Fakultas the Scientific Study of Population, Psikologi, Universitas Indonesia. Liège, Belgium. We would also like to thank Susan Kippax and Werasit The studies described in this report Sittitrai for their valuable advice in rela- were developed from a general tion to protocol development, and research protocol designed by Annie George for technical assistance Purnima Mane who is now Programme during the studies. Administrative and and Planning Coordinator with the secretarial support was provided by Joint United Nations Programme on Eva Rodrigues, Penny Ratcliffe, Sylvie HIV/AIDS (UNAIDS). We would also Schaller, Paula Soper and Julie Dew in like to thank Anne Malcolm, Annie Geneva, and by Helen Thomas, Maria George and Jane Verrall who provided Harrison and Paula Hassett in London. technical assistance in the course of the research; Eva Rodrigues, Penny Finally, thank the individuals, families Ratcliffe, Sylvie Schaller and Paula and households that participated in the Soper who provided secretarial and local studies. Without their involve- administrative support in Geneva; and ment and willingness to speak openly Helen Thomas, Maria Harrison and about their experiences, the research Paula Hassett who did the same in described here could never have been London. Finally, we would like to thank completed. Chartex International for providing the female condoms distributed in each Peter Aggleton and Ian Warwick study site. London Peter Aggleton, Kim Rivers and Sue Scott Part 3 London and Stirling Use of the female condom: gender relations and sexual negotiation

We thank the research teams in Costa Rica, Mexico, Senegal and Indonesia without whom these studies could never have been completed. In Costa Rica the investigations were coordinated by Jorge Elizondo of the Instituto Costarricense de Investigación y

2 CHAPTER X

INTRODUCTION nerability was first being recognized. All were commissioned by the World Health Organization’s Global Since early in the epidemic, enquiry Programme on AIDS (WHO/GPA), into the factors influencing HIV-relat- were supported by the GPA Steering ed vulnerability has been recognized Committee on Social and Behavioural as essential for prevention efforts. Research, and were completed with While much early work focused on the the active support of UNAIDS. They individual determinants of sexual and vary in their focus from investigations drug-related risk-taking, increasingly examining the contextual factors the contextual factors which render affecting risk-related sexual behaviour some groups more vulnerable than among young people, through others has come to be recognized. enquiry into the manner in which Factors as diverse as age, gender, households and communities respond social position, economic status, cul- to HIV and AIDS, to studies of gender tural norms, beliefs and expectations relations, sexual negotiation and the determine the risks faced, and enable female condom. and constrain individuals in their actions. WHO Regional Offices, WHO representatives and WHO/GPA in-country It is now widely recognized that both staff played an important role in iden- individual persuasion and social tifying the principal investigators and enablement are essential for pro- research teams which conducted the gramme success, and increasing num- studies. Each investigation was con- bers of prevention programmes and ducted by local researchers who were activities are designed on this encouraged to work closely with assumption. The recent UNAIDS national HIV/AIDS programmes and report Expanding the global response nongovernmental organizations to to HIV/AIDS through focused action maximise the potential usefulness of recognizes the importance of such an their findings. Technical assistance approach and seeks to encourage its and support was provided to the stud- application in countries across the ies both by the staff of UNAIDS and world. Yet knowing how to develop by advisers. The preparation of the and fine-tune programmes requires comparative analyses of findings was insight into the often complex deter- undertaken by Peter Aggleton and minants of behaviour in specific cul- colleagues at the Thomas Coram tural settings and contexts. Good Research Unit in London working quality social enquiry has a key role to closely with the consultants and play in providing this information and authors named in each individual in supporting the development of report. work that is attuned to the needs of particular groups. The reports identify the principal findings from each of the three sets of The three sets of studies described studies along with policy and practice here were begun at a time when the implications. They highlight the value importance of contextual enquiry into of close focus research that remains the determinants of HIV-related vul- sensitive to the intentions, beliefs, life

3 projects and experiences of individuals, families, households and communities. They provide stimulus for further work in each of the fields described using complementary methods and approaches. We hope that national authorities and programmes, researchers, programme developers and intervention specialists find them interesting, and that they act as an important source of impetus for future work.

Michel Caraël Team leader: Prevention UNAIDS

4 CONTENTS

ACKNOWLEDGEMENTS 1 INTRODUCTION 3

PART 1 Young people and risk-taking in sexual relations

Chapter 1 INTRODUCTION 10 Existing research frameworks 11

Chapter 2 BACKGROUND 13 Preliminary literature review 13 Subsequent literature review 14 Conceptual framework 16

Chapter 3 STUDY DESIGN AND METHODOLOGY 20 Research Plan 20 Rapid assessment 20 Main fieldwork 21 Analysis and write up 21 Principal research methods 21 Additional research methods 22 Fieldwork sites and sampling 22 Sample sizes 24 Time scale 25

Chapter 4 COMPARATIVE ANALYSIS 26 Key concepts 26 Sexual culture 26 Sexual identity 27 Sexual meaning 28 The purpose of the three concepts 30 Similarities 31 Concepts of youth 31 The challenge to traditional sexual culture 33 Gender 34 Modernization 39 Urban, rural and provincial aspects 42 Other effects of larger social forces 42 Comparisons 42 The impact of dominant understandings of sexuality 43 Onset of sexuality activity 44 Embodiment 44 Mass media 45 Space 45

5 CONTENTS

HIV risk assessment, condom use and ideas on safer sex 47

Conclusions 47

Chapter 5 RECOMMENDATIONS 49 Conceptual leadership 49 Programmatic responses 49 Gender 50 Dissemination 52 Research 52

COUNTRY REPORTS 54

REFERENCES 55

PART 2 Community responses to AIDS

Chapter 1 INTRODUCTION 58

Chapter 2 BACKGROUND 60 Societal impact and responses 61 Community impact and responses 62 Household impact and responses 63 Concluding comments 65

Chapter 3 STUDY DESIGN AND METHODOLOGY 66 Research questions 66 Study design and samples 66 Study sites 68 Dominican Republic (Santo Domingo) 68 India (Mumbai) 68 Mexico (Netzahualcóyotl and Mexico City) 69 Thailand (Chiang Mai Province) 69 United Republic of Tanzania (Kyela District) 70 Data analysis 71

Chapter 4 COMPARATIVE ANALYSIS 72 Local beliefs 72 Beliefs about HIV and AIDS 72

6 CONTENTS

Beliefs about transmission 73 Other beliefs 76 HIV transmission: children and young people 78 Household responses 79 Becoming aware of AIDS 79 Family support 80 Income and household care 82 Illness progression 83 Community responses 84 Relatives 84 Neighbours and other community members 84 Volunteers 85 Care by friends and community members 86 The household and the community 87 Death and mourning 87 Care of bereaved children 88 Treatment and health care 89 Traditional and complementary therapies 90

Chapter 5 CONCLUSIONS 92 Paying your way... 92 ...in a man’s world 93 Stigma and discrimination 93 Public, private, voluntary and informal sector care 94 Bereavement and funerals 95 Recommendations 95 Policy development 95 Programme development 96

COUNTRY REPORTS 99

REFERENCES 100

PART 3 Use of the female condom: gender relations and sexual negotiation

Chapter 1 INTRODUCTION 104

Chapter 2 BACKGROUND 106 Context 106 Gender, power and sexuality 108 Sexual communication and ‘negotiation’ 108 The female condom 110 Initial responses 110 Acceptability 111 Effectiveness 111

7 CONTENTS

Cost 111 Sexual communication 111 Women’s empowerment 112 Concluding comments 113

Chapter 3 STUDY DESIGN AND METHODOLOGY 114 Research questions 114 Gender relations 114 Communication and negotiation 114 Risk awareness and protective behaviour 114 Study design and samples 115 Study sites 117 Costa Rica 117 Indonesia 118 Mexico 119 Senegal 120 Data analysis 121

Chapter 4 COMPARATIVE ANALYSIS 122 Prevailing gender roles and relations 122 Sexual communication 125 Perceptions of risk 129 Negotiating safer sex 130 Responses to the female condom 131

Chapter 5 CONCLUSIONS 135 Recommendations 137

COUNTRY REPORTS 139

REFERENCES 140

8 Part 1 Young people and risk-taking in sexual relations

Gary Dowsett1 and Peter Aggleton2

1 Centre for the Study of Sexually Transmissible Diseases, La Trobe University, Melbourne, Australia 2 Thomas Coram Research Unit, Institute of Education, University of London, UK CHAPTER 1

INTRODUCTION The category ‘young people’ is a recent one. Like its predecessors, the term attempts to distinguish between childhood and full adulthood, and historically the gap between these two has widened. In the West, and particu- Those attending any major international larly since World War II, increased or national conference on HIV/AIDS retention rates in secondary schooling cannot help but notice the concern and rising enrolments in postschool about, and emphasis on, young people. educational institutions have increased The XI International Conference on young people’s dependency on their AIDS in Vancouver saw abstracts for no families, communities and the state. In less than 319 posters and papers about contrast, more than half of all 14–16- aspects of young people’s risk of HIV year-olds (depending on the country) infection, or responses to it. As a major were likely to be in the workforce only sector in almost every country’s popula- a few generations ago, contributing to tion seen as at enhanced risk in this pan- family incomes or working for family demic, young people are viewed as a businesses, farms and/or in domestic ‘problem’ — a problem for prevention labour in the home. This last point also education, and a problem for potential highlights the fact that gender differ- HIV transmission. ences in relation to paid and unpaid work, and in relation to school reten- This problem is captured by the very tion and further workforce training, words so often used to describe young have changed rapidly in the West in people — youth, youngsters, teenagers, the last 50 years. juveniles, minors, adolescents. However, the term ‘young people’ is itself ambigu- In developing countries, ambiguity ous. The term is often used to refer to concerning young people is even more those between the ages of 12 and 26, pronounced. The process of although in many cultures the upper Westernization occurring in many level of this cohort should properly be developing countries brings with it new regarded as fully mature adults. Were ideas, new influences and novel ways the age of onset of sexual activity to be of understanding. It is important that used as a starting point for the lower Western concepts of young people, as limit, then we would have to recognize an age cohort dependent on family, that in some parts of Africa about 40% of community and/or the state in an 15-year-old young men have experi- extended process of preparation for enced intercourse, whereas in some adult and productive life, are not parts of south-east Asia, over 80% of 20 applied uncritically to developing year-olds are virgins, still waiting to trans- countries. It is important too that form themselves from ‘children’ into young people are not seen globally as ‘young people’ (Cleland & Ferry 1995). It some artificial homogeneity. This is is this difference between the cultural especially important in the formulation experiences and understandings of dif- of public health strategies to limit the ferent groups of young people that effects of HIV. Local specificity in public makes direct comparisons between health responses indicates a recogni- countries and cultures so difficult. tion that different epidemics of HIV

10 CHAPTER 1

(and sexually transmissible diseases sity students. Second, there is a distinct [STDs]) exist and that there is a need absence of young people themselves in for cultural appropriateness. most research. Young people constitute a population most usually researched It was the awareness, both of the wide- upon, a population seen as being ‘at- ly acknowledged vulnerability of young risk’, almost by definition. They are a people to HIV infection globally, and group regarded in most societies as yet the need for better informed under- to become real persons, full citizens standings of cultural differences and and to be accorded full human rights. forms, that led to the development of They are a group whose behaviour, par- the multi-site study described here. ticularly sexual behaviour, is often The title of the research programme — regarded as premature if not immature, Studies of contextual factors affecting immoral or at least unfortunate, and risk-related sexual behaviour among whose own ideas, experiences and young people in developing countries concerns about sexuality are mostly — hereafter the Contextual Factors neglected by society at large. study, brings this concern with cross- cultural difference to the fore, and the The studies described here aim to con- seven research teams who carried out tribute to new thinking about young these studies successfully document people and sexual health. They such differences. Yet as we will see demand that we grapple with the there are striking similarities too, the changing situation for young people in most apparent being the situation of Cameroon, for example, as continuing young women when it comes to urbanization radically transforms the attempting to ‘negotiate’ sexual rela- rules for sexual conduct. They encour- tions with men. The value, then, of this age a recognition that some middle- cross-cultural comparison lies both in class young women in Costa Rica are its capacity to reveal differences and competently handling significant similarities, and in its ability to draw changes in gender roles, while less lessons and ideas from these seven affluent young men are struggling to countries that might apply elsewhere. maintain their sense of self through sexual success. They suggest we should take very seriously the fact that Existing research young women experience consider- frameworks able coercion in sex in Papua New Guinea, yet these same women claim Research on young people and the right to physical sexual pleasure, HIV/AIDS has changed little throughout following their Western counterparts, the pandemic. The global literature on and distancing themselves from their this topic (which is reviewed briefly in mothers’ and grandmothers’ experi- Chapter 2) is familiar in two respects. ences and understandings of sex. And First, a large number of surveys have they encourage us to stop ignoring the been undertaken in many countries to same-sex pursuit of sexual pleasure examine young people’s knowledge, reported by young people in every attitudes, practices and behaviours. country in this study, thereby recogniz- These have usually been conducted ing the very real diversity of sexual among accessible populations of experience among young people young people such as school or univer- across the world.

11 CHAPTER 1

Health promotion successes in HIV It is in relation to those broader con- prevention share certain common cerns that the studies reported here characteristics, and a central one of were intended to make their mark. these is the notion of community inclusion in processes of intervention The material reported from each development and deployment. A key country is rich. Similar methodolo- message from the seven countries gies assured that certain issues were that participated in this multi-site well covered by all seven studies, study is that young people them- and these will be discussed in detail selves must be included in the devel- in Chapter 3. Differences in opment of health promotion pro- approach between countries, how- grammes if prevention approaches ever, also enabled the studies to are to be relevant. Young people’s provide quite specific insights into sexual conduct, the meanings they young people’s sexual culture in par- generate concerning sex, and the ticular contexts. Although we cannot individual and shared investment they generalize these findings to all have in sexual exploration, pleasure young people, they are valuable in and activity, must be represented in a helping us rethink notions of sexual- plausible and respectful manner if we ity. They enable us to think about are to convince young people every- sexual expression as a set of mean- where that HIV/AIDS is a real danger ingful acts, not just as a biological to them. urge. In this way, sex can be seen as a deeply inscribed process of self- To achieve this accurate and fitting construction, pursued in the context representation of young people’s sex- of changing social expectations and ual conduct in health promotion often rapid economic change. material and programmes, it is important to know more than their specific The report has been written for a sexual practices, the frequency of wide range of people working on coitus, the age of sexual initiation, or issues relating to young people and the extent of sexual experimentation HIV/AIDS. It describes work initiated of various kinds. While this kind of by the World Health Organization’s information is important in gauging former Global Programme on AIDS the size of the ‘problem’, rarely does (WHO/GPA) and completed with the it help us know how to educate, what support of the Joint United Nations curricula to use and how to teach. Programme on HIV/AIDS (UNAIDS).

12 CHAPTER 2

BACKGROUND ‘contributed little to recent attempts to plan and introduce appropriate interventions to encourage risk-reducing behaviours’3. As for the others, the major weakness was their lack of, and need for, deeper understandings of The Contextual Factors study was social and cultural contexts as the basis conceived of as a multi-site compara- for the development of perceived nor- tive study guided by a General mative attitudes and behaviours. Research Protocol that outlined the overall research questions, objectives In reviewing this literature on young and methodology. people’s sexuality, a number of emphases in existing social and behavioural research were identified. Preliminary literature Much previous research had focused review on: (1) measuring patterns of sexual Before this General Research Protocol behaviour (e.g., age at first was developed, a literature review intercourse; levels of sexual was commissioned to provide an up- activity; experience of anal to-date picture of young people’s sex- intercourse); uality at that time, in relation to risks (2) ascertaining the meanings of associated with HIV infection and sexual activity for young people AIDS.3 In this review, various method- (e.g., familiarity with sexual ological approaches to investigating partners before intercourse; young people’s sexuality were dis- characteristics of first sexual cussed, and their comparative partner, such as age, level of strengths assessed. Issues such as cul- experience; category of first tural appropriateness, flexibility in sexual partners, such as sex use, terminology, sequencing of ques- work client, peer, spouse; rea- tions and issues, implied values, and sons for, and reactions to, first the divergence between medical/sci- sexual intercourse; reported entific and lay persons’ understand- condom use); ings of HIV/AIDS were examined. (3) describing community understandings of sexual activity Four main theoretical frameworks regu- (e.g., generational differences; larly used in investigating young peo- traditional values; understand- ple’s sexuality were identified: the indi- ings of sex work); vidual differences approach; the behav- (4) identifying reported sources of ioural modelling approach; the sexual information about sex and ‘scripts’ and discourses approach; and HIV/AIDS (e.g., families; over- structural explanations. The review con- seas influences; peer groups; cluded that, while each of these had schools; family planning pro- particular strengths, the first generally grammes; mass media);

3 Ingham R (1992) Personal and contextual aspects of young people’s sexual behaviour: A brief methodological and theoretical overview. Unpublished paper prepared for WHO/GPA/SSB, Geneva.

13 CHAPTER 2

(5) exploring the impact of conduct an exhaustive review of the HIV/AIDS on young people’s current literature on young people sexual activities (e.g., knowl- and HIV/AIDS, but to identify the edge of HIV transmission; overall direction of social research and issues of personal vulnerability; any more recent concerns. dynamics of condom use); (6) analysing the structural and One of the enduring types of research contextual aspects of sexual on young people, particularly among behaviour (e.g. gender, socio- school and college students, was economic issues). found to be the standard and now very familiar Knowledge, attitudes, The review concluded by identifying beliefs and practices (KAPB) survey. future research priorities in relation to These studies generally and consis- young people and HIV/AIDS. First, tently reveal the well-known disparity the cultural bases of normative atti- between levels of knowledge about tudes and behaviour were regarded AIDS and HIV transmission (usually as needing further investigation, with good) and changes in risk-taking prac- particular focus on socioeconomic tices in response (usually less good). issues and the relationship between Recent studies in Ethiopia (Mulatu & the individual and society. A second Haile, 1996; Teka, 1993), Zaire (Binda priority was the need to look closely at ki Muaka, Ndonda-Kumba & Lufua- how to make interventions sensitive to Mananga, 1993), Zimbabwe local situations and conditions. Third, (Campbell & Mbizvo, 1994), the sociopolitical context of research Nicaragua (Egger et al., 1993), was identified as important to the Namibia (Zimba, 1995), Kenya construction of longer-term research (Pattullo et al., 1994), and Mexico planning. (JOICEFP, 1994) offer examples of the continuing deployment of this type of Subsequent literature research. The findings of such surveys review are remarkably consistent and encourage some judicious rethinking of the A subsequent review of the literature need to sponsor more of this kind of on young people and HIV/AIDS was enquiry. While local data may be use- undertaken after the studies them- ful to confirm local specificities, the selves and prior to the preparation of reliance on college students in partic- this report. This revealed few signifi- ular, and the substantial sampling bias cant shifts in the type of research this produces in developing countries, being undertaken, although there must call into question the validity were some indications of a growing and usefulness of this type of enquiry. concern for contextual and cultural issues. This supplementary review A second issue in the more recent lit- involved examining abstracts from the erature concerns children living on the Berlin, Yokohama and Vancouver streets. While recent studies have International HIV/AIDS Conferences, reported on the prevalence of STDs, and conducting searches on the poor knowledge levels about Medline, Popline, Sociofile and HIV/AIDS and the part played by sex Psychlit CD ROM data bases for the in exchange for favours, goods and years 1992-96. The aim was not to money (Richens, 1994), there has

14 CHAPTER 2 been a notable shift in emphasis An added dimension to these reports toward a more cultural and contextual on cultural context and sources of understanding of the circumstances in meaning for young people has been which street children pursue and the documentation of significant gen- engage in sex. In South Africa, for der differences. Bhende (1993) has example, Richter and Swart-Kruger highlighted the effects of gender (1996) have highlighted the inadequa- inequalities on young women in cy of rational choice models of risk Bombay, India, arguing that women’s assessment and decision making in unequal status is an important factor the face of coercive sexual contexts, to be taken into account in health pro- pointing to social conditions of risk- motion. Ford and Kittisuksathit (1994) taking beyond the individual. Ruiz have noted the gendered nature of (1994) has stressed the importance of sexuality and sexual expression understanding street life as a culture among young Thai factory workers. that contextualizes risk-taking behav- Castaneda, Allen and Castaneda iour in Colombia. Raffaelli et al. (1993), (1996) have pointed to the contrast using a combination of qualitative and between young women’s and young quantitative methods, have described men’s expectations of sex and the ten- the integrated nature of sex in sion between cultural approval for exchange for goods, services and sexual experience among young men, (important and often neglected) plea- despite demands for sexual absti- sure in street life in a study of the early nence among young women. These and diverse sexual experiences of are important findings since they sug- street children in Belo Horizonte, gest that gender differences are Brazil. Together, these studies under- deeply structured and culturally line the importance of understanding inscribed sets of meanings and under- sexual behaviour among street chil- standings. dren not as isolated and individual risk-taking but as aspects of collective A number of other studies have behaviour deeply embedded in a sex- expanded our understanding of the ual culture. cultural factors affecting young people’s sexual conduct, arguing for the Other work concerning the meanings extended use of a range of qualitative of sex builds on this idea of sexual cul- research methodologies. Focus ture. Domingo (1995), for example, has groups and in-depth interviews fea- stressed the importance of motivation ture strongly in a 13 country study of in assessing sexual activity among young women’s sexuality described young Philippinos in order to develop by Weiss and Rao Gupta (1996) at the prevention interventions. There has XI International AIDS Conference in also emerged a concern with tensions Vancouver. Similar work on young in social life as a result of HIV/AIDS, men’s use of condoms in Zambia notably in relation to religious attitudes employed in-depth interviews and teachings on condom use (e.g., in (Feldman et al., 1993). Paiva (1993) Zaire - Haworth et al., 1996; in Mexico used action-research methods among - Castaneda, Allen & Castaneda, 1996). Brazilian young people, finding that This points to structural sources of con- traditional sexual culture was an flict for young people in determining important factor influencing expecta- decisions about sexual conduct. tions about sex. A combination of

15 CHAPTER 2 quantitative and qualitative methods In order to develop better education- in Thailand allowed a more complex al interventions to prevent HIV infec- and culturally contextualized picture tion, greater emphasis needs to be to emerge of young people’s commu- given to the ways in which young nication and interpersonal concerns people understand their social and (Pattaravanich, 1993). physical worlds, and to the social and cultural processes that help them This brief overview of the more recent make sense of sexual desires, feelings literature on young people, HIV/AIDS and interests. This theoretical and sexuality therefore reveals a approach emphasizes the social and growing focus on sexual culture and cultural rules and expectations that the social contexts within which differ- operate in particular societies. It ent kinds of sexual conduct occurs. It focuses on the processes through reinforces a sense of the timeliness which identities (including sexual and focus of the Contextual Factors identities) are acquired, developed studies described here, and encour- and maintained, and on the way in ages an appreciation of the value of which meanings about sexual behav- in-depth qualitative enquiry in the iour and sexuality are communicated pursuit of a better understanding of and learned. young people’s perceptions, beliefs and needs. A central theoretical assumption here is that sexual activity is, to a large Conceptual framework extent, socially constructed, in addition to having its biological compo- Taken together, findings from the two nents, and that all societies mould reviews identify a number of princi- basic sexual urges for their own social ples that informed the conceptualiza- or cultural purposes. A good example tion and development of the of this direct social moulding of sexu- Contextual Factors study and its qual- ality can be seen in the marked differ- itative methodology, using individual ences between countries in their legal and group interviews. Theoretically, ages of consent. It was important, the study was based on the belief therefore, that the research methods that, in order to intervene effectively used were sensitive to the processes in a moment of sex-related risk to through which various societies prevent HIV infection, it is essential to actively construct ideas about what is understand how those participating sexually appropriate (or not) for young view what is happening — in particu- people. It was also important that all lar their intentions, their interests and the relevant ‘players’ or participants in the possible outcomes of the event. determining the appropriateness or For many young people, sexual moti- otherwise of young people’s sex lives vations are complex and may even be were involved in the studies. unclear or largely unformulated. The pursuit of sex may be bound up in Attention was also directed toward confusing expectations and fears. other factors that may affect sexual These expectations and fears are like- understandings and conduct (such as ly to be couched in identifiable ideas, historical and economic conditions), terms and frameworks and formed in as well as to the influence of particu- local or immediate cultural contexts. lar institutions such as legal and reli-

16 CHAPTER 2 gious systems. Wherever possible, three areas (although there is consider- efforts were to be made to link with able overlap between these): questions other studies that had already collect- relating to the meanings of sexual activ- ed data on young people’s sexual ity and sexuality (ies), including the dis- conduct using other methods (e.g., courses of sex; questions concerning existing KAPB and Partner Relations the contexts of sexual activity; and surveys). questions about specific sexual behaviours and activities. The overall objectives of the research programme, therefore, were to: In terms of the meanings of sexual activity and sexuality(ies), the central • examine the personal and social questions were: contexts in which sexual activity among young people takes place • how do young people learn, and and the ways in which such activi- what do they learn about sex and ties are explained and justified; sexuality; • analyse the range and shaping of • what are the sources from which meanings of sexual activity they acquire knowledge, values among samples of young people and attitudes, and what is the rel- in particular contexts and cul- ative importance of each of these; tures; • how are feelings and thoughts • identify the different sources and about sexual activity shaped and contexts from which young peo- formed, and how do the mean- ple acquire their knowledge and ings of sex and sexuality alter in understanding of sexual issues; different age groups over time; • explore the dynamic processes and through which young people • what do young people expect learn about sexual issues and from or seek in their sexual appropriate sexual meanings, and relations? how alternative meanings develop and change; and • work toward developing culturally Potential sources of information includ- relevant explanatory models of ed the mass media, parents and other sexual behaviour among young kin, and schools and peers, as well as people, and assess the implica- less direct influences such as religion, tions of these models for the legal systems or other cultural traditions. design of appropriate preventive Examples of meanings regarding sexual interventions. conduct might include general ideas about sex and procreation. Other sexu- The key research questions were out- al meanings might be revealed by lined in the General Research Protocol answers to the following questions; is and formulated to ensure that findings sex always associated with physical from the different countries undertak- pleasure? Are sexual desires publicly ing the study had some common points acknowledged or hidden? How does of reference. Research questions were each society understand and recognize of two types: principal research ques- gender identity for males and females? tions, and additional research ques- Are transsexuals and transvestites rec- tions. The former were grouped into ognized cultural categories? Are there

17 CHAPTER 2 recognized rituals for sexual initiation, or • socially approved or disapproved for achieving sexual maturity? What are sexual activities, and their relation each society’s ideas about sex and love, to gender, age, sexual prefer- power, rebellion, curiosity, and so on? ence, and other factors.

In addition to obtaining answers to Included in these considerations these questions, data were also to be were to be the actual or potential collected on the way sex is talked impact of aspects of specific cultures, about, the hierarchy of meanings, the such as legal constraints (e.g., the sources of alternative meanings, the laws that govern sexual conduct and resolution of conflict between mean- their enforcement or contravention) ings (at individual and relational lev- and economic factors (e.g., the els), and the development of mean- effects of poverty on sexual conduct). ings over time. In seeking to identify The term contexts encompassed and explore the contexts of sex, general norms and regulations within researchers’ attention was directed societies, those in specific locations, toward: and those affecting interpersonal behaviour in localized settings. • the social, emotional and physical contexts in which sexual activity Information was also to be gathered occurs (e.g., the extent to which on the sexual practices of individual early sexual contacts take place respondents, with particular atten- with relatively long-term partners, tion being given to the details of or with sex workers); events and contexts, and of personal • respondents’ expectations of par- explanations, interpretations and jus- ticular contexts (e.g., the implica- tifications. The relational aspects of tion that drinking and socializing sexuality were seen as particularly at particular places implies that important. These included such one might be looking for a sexual things as the emotional (or affective) partner, or being bought presents aspects of sexual activity, the kinds of by an older person implies that partnerships entered into, and how sexual favours should be given in these are understood, valued and return); talked about. Since the direct obser- • perceived ‘rules’, such as report- vation of sexual activity was not posed obligations and expectations sible, personal accounts and expla- of self and others (e.g., that all will nations of sexual activity were to be marry); sought. Social settings in which • the integration of alternative young people actively pursue the meanings into different contexts development of sexual interests and (e.g., there may be conflicting relations e.g., at discothèques or in definitions and pressures from leisure activities), were to be target- parents, the media, peers and ed for systematic observation by others); researchers. • differing perceptions of vulnerability (e.g., how perceptions of As stated above, research teams par- risk of HIV infection and/or preg- ticipating in the programme were nancy alter with age, gender, encouraged to add to the study ethnicity); and design additional research themes

18 CHAPTER 2 and questions that arose for their par- • implications for health promotion ticular country, for the areas chosen as interventions to minimize their research sites, and/or which risk-related sexual behaviour. related to the samples chosen. Possible additional themes included Throughout 1993, site visits were the migration of young men from rural undertaken to over 12 countries to to urban areas, changes in traditional identify potential principal investiga- sexual values and meanings as a result tors and to assess institutional capaci- of economic development, and the ty to undertake studies of this com- local impact of new sex education plexity. Potential principal investiga- programmes on contemporary sexual tors were invited to attend a briefing identities. meeting to familiarize themselves with the General Research Protocol, and to It was hoped these studies would lead receive guidance on development of to: the country-specific local research proposals. Draft local research pro- • a better appreciation of dominant posals were subsequently submitted sexual understandings, sexual for assessment, returned with sugges- meanings, sexual identities and tions for improvements, and were sexual cultures in seven contrast- then reviewed formally by the GPA ing sites; Steering Committee on Social and • a sounder understanding of the Behavioural Research in October implications of such understand- 1993 and March 1994. ings, meanings, identities and cultures for self-understanding and The local research proposals from individual sexual behaviour; seven countries were chosen to par- • a comparison between the ways ticipate in the programme: young women and men come to Cambodia, Cameroon, Chile, Costa understand themselves sexually, Rica, Papua New Guinea, the and the consequences of this for Philippines and Zimbabwe. These patterns of sexual behaviour; seven countries received funds from • analyses of sexual behaviour per- WHO/GPA/UNAIDS to undertake the ceived as ‘risky’ (in relation to HIV studies, most of which took an aver- and STDs) within a specific cul- age of two years to complete. The ture; and final study (that in Cambodia) finished in 1996, and this comparative analysis of findings was finalized some nine months afterwards.

19 CHAPTER 3

STUDY DESIGN AND institutional imperatives in one population (e.g., school rules, religious pro- METHODOLOGY hibitions and so on) that are likely to produce the same effects, actions and understandings in other similar popu- The overall research methodology lations. In other words, by uncovering employed in the studies was qualita- how social forces and cultural compo- tive, with an emphasis on an in-depth nents affect a certain population, it is comprehension of people’s own expe- often possible to generalize these riences, rather than the use of statistical findings to other populations affected analysis for the purposes of generaliza- by the same forces and components. tion. The aim was to identify the social As we will see, in these studies’ find- processes that need to be understood ings there are similarities that cross when thinking about HIV-related public national boundaries and even span health policies and programmes. The continental divides, and these can multi-site study aimed to theorize the judiciously be generalized to other relation between individuals, their countries. There are also country-spe- immediate milieux, and broader social cific practices and understandings, and cultural forces. We can understand however, that relate directly and only how a society works best when we to a particular country itself. understand how its citizens interact on a daily basis with each other, with Research plan groups of people, and with others in institutions and organizations. In this The studies were undertaken in three kind of enquiry, therefore, individual stages: a rapid assessment process; a experience is investigated, not for the period of main fieldwork; and an purpose of accounting for majority or analysis and write-up period. minority patterns of behaviours, but in order to explain experience and Rapid assessment behaviour. The aim is not to produce information on population characteris- The purposes of the rapid assessment tics, but to provide the sometimes were to enable local research teams to detailed and complex ideas needed fine-tune their original research for policy and programme develop- design, pinpoint more exactly field- ment and implementation. work sites, and assess and clarify the sampling frame. This period allowed Generalization is possible in qualita- for corrections to the original propos- tive research, but not at the level of als, designed as they had to be in majority or minority experience and abstract, and proved an important and their correlates. Instead, it is possible useful step. For example, one study to identify and characterize the social (Chile) added a third group of young forces or cultural components under- people to its planned two-part sample lying individual or group behaviour. after their initial rapid assessment. This Generalization can also occur when enabled the inclusion of young people assessing the contribution of particu- marginalized from mainstream society lar cultural and historical forces to pat- and whose sexual behaviour was terns of behaviour, and by revealing found to be somewhat different.

20 CHAPTER 3

Rapid assessments also allowed the tunity to sharpen detail, clarify argu- research teams to test out both the ment, and focus recommendations. principal research methods common to all sites (see below), their chosen additional research methods, and to Principal research methods undertake some initial key-informant interviewing to sharpen the focus on Three principal research methods principal and additional research were employed in all studies: individ- questions. Reports from the rapid ual interviews, focus groups, and par- assessments were sent to WHO/GPA ticipant observation. The standard- for technical assessment and individ- ized use of these three principal ual research teams were assisted, research methods was intended to where necessary, in adjusting the generate data that could be used for final design for the main fieldwork comparative purposes across the period. seven countries.

Main fieldwork As the term individual interviews suggests, this research method involves During the main fieldwork period, interviewing people in a one-to-one technical assistance was made avail- situation. Such interviews were con- able to the research teams. This ducted primarily with young people, included site visits by staff, consul- but also with selected ‘key players’ tants and temporary advisers; regular identified during the rapid assess- communication and information; and ment. In relation to young people, the informal support from other interest- intention was to explore in a confi- ed parties. Various conferences and dential way issues related to sexual meetings, and later e-mail corre- conduct and understanding. In gener- spondence to some sites, facilitated al, such interviews lasted between more regular communication and one and two hours (although longer in provided additional opportunities for some places), and the researchers sharing ideas, discussing fieldwork explored the principal and additional problems and keeping the seven research questions in a flexible man- studies in touch with each other. ner. Wherever possible, interviews were tape-recorded with the informed Analysis and write up consent of the interviewee.

A meeting in Geneva in 1994 of all The term participant observation cov- principal investigators provided an ers a range of research techniques important first opportunity to share from ‘observer as participant’ to ‘par- preliminary findings, identify major ticipant as observer’, the main pur- themes, explore similarities and dif- pose in this case being to allow close ferences in the evolving analyses, observation of the everyday behav- and established a common frame- iour and contexts that relate to sexual work and process for producing indi- activity. The extent to which an vidual country reports. Finally, an observer can become fully immersed assessment of individual countries’ in the activities under observation draft final reports by the authors of depends on many things, including this report provided a further oppor- the nature of the activity itself, and the

21 CHAPTER 3 relationship between the observer Additional research and those being observed. Direct methods observation of sexual activity was not possible in these studies (nor is it in All research teams were encouraged most HIV/AIDS research irrespective to add additional research methods to of methodology), therefore self-report their design if this was thought to be is usually the source of most data. useful. This was done in the recogni- However, much can be gained from tion that country-specific research careful study of the social contexts in questions might require different which potential sexual contacts are methods of investigation, or that cer- made, and which provide, shape and tain sections of samples chosen might give meaning to sexuality in specific prefer other methods. Most research cultures. Suggested participant obser- teams took advantage of this oppor- vation contexts included bars, clubs, tunity to expand the scope of their discothèques, parks, streets and other research. Additional research meth- places where young people congre- ods chosen included the use of gated or met regularly. Such observa- ‘vignettes’ in Papua New Guinea, an tions were intended to provide useful ethnographic case study of a dis- information on the manner in which cothèque in Chile, linguistic analysis relationships are formed, the patterns of terms and meanings of words for of interaction between people, the sex and sexual relations in the rules that govern behaviour in such Philippines, discourse analysis in contexts, and the range of ‘key play- Costa Rica, peer-led focus groups and ers’ in each setting. peer participation in preliminary analysis in Cambodia, and attendance The basic aim of focus groups is to at certain ritual events in Cameroon. encourage a group of between five and ten people to focus on specified Fieldwork sites and topics through a set of facilitated sampling activities and to gather data through an interactive process of collective In their local research proposals, reflection.4 The composition of the research teams were asked to nomi- groups selected in these studies nate at least two sites in which to depended on the sites in which the undertake their research. The purpose fieldwork was undertaken, but were of having at least two sites in each expected to be constituted largely of country was to encourage possibilities young people, although the method for comparison or contrast, thereby could in principle be used with other preventing an artificially narrow view interested parties such as parents and of young people from unwittingly youth leaders. emerging. Various combinations of site-selection criteria were used. Some teams chose three sites (Papua New Guinea, Zimbabwe). Some drew a rural-urban distinction (Cambodia,

4 Focus groups are a valuable method for obtaining a wide range of understandings, views, opinions and attitudes on particular social and cultural activities. Focus groups are not the same as group interviews, where participants in turn are asked similar questions and individual experience is reported in front of others for comment or reflection.

22 CHAPTER 3

Papua New Guinea, Cameroon), and In cases where existing data on young two selected contrasting sectors of the people were available, such as infor- same cities using social class and mation from surveys conducted for socioeconomic status as the distin- WHO or other agencies, research guishing factor (Chile, Philippines). Still teams were asked not to replicate others chose a capital city/provincial such data collection but to use exist- town difference (Costa Rica, ing data as a starting point to ensure Philippines, Papua New Guinea, that the relatively small numbers of Zimbabwe). The attempt here at con- young people recruited for detailed trast and comparison was not one that investigation in the study were rea- argued that complete coverage of all sonably typical of the larger popula- young people could thereby be tion of young people in at least some achieved. Rather, the aim was for high respects. In some countries, existing contrast, which would maximize the data included social descriptors such capacity for clear comparison so that as age, education level or ethnic back- social processes could be uncovered. ground, and sexuality indices such as The details of each country’s site selec- age at first intercourse, numbers of tion are contained in the final country partners and patterns of contracep- reports, and these are summarized in tive use. Similarly, the selection of the Appendix to this comparative members of focus group discussions analysis. was to be guided by data from these other sources where applicable. The General Research Protocol provided research teams with guidelines on It should be noted that representative- the sampling frame to be employed. ness was not required for this study. The studies’ samples were constituted, Instead, the aim was to develop a het- in the main, by young people. Within erogeneous grouping of respondents these parameters, it was important to in terms of their relation to the research specify the involvement of particular questions. In this sense, the sample age ranges in each country, since it was recruited at any site was constructed anticipated that these would vary con- theoretically in relation to the principal siderably across cultures. So that and additional research questions. national comparisons would be possi- Theoretical sampling can be defined as ble, sampling covered fairly similar ages a systematic process of selection in all of the studies. The age range for developed from an understanding of young people in each study was: the theoretical field, a recognition of Cambodia 15-26; Cameroon 15-30; the social circumstances surrounding Chile 18-22; Costa Rica 12-19; Papua the issue to be researched, and an ini- New Guinea 12-24; Philippines 16-24; tial estimation of which populations and Zimbabwe 13-24. might best provide useful data on these particular research questions. As In order to analyse gender issues and a result, who is interviewed, when, and their relation to sexual matters, it was how will be decided as the research important that young women and progresses, according to one’s assess- young men be equally selected, except ment of one’s current state of knowl- in circumstances where an over-sam- edge and one’s judgement as to how it pling of one sex or the other was nec- might be developed further essary for a particular reason. (Hammersley & Atkinson, 1989).

23 CHAPTER 3

Comparability across studies was to employing the principles of theoretical be achieved not only through the sampling. In Cambodia, 341 respon- basic sampling frame involving com- dents in two sites were involved (146 mon social descriptors and sexuality young men, 135 young women, 60 indices, heterogeneity, and sample others—adults, key informants or offi- weighting (if used), but also through cials). In Cameroon, a total of 550 par- the relationship between the groups ticipants were interviewed, 301 men of young people selected and the and 249 women in three sites. Of issues being investigated. For exam- these, 108 were key informants. An ple, if migration between rural areas additional 22 focus groups were con- and cities were regarded as an impor- ducted. In Costa Rica, 56 people in tant issue, then young people who are two sites were interviewed (25 young involved in this migration would obvi- men and 31 young women). This study ously be a starting point for sample did, however, involve interviews of recruitment. much greater length than the others did. An additional 48 people partici- Sample sizes pated in focus group discussions, with approximately equal numbers in each The methodology used in this study site. In Chile, the sample was divided was never intended to generate data into three different socioeconomic cat- on the relative frequencies of particu- egories investigated through the fol- lar sexual or social behaviours. Rather, lowing techniques: 21 group discus- it was designed to explore the range sions involving approximately 250 of sexual meanings, identities and cul- young people; 36 individual interviews tures. Each research team was expect- on general discourses of sexuality; 26 ed to obtain a heterogeneous sample individual interviews on selected sexu- of young people in the particular ality topics; 48 individual interviews on country, community or site, developed information systems; and innumerable from the theoretical sampling model. observations in the ethnographic Exact sample sizes, therefore, were not study of the discothèque. The sam- specified in the General Research pling in almost all categories was Protocol. Instead, general guidelines divided equally between the sexes. were developed to assist each The Philippines study involved 78 research team in developing their young men and women in the initial sampling frame. It was planned that rapid assessment stage, 120 in in- each study would complete between depth interviews and 70 in focus 100 and 200 individual interviews, and groups in the main fieldwork period, between 10 and 30 focus group dis- equally divided between the sexes, cussions. The exact allocation of these and 33 others in both phases — a total numbers between sample categories of 301 respondents in two sites, divid- (e.g., young people versus other ‘played into two categories in each site by ers’) was to be decided after the initial socioeconomic class. In Papua New rapid assessment process. Guinea, the research team held 64 focus group discussions, 174, 331 and The various studies produced differ- 493 ‘media, ‘personal’ and ‘secret’ ently structured samples as a result of interviews respectively,5 and 77 key

5 See the Papua New Guinea research report summary in the Appendix for an explanation of these interview types.

24 CHAPTER 3 informant interviews. Fifty-two respon- Time scale dents were observed. This large number of 1,191 study participants (560 The first studies were funded at the males and 631 females) across three end of 1993, and a second set was sites must be understood within the funded in March 1994. All studies framework of the much shorter inter- were under way by mid-1994. All final view methods used in the study. reports had been received by the end Finally, the Zimbabwe study involved of 1996, although it was necessary to 82 young men and 91 young women seek clarification on aspects of in individual interviews. A further 21 methodology and study findings focus groups and 5 key informant throughout the early part of 1997. interviews were conducted with mem- This comparative analysis was com- bers of both sexes in three sites. pleted some six months later.

25 CHAPTER 4

COMPARATIVE historically formed and trusted patterns ANALYSIS of conduct, and ways of relating. Key concepts

There are a number of levels at which The concepts of sexual culture, sexual the findings from this study have rele- identity and sexual meaning allow us vance beyond the individual countries to cluster superficially different and involved. The first derives from the culturally distinct experiences of sexu- closeness gained to young people. al activity among young people into Throughout the analyses, there is a understandings that cut across the clear sense of contact with, respect seven countries involved. for, and knowledge gained from an extended engagement with young Sexual culture people. There were no one-off, brief and carefully distanced (‘objective’) Sexual culture is a concept that recog- moments of contact. Instead, sex is nizes that there are systems of sexual presented as an embodied, meaning- behaviour among any group of peo- ful, collective act for young people. ple. The term suggests that sexual Sensation is felt, bodies matter, minds activity is not simply a manifestation cogitate, and real young people of biological drives or ‘natural’ struggle with the meaning of their processes. As used in this study, the desire. The cultures studied are clear- concept of sexual culture allows us to ly in conflict over young people’s sex- understand the origin and source of ual conduct and no amount of denial information about sex and specific or refusal will counteract that. As a expectations of it. It also helps us rec- number of country reports note, it is ognize the specific shaping of erotic not young people who are the prob- procedures and sensations, and why lem, but the way they are (mis)under- one form of sexual expression may be stood by many who seek to work with taken for granted in one culture but them. eschewed in another.

This brings us to a second purpose In this sense, sexual cultures are to be behind this analysis — the desire to regarded as social products, made develop new thinking and better and re-made in daily life, often locally, HIV/AIDS health promotion for young and therefore adaptable to changing people in developing countries. pressures and circumstances. For Marked levels of sexual risk-taking are example, notable findings across all noted in all reports. This should not sur- countries in this study were the rapid- prise anyone. What should surprise is ly changing expectations and prac- the complexity of personal and cultural tices of sex among young people as a investment in forms of conduct that — result of modernization and rapid at least in the minds of health profes- urban growth. A second example sionals, epidemiologists and health might be different understandings of educators — are so easily reduced to the importance of virginity to young ‘risk-taking’. What appears on the sur- women and men expressed by differ- face as intractable behaviour is in fact ent groups of young people even

26 CHAPTER 4 within a single country. And if there is quences of the differences between a single key overall finding from these sexes included the common experi- studies, it is that there are many sexu- ence of sex exchanged for favours al cultures within a particular setting, among young women, and the belief and no individual country has merely in the urgent ‘need’ for sexual experi- one. ence among young men. In this analysis, sexual identities will be used sole- Sexual identity ly to deal with matters of sexual activity and interests; the term gender Sexual identity is a concept that pro- identity will be used to refer to other vides a psychological (or psychic) issues of masculinity and femininity. place for situating the self in sexual Sexual identity and gender identity activity. This recognition of the self in are not just about personal relation- sex (how one behaves, what one ships and a sense of the self, but also wants, what one expects of oneself about finding one’s place in families and others) is crucial to any individ- and communities, and in economic ual’s accurate assessment of risk in and cultural life. For example, in relation to potential HIV or STD infec- Papua New Guinea personal identity tion. is linked traditionally with land, its use and with certain areas or homelands. Sexual identity is a recent and very This link has been weakened by Western term, receiving significant migration, mobility and the undermin- attention since the rise of second- ing of traditional modes of social con- wave feminism, often in relation to, trol related to membership of, and and sometimes confused with, gender participation in, that traditional social identity. Gay and lesbian theory has and economic order — in a sense, one also addressed the issue of sexual is what one does within the broader identity, particularly in the personal community. Older people in Papua consequences of same-sex sexual New Guinea lament this weakening of partner choice. The distinction traditional life, particularly in sexual between sexual identity and gender matters, and see this as a source of identity can be difficult to maintain, as permissiveness and sexual experi- gender plays a crucial part in structur- mentation. In other words, they ing the sexual identities of young peo- observe the shift from identity based ple in all countries in the study. on kin or territory to a more psycho- Indeed, larger structural imbalances logical sense of self, acted out also in between the sexes have profound sexual conduct.6 consequences for the expectations and experiences of young men and This shift signals a major, unnoticed women in sexual matters and emo- impact of modernization on develop- tional relationships. Among the finding countries, one that is occurring ings of this study, worrying conse- rapidly and unsteadily. One conse-

6 This shift from collective to personal individualized identity, as modernization proceeds, is largely unnoticed by many identity-obsessed Westerners, but is worthy of greater attention. It registers a breaking-up of local cultures, which education programmes, particularly those charged with personal or health education, may need to tackle. These are longer-term issues that need to be taken into account in development and modernization programmes, health and education initiatives, and in HIV/AIDS and sex education.

27 CHAPTER 4 quence of this rapid and unsteady people. The meanings of sexual con- change was noted again in the Papua duct and interests can vary greatly: New Guinea report. There, in the meaning can be derived from collec- somewhat lost and marginalized situa- tive and traditional rituals, or from a tion for some young people in Port more modernized personal desire. Moresby, landlessness and lack of Other sexual meanings can be gener- employment prospects (and therefore ated in close interpersonal relation- no cash substitute for wealth) creates ships, or are derived from more gen- raskol (hoodlum) activity. This is con- eral societal expectations about sex. tributing to the widespread rape of Meanings change, and the influence young women, which as the Papua of an increasingly internationalized New Guinea report notes is ‘seen as framework for understanding sexual theft’. These shifts in identity from its activity has been greatly extended historical and traditional roots were with the advent of the HIV pandemic. noted also in Cameroon and This has already been noted in rela- Zimbabwe in strikingly similar ways, tion to sexual identity. Beyond that and in each case were seen as a major issue, Western frameworks, notably contribution to changing sexual con- the binary opposition between ‘het- duct among young people. erosexual’ and ‘homosexual’ people, imposes an interpretative framework Although some older respondents in at times at odds with indigenous and these studies saw this in a cause and local understandings, yet one that is effect way, this is not the important increasingly pervasive. This binary point. Rather, it is vital to recognize opposition is barely 100 years old in that sexual conduct is not uniformly its dominant position as the interpre- regarded as ‘naturally’ constitutive of tative sexuality framework. It is also the self in all cultures, as it is in the increasingly subject to critique in cur- West, but is becoming so as social rent sexuality theory in the West. and economic modernization pro- Argument about its limitation as a ceeds. We are all now familiar with framework for understanding sexual projective environmental impact stud- conduct in HIV/AIDS work has been ies, increasingly regarded as neces- already clearly established in earlier sary before any economic develop- debates on sexuality and HIV/AIDS in ment activity begins. It may be that Latin America, and are again pertinent future development and moderniza- in these studies. tion initiatives should assess ‘sexual impact’ in a similar way by looking, for However, mass communication is pro- example, at the way in which a new moting a ‘heterosexual’/‘homosexual’ mining town might produce increased conceptualization of sexuality as prostitution. This would acknowledge almost the single global understand- the broader social and cultural conse- ing of many young people’s sexual quence of development initiatives. experiences and expectations. The effects of this are still partial as noted, Sexual meaning for example, in Cambodia where young people reported themselves as Sexual meaning as a concept offers a indisputably ‘heterosexual’ but did way of getting a firmer grip on the sig- not regard those men who engaged nificance of sexual activity to young in same-sex activity as ‘homosexual’.

28 CHAPTER 4

Similarly, same-sex activity in Papua niques of analysis, the central signifi- New Guinea is rarely regarded as cance of sex to young people can no homosexual (and almost never as longer be in dispute. Sex is not an area ‘gay’). Among working class and mar- of conduct that does not belong to ginal youth in Costa Rica and to a less- young people, even if some adult er extent in the Philippines and members of each society wish it were Cambodia, same-sex sexual conduct so. Clearly, the young people in these was largely regarded as gender-relat- studies have their own sexuality, what- ed. The playos of Costa Rica, the kteuy ever the rules, whatever the traditions, of Cambodia and the bakla of the and whatever the dangers. Any notion Philippines (all effeminate men who of young people as somehow still pre- engage in same-sex activity) were sexual, or simply play-acting at sex, or seen mostly as products of gender practising a little in preparation for inversion — in essence more transsex- later, must be rejected for a framework ual than homosexual. This example that now recognizes young people as offers just one obvious moment of fully sexual beings, logically and legit- seeing how sexual meanings are not imately so. Whether adults like it or universal or uniform, even when they not, the young people in these seven describe superficially similar kinds of countries own their sexual conduct behaviour. already.

Contrasts between countries in rela- This recognition profoundly affects our tion to sexual meanings were most assessment of the importance of sex marked with respect to gender. There and sexual relationships for young is no doubt that relationships are more people, and must force a reckoning important to most young women than with their meanings about sex. In they are to most young men, and that other words, although there is indeed the physical sensations of sex acts conflict and contradiction between tra- dominate most young men’s minds. ditional sexual mores and young peo- There were, however, exceptions to ple’s current sexual conduct in all this general point about differences countries studied, any new set of between young men and young meanings attached by young people women, and these must be taken into to current sexual activity must be taken account in any attempt to insert a bio- seriously and dealt with respectfully if logical or genetic explanation for such we are to assist young people in cop- differences here. Any simplistic idea ing adequately with HIV/AIDS and that maleness or femaleness is the fun- other sexual health issues. It is clear damental source of these differences that young people can and do simply fails to take into account the powerful ride out the contradictions between effects of social class, regional contrast their ideas and those of their parents, and modernization. In other words, and conflicts between traditional and there is much more than biology more modern values. Sometimes this involved here. might be done through cognitive strategies of ‘compartmentalization’, The importance of sexual meaning as as the Costa Rica study suggests. a theme cannot be underestimated. Sometimes new sexual possibilities are Irrespective of somewhat different opened up as a result of migration and methods of investigation or tech- mobility as suggested in Cameroon

29 CHAPTER 4 and Zimbabwe, or by the creation of a strategies. A few examples will illus- new language for sex and sexual rela- trate the importance and subtlety of tions as noted in the Philippines and sexual meanings and their shifts, Cambodia. There are also indications noted in all studies as vital to accurate of countervailing discourses of gender representations of young people’s sex- transformation, e.g., transsexualism, ual experience and their understand- or modernization and individual plea- ing of sexual activity and its deeper sure, as evidenced in Chile and Papua relational contexts. New Guinea, which provide young people with new meanings for sex and Example one. It is important to new ideas about what sex is for. recognize pror chuk (to pump the penis), the new word that young These all become strategies of legiti- Cambodian people have given mation for new sexual ideas among the Khmer language to describe young people, and can be effectively masturbation for men. There were used as resources in health promotion many other terms noted in that and prevention education. All the study for oral and anal sex, mas- Contextual Factors studies argue for turbation, and so on that young the need for HIV/STD prevention to be people use. couched in the terms of young people, in their own words and meanings. Example two. In this same country, These strategies of legitimation pro- it is important to understand the vide the resources for an effective edu- significant difference between kou cational approach, as noted in the songsar (a relationship without Philippines report: sexual activity) and kou sne’har (a relationship implying sexual activi- The present research project shows ty). This distinction registers a shift active, discerning young adults, not in young people’s relations with necessarily well informed, but in boyfriends or girlfriends to some- spite of these problems, managing to thing deeper, something poten- negotiate with the world around tially involving sexual love. This is them. The inquisitiveness of young similar to the idea of a ‘person with adults needs to be encouraged, and a good place to start is handling this rights’ in Costa Rica, a term which curiosity as it relates to sex and sex- allows young women to permit uality, allowing young adults to young men greater exploration of question norms, and in the process, their bodies. recognizing their self worth. Example three. In the Philippines, The purpose of the three concepts the idea of madala implies being swept along erotically and rela- These three themes, sexual culture, tionally, not necessarily but often sexual identity and sexual meaning, in towards, sexual activity. Similarly, their distinct ways, are useful tools for in Costa Rica the notion of tem- examining young people’s sexual con- plado, in both women and men, duct in other countries beyond these registers a kind of irrefusable seven, and for doing so in a way body ‘heat’, a burgeoning sexual directly related to harvesting material desire that permits, and even for developing health promotion demands, satisfaction.

30 CHAPTER 4

These are not just semantic differ- sexual experience makes no sense ences or merely local slang. Nor is a when there is a commonplace of five simple argument being mounted here ‘sexual baptisms’ for a young man: merely to encourage the use of local sex with an agi (a cross-dressing language or colloquialisms in health homosexual man), a female sex work- education. That all studies offered er, an older woman, a girlfriend, and detailed accounts of the subtle shifts finally his wife. Similarly, in Costa Rica in terms used by young people for sex among more affluent and educated acts and sexual relations, indicates the young people, virginity is widely importance of such detail to young regarded more as psychological state people themselves. In all the coun- and therefore both young women and tries studied, young people employ men ‘lose’ their virginity. This con- specific and finely graded terms and trasts with a more direct physical shades of meanings both to under- understanding of virginity (and its loss stand what is happening to them- for young women alone) among work- selves and their bodies, and to locate ing-class Costa Ricans, thereby regis- themselves within their culture and tering also that meanings shift not just age group. These terms and mean- from country to country but within ings help young people to measure countries themselves. themselves against adults and their immediate milieux. They verify sexual Similarities progress and assuage fear, doubt, guilt and any sense of being lost to an A set of recurrent themes and issues incomprehensible sexual desire. In the emerge from the seven countries. adult world’s deafening silence that These are: concepts of youth; the surrounds young people’s sexual con- challenge to traditional cultures; gen- duct in all countries studied, and in der; modernization; urban, rural and the face of the appalling lack of infor- provincial aspects; and other effects mation with which young people in of larger social forces. every country in the study seek to understand sex, these terms and the Concepts of youth culturally specific concepts they embody enable young people to find Not surprisingly, all countries reported their sexual ‘feet’. The terms encode a special category for young people, and make possible sexual interac- although the definition and bound- tions. Without them, confusion reigns aries changed from country to coun- for young people and their bodies try. There is an important lesson here: become their only teachers. undoubtedly the category ‘young people’ offers a way of defining this Health promotion that either relies on particular population, which can be clinical lists of sex practices or con- generalized so as to make global cepts of sexual development that uni- comparisons among all young peo- versalize the specific to the point of ple. There was a remarkably uniform misrecognition (e.g., theories of ado- concept of young people in relation lescent development) will fail to reach to age cohort and maturation, even young people. Again, a simple exam- when the onset of sexual activity and ple comes from one site in the the time to marry (the parameters in Philippines. Here, the notion of first many countries) differed between age

31 CHAPTER 4 cohorts quite markedly. So, a focus on at times, carries with it an explicit young people is understood every- intention to delay or prevent young where as valid. people from engaging in sexual activity that, in some cases only a genera- Yet herein lies a danger: a kind of tion or two ago, was seen as perfectly global, essentialist category called ‘natural’ or understandable. ‘young people’ tends to mask quite diverse cultural definitions. What is Yet it is also clear that young people in also clear is that definitions of young all seven countries (and likely every- people are changing in relation to where else) are contesting such creep- sexual activity. In Zimbabwe and ing ‘infantilization’ with their bodies Cameroon, and also in Papua New and in their sexual pursuits. Indeed, Guinea and Cambodia in slightly dif- the Costa Rica report notes that ferent ways, young women were once delayed marriage and extended edu- regarded as ready for sexual activity at cational training, mainly at universi- or not long after menarche. In the ties, among middle-class young peo- past, in some of these countries these ple has actually enabled premarital young women were then ‘traded’ as sex to be more permissible among wives through various methods and those young people, as they increas- customs of exchange, dowry or bride ingly regard the delay as unaccept- price. In almost all countries in the able. All seven country reports argue study (though with different kinds of that exhortations to abstinence or explanation) young men were regard- delayed sexual activity are resisted by ed as ‘needing’ sexual experience many young people and seen for what once they were pubescent, and sex they are — moral agendas dressed up with sex workers, male peers or older as health promotion. women was sought, encouraged tacit- ly or facilitated directly by older Second, the age ranges chosen by the males, families or peers. Although various studies reported here are suffi- these ideas are often couched in ciently different to force a recognition questionable ‘naturalized’ terms, i.e., that ‘young people’ is a category that related to some biologistic character- must be culturally defined first and istic or ‘hydraulic’ proclivity,7 the foremost, even if it is under increasing important point is that these practices international pressure to be all-inclu- all register a widely held, if unstated, sive and consistent for the sake of pro- view of young people as sexual grammatic and bureaucratic neatness. beings from puberty. The very notion of young people as being that cohort, for example, from Undoubtedly this view is increasingly 12 to 26 should be regarded as a dubi- undermined by processes of delaying ous fiction, and the recent broader young people’s accession of full adult acceptance of the definition of the status and rights, and the extension of ‘child’ as having an upper age limit of their dependency through longer 18 is similarly questionable in this schooling or high unemployment, regard. In the Philippines, where 18 is noted earlier. This shift in the defini- regarded as the minimum age for mar- tion of young people up to ages of 26 riage, such a definition may seem

7 Notions of biological or hormonal ‘pressures’ that build up and demand ‘release’.

32 CHAPTER 4 appropriate for a country where the health education. In a sense there is an average age at first sexual activity was increasing tension between changing reported as being between 18 and 19 gender roles, and the potential sexual for young men and between 21 and consequences of them, that young men 22 for young women. But this age are handling less well than women. marker is irrelevant in Papua New Guinea where the period between the In contrast, some reports note that appearance of secondary sex charac- young people are also becoming sex- teristics and marriage is relatively ually active earlier (e.g., in Cambodia), short, in direct contrast to its lengthen- and are engaging in sex with a larger ing in the West. Elsewhere, the onset number of partners (e.g., in Chile), and of sexual activity was reported as a wider repertoire of practices (e.g, sometimes being as early as 9 in anal and oral sex in Costa Rica and Zimbabwe, and in Chile 32% of young Cambodia). All reports note same-sex people reported having had sex activity and use of prostitutes among before age 15. young men. The reports from Zimbabwe, Papua New Guinea and Third, there is certainly some evidence Cameroon note a significant and that extended schooling and actively pursued (sometimes encour- increased retention rates in some aged) culture of sex for favours, goods countries in this study are prolonging or cash among young women. The the period in which young people prospect is slim for this progressive remain in this category as a cohort ‘infantilization’ of young people pro- before being regarded finally as ducing a significant delay in onset of adults. Indeed, there were accounts sexual activity, or a widespread shift to from more affluent young women in monogamy. Zimbabwe who are consciously delaying sexual intercourse (but pursuing Indeed, the only prospect for such a other sex practices such as ‘fingering’ delay would appear to lie with com- and ‘smooching’) so as not to affect prehensive sex education, noted in the their schooling. Similarly, young research literature as definitely not women in Papua New Guinea argued producing earlier onset of sexual activ- that were there more schooling on ity, and in some places achieving offer, they would delay sexual activity if delayed onset. This determination of they were likely to obtain better job young people to have a sexual life of prospects as a result. their own would seem to reinforce the finding in all countries in this study that This effect, however, may only apply to sex education is wanted and much young women. Young men in those needed by young people. These parts of the various samples where the seven countries report sex education most change was happening to young as woefully inadequate everywhere, as women’s prospects, as noted in Costa do their young people. Rica, Chile and to a lesser extent the Philippines, seem to accept certain The challenge to traditional sexual changes in gender role, but this seems culture scarcely to affect their pursuit of penetrative sex when and wherever possible. There is no doubt that all countries are This is an important issue for sexual experiencing a waning of ‘traditional’

33 CHAPTER 4 sexual values and cultures coinciding needs are beyond their control and with this growing pressure to identify demand immediate satisfaction. Held young people as a special category. In as true by men and women in every some countries where sexual activity country in this study, this pervasive among this cohort was less problemat- myth can either be read as confirma- ic in traditional cultures, young people tion of modern feminism’s concep- are using aspects of traditional culture tion of a universal patriarchy, or as to fashion a new one. The increase in a proof of the particular interpretation culturally transformed group male cir- of biomedicine that sexual interests cumcision in Papua New Guinea, using are merely genetically or hormonally far from hygienic means, represents a driven. More telling is the confusion determination to create, anew, sexual expressed by so many young men in meaning where it once was. Similarly, these studies as to the exact mean- the use of sex workers by young ing of the things they feel, and their Cambodian men receives a perverse ineptitude in translating those kind of approval from the still-domi- embodied feelings and experiences nant traditional value placed on the vir- into any explanation other than those ginity of its brides. No doubt, traditions handed down to them by their of dry sex in Cameroon, for example, fathers, brothers and other males. If also create current problems in relation young people are experiencing to condom use and damage to geni- dilemmas in relation to their sexual tals, countering claims that it is the loss instincts and conduct, these country of sexual traditions that has led to reports confirm that it is largely the increases in STDs. Traditional sexual failure of traditional sexual cultures to culture is often involved in the produc- offer meaningful, actionable rules for tion of the very ‘deviancies’ it deplores. conduct and sensible guidance as to sexual relations which is at fault. More worrying is the transformation of Before fault is so easily found with traditional patterns of exchange, in young people themselves, it is which sex was once unproblematically important to note how often we ask a part, into a widespread cash-for-sex that they grow to sexual maturity in or goods-for-sex sexual economy silence, amid misrepresentation, and noted in Zimbabwe and Papua New on their own. If they make mistakes Guinea. Ironically, this transformation (according to adults’ terms), it is most has also enabled young women to probably because they are inexperi- fashion for themselves a kind of sexu- enced, ill informed and unseasoned, al independence in which body plea- not because they are, by definition, sures have found legitimacy. These reckless. two reports also noted that families and kin are often involved in either Gender encouraging or setting the terms of this exchange; so, again, any appeal There is a striking convergence in all to tradition would appear less than seven country reports in the experi- unsullied by its own history. ences of young women and their difficulty of handling young men’s expec- Another aspect of the challenge to tation and understanding of sex acts, traditional culture can be found in sexuality and sexual relations. There the idea that young men’s sexual are a number of related issues here:

34 CHAPTER 4

• differential cultural understand- temples questions the extent of absti- ings of young women’s and men’s nence achieved. sexuality; • different expectations of young In the pervasive formulation of their women and young men in rela- ‘needing’ sexual experience, the ori- tion to sexual conduct; gin or nature of the sex partners • the centrality of virginity to young whom these young men are so ener- women’s sexuality and challenges getic to find and actively penetrate to that centrality; remains tacit. Unless they are animals, • the experience of coerced sexual male peers, sex workers, or older activity and rape among young women and men (all reported in these women (and some young men); studies), these unnamed sex partners and can clearly only be young women. So • the effects of modernization on this double standard also masks an gender roles. underlying contradiction within understandings of femininity itself: an Individual country reports all highlight implicit recognition of the sexual a similar differential in cultural under- potential of young women and of standings of young women’s and their silenced engagement in sexual young men’s sexuality. Although there activity with young men. Another are some differences between coun- result of this double standard is that tries (and some divergence within sexually active young women, or countries themselves), there is a those assumed to be so, are subject marked perception that young men to derogatory classification as sluts, are sexual beings and young women whores, cheap, fallen, wayward, sick ought not to be. This is recognizable or shameful. at the simplest level as the ‘double standard’ noted by feminist writers In these studies, one example of over the last century or more.8 The those contradictions in action to the only country to report a formal tradi- detriment of young women is also a tional pattern of young men’s sexual Cambodian one, where the deflower- abstinence was Cambodia, where ing of young women is regarded as novitiates in Buddhist monasteries the ultimate sport by young men once paralleled young women’s peri- (although it would appear that fewer od of seclusion prior to (quite early) have actually experienced it than marriage. The practice of young men report the desire to do so), replete entering temples for a period was with mythologies about the act itself almost destroyed during Cambodia’s and its consequences. Any (and many) tragic recent history and is only now young women so ‘deflowered’ are dis- being encouraged anew. However, honoured thereby, as well as by the parallel evidence in the Cambodia dishonourable action of many young study of male-to-male eroticism in the men in rejecting them soon afterwards

8 The essential meaning of the double standard is that women ought not have sex outside of culturally approved boundaries, almost uniformly marriage, and that young men must ‘sow their wild oats’ and gain sexual experience whenever and wherever they can. There is usually a begrudging deference to some vague notion of young men remaining sexually inexperienced as well, but this carries little weight at the collective behavioural level, even if it is enforced and valued at the individual level for some or as doxy.

35 CHAPTER 4 for their acquiescence to sex. This vigorously by young people and their double bind for young women oper- elders. Indeed there, as in Zimbabwe, ates hand-in-hand with a high value young women’s claims to sexual plea- placed on virgin brides. sure clearly help invalidate notions of abstinence and opposition to premar- The issue of women’s virginity arose in ital sex. all studies and remains a dominant framework within which young women This challenge to virginity is a com- are forced to understand their bodies, plex one. In Cambodia most young their sexual interests and any sexual men still demand a virgin bride, and experience. As the best marker of the this is the case in other country double bind for young women, virgin- reports. But young men’s avid pursuit ity is still held as important in two of penetrative sex with young women ways: as a guarantee of the value of a almost guarantees a growing short- potential marriage partner, and as age of virgins. That they cannot see proof of the character and worthiness the paradox in the position in which of each young woman in the eyes of they place young women, and find her partner, family and community. themselves, remains one of the great Yet, the value placed on young conundrums in sexual culture. women’s virginity is ambiguous for Moreover, the global influence of many espousing it. Young women youth culture, despite an emphasis on clearly find it an uncertain barrier to romantic love (which might support sexual activity, and in Costa Rica and waiting for marriage), increasingly val- Chile among middle-class young peo- idates a ‘surrender to love’ (as the ple it has lost its once considerable Philippines report noted), as one of status. The Chile report argues that the new understandings and expecta- middle-class young women and men tions of sexual activity available to now see virginity as a largely symbolic young people today. In the face of the and psychological state, a hurdle growing, if uneven, dissolution of vir- between sexual inexperience and sex- ginity as a criterion of value, there is a ual experience to be positively leapt powerful cry from young people in over. In this sense, loss of virginity for these reports for some clearer and both is a rite of passage not a terrible less hypocritical guidance on this sin. But more working-class respon- issue. dents in these studies still place a wavering emphasis on virginity at Nevertheless, there is still a wide- odds with their more middle-class, spread framing of young women’s educated colleagues. This may reflect sexuality as an absence, or at least both some impact from modern femi- abstinence, in all countries studied. nism and a weakening of the utility of There is considerably more attention virginity in guaranteeing anything. paid to preparing young women for The Philippines study noted that their (hopefully delayed) reproductive births out of wedlock to famous polit- roles. Striking in the country reports is ical and artistic figures were defended the level of energy put into explaining as being ‘modern’. In Papua New to young women the ins and outs of Guinea, virginity is less valued in prac- menstruation and reproduction. There tice than in principle, and the sexual were certainly findings in many coun- initiation of young women is pursued try reports that this education was

36 CHAPTER 4 often patchy and partial, delayed or older sons, sex workers or pornogra- deferred, and generally inadequate. phy, and ultimately entrusting their Certainly, little is said about sex or son’s sexual progress to happen- sexual pleasure, and many country stance, peer-based mythology and reports noted a breakdown in tradi- their own adventuring — a potent mix tional methods of educating young of omission and error. women in these issues — from seclusion, the guardianship of various There are various versions of this female kin, to the guarantees incorpo- same story in all country reports, rated in bride prices or dowries. A which indicate that neither sex is repeated theme was the failure of the adequately prepared by its older often presumed central relationship mentors for a confident sexuality. Yet between mothers and daughters as a there is evidence of a growing dis- source of education about young course on young women’s sexuality women’s bodies and sexuality. Some that is productive, not merely repro- country reports (Cameroon, ductive, and speaks positively of Zimbabwe, Papua New Guinea) noted women’s sexual desire in terms traditional taboos operating for this approaching the notion of a sexual issue as well as other, more modern, drive not dissimilar to the more com- pressures increasing the difficulty. mon hydraulic notion of young men’s irrepressible sexual instincts. In Yet, even more striking was the almost Papua New Guinea young women complete absence of any similar, for- talked of their body responses to sex mal, educational processes for young and to pleasure (kisim piling or get- men. Physical responses to male ting feelings), which contrasted with puberty — inexplicable body growth, older women’s sense of sex as duty unsightly secondary sex characteris- or forbearance. Young Cambodian tics, unexpected and untimely erec- women understand the notion of tions, perplexing first ejaculations, the ruam lob (sex for pleasure) and unanticipated sensations and conse- offered at times graphic accounts of quences of sexual activity — were their sexual arousal and desire, experienced by all young men in an although sex for pleasure is less prac- absence of information other than tised among young women than from peers and siblings. The encour- young men. Even young men in agement of boys to be sexually active Cambodia recognize desire and is doubly endowed with a desperate arousal in young women in the quest for knowledge and information phrase rormuol khloun (bend and roll in the face of silence about their sexu- like a rat). Young women from Costa ality. This is the double bind for young Rica and Chile, particularly from the men that then intersects that for middle-class parts of those samples, young women, noted above, and pro- reported a loosening of constraint in duces sexual confusion and distress. regard to premarital sex (Costa Rica) Families emerge as almost peripheral and casual sexual encounters (Chile), to the process whereby young men with notions of female arousal and obtain any information or learn to desire informing this changing sexual understand their bodies and sexuality. order. Importantly, young people in Fathers appear irrelevant to the the Philippines report a coming- process, passing their responsibility to together of bodies within relation-

37 CHAPTER 4 ships that acknowledges pleasure for child-bearing have barely been dent- both before marriage. ed. A more disturbing finding from Papua New Guinea noted that the In young people’s comments there is developing of young women’s desire sometimes a distinct sense that and sexual interests, in conjunction changes in these understandings of with the ascendancy of the cash econ- sexual pleasure are patchy and often omy, has left young men, many of offered in defence of sexual ‘indiscre- whom are unemployed, without the tions’, loss of virginity, and distance cash to pay for sex in a society where travelled from traditional practices. sex in exchange for goods or favours Yet here again, the Cameroon report has a long, traditional, if rapidly trans- noted that sex was not regarded as forming, cultural pattern. One bad or unnatural traditionally, and in response to being cut out of this some places young women were once rapidly emerging sexual economy was prepared for and initiated into sex a widespread, rationalized resort to immediately post-puberty by female rape. kin organizing a young male partner to assess her readiness. That said, the These studies were not quantitative in extent of this current renovation nature and cannot apportion this should not be overstated, having experience of coerced sex to any per- been most noticeable among young, centage of young women (or men) middle-class Costa Rican and Chilean with confidence, but that does not women, responding in part to a long- diminish the seriousness of the issue standing engagement with Western and the necessity for programmes feminism in their countries. and interventions to minimize this very serious threat to sexual health. The Some country reports noted a good widespread experience of coerced deal of confusion among some young sexual activity among young women men as a result of the change in noted in many country reports is young women’s sexual interests and gravely distressing. The most graphic expectations. This confusion appears account of the institutionalization of similar to the ‘crisis’ in masculinity in rape and also pack rape of young the West. The Chile report noted a cri- women was, undoubtedly, reported in sis in machista culture, premised as it the Papua New Guinea study. There, is on sexual adventuring, initiation the suggestion of a connection and conquest, brought about partly in between traditional definitions of response to young women’s new sex- gender and sexuality should not ren- ual assertiveness and a rupture in the der the experiences of Papua New salience of virginity. The Costa Rica Guinean women as exotic. The Papua study found evidence of some accom- New Guinea report argues that some- modation among young men to this thing more profound is occurring — a change in sexual roles at the level of change in sexual culture that is regis- gender roles (e.g., doing more house- tering pressures and distresses from work, defending women’s right to larger social forces, such as economic work), but prevailing discourses of modernization. ‘good/bad’ women, men’s ‘natural’ sexual assertiveness, and women’s The data in other country reports, and ‘natural’ inclination to marriage and the upsurge in reports of sexual

38 CHAPTER 4 assault, pack and date rape, rape in them, is to reinforce the delusion that marriage, incest and sexual abuse, sexuality is an individual, freely chosen and rape of men in the West signify a or merely interpersonal experience. larger, similar arena of crisis. The stud- Fostering an awareness of the structur- ies conducted here do not attempt an al, larger, socioeconomic forces that explanation for that larger crisis; nor impact upon sexual conduct in all soci- can this comparative report. But, com- eties may prove to be a more rounded bined, the disturbing evidence of context for helping young people coercion in sex in these reports war- assess their risk in the face of HIV/AIDS. rants an urgent, considered inclusion In other words, sexual and drug-related of this issue in any conception of sex- safeties may amount to knowing not ual health and HIV/STD prevention. just the rules of a game, but also the arena in which it is to be played. This overview of young people’s sexual culture has highlighted already a set Modernization of tensions in the gender differentiat- ed pattering of young people’s sexual- Social and economic development ity, with quite common patterns and modernization do not simply emerging in all countries, if in different change the day-to-day conditions and ways. These tensions are influenced by circumstances in which people find longstanding or traditional gender dif- themselves. They also affect personal ferentials, themselves under pressure. and interpersonal life at the level of There are both encouraging findings the body, the sexual development of and disturbing consequences at the the self, and understandings of desire level of young people’s sexual culture, and pleasure, and expectations of of the destabilization of traditional both. All studies reported the marked understandings of sex and sexuality, effects of economic modernization and of the impact of gender on sexu- and the pressures caused, in particu- ality. It is important to note that these lar, by rural transformation and findings reveal processes of both increased urbanization. In addition to destruction and construction, not the changes at the personal level list- merely some descent into chaos. ed above, these pressures assist, Change is occurring in sexual cultures among other things, in providing the in all the countries studied with infrastructure (bigger cities, bars and uneven effects and at various paces. clubs, gymnasiums, etc.) for greater There is a strong sense of young peo- sexual opportunity among young ple being at times somewhat rudder- people (and adults as well). less, yet also active in creating new Modernization brings with it new meanings and giving themselves often understandings and ideas to justify quite adequate directions. The glaring how life might be lived differently omission in this active response to from traditional ways. These under- change is young people’s distinct lack standings and ideas facilitate sexual of understanding beyond the person- experimentation and the breaking of al, interpersonal and milieux levels of traditional sexual rules. the impact of large-scale social change on their sexual conduct. One of the The Cameroon study in particular greatest dangers young people face, noted how urbanization had undone and one of the great disservices we do the patterns of exogamy that pre-

39 CHAPTER 4 vailed in traditional village life. A sig- (UNTAC) in 1992 greatly increased the nificant increase was reported in the opportunities for young women to range and number of sexual partners pursue a social life with acknowl- now living in close proximity. Simple edged sexual aspects (e.g., as bar changes, like the arrival of a road or girls), beyond that previously avail- highway, vastly increase the sexual able. In Chile, the ‘party’ ethnography traffic as well. This was noted in Papua in a discothèque pointed to the simi- New Guinea and Zimbabwe, partly larly important place of new social through the arrival of outsiders (truck venues, in this case a sophisticated drivers, military personnel) but also by discothèque that encouraged, facili- encouraging settlement at points on tated and legitimated seduction. the roads as well. In Zimbabwe, the growth of the dormitory town of A discussion of the importance of Mbare, both as a result of previous spaces for young people is presented colonial government’s policies of later, but it is important here to note racial separation, but also as a result the increasing size and availability of of more recent migration to the eco- commercial infrastructure and its nomic opportunities seemingly impact on the changing patterns of offered by Harare, has led to a new sexual engagement for young people. sexual culture in its own right. Here, The significance of this escalation in new patterns of sex work, an intensity commercial infrastructure has yet to in habitation leading to losses of pri- be adequately recognized in HIV/STD vacy, and an expanding number of prevention strategies. commercial entertainment venues provide increased sexual opportunity. In Papua New Guinea, modernization This, at one level, may seem unprob- has produced dubious benefits as the lematic, but it clearly provides a fertile economic gains made in the past 50 ecology for increased sexual activity, years are destabilizing just as the HIV increased numbers of partners and epidemic is increasing rapidly. The net the increased likelihood of STDs. effect of the dramatic shift to a cash economy has in part transformed With this kind of mobility and urban- many traditional patterns of exchang- ization comes a kind of new sexual ing sex for goods (and the more infrastructure in the form of bars, dis- enduring exchange of bride price in cothèques, and other venues. The marriage) to the point of increased Zimbabwe study also noted the devel- participation of women and men in a opment of similar infrastructure kind of commercially organized sex around coach terminals. This infra- work. Sex can be traded for cash, and structural provision not only applies to since cash is hard to come by for large urban centres. The Papua New young people, this has led to an Guinea study, for example, noted the increase in both female and male sex existence of similar venues on a small work. scale at village level and suggested a direct link between such venues and The impact of economic development pack rape. The dramatic rise of bars was not confined to countries such as and similar establishments after the Zimbabwe, Papua New Guinea and arrival of the United Nations Cameroon. Costa Rica reported Transitional Authority in Cambodia uneven effects of economic develop-

40 CHAPTER 4 ment in its comparative analysis of social and economic terms — the two sites, ‘Villa del Mar’ and ‘Villa del need to disassemble the singularity Sol’. ‘Villa del Mar’ is a small, margin- of the term ‘young people’. All al, largely working-class town with a reports noted the significant differ- deteriorating economic base and high ences among the young people in levels of unemployment and under- and between the various sites stud- employment. ‘Villa del Sol’ is an inly- ied. These differences were not ing suburb of San José with a modern merely the product of the research middle-class culture and brighter eco- design, for no country used the same nomic prospects for its well-educated axes of comparison except for gen- young people. There were notable der. In other words, these patterns of differences between young people in differentiation are likely to be readily these two sites in relation to decision- translatable from country to country. making about sexual activity, educa- In utilizing the social class differential tion and prospective professional used in Costa Rica, the Philippines, imperatives, notions of sexual rights, Zimbabwe and Chile in Cameroon, such as monogamy, virginity and sex- Cambodia or Papua New Guinea, ual initiative, and in understandings of similar patterns of difference might sexuality, which were definitely related be found; alternatively, employing to access to educational resources the rural/urban site selection used in and other social and cultural capital. Papua New Guinea and Zimbabwe in In Chile, similar striking differences Chile and Costa Rica would throw were found between middle-class stu- light on different patterns of differen- dents and young workers, and that tiation there as well. One conclusion research team was even motivated to to be drawn from this study is that add a third sector to their sample, ‘the any simplistic characterization of inactives’, to register the significant young people as a singular homoge- differences in sexual activity and neous population is a serious error of understandings found among these judgment if it dominates the design impoverished young people on the of health promotion and HIV/STD fringes of the youth labour market, public health programmes. The rec- with scant education and few foresee- ommendation from all country able life prospects. In Zimbabwe, this reports, that single nation-wide differential impact of economic devel- approaches to prevention education opment was noted among more afflu- are inappropriate, and that local, ent high-school students from a mid- context-specific health promotion is dle-class area of Harare, where there by far the preferable strategy, is high- was some evidence of young women ly significant. delaying vaginal intercourse based on their judgements about their social When the middle-class sections of and economic prospects. the samples in these studies are set aside, a common feature for all other Recognizing the socially stratified young people is their wretched effects of modernization on sexual socioeconomic situation, facing non- activity and culture is not a simple or existent or degraded youth labour obvious task. All country reports markets, poor prospects for educa- stress — as a result of the differentiation and training, and little social tion noted, among other things, in support at community level. The

41 CHAPTER 4

Costa Rica study argues that it should Other effects of larger social come as no surprise that young peo- forces ple’s bodies become their only resource in these circumstances, and All the reports indicate the impact of the pursuit of sex remains one area of larger social forces. Development and activity in which success, pleasure, cultural tensions, immediate histories, self-worth and emotional depth are or processes of modernization are not experienced. The importance of sex simply context as ‘backdrop’, but con- to young people in the face of such text as ‘structure’. If we research sexu- unyielding socioeconomic marginal- ality without reference to these larger ization can never be overstated. forces, but simply at the level of individual or even aggregated group Urban, rural and provincial aspects behaviour, we shall fail to understand the motivation for sex. We shall also Migration to urban areas is not a new fail to see the constraints and phenomenon, but the Zimbabwe, enabling dimensions of these larger Cameroon, Papua New Guinea and forces underlying sexual culture and Cambodia country reports all com- its changing emphases and impor- mented on this issue and noted the tance not only for young people but significance of urban/rural differences. for all sectors of society. HIV/STD pre- Papua New Guinea in particular vention cannot ignore these larger reported circular forms of migration forces. They are the constituents and with relatively frequent movements determinants of health as much as between village, town, city and back, individual behaviour is. There is a rather than the semi-permanent form need to move beyond a narrowly noted in Cameroon, or the seasonal behavioural model of health and indi- movements noted in Zimbabwe, vidual risk to take seriously the broad- linked to harvesting, mining and other er structural understanding of vulnera- forms of seasonal work. bility (through class distinction, race, gender, age, socioeconomic opportu- These different forms of migration nity, social and cultural factors) report- have a number of consequences. ed here. First, there is a marked weakening of traditional ties with family, with kin and with traditional sexual cultures. A Comparisons second consequence is the enhanced availability of sexual partners (noted We will now look at those aspects of above). The third is the expansion of young people’s sexuality that have various kinds of sex for favours and implications for the design of future other types of informal exchanges, in prevention efforts. The main com- addition to the institutionalization of parative frameworks that emerged various patterns of more directly com- from the country reports were as fol- mercial sex work. Cameroon in partic- lows: the impact of dominant sexual- ular reported periodic sex work tied to ity frameworks; onset of sexuality wage payments as a feature of this activity; the centrality of bodies; form of migration, with influxes of mass media; space; HIV risk assess- young women to towns or other sites ment, condom use and ideas on where men migrate in search of work. safer sex.

42 CHAPTER 4

The impact of dominant under- sures of modernization, the interna- standings of sexuality tionalization of youth culture, and the clearly proven sense of legiti- All country reports recognized that mate ‘ownership’ of sexuality that dominant understandings of sexuality young people claimed in all coun- emanating from the churches, the tries studied. state and its laws, and from medicine and public health, are still powerful in The Cambodia report in particular shaping sexual understandings and, urged key adults working with young sometimes, activity among young Cambodians not to continue to use people. They do so, however, in ways myths about sexuality drawn from that often conflict with young people’s their own experience alone, or from a experience and expectations of sex Cambodia of a near-mythical past. itself. Strong concern was also The dramatic and often tragic recent expressed by a number of research history of that country demands a teams at the apparent failure of these recognition of a deep and immutable dominant understandings, first, to transformation in sexual life, the grasp the complexity of young peo- report argues, requiring a somewhat ple’s sexual culture and, second, to hard-nosed approach to present dan- comprehend the impact of that reality gers rather than a reliance on a kind on HIV/AIDS. of romanticized Khmer history. This argument about working with the pre- Churches were singled out in Costa sent offers advice of value beyond Rica, Papua New Guinea, the the boundaries of that particular Philippines and Cameroon for their country. failure to endorse condom use as an HIV/STD prevention strategy in A third area where dominant under- those countries. The Cambodia standings are not applicable to the report noted the need for a stronger current sexual conduct of young peo- and more active role for the domi- ple is in relation to same-sex activity. nant Buddhist religion in similar The General Research Protocol of the terms. These are not simply anti-reli- study did not direct country research gious calls for action on the part of teams to explore this issue. Those the research teams; they are ground- research teams that pursued the issue ed directly in the empirical evidence found it relevant to their design or in these studies that young people’s salient to their analysis as fieldwork sexual lives and sexual culture have progressed. The findings indicate that been transformed in ways that are many young men (and some young deep and complex, and that such women) in all countries are willing to change will continue. Exhortations to try, and find pleasure in, same-sex abstinence and celibacy, fidelity or activity. It is also clear that the defini- delayed sexual activity, or to return tion of these activities within the to ‘traditional’ sexual transactions Western model of ‘heterosexuality’/ will definitely be disregarded, ‘homosexuality’9 is culturally inapplica- ignored, brushed aside or transcend- ble to most countries, and confounds ed by young people under the pres- prevention efforts on this issue.

9 Often read as ‘majority’/‘minority’

43 CHAPTER 4

In circumstances where these domi- For example, not only do peers, par- nant understandings might offer ade- ticularly male peers, provide a social quate and accurate guidance and arena for experimentation and information about sexual matters encouragement for initiating sexual including HIV/AIDS and STDs to activities, with girlfriends or young people, most notably in boyfriends, sex workers or in group schools and other educational institu- sex, but young people also provide tions, all country reports bemoan the each other with special words, phras- inadequacy of available programmes. es and a quite sophisticated sexual Either these are restricted to the biol- discourse to employ in pursuing, ogistic or the moralistic, or glaring in enacting and understanding their sex- their omissions (e.g., women’s sexual ual conduct. This discourse, docu- pleasure and same-sex activity) or par- mented particularly well in the tiality (e.g., sex equals reproduction). Philippines study for example (and Indeed, the Costa Rica report notes offering a strong argument for the that young people are often left to research method employed), is in fact survive on myth, ‘magic’ and muddled a powerful structuring of sexual expe- thinking instead of clear scientific rience, in a sense providing an already information about human sexuality. well-trodden, culturally specific path Other countries reported similar, often toward sex rather than away from it. wildly inaccurate, ideas about sex and Not all young people follow that path the body used by young people in to premarital sex — there are consis- their efforts to make sense of what is tent, if few, accounts of restraint in all happening in their lives. Although country reports — but the prevailing some of these ideas can be dressed sexual culture of young people in this up as alternative explanations or local study seems to be increasingly mov- traditions and understandings, many ing toward a wide acceptance of pre- others are just myth, as a close read- marital sex. This acceptance is similar ing of the individual country reports to the rapid changes in young peo- reveals. ple’s sexual conduct in the West following the evocatively misnamed Onset of sexuality activity ‘permissiveness’ of the 1960s and the invention of the contraceptive pill. The great variation in ages for onset of sexual activity confirms previous Embodiment findings that there can be no generalized starting point for young people’s One of the richest sets of material introduction to sexual activity based obtained by the country research on the arbitrary distinction of young teams concerned the body’s respons- people as a universal category or age es to sexual activity. This is an area of cohort (Cleland & Ferry, 1995). Also, sexuality research that is often the seemingly accidental initiation ignored empirically and yet is crucial into sex in most young people’s to understanding the experience of accounts must be interpreted as com- sex. Young people in most country plex culturally specific moments, studies talked of the experiences of which will necessitate a significant their bodies, the sensations of arousal, rethinking of prevention education ejaculation, orgasm, pain, pleasure, among the presexual. and the physicality of the partner’s

44 CHAPTER 4 body and their own in the midst of concerns the impact of new media sex. There are quite honest accounts technologies, such as videotapes and in some studies of the sensations of magazines, which in a number of various practices such as masturba- country reports were listed as sources tion, first penetration (vaginal and of information on sex for young peo- anal), oral sex, and group sex that are ple. These are sources of information encountered by young people in a valued by young people. Various completely unprepared state. Not countries reported the impact of only are the basic knowledges about ‘romance’ literature, popular music, sex and reproduction quite often film and the like, and it is important to absent from the understandings, note that although pornography was meanings and ideas young people more readily available through these bring to their growing sexual desires, new technologies, not all such materi- but their bodies once engaged in sex al enjoyed by young people was reveal what they have not been told pornographic. Care should be taken or prepared for at the level of physical to record the differential impacts of actions and responses. new technologies and the type of messages they disseminate. Increased This should be of vital concern to sex access to globalizing technologies is educators and health promotion pro- not just going to affect young people fessionals. In the absence of any ade- in developing countries, but every- quate information about the nature of where, and adults will not remain the body’s capacity for sexual arousal untouched by these developments and response, simple information either. about the mechanics of sex becomes lost, forgotten or irrelevant when sex Space is actually happening. The dangers here for HIV/STD transmission are The country reports all register the obvious. Getting ‘carried away’ is importance of space to young people, shorthand for the physical responses not only as places where the social in sex overwhelming judgement, and interaction that precedes sex occurs, this warrants an inclusion of explicit but also in relation to the real possi- information on the body-in-sex in bilities for sexual activity such spaces preparing young people for the sexu- create, and in which young people al lives they are exploring. The often create specifically sexual oppor- absence of such information will tunities. The most dramatic example inevitably mean that the body’s sensa- of this was offered in Chile’s ethno- tions alone will lead young people graphic account of ‘parties’ and ‘par- toward pleasures that will not be tying’. This account focused on the denied. significant difference between peer- organized, informal, social events that Mass media offer space in private homes for getting together, drinking, listening and Wider social and economic forces in dancing to music, flirting and maybe each country are certainly altering having sex, and a more formal space sexual possibilities for young people, created by the discothèque, where but in ways largely invisible to daily young people rely less on couples and experience. One simple example here peers, where music and dancing are

45 CHAPTER 4 erotically enhanced by a deliberately ticularly these spaces grant licence in designed atmosphere, and young a way not permitted with girlfriends. people are able to pursue almost The Cambodia report is again instruc- anonymously a more casual kind of tive on this issue. It notes the possibil- sexual encounter. There may even be ity of more varied sexual practices sections of such venues that facilitate with sex workers, including anal inter- preliminary sexual activity. This is not course and group sex, and a very an unfamiliar scenario in any major important distinction being made by city in the world today, and such young men between sex workers venues in some countries are gov- (most often Vietnamese women), what erned by various laws on alcohol and is permitted sexually with them, and drug use or rules of proper sexual how they are treated and regarded. behaviour in public. But the Chile The ‘good/bad’ (‘virgin/whore’, example warns that any reading of the ‘Madonna/Magdalen’) distinction, idea of ‘sex-on-premises’ venues as used against sexually active women being simply restricted to brothels, and noted in most countries, is doubly massage parlours, bathhouses and cruel when it is linked to forms of the like, is clearly mistaken. And the racial or ethnic distinction. The idea that all such venues are ‘adult- mythology in Cameroon that Pygmy only’ spaces is no longer sustainable. women must be forced to have sex is a similar example. Even if some of these places can be regulated in some way by laws or The common acceptance of, and rules of public behaviour, not all can. often outright support for, young men In Papua New Guinea, it is clear that using the services of sex workers and many dance halls and related alcohol going, often in groups, to visit the or drug use are rarely policed or places set aside for them, create a policeable. The attendant sexual reasonably observable sexual space. activity noted in the Papua New More readily observable, but less Guinea report, including significant understood, is ordinary street life read amounts of sex for cash or favours, as erotic space. The Costa Rica group sex and rape, is directly facili- research team effectively used street tated for young people specifically in observation to uncover the eroticism such spaces. In Cambodia, places of the street, and their report argues such as Khren Sray (a recreational that street life is crucial to young peo- resort on the river outside Phnom ple’s sexuality, particularly poorer Penh) are well known and understood young people. In the street, young to be sites of seduction and lovemak- men (who often and usually have ing — not just for the young at heart. more freedom to traverse such spaces There is a kind of humorous and toler- than young women) meet peers, older ant acceptance of the existence of men, homosexuals, sex workers and such spaces, for Cambodian culture is potential sex partners. They see not anti-sex. seduction and sometimes even sex acts. Spaces for sex to occur — parks, Beyond such recreation spaces there beaches, alleys, in cars, behind fences are also the specific sectors of cities — become inducements, enhance- and towns set aside for sex work ments, adding frisson (as the Costa encounters, and for young men par- Rica report argues) to the illicit, turn-

46 CHAPTER 4 ing sexual adventuring into a repudia- is clear that HIV itself is not sufficiently tion of the private and narrowly prioritized to override other considera- approved kind of sex young people tions in relation to young people’s sex- are expected and hoped eventually to ual expression. This conclusion war- pursue. rants more respectful consideration in any rethinking about the character of The central argument from this discus- young people’s risk assessment. sion of sexual space is that young people inhabit a number of differently There is clearly strong evidence that structured spaces: spaces they create; many young people have dangerous- spaces others create for them; spaces ly poor levels of knowledge about created for others but which young HIV and STDs, and some countries people traverse; spaces specifically have evidence from previous studies for sex; spaces that are sexually of high levels of risk-taking, e.g., charged. Indeed, it is clear that young Papua New Guinea, Costa Rica, the people are not and cannot be quaran- Philippines. These poor levels of tined from sexual spaces, unless knowledge and worrying levels of adults are prepared to ‘de-sexualise’ risk-taking are not uniformly spread the spaces they themselves currently across all samples of young people enjoy — and that is highly unlikely to studied, demanding again that the happen. Such spaces cannot be tar- simplistic declaration of all young geted for health promotion in uniform people being ‘at-risk’ by definition, ways; they need thoughtful and simply because they are young, be appropriate techniques of interven- set aside for a more sophisticated tion. Developing interventions for recognition of the social and cultural these non-institutional and informal contexts in which risk is constituted. sexual spaces will require more For example, the Papua New Guinea grounded and localized approaches report noted that condoms are very than can be achieved in the formal favourably accepted by young peo- curriculum of a sex education pro- ple there; how to access and distrib- gramme for instance, or for that mat- ute them in the context of a rapidly ter in large-scale, mass-media cam- deteriorating public health system is paigns. Young people themselves are the real problem. Social marketing of vital to the development of such pro- condoms needs a strategy in this con- grammes geared to informal sexual text quite different from that spaces, for without their guidance and employed in places where condoms disclosure adults will be refused are readily available but not used access to these carefully guarded (Philippines), or disliked for their secrets. interference with machista concepts of sexuality such as in Chile. HIV risk assessment, condom use and ideas on safer sex Conclusions All reports revealed a marked convergence in actual risk-taking among The similarities between countries young people, even in the context of involved in this study provide com- significant and serious consideration pelling evidence of a complex and by young people of safety and risk. It heterogeneous situation for young

47 CHAPTER 4 people that can be generalized to becomes much more meaningful, many other countries on reflection sought-after and even legitimate in and with due care. There is little to be the face of a rapidly changing but gained in developing HIV/STD pro- uncertain world. grammes among young people from homogeneous, simplistic, universaliz- This rethinking of our understanding ing and deeply romanticized visions of of young people’s sexuality requires young people as one inevitably at-risk a shift in emphasis from the merely population. There really is no one behavioural and descriptive, to the population called ‘young people’ and more sociocultural and interactionist. no one strategy to be developed to Young people’s sexual conduct provide for them. Naïve understand- needs to be understood as directly ings or any oversimplification of the related to the quality of their rela- complexities underlying terms such as tionships with each other and the ‘vulnerability’ or ‘at-risk’ can easily place sex has in developing those lead these to become catch-all clichés relationships. The daily social life of that distort rather than assist our young people is deeply bound up understanding of the situations in with sex acts and sexual meanings. which young people find themselves This interpretation requires that we in relation to sexual health. It is far reject notions of young people as more useful to pursue the under- victims, as passive, as vulnerable lost standing of young people’s sexual souls on the brink of self-destruction behaviour in the context of their (pervasive ideas about them). We immediate peers and surroundings. need to recognize the rich resources There, we must take into account local that young people actively draw cultural forms and expectations of upon, create and then bring to their sex. We must also include the physical own sexual pursuits and interests. It attractions and sensations of sex, and is only by recognizing young peo- must reckon with any dispute ple’s capacity to act on their own between all of this and the dominant behalf that we will be able to devel- understandings of sexuality that op a realistic understanding of their would appear largely to fail to inhibit sexual lives. This understanding will or alter young people’s interest in sex provide the most effective and reli- and sexual relationships. But we must able stance from which to work with also not forget the larger, socioeco- young people to help them develop nomic contexts implicated in the active ownership of their health and, potentials in, and constraints upon, consequently, to take responsibility young people’s daily lives. Sex for managing their own HIV/STD risk.

48 CHAPTER 5

RECOMMENDATIONS In the light of this, we urgently need to shift the frame of reference from the hitherto naïve and simplistic representations of young people and HIV/AIDS toward this more sophisti- The country reports contain a number cated and directly useful framework. of recommendations for the specific The starting point is a significant re- countries that took part in the working of the concept of vulnerabili- Contextual Factors study, and many of ty and a prioritizing of young people these offer useful ideas to other coun- in relation to risk-in-context, as dis- tries as well. These individual country tinct from merely as ‘at-risk’ by defini- recommendations will not be present- tion. ed here; they are available in the full country reports and are noted in the country summaries in the Appendix to Programmatic responses this report. Instead, we will concen- trate on key themes and issues of rel- A marked upgrading of effort in evance to those involved in pro- HIV/AIDS prevention and sexual gramme development and implemen- health promotion among young peo- tation with and for young people ple is also needed. This means intro- under the following headings: con- ducing, developing and upgrading of ceptual leadership; programmatic sex education programmes in educa- responses; gender; dissemination; tional institutions. This requires dis- and research. tinctively educational expertise, different from that of medical or health pro- Conceptual leadership motion professionals, and will involve including new bodies of expertise The similarities and comparisons dis- from the educational specialities (cur- cussed in Chapter 4 offer new starting riculum development specialists, points for rethinking issues related to teacher trainers, classroom materials young people, HIV/AIDS and sexual producers, pedagogy evaluators, health. They promote a more sophisti- community and adult educationalists, cated understanding of sexuality and those skilled in school management, social relationships, which contextual- and so on). The intellectual discipline izes sexual practices and relations with and professional field of education is their meanings and intentions, their large and specialized, with enormous origins and foundations, and includes resources for enhancing countries’ their capacity to transform culture. responses to HIV/AIDS. The formal This kind of understanding of young education sector, intellectually and people’s sexual conduct is more imme- professionally, has mostly been diately useful, more easily apprehend- ignored in the narrow definition of ed, and more directly translatable into health that presides in most countries programmatic responses than descrip- and in many international responses tive statistics of sexual practices, social to date. epidemiological modelling of risk-taking, or standard discourses of young Young people in these studies value people’s vulnerability. any information they have received

49 CHAPTER 5 from their schools and university- often destabilizing whatever capacity based programmes, despite the fact families have to deal with their sexual- that they regard it as woefully inade- ly maturing offspring. There is a need quate most of the time. The willing- for programmes of support for adults ness of young people to engage in who work with young people, includ- formal sex education (governmental ing families, to take advantage of the and religious opposition notwith- declared willingness of young people standing) is a tremendous resource to learn and their desire for assistance not to be wasted. There is important revealed in these studies. Simply rein- international work to be done devel- forcing traditional educational rela- oping generic frameworks for tionships will not work; new approach- HIV/STD sex education at various es are needed and this may require institutional levels. These clearly need further research, but there is a valu- to outline the coverage of human sex- able health promotion resource avail- uality to be addressed and the unde- able in both the concern of adults and niable need for basic information on the willingness of young people at HIV/STD prevention, such as condom least to listen to and hope for infor- use. mation from those adults.

Providing technical educational guid- Many young people in developing ance in the development of sex edu- countries are not in educational insti- cation programmes will require more tutions and even more are not in the than a shift in expertise toward educa- workforce. Out-of-school health pro- tionalists; it requires tackling those motion is therefore a priority. quite powerful discourses and institu- Ethnographic research methods can tions that refuse young people be of value in needs assessment, local respectful recognition of their sexual area assessments, in the observation reality. Condom use is very patchy of local sexual cultures (spaces, among young people, and premarital processes, language and intrigues), in pregnancy would appear on the rise in the rapid assessment of sexual net- most countries in this study. works, and so on. There is a need for Encouraging condom use by young skills training for those working with people cannot be other than a first young people out-of-school to assist priority. Health promotion that them to obtain efficiently, and utilize ignores the realities of an authentic effectively, the kinds of information sexuality for young people and there- gained. by denies them adequate access to condoms is condemning them to Gender enhanced HIV/STD risk and rising rates of premarital pregnancy at Renewed focus is needed on the younger ages. enhanced risk many young women face with reference to HIV/STD, Parents are struggling to deal ade- noted in every country report in the quately with the rapidly changing sex- Contextual Factors study, and con- ual culture of young people. firmed by other recent research from Traditional forms of sexual training are the International Center for Research often partial or no longer working, on Women (Weiss, Whelan & Gupta, and larger socioeconomic forces are 1996). The growing practice of

50 CHAPTER 5 exchanging sex for cash, goods or women themselves. There is a very favours; the increase in intermittent positive image of young women’s sex work related to (sometimes tran- sexuality emerging from these stud- sient) economic need and migration; ies, and it must be respected. an increasing recognition of women’s sexual pleasure; the widespread and But gender does not simply equate undeniable experience of young with women; it concerns men as well. It women of coerced sex; the impact of is clear from every country studied that globalizing sexual cultures: these are young men are sadly neglected by fam- as much an urgent part of the ilies and societies when it comes to HIV/STD agenda as is condom pro- their sexuality and sexual development. motion and sex education. Young For all their sexual activity, for all the women are not, by definition, merely instances of sexual distress and anguish ‘vulnerable’, they are actively they inflict on young women, young engaged in the production of their men pursue sex and are left to pursue sexual cultures, albeit with fewer cul- sex and their understanding of it in turally approved ways of controlling almost total silence and the absence of their own bodies and exploring their support. It is not surprising therefore own desires than men. Yet, there is that they get it wrong so often. There is solid evidence in this study, particu- a clear need to demarcate a specific larly from young women in Papua agenda for young men in addition to New Guinea and Cambodia, which that for young women which is already indicates that it is foolish to ignore established. Growing international young women’s active pursuit of sex- interest in men’s health, increasing evi- ual interests and pleasures. Young dence of poorer health outcomes for women in Chile, embracing their men in many countries, and evidence in desire for occasional or casual sexual these studies of marked same-sex and encounters, are clearly determined group-sex activity and increasing use of that they are not to be regarded as sex workers, argues for a speedy devel- wayward, wicked or sexually loose, or opment of effort in this area. Young in any other terms that have previous- men’s sexual health is not just about ly accompanied young women’s pre- STDs and HIV prevention, or a focus on marital sexual activity. It is important men as clients of sex workers; it is about also to note, for example, that young recognizing the sexual needs of young women in the Philippines and men, their search for information about Zimbabwe regard young men’s sexual their desires, experiences and bodies, attention and activity as necessary and the help they need in facing the parts of developing sexual relations; impact of those same, larger, socioeco- lack of such attention is often read as nomic forces that their female counter- serious lack of interest and threatens parts have to contend with. This argu- the development of the relationship ment is not a claim to equal time and toward marriage. There are many equal space. It is a recognition that other instances in the findings of this unless attention is paid to young men’s study that indicate that any conceptu- sexuality, most efforts to help young alization of young women as sexual women will be largely ineffective. The victims, and of their sexual experi- interdependence of the situations for ences as unfortunate and inauthentic, both sexes is clearly revealed in these will be soundly resisted by young studies.

51 CHAPTER 5

A final caution must be issued here: countries for its lack of foresight in the owned, authentic and pursued relation to dissemination of findings sexual life of young people in these and their implementation within pro- studies demands that attempts to grammes. Within the countries intervene and assist them are not involved in this study, it would be a done at the expense of one sex for great shame not to see the findings the other. Representing young men as widely discussed and implemented. ‘predators’ is as counterproductive as representing young women as ‘vic- Research tims’. Ultimately, the importance to young people of their sexual lives and The findings from these seven coun- relationships with each other will lead tries and the comparative analysis them to reject such representations reveal the strength of this kind of and the health promotion messages research in getting to the heart of accompanying them. complex social problems that com- pound the impact of HIV/AIDS. A sim- Dissemination ple example of this is the qualitative difference in utility and depth Some of the findings of the various between survey findings on frequen- country reports in the Contextual cies of unsafe sexual practices and the Factors study are country-specific and notion of sexual cultures explored in are not generalizable in any conven- these studies. Although the former tional research sense. But each is focuses attention on patterns of risk- instructive both methodologically and taking and potential population tar- substantively. There is significant utili- gets, the latter is more able to offer ty to be gained by disseminating people developing interventions the these findings to ministries of health, wherewithal to generate them. Thus, ministries of education, national AIDS the methodologies are not (and have programmes, peak and local non- never been) in competition; they com- governmental and community-based plement each other in very important health promotion organizations, ways and both are needed to provide HIV/AIDS resource centres, and to proper research advice. community educators themselves. Education departments responsible It might be argued that many coun- for sex education, personal develop- tries, however, do not have these ment programmes and the like will research capacities and few would be also be a receptive audience for these able to undertake research of the findings, as are professionals and vol- complexity of this study. Yet, even unteers working with out-of-school though the research teams who young people. Researchers too will undertook this study were not all benefit from understanding the utility experienced in qualitative research of the methodologies used by the var- methodology, the research process ious country research teams in this itself provided remarkable training in study. new skills and the consolidation of existing skills for all teams. The theo- There has been significant and ongo- retical shifts were also significant. The ing criticism of HIV/AIDS research in findings of the country reports clearly many developed and developing indicate the sophistication of these

52 CHAPTER 5 teams and the work they undertook. final reports. This suggests the devel- This argues for the development of opment of social research models that programmes of in-country research combine clear research design princi- that include components for strength- ples, methodological innovation, and ening research capacity. Close-focus flexible research management and sup- research is ideally suited to this pur- port techniques. It is important not to pose and, as the findings of this study set research priorities on the basis of reveal, also offer rich materials for pressing research concerns or issues immediate use in developing pro- alone; there is as much to be gained by grammes for the target populations. stimulating and developing social research ‘processes’ and relationships as there is in concentrating on ‘content’ Part of the Contextual Factors study’s and research agendas. The spin-off success is due also to the overall design from this approach is considerable in of the project, from the development of terms of in-country research planning, the General Research Protocol (which resource allocation and capacity held up remarkably well as the ground strengthening, and increases the focus work for all the studies), through the on health promotion as a research- technical support provided at a number informed activity, both for prevention of stages in-country and from Geneva, and for health management for people to detailed guidance for the writing of with HIV and AIDS.

53 CHAPTER 5

COUNTRY REPORTS behaviour amongst young people in urban and rural Zimbabwe. University of Zimbabwe, Harare, Zimbabwe.

Abega S-C (1996) Learning and the Contact Ms Agnes Runganga, Human real life experience of sexuality among Behaviour Research Centre, PO Box young Cameroonians aged from 15 to BW 1517, Borrowdale, Harare, 30. University of Yaoundé I, Yaoundé, Zimbabwe. Cameroon. ******* Schifter J, & Madrigal J (1996) Sexual culture and rebellion in Costa Rican Contact Dr Sévérin-Cécile Abega, youth: impact on the risk of HIV/AIDS Faculty of Social Sciences of the infection. Instituto Latinoamericano de Catholic University of Yaoundé, BP Prevención y Educación en Salud, San 11,Yaoundé, Cameroon. José, Costa Rica. *******

Jenkins C (1996) Report on youth, Contact Dr Jacobo Schifter, Director, urbanisation and sexuality in Papua ILPES, PO Box 561-1002, San José New Guinea. Papua New Guinea 1000, Costa Rica. Institute of Medical Research, Goroka, Papua New Guinea. ******* Tan ML, Batangan TU, Espanola HC, Contact Dr Carol Jenkins, Section Guiang L, Quintilla L, Yacat J, Banaag Coordinator Health and Population, R, Jusayan S & Heredia C (1996) CARE Bangladesh, Rd 7/A, Hse No. Contextual factors affecting risk-related 60, Dhanmondi, Dhaka 1209, sexual behavior among young adults in Bangladesh. the Philippines. Health Action Information Network, Quezon City, ******* Philippines.

Marshall MT, Canales M, Lopez G & Scombich X (1995) Chile: youth, culture Contact Dr Michael Tan, Executive and sexual behaviour. Corporación de Director, Health Action Information Salud y Políticas Sociales (CORSAPS), Network, 9 Cabanatuan Road, Philam Santiago, Chile. Homes, Quezon City 1104, Philippines.

******* Contact Dr Maria Teresa Marshall, Universidad Càtolica Vicuña Mackenna, Tarr, CM (1996) People in Cambodia 4860 Campus San Joaquín Escuela de don’t talk about sex, they simply do it! Trabajo Social, Santiago, Chile. University of Fine Arts, Phnom Penh, Cambodia. ******* Runganga A (1996) Social and contex- Contact Dr Chou Meng Tarr, PO Box tual factors affecting risk-related sexual 1010, Phnom Penh, Cambodia

54 CHAPTER 5

REFERENCES Egger M, Ferrie J, Gorter A, Gonzalez S, Gutierrez R, Pauw J & Smith GD (1993) HIV/AIDS-related knowledge, attitudes, and practices among Managuan secondary schools. Bulletin of the Pan American Health Bhende A (1993) Evolving a model for Organization, 27(4),360-369. AIDS prevention education among low-income adolescent girls in urban Feldman DA, O’Hara P, Baboo KS, India. International Center for Chitalu WN & Lu Y (1993) Sexual Research on Women, Washington behavior among male adolescents in DC, Women and AIDS Research Zambian secondary schools. IX Program, September: 3 (Report-in- International Conference on AIDS, Brief). Berlin, July (abst. no. PO-D02-3433).

Binda ki Muaka P, Ndonda-Kumba R Ford NJ & Kittisuksathit S (1994) & Lufua-Mananga G (1993) Impact of Destinations unknown: the gender HIV/AIDS educational programme on construction and changing nature of sexual behaviour of young students in the sexual expressions of Thai youth. harbour town of Boma, Zaire. IX AIDS Care, 6(5),517-531. International Conference on AIDS, Berlin, June (abst. no. PO-D14-3818). Hammersley M & Atkinson P (1989) Ethnography: Principles in Practice. Campbell B & Mbizvo MT (1994) Routledge, London & New York. Sexual behaviour and HIV knowledge Haworth A, Mataka EN, Muzizi L & among adolescent boys in Poulter C (1996) Religious beliefs and Zimbabwe. Central African journal of sexual behaviour of tertiary level stu- medicine, 40(9),245-250. dents in Zambia. XI International Conference on AIDS, Vancouver, July Castaneda X, Allen B & Castaneda I (abstr. no. Th.C.4466). (1996) Migration, virginity and sexual initiation: factors associated with JOICEFP (1994) Gaining a clearer pic- STD/AIDS risk perception among ture of youth behavior, Latin America. rural adolescents in Mexico, IX JOICEFP news, April, (238),4. International Conference on AIDS, Vancouver, July (abstr. no. Tu.D.2703). Mulatu MS & Haile DN (1996) Sexual risk behaviour and condom use Cleland J & Ferry B (eds.) (1995) among Ethiopian high school adoles- Sexual behaviour and AIDS in the cents. IX International Conference on developing world. Taylor & Francis AIDS, Vancouver, July (abstr. no. and World Health Organization, Mo.D.1692). London. Paiva V (1993) Sexuality, condom use Domingo LJ (1995) Youth under and gender norms among Brazilian threat of HIV/AIDS. Young adult fertil- teenagers. Reproductive health mat- ity and sexuality study II, University of ters, 2,98-109. the Philippines, Population Institute, Quezon City, November.

55 CHAPTER 5

Pattaravanich U (1993) Effect of cultur- Richter L & Swart-Kruger J (1996) al factors on safer sexual behaviour AIDS risk of South African street chil- among adolescents in Thailand. In: dren: the need for social and individ- Ford N & Chamratrithirong A, eds. ual interventions. XI International UK-Thai Collaborative Research Conference on AIDS, Vancouver, July, Development in Reproductive and (Abstr. No. Th.D.4912). Sexual Health: Proceedings of the Symposium on the Mahidol-Exeter Ruiz J (1994) Street youth in British Council Link, Thailand, Mahidol Colombia: lifestyle, attitudes and University, Institute for Population and knowledge. AIDS health promotion Social Research (IPSR), exchange, 1,12-14. Nakornpathom, 73-83. Teka T (1993) College students’ atti- Pattullo AL, Malonza M, Kimani GG, tudes and knowledge of AIDS. Muthee A, Otieno PA, Odhiambo K, Ethiopian medical journal, 31(4),233- Moses S & Plummer FA (1994) Survey 237. of knowledge, behaviour and attitudes relating to HIV infection and AIDS Weiss E & Gupta RG (1996) Lessons among Kenyan secondary school stu- learned from conducting sexuality dents. AIDS Care, 6(2),173-181. research. IX International Conference on AIDS, Vancouver, July (abstr. no. Raffaelli M, Campos R, Meritt AP, Tu.D.2720). Siqueira E, Antunes CM, Parker R, Greco M, Greco D & Halsey N (1993) Weiss E, Whelan D & Gupta RG Sexual practices and attitudes of (1996) Vulnerability and opportunity: street youth in Belo Horizonte, Brazil. adolescents and HIV/AIDS in the Social science and medicine, developing world, International 37(5),661-670. Center for Research on Women, Washington DC. Richens J (1944) Sexually transmitted diseases in children in developing Zimba RF (1995) Secondary school countries. Genitourinary medicine, students’ risks that may promote HIV 70(4),278-283. infection and the spread of AIDS. School psychology international, 16(1),67-78.

56 Part 2 Community responses to AIDS

Peter Aggleton and Ian Warwick1 (from an original analysis by Ian Warwick)

1 Thomas Coram Research Unit, Institute of Education, University of London, UK CHAPTERCHAPTER X1

INTRODUCTION It is against this backcloth that the studies described in this report were undertaken. Knowing more about how households, families and com- The global epidemic of HIV and AIDS munities cope with HIV disease is of has brought with it some of the great- great importance if we are to develop est devastation known this century. programmes and interventions to There are few places in the world reinforce positive responses and to untouched by HIV and its ability to counter more negative reactions. Yet irrevocably alter the lives of individu- as the studies described here make als, families, communities and soci- clear, local responses to infection have eties. But as numerous authors have their origins in often long-standing pointed out (see, for example, Mann, community structures and beliefs. Tarantola & Netter, 1992; Mann & These include how HIV and AIDS are Tarantola, 1996; Piot, 1997), AIDS is perceived and understood locally, tra- not simply a biomedical phenome- ditional ways of coping with illness non, it has profound social and per- and disease, assumptions about the sonal consequences. It has the poten- responsibilities of families and neigh- tial to inflict the most terrible pain and bours in times of crisis, and stigma suffering, to end all hope for the and discrimination in relation to dis- future, and to rearrange lives in ways eases acquired sexually and/or that could never have been anticipat- through drug-related activities. ed. It separates parents and children, husbands and wives, brothers and sis- The principal aim of the research ters, lovers and friends. It can destroy described in this report was to identi- economic systems, and the capacity fy some of the ways in which house- of towns, villages and communities to holds and communities in different provide for themselves (World Bank, parts of the world have responded to 1997). And it can inflict an intolerable the AIDS epidemic, and to draw con- burden on hard-stretched health ser- clusions about the kinds of activities vices in both developing and devel- and programmes that might usefully oped countries. Yet AIDS also has the strengthen existing responses for the potential to trigger responses of benefit of people living with HIV and tremendous courage. Like many cata- AIDS, their carers and their communi- strophes, it can bring out the best in ties. Following an initial assessment of people — the capacity to care for oth- potential study sites, the countries in ers, the desire to show understanding which the studies were conducted — in times of crisis, and the ability to the Dominican Republic, India, offer support in extreme adversity. Mexico, Thailand and the United And here lies the paradox, for in the Republic of Tanzania— were selected midst of this damage and destruction by the Steering Committee for Social lies the means to teach us what we are and Behavioural Research of the — a global community united by a World Health Organization’s former common desire to deal with HIV and Global Programme on AIDS. In each AIDS, and its impact on households case, and with the support of national and communities worldwide. authorities, local principal investiga-

58 CHAPTER 1 tors and teams undertook the studies issues of relevance to the planning of which are described, often in close future programmes and interventions. collaboration with nongovernmental A series of individual country reports organizations and groups of people were developed and are available living with HIV and AIDS. This was a direct from the principal investigators deliberate strategy since it was (see p. 99). This comparative analysis believed that only by being close to of findings identifies the major themes the ground would insights of real and issues emerging across all five value emerge. countries. In keeping with the methodological commitments of The research methods adopted in qualitative and exploratory enquiry, each case were largely qualitative and no effort has been made to quantify exploratory — the aim being to iden- the responses given. The themes tify recurrent themes and patterns of highlighted, however, are those which response, perhaps as the prelude to local investigators identified as typical later more quantitative enquiry. A and recurrent within their sites. general research protocol for the Idiosyncratic and unusual responses studies provided an overall framework are always cited as such. within which the research was to be conducted, but principal investigators The studies themselves were carried and their teams were invited to refine out in Santo Domingo the capital city and extend this basic framework in of the Dominican Republic; in Mexico accordance with local needs. Two City and Netzahualcóyotl in Mexico; contrasting sites were chosen within in the Kyela district of the United each country, and in each of these an Republic of Tanzania; in Mumbai in initial rapid assessment process, or India; and in two villages in Chiang RAP, preceded a period of in-depth Mai Province, Thailand. Following a data collection. The latter most usual- description of the background to the ly involved interviews with key infor- studies (Chapter 2) and of the study mants, people with HIV and AIDS, design and methodology employed affected families and households, (Chapter 3), this report contains a focus group discussions, and observa- comparative analysis of findings tion. Data was analysed thematically across all five sites (Chapter 4). A final in relation to a series of research ques- chapter outlines some of the key tions outlined in the general research implications for policy and practice protocol and to identify site specific (Chapter 5).

59 CHAPTER 2

BACKGROUND The theoretical rationale behind the programme of work was

‘... (that) human beings, either as individual or collective bodies, act in response to events and crises that chal- The impact of HIV and AIDS on lenge their structured patterns of exis- households and communities, and tence and survival and, in the process, the responses this gives rise to, are their lives move on and societies of considerable policy and practical evolve’. (World Health Organization, interest. Knowing more about these 1993). responses may enable us to identify the reactions of individuals, families AIDS was therefore viewed as one of and affected communities that war- a number of challenges facing rant reinforcement. It may also lead households and communities, one to to a better understanding of the which they must respond if they are more negative consequences of the to continue to function well. The epidemic such as discrimination, diversity of responses witnessed stigmatization, social exclusion, and globally results from the impact of individual and household distress. In the epidemic on different communi- order to learn more about these ties at different stages of the epi- positive and negative responses, demic as well as prevailing econom- and their implications for HIV pre- ic, sociocultural and institutional vention, care and support, a series conditions and traditions. Numerous of two-year studies was initiated in studies have now shown how social 1993/94. context influences individual and community responses to HIV and These studies set out to examine AIDS (see, for example, Dowsett & household and community respons- O’Brien, 1994; Leonard & Khan, es to HIV and AIDS in five different 1994; Dodd, 1996; AIDS Analysis countries —the Dominican Republic, 1997). Social context includes influ- India, Mexico, Thailand and the ences due to society, community, United Republic of Tanzania. The households and the individual. aim was to examine the relationship Society is usually taken to refer to between sociocultural, political, the broad civic, cultural and perhaps economic and demographic factors sub-cultural features of a population, determining different responses at whereas community describes geo- household and community levels. graphical, ideological or behavioural Through in-depth study in a range commonalities. Households may be of different sites, it was hoped to construed as the relationships identify policy and intervention between a number of people, but implications of value to international may also consist of just one person and national agencies and those living alone. Individuals may be working at grass roots level. The understood in reference to their bio- majority of studies began in 1994 logical nature, their emotional and and ended in 1996, enabling this social characteristics, their cognitive comparative cross-site analysis to be capabilities and/or their spiritual completed shortly afterwards. qualities. Responses to HIV and

60 CHAPTER 2

AIDS depend not only on this range increasingly travel far from their part- of societal, community, household ners in search of work, and women and individual factors, but also may have little option other than to change over time. sell sex (occasionally or regularly) in order to obtain income. Societal impact and responses The development of transport systems may also have facilitated the Opinions differ concerning the growth of the epidemic. As impact of HIV and AIDS in different Outwater (1997) has recently noted societies. There can be no doubt in her discussion of the social and that AIDS has had a serious impact economic impact of AIDS on women on individuals, their families and in the United Republic of Tanzania, friends, and there is widespread there is often a lower incidence of acceptance that its effects on health HIV infection among communities care systems can be substantial furthest away from urban areas and (Barnett & Blaikie, 1992; Ainsworth & transport routes. Indeed, the offer of Over, 1994; Cohen & Trussell, 1996). a renovated road to one group of vil- It is also well established that HIV lagers was reportedly turned down and AIDS have effects on house- for fear that it would ‘bring AIDS’ holds including, at the aggregate into the community. In some coun- level, the extent and depth of pover- tries in south-east Asia, and prior to ty (World Bank, 1997). While house- recent economic crises, rapid eco- holds may attempt to compensate nomic expansion has caused a for the loss of economically active decline in traditional forms of adults by, for example, working employment. In parts of Chiang Mai longer hours, selling assets and with- province in Thailand, for example, drawing children from school, the traditional rice farming has given extent to which they can cope with way to construction labour, caddying the effects of HIV and AIDS varies and work in the hospitality and dramatically. The economic impact entertainment industries, as rice of AIDS has been documented as fields have been transformed into being larger on poor households, golf courses and new housing affecting food consumption, child estates (Singhanetra Renard cited in nutrition and school attendance World Health Organization and among other factors (World Bank, UNAIDS, 1998). 1997). Paralleling these economic and infra- There is evidence to suggest that the structural transformations have been structural adjustment programmes the responses of national and local instituted in many African countries governments. Internationally, gov- since the 1970s may have inadver- ernment responses to the epidemic tently contributed to the growth of have varied from the ‘inspirational the epidemic by increasing migra- and the compassionate’ to the ‘igno- tion and urbanization, and by caus- rant and discriminatory’ (Grunseit & ing a reduction in spending on Kippax, 1992). Punitive government health and social services (Ankrah, responses can affect both the quality 1996). Men, for example, may and availability of care as well as the

61 CHAPTER 2 efficacy of prevention initiatives. While the impact of AIDS on women They can also result in the apparent is increasingly recognized and some invisibility of people with HIV and CBOs are beginning to respond AIDS, thereby reinforcing the belief specifically to their needs (Shrestha, that AIDS is an epidemic of ‘out- 1996), the self-help ethos of these siders’. The factors influencing differ- organizations, and ironically their ent kinds of government response successes, can conceal deficits in are poorly understood, and more more formal health care provision. information is needed about how to Moreover, if women are cared for pri- transform negative responses into marily through the voluntary and more supportive ones. While the informal sectors, their needs can studies described here did not become further marginalized by the attempt to explore these concerns in formal health care system. depth, it is important to recognize that household and community CBOs are vital for the provision of responses always occur within a care in communities affected by AIDS broader national economic and (Mercer, Mariel & Scott, 1993; political context. Leonard & Khan, 1994; Fernandes et al., 1996; Jayaseelan, 1996). While a Community impact and number of factors, including limited responses resources (Leonard & Khan, 1994), constrain the effectiveness of such Communities can be thought of in a organizations, they are important number of ways: geographically, in sources of support for groups as terms of the sharing of a common diverse as children, women, drug identity, or linked by a set of com- users and homosexually active men mon practices. In relation to care (Mello, 1996). The effects of their and support, communities consist of actions extend beyond support, how- at least three interlocking systems: ever. Grunseit & Kippax (1992), for the formal health care system; the example, have highlighted how CBO- voluntary care system including sup- supported community mobilization port provided through community among gay men in the Dominican based organizations (CBOs) and Republic helped encourage safer sex- nongovernmental organizations ual practices and provided the oppor- (NGOs); and the informal care sys- tunity for significant changes in tem consisting of the support pro- knowledge and behaviour. vided by friends, families and other relatives. Even so, community organizations and affiliations are not always successful in Mitigating the effects of HIV and bringing about positive changes in AIDS on individuals has often been people’s behaviour. Even in well-fund- given low priority by national HIV and ed CBOs, resources can be wasted as AIDS control programmes (Cohen & a result of poor management Trussell, 1996; Ankrah et al., 1997). In (Mannemplaven in Grunseit & Kippax, developing countries, day-to-day 1992). Which organizations are best care and support is usually provided placed to provide HIV- and AIDS-relat- by unpaid household and community ed services has been examined by members, who are usually women. Cohen & Trussell (1996) in their call for

62 CHAPTER 2 further research into the optimal roles of these areas are likely to require special- national governments, international ist training. donors and NGOs. They argue for a closer examination of the efficacy of Whilst there is an increasing awareness AIDS-specific, as opposed to non-AIDS- of the HIV- and AIDS-related needs of specific NGOs, in mitigating the impact women, and a recognition of the of AIDS. necessity of making services more gynocentric, there is little to suggest Ankrah, Schwartz and Miller (1997) how to develop services so as to meet have identified some of the main ways both women’s and men’s needs. Seidel in which health care and social support (1996) has recently pointed to the systems can meet HIV- and AIDS-relat- importance of addressing gender as ed needs at community level. Whereas opposed to ‘women’s issues’. Such an health care systems encompass formal approach re-focuses policy and inter- structures, services and activities vention attention on‘... the difference ‘designed primarily to deliver clinical between women’s and men’s interests management of the disease and to even within the household...’ ensure health maintenance’, support systems provide various forms of Household impact and social assistance to lessen the effect of responses AIDS-related stresses. These might include tangible forms of aid such as HIV and AIDS impact significantly on money, food and psychological assis- households and families, triggering a tance, as well as emotional support variety of responses. Households and and information. kinship networks have an important role to play in caring for bereaved children However, these same authors also high- (Forsyth & Rau, 1996; Levine, 1996) and light the androcentric nature of much in providing support and care for peo- existing health care provision. Fees and ple with AIDS (Chuaprapaisilp & the geographical location of services Parsons, 1996; Forina et al., 1996). They can make access to health care prob- may also offer a forum within which lematic for women. Furthermore, people can appraise and evaluate the women’s AIDS- and HIV-related med- information they receive about HIV pre- ical problems may not be recognized vention (Ruben, 1996). In circumstances and treated, so excluding them from where household members provide access to services. To address such most of the care for people with AIDS, bias, existing health care systems need stress can take its toll. Family members to be re-oriented. Ankrah, Schwartz and themselves may require support in a Miller (1997) suggest that traditional form that takes into account their emo- healers, who in Africa and parts of Asia tional, physical, financial and spiritual treat 70-80% of the population, could needs (Kadhumbula, Tibatemwa & be more fully utilized in AIDS-related Kalemba, 1996; Machinjili, 1996; treatment and care. Providing AIDS- Mawejje et al., 1996). Once again, related medical care through maternal women are often the main providers of and child health facilities, as well as care (Grunseit & Kippax, 1992). through primary health services, could make access for women easier. The economic impact of HIV and AIDS However, health care staff working in on households varies according to the

63 CHAPTER 2 severity and length of the illness, as Where children have been bereaved well as available financial and human through the death of their parents or resources. Cohen and Trussell (1996) carers, the reconstitution of house- have identified three types of costs holds to provide them with care can facing households: direct costs such as take a number of forms. In most coun- those associated with medical expens- tries there are formal or, more often, es; indirect costs such as foregone informal arrangements within families earnings; and costs to other house- for the care of bereaved children. holds connected with the payment of However, despite the tradition of funeral expenses or the care of extended family members taking care bereaved children. The impact of AIDS of bereaved children who have lost also depends on who within the their parents, the costs of providing household is affected. The most care can discourage this kind of care- severe impact occurs when the head taking (Auer, 1997). In some non- of household becomes unwell and is industrialized countries, CBOs have unable to work. responded by setting up foster care placements (Auer, 1997). A diagnosis of HIV can lead to the dissolution of marriages and partnerships Where the impact of AIDS has been and, through fear of this, discourage greatest and there are few if any people from disclosing that they have appropriate adults to care for HIV infection. For women especially, bereaved children, some households being HIV seropositive can result in may be made up of children alone. If being unable to find a marriage part- they have a plot of land, shelter, and at ner or, if married, in forced separation least one girl or young woman, these or divorce. Given the low economic households seem to have a better status of unmarried women in many chance of surviving than boy-only developing countries, HIV may cause households (Auer, 1997). Again it is not only the breakup of relationships usually women, in this case young but also the loss of the means of sur- women or girls, who are able to pro- vival. Without adequate support, both vide an affected household with the men and women can be placed in the means of survival by bringing food and situation of either having to withhold preparing it, and by generating information about their HIV serostatus income through small scale trading or put their partner’s health at risk activities. Members of boy-only house- (Grunseit & Kippax, 1992). holds appear more likely to resort to living on the streets and gaining an Where children living with AIDS are income through begging. unwell, there may be immediate care needs associated with HIV. As with Group homes may also have a role to adults, the severity of illness can vary play in mitigating the effect of AIDS on over time and differs between indi- bereaved children. In Zambia, for viduals. Unlike adults, however, example, alongside support and chronic illness can lead to delays in involvement by community leaders and physical and social development extended family networks, it has been (Auer, 1997). This can create longer- possible for adults to share the respon- term challenges, which will need to sibility of providing for the emotional, be addressed. physical and spiritual needs of rela-

64 CHAPTER 2 tively large numbers of children (Auer, gate its effects can obscure the multi- 1997). faceted nature of community level AIDS-related care and support. They Where parents or care takers are suggest that interventions are better themselves HIV seropositive, deci- analysed along three dimensions: the sions have to be made about level of social organization of the whether or not children should be intended beneficiary (for example, a told. These decisions can be influ- society, a community, or an individ- enced by the shame sons and daugh- ual); the type of provider (for example ters may feel about their parents. a government, CBO, or household Where children have lost their par- members); and the kind of support ents to HIV-related illnesses, they offered. The latter might be tangible may feel that their parents were bad (such as physical resources) and/or or immoral. This stigma may have a psychological (such as emotional sup- substantial impact on the child’s or port), and can be either fixed or vari- young person’s subsequent sense of able. self-worth or esteem. The mothers of children infected through vertical Given the issues discussed, the fol- transmission may feel a sense of guilt lowing factors are likely to be among and find it difficult to let their chil- those determining whether appropri- dren know about their common HIV ate forms of support to lessen the status (Auer, 1997). These findings impact of HIV and AIDS are likely to emphasize the need to develop emerge locally: interventions which dispel feelings of stigma, discrimination, shame and • the severity of the impact of AIDS blame, of which self-blame is a sig- at a community and societal level; nificant factor (WHO/UNAIDS, 1995; • the level and nature of govern- Bor, 1997) ment response(s); • the macro-economic context; Concluding comments • existing practices and meanings linked to gender, sexuality, illness Despite an increasing number of and disease, as well as the care of studies examining the impact of AIDS children; at community and household levels, • existing health care and support recent reports often take the form of systems; local descriptions of service provi- • opportunities for community sion, with a focus either on one ser- mobilization, especially among vice (such as a particular CBO), one women; and community (such as ‘orphans’ or • the social background of the women), or one issue (such as coun- affected household (including selling, or the use made of health available economic and human information). As such, they are unlike- resources). ly to offer an adequately holistic understanding of local responses to The challenge lies in identifying how the AIDS pandemic. As Cohen and these among other variables influ- Trussell (1996) have argued, too sin- ence local patterns of response, and gular a focus on interventions to sup- the provision of HIV- and AIDS-relat- port responses to AIDS and to miti- ed care and support.

65 CHAPTER 3

STUDY DESIGN AND social reproduction, and health and illness; METHODOLOGY • relating these responses to socioeconomic structures and dominant formal responses (if any) to The programme of research described HIV/AIDS; here was guided by a general • identifying the most constructive research protocol developed by the response strategies and their World Health Organization (1993). determinants, and thus relevant The primary focus of the enquiry as modes of intervention in contend- outlined in this protocol, was to: ing with HIV/AIDS locally; • identifying mechanisms and princi- ‘ ... explore the reactions and respons- ples that give rise to and sustain es of households and communities the strategies that enable people, affected directly or indirectly by individually and collectively, to HIV/AIDS, and of those who are prevent the transmission of becoming increasingly aware of their HIV/AIDS, to discourage discrimi- vulnerability’ (WHO, 1993: 6) nation, and to care for and support those affected by the epidemic. Subsequent to the preparation of this general research protocol, potential principal investigators from a range of Study design and samples countries were invited to prepare local research proposals. These proposals All five projects followed the design were reviewed by the former of the general research protocol, but WHO/GPA Steering Committee for variations were allowed to meet local Social and Behavioural Research and needs and contingencies. A two five studies — in the Dominican stage design was followed in each Republic, India, Mexico, Thailand and country. Following the preliminary the United Republic of Tanzania — identification of potential study sites, were recommended for funding. a Rapid Assessment Process (RAP) took place to select the areas and Research questions communities for later in-depth study. Methods used during the RAP includ- Potential principal investigators were ed documentary and archival expected to address a range of research, and interviews with key research questions in their studies. informants. The subsequent selection These included: of communities took place with regard to geographical and/or social • describing household and commu- characteristics, these being defined in nity responses to HIV/AIDS and relation to the local study’s principal exploring patterns of change aims and objectives. which affected such responses; • relating such responses to domi- In the second stage of work, house- nant sociocultural patterns of holds were purposively selected from meaning attached to HIV/AIDS, within each community according to sexuality, survival, biological and their potential to shed light on the fac-

66 CHAPTER 3

tors influencing local responses to HIV identified as appropriate ways of and AIDS. Information was elicited by obtaining additional information. means of individual interviews with These general guidelines were inter- affected individuals, couple interviews, preted in each study site as follows, interviews with other household mem- and a comparison between the sam- bers and focus group discussions. ples characteristic of each site can be Observation and case studies were found in Table 1.

TABLE 1

Country Site characteristics RAP Main data collection

Dominican One high and one low Analysis of existing Individual, household and group interviews Republic prevalence site in records and studies, with 43 households — comprising 12 family Santo Domingo, the observation, key infor- groups directly affected by HIV/AIDS (6 in capital city. mant interviews. each site); 14 family groups indirectly affected by HIV/AIDS (7 in each site); 17 occupational families in sex industry (8 low prevalence area, 9 high prevalence area). India High and low preva- Analysis of existing Individual interviews with 26 people living with lence areas in Mumbai, records, media reports, HIV/AIDS (PLWA), couple interviews with 4 plus other areas in key informant inter- couples (both partners infected), interviews in same city. views. 25 households. Community based focus group discussions.

Mexico Two contrasting com- Analysis of existing (i) Neztahualcóyotl: individual or group inter- munities: records and documen- views with 19 PLWA, with 37 relatives, and with (i) Neztahualcóyotl and tary evidence, observa- 14 individuals indirectly affected by HIV/AIDS. (ii) Mexico City gay tion, key informant (ii) Mexico City gay community: individual or community interviews group interviews with 10 PLWA, with 25 volunteers and members of social networks, and with 8 relatives of people affected.

Thailand One high prevalence Analysis of existing Individual interviews with 60 household mem- community (HPC) and records and studies, bers (HPC) and 53 household members (LPC); one low prevalence observation, key infor- focus group discussions with 4 villager groups community (LPC) in mant interviews. (HPC), 3 villager groups (LPC); informal inter- Chiang Mai province. views with 80 others (HPC), 49 others (LPC). Observation. Case studies.

United Two villages with high Analysis of existing Individual interviews with 53 people, house- Republic of AIDS-related mortality, records, observation hold interviews with 40 households (10 per vil- and two villages with and key informant lage), focus group discussions with 16 groups Tanzania low AIDS-related mor- interviews. (one involving elders and health workers in tality in Kyela district. each village). Observation.

67 CHAPTER 3

Study sites At a household level, this included an analysis of perceptions of vul- Dominican Republic nerability, sexual behaviour, gen- (Santo Domingo) der relations. At a community level, it involved the study of per- Site selection — The capital city of ceptions of vulnerability, discrimi- Santo Domingo was chosen as the nation and stigmatization as well area within which study areas would as sexual practices and norms. be selected. Two comparable lower class communities were chosen, one India (Mumbai) with a higher than median number of AIDS-related deaths (as reported to Site selection — Two study areas were the National STD/AIDS Programme), chosen following an initial assessment the other with a lower than median of levels of HIV infection in different number of reported deaths. areas of Mumbai as assessed from public hospital records. One was an Rapid assessment process — During area of high reported incidence of HIV the RAP, key informant interviews, infection, the other had a proportion- observation and an analysis of existing ally lower incidence. As the research records were undertaken. This progressed, additional data were col- enabled the identification of the lected outside of these areas to demographic profile of local house- enhance the quality of the study. holds; dominant forms of economic activity; leisure activities; institutional Rapid assessment process — During responses to AIDS; traditional rituals the RAP, information was collected involving sex, partnership, marriage, and analysed from key informants and death, burial; and local beliefs about secondary sources. Key informants health. included public health officials and representatives of the State AIDS Cell, Main data collection — Following the social workers, counsellors, medical RAP, focus group interviews were professionals and NGO staff. The RAP completed with community leaders in aimed to: each study area. Household group and individual member interviews • provide an understanding of the were also carried out in selected sociocultural context of the epi- households. Together, these sought demic in Mumbai and of major to identify: institutional responses to it; and • develop an understanding of the • responses to illness, suffering and magnitude and sociodemographic death. At a household level, this profile of the epidemic in Mumbai. included enquiry into household composition, roles and economic Main data collection — The rapid activity, issues of care, support and assessment was followed by data col- bereavement. At a community lection from household and communi- level, it included study of ways in ty respondents using individual and which support was, or was not, couple interviews, as well as focus sought. group discussions. Household inter- • issues relating to HIV transmission. views aimed to identify:

68 CHAPTER 3

• experiences of living with, and Main data collection —The RAP responses to, HIV and AIDS; enabled potential key informants to • the nature of household care in rela- be identified for the second phase of tion to AIDS; and work. During this second phase, infor- • problems associated with the provi- mation was gathered from household sion of care and ways of coping with and community members through AIDS-related issues. individual and group interviews as well as focus groups. Group interviews Mexico (Netzahualcóyotl and enabled the identification of: Mexico City) • household structure(s) and ‘eco- Site selection — In Mexico, two com- nomic survival strategies’; munities were selected. The first of • shared values and beliefs about these — Ciudad Netzahualcóyotl, a AIDS, health, disease and death; municipality in the State of Mexico — and had a distinct ‘geographical and social • household problems resulting unity’ and was reported as having a from AIDS. very high proportion of AIDS-related deaths. The second community consist- Individual interviews enabled an in- ed of the ‘network of specific social and depth exploration of: cultural relations’ that make up the homosexual community(ies) in Mexico • positive and negative reactions and City. Behaviourally homosexual and responses to AIDS (e.g. support, bisexual men currently constitute the solidarity, rejection, discrimination). largest number of those reported to be living with HIV or to have died from Thailand (Chiang Mai province) AIDS-related illness in Mexico. Site selection — Chiang Mai province Rapid assessment process — The RAP was chosen as the study site since it involved collecting and analysing infor- had a high reported prevalence of HIV mation from key informant interviews, and AIDS and because of the observation and documentary evi- researchers’ prior familiarity with eco- dence. Key informant interviews were nomic and social factors affecting undertaken with community leaders household and community responses. and health workers in Ciudad Within the province, two contrasting Netzahualcóyotl, and with leaders in study areas were chosen according to the gay community. The rapid assess- reported AIDS prevalence as reported ment aimed to identify: to Village Health Stations — the first had a high reported prevalence of • the demographic, socioeconomic AIDS, the second had a lower report- and cultural context of the two com- ed prevalence. munities, with a particular focus on occupational structures, income Rapid assessment process — During (re)distribution, and institutional sta- the RAP, information was gathered via bility; and interviews and observation, as well as • values and belief systems, particu- from written records. Key informants larly those pertaining to sexuality, included village health volunteers, tra- health and illness prevention. ditional healers, monks, village offi-

69 CHAPTER 3 cials, leaders of the village senior’s provision of care and support to club, ‘housewives’ groups’ and youth people with AIDS. groups. The RAP provided: United Republic of Tanzania • information on economic and (Kyela District) leisure activities, social groupings and support systems; Site selection — Kyela district in the • information on the demographic Mbeya region of the United Republic profile of the community and of Tanzania was chosen as the study households; and site. Within this district, two areas • an overview of prevalent institu- were chosen with relatively low tional and community responses reported AIDS-related mortality, as to HIV and AIDS. well as two areas with high reported AIDS-related mortality rates. Main data collection — During the next phase of the study, households Rapid assessment process — As many were categorized according to socioe- local primary health care centres were conomic status and age of household reported as having no facilities for HIV head. Information was obtained from antibody testing, some of those with individual and group interviews, as HIV-related disease might not know well as via participant observation in the nature of their illness (or even that community activities. Interviewees they were unwell). During the RAP, included community members living therefore, households were selected with and affected by HIV and AIDS, as in consultation with village commit- well as what were termed ‘vulnerable’ tees if they had a member who had individuals. The latter included sex been ill for more than three months, workers and their clients, singers and or if a member had died of a chronic female golf caddies, and teenage illness in the year immediately prior to gang members. This phase of activity the study. aimed to: Main data collection — Individual and • collect information on sexuality, group interviews, as well as focus class, gender, and economic and group discussions, were conducted family relations as they relate to with household and community mem- responses to HIV and AIDS; bers in two villages in each of the • gather information on household selected study areas. Focus group dis- composition as well as family and cussions were undertaken with kinship networks of households women, young people and teachers. within which someone had died Data collection aimed to: from an AIDS-related illness; and • identify household and community • identify beliefs about HIV and responses to AIDS, including AIDS; issues relating to the stress of car- • examine household, clan and coming, attitudes towards people with munity responses to HIV and AIDS, HIV, the influence of media cam- including individual and communi- paigns, the role of traditional heal- ty coping strategies, and clan sur- ers, and traditional ways of coping vival strategies; with illness and their impact on the • explore issues relating to sexuality

70 CHAPTER 3

and changing socioeconomic cir- observed locally was also explored, as cumstances; and were any similarities and differences • explore the circumstances which between the views of particular facilitated individuals and groups groups. taking preventive action against HIV, which discourage discrimina- Given the preliminary and exploratory tion, and which promoted care nature of the investigations, differ- and support for those affected by ences in sampling frames and sam- AIDS. ples, and the qualitative nature of the data collected, it would be inappro- Data analysis priate to report on the frequency with which particular views were expressed In each local study, data from individ- or actions observed. Instead, the ual interviews and focus group discus- emphasis is on themes and issues that sions were analysed thematically in recurred throughout the data sets relation to the aims, objectives and amassed at each site, as well as those research questions prioritized by the prevalent across different countries. general research protocol, and the Unless otherwise stated, the views research concerns identified in each stated are those identified by the prin- local research proposal. Thematic cipal investigator and their team as analysis involved an examination of being typical of respondents at that prevailing patterns of positive and site. Minority views are always report- negative response at household and ed as such. In order to enhance the community levels, the interrelations reliability and validity of interpreta- between these, and identifiable eco- tions in this comparative analysis, its nomic, social and cultural determi- contents have been reviewed by all nants. The relationship between what the principal investigators involved in people said in individual and group the multi-site study and appropriate interviews and what could be corrections have been made.

71 CHAPTER 4

COMPARATIVE ty rates in the two villages selected for the study, the majority of informants ANALYSIS were of the opinion that HIV/AIDS was not a serious problem in their locali- ties, and that it was a disease affecting emigrants to urban settings from the We will begin our analysis by examin- village.’ (Ta, p.78).1 ing local beliefs about HIV, AIDS and transmission, since these establish the Efforts to distance AIDS from the self backcloth against which to make were also apparent in respondents’ sense of the ways in which individuals, accounts in Santo Domingo where households and communities respond interviewees from areas of low AIDS to the epidemic. Following this, we impact had difficulty in referring to will look in turn at household level AIDS as AIDS.1 responses, community responses, and interrelations between the two. ‘During interviews it was noticed that many participants failed to call AIDS Local beliefs by its name, and referred to it as “that”. This was understood as an indicator of the degree to which AIDS Beliefs about HIV and AIDS may be invested with emotional over- tones, regarded as a taboo topic, or as In all countries and at all sites within an open secret.’ (DR, p.13). them, respondents failed to draw clear distinctions between HIV and Local beliefs about the geographical AIDS. This may have been due to the distribution of AIDS were manifest in belief that there was little need to dif- all studies. In Kyela district, the belief ferentiate between the two or, as was that AIDS was a disease of towns the case in the Kyela district of the seemed to have been reinforced by United Republic of Tanzania, the local men’s earlier migration to urban areas unavailability of diagnostic tests. in search of paid work. Some had sub- Whatever terms were used, one over- sequently become ill and returned to arching theme was apparent. their villages for care and support. This Respondents either talked of, or acted had bolstered local perceptions that as if, AIDS was a disease of ‘others’. If AIDS was a disease of the big cities. you lived in a village, AIDS was seen as a disease of towns if you lived in a At this same site, respondents city, AIDS was seen as affecting only showed a marked reluctance to call an certain ‘hot spots’; and if you had sex, illness AIDS, preferring to describe it you were protected so long as you in less stigmatizing and perhaps more ‘knew’ the other person (even if she or treatable ways. This useful imprecision he was a sex worker). was reinforced by talk of local men’s inability to meet their financial obliga- ‘Despite current reports of high HIV tions to elders after migrating to sero-prevalence...and the high mortali- urban areas in search of work: not

1 Individual site reports are cited by the name of the country followed by the page of the local report on which the relevant reference is located. The countries are identified as: DR-Dominican Republic; I-India; M-Mexico; Th- Thailand; Ta-United Republic of Tanzania

72 CHAPTER 4 meeting such obligations was under- and “kigune” when it affects adults. stood as being the cause of tradition- The illness is a traditional wasting ill- al illnesses treatable by local healers. ness. People find it difficult to accept that a wasting illness may be AIDS. ... In all countries, AIDS was equated The researchers noted that “kigune” and “ngoto” are used interchangeably with death. Not only was it seen as in some focus group discussions and incurable, but it was perceived as individual interviews without the above involving a great deal of suffering. In distinction.’ (Ta. p. 55-56). Mumbai, for example, AIDS was described as a ‘thinning disease’ in Beliefs about transmission which people lost their resistance to other illnesses, became weak and Respondents subscribed to a range of died. Among gay community respon- beliefs about how HIV and AIDS were dents in Mexico City, a diagnosis of transmitted. These beliefs about trans- HIV was said to be met with despair. mission accounted for how infection As one respondent commented: occurred, the types of people affected, how people were affected, and the ‘ ... at the beginning, I felt as if I lost special vulnerability of some children. control completely, because through The main way in which HIV and AIDS the experience of friends I thought that there was nothing to do and that death were said to be transmitted was was going to be very quick.’(M. p.175). through sex. Where mentioned, infection through injecting drug use, and At all sites except Kyela district in the transmission from an infected mother United Republic of Tanzania, it was to foetus or through blood-to-blood generally known when someone, at contact appeared to be of more minor least in the later stages of illness, had concern. AIDS. In Kyela district, however, where a definitive diagnosis was difficult to Beliefs about sexual transmission make due to the lack of antibody were influenced by local ideas about detection kits, there remained much what was, and was not, appropriate uncertainty as to whether a serious ill- conduct for men and women. In all ness was AIDS. While certain symp- study sites, expectations of appropri- toms were suspected as indicating the ate sexual conduct differed for men condition — including weight loss, an and women. Men were generally per- itchy bodily rash, recurrent fevers and ceived as having a greater degree of chronic diarrhoea — assuming that an sexual freedom and, although sex illness was not AIDS enabled the indi- was often expected to be limited to vidual concerned (and other house- within marriage, young men were hold members) to avoid stigmatiza- encouraged to visit sex workers as tion and shame. part of the process of becoming adult. As adults, men’s perceived ‘A large proportion of informants claimed that a cultural condition called need for sex was seen as greater than “kigune” had existed for a long time that of women. Men were generally and has symptoms similar to AIDS. expected to have sex outside mar- This condition results from not fulfill- riage — sex being seen as a pleasure ing ones obligations to elders or the and a way of satisfying their desires. clan ... This disease we have here is As the father of a man with HIV in called “ngoto”, when it affects children Mumbai put it:

73 CHAPTER 4

‘It is possible that he would have home. Men on the other hand were done something like that (visited sex able, and indeed expected, to travel workers). Who wouldn’t have done it greater distances in order to provide at that age? ... this is a thing (that) no financially for the household — one can resist. Who is such a perfect migrating from villages to urban person?’ (I, p.29). areas, and from one urban area to A similar theme was highlighted in another, in search of work. But with the report from Chiang Mai many households being in a state of province, permanent economic need, and with ‘The community’s socialization of traditional work patterns changing, gender inequality and values ... these geographic and employment results in villagers believing that patterns were in transition. To pro- males should have sexual experience vide for family members, women before marriage. By engaging in pre- were increasingly involved in trading marital sex, neither his reputation, (such as selling commodities at mar- nor his virginity is affected ... For kets) or in various service industries sons, norms are inculcated less rigid- (including bar work, entertainment ly than daughters. Sons, for example, and sex work). are normally free to go out at night, which often involves drinking or even visiting sex workers, whilst daughters Beliefs about transmission were are not.’ (Th, p.69). linked to beliefs about those people most likely to be infected. Generally Women, on the other hand, were three kinds of people were identified perceived as needing to confine as particularly likely to be infected: their sexual activity to marriage, and women who earned money through their status was dependent on com- selling sex, men and women who had plying with such prescriptions. A sex outside of marriage, and ‘homo- woman’s reputation, her physical sexuals’. In many cases such images safety and her access to a home and seemed to have been reinforced by shelter could depend on her faithful- national and local HIV prevention ness to her husband. In campaigns. In Chiang Mai province, Netzahualcóyotl, for example, for example, it was reported that respondents stated that women attempts had been made to educate should be obedient to men. If not, female sex workers about the risks they should expect abuse and associated with AIDS, almost to the recrimination — which could be sex- exclusion of interventions with other ually as well as physically violent. groups. Because of this, sex workers While in all sites women were report- were said to be very aware of the risk ed as having extramarital sex, includ- of AIDS and encouraged clients to ing occasional sex with younger use condoms. In Santo Domingo, it unmarried men, such women tended was said that most sex workers to be in a minority. abstained from unprotected sex, and sought to negotiate condom use with These views about men and women clients. impacted upon their freedom of movement. Women’s movement was Men, it appeared, seemed almost traditionally restricted to undertaking compelled to have sex outside mar- tasks in and immediately around the riage. In Mumbai, for example, men

74 CHAPTER 4 were seen as having a ‘natural’ need mistic, however, suggesting men for sex. This, along with a lack of pri- could change through self-control vacy in which to have sex with their and indeed had changed by having wife and the wife’s possible exhaus- fewer extramarital affairs and, when tion after carrying out her routine doing so, using condoms. chores, meant that men used sex workers as a source of ‘relief’. As Women’s ability to negotiate con- one married woman put it, dom use, particularly with husbands, was particularly problematic. As ‘Lovemaking is for a husband’s plea- already indicated, women were gen- sure. It’s a duty for his happiness.’(I, erally expected to have sex in accor- p.49). dance with their husband’s wishes. There was evidence, however, that In Mumbai, although preferring their men were beginning to take notice of husbands not to do so, wives felt safer sex messages, at least in the that if a husband were to visit a sex context of sex work. In Chiang Mai worker he should be responsible province and in Santo Domingo, for and use condoms. Men in this same example, they were reported as more site, however, reported feeling that likely to use condoms when having it was the sex worker’s responsibility sex with sex workers than with other to ensure that a condom was used. partners, unless they had a particular- Moreover, the belief that condoms ly close relationship with the sex work- were only for use outside marriage er when the use of protection dimin- was reported as having led to hus- ished. Where the emphasis in public bands infecting their wives. information campaigns had been on ‘My husband knew of his condition the risks associated with sex with sex and how it affects others. Even then workers, men reported routinely using he put me into danger. Every one of condoms with sex workers, but rarely us is surprised that he was educated with other extramarital partners. and yet he didn’t tell. He was asking me to take out the Copper T [an ‘ ... evidence from focus group dis- IUD], he said it hurt him. He wanted cussions revealed that condom use another child!’ (I, p.36). was practised by men when visiting sex workers (albeit not all the time), Respondents in Santo Domingo but not with (other) women whom stated that wives should be entitled they believed did not have HIV/AIDS. to protect themselves if their hus- This irregular and selective use often bands had sex with other women. has led to false confidence.’ (Th, p.78). Nonetheless, this remained problemat- Among respondents from families ic since while women might be expect- affected by AIDS in Santo Domingo, ed to demand condom use of men, there was pessimism about the they were reported as being powerless extent to which men would change to do so — with their fate (and health) their sexual behaviour — even being entrusted to God instead. regarding condom use. ‘Women married to (unfaithful men) felt Respondents from families unaffect- powerless to avoid infection and put their ed by AIDS were rather more opti- fate in God’s hands.’(DR, p.14).

75 CHAPTER 4

Similar concerns were expressed by sex workers in Santo Domingo, for respondents in Chiang Mai province example, reported predestination as who felt that not enough attention one of the reasons why people might had been given to the negotiation of get AIDS. condom use by women. Respondents at this same site also noted that a wife ‘A few sex workers in (the low impact was more able to negotiate condom area) did not know that sperm was use within marriage if she was educat- infective, ignored the length of the ed about the need for protection and incubation period, and attributed infec- had money of her own with which to tion to predestination or bad luck.’ (DR, p.12). engender a degree of bargaining power. Access to information about Similarly, a number of respondents in HIV and AIDS was also seen as a pre- Netzahualcóyotl felt that AIDS had a requisite for women’s empowerment. supernatural cause. Whether or not a person got AIDS was said to be in ‘ ...women who can convince their hus- God’s hands, and could be a form of bands or sexual partners to take blood divine punishment. In Mumbai (and in tests or use condoms are those who either have better socioeconomic back- line with eastern philosophies of life it grounds, are better educated, or have was said), those with ‘weak minds’ had some training and access to infor- were believed most likely to fear mation about HIV/AIDS...’ (Th, p.28). becoming infected by AIDS. In Kyela district, those who had not fulfilled The influence of earning power on their obligations to elders were seen changing relationships between men as especially vulnerable. and women was noted in respondents’ accounts in the Kyela district of the In Chiang Mai province, respondents United Republic of Tanzania. Money suggested there was less of a need to was seen as affording women greater use condoms when partners independence, and this needed to be appeared ‘clean’ and ‘healthy’. taken into account when developing prevention work. ‘Only in the past four decades have rural Thai people become familiar with ‘ ...amongst women, strong links “germs” (as causes of disease) in addi- (were) made between increased eco- tion to the belief in former causes such nomic power of women and increased as spirits or imbalance of body con- social freedom. If one is to think of stituents of the body and the mind. In (HIV) risk reduction strategies in this the case of HIV/AIDS, many men said population, it is clear that much cau- they did not use condoms with sex tion will be needed for such workers or with non-sex workers, if, in approaches not to impact negatively their judgement, their partners on the income generating activities appeared clean and healthy. In other women perform outside the home.’ words, unless the germs of viruses (Ta, p.80). manifested themselves through their partner’s physical abnormality or ill- Other beliefs ness, men felt they were having safe sex.’ (Th, p.54). In Netzahualcóyotl and Santo Domingo, AIDS was said to be caused ‘Knowing’ (or not knowing) a potential by fate or the wishes of a deity. Some sexual partner helped people make

76 CHAPTER 4 judgements, albeit not necessarily of media messages. accurately, about risk. In Chiang Mai province, it was reported that safer ‘Advertisements say that it is not a con- sex was equated with ‘knowing’ your tagious disease then why do doctors partner. There was less of a need to and nurses not touch them...?’ use a condom when having sex with ‘If it is not a contagious disease then someone in your own village than with why was ... (somebody’s) wife asked to a stranger. Similarly, and according to wear gloves and mask when entering gay community respondents in his room?’ (I, p.52-53). Mexico City, men had unprotected sex because they cared for each other Connections were often made and did not want to have a condom between other serious illnesses and ‘in the way’. concerns about HIV and AIDS. For example, community interviews in ‘ ... if love is real, it isn’t needed (a con- Mumbai revealed that links were dom), no matter what happens. If (the made with local beliefs about tuber- other) lives with the virus, maybe I’ll culosis, and in Santo Domingo local say: “I live with the virus, maybe I’m beliefs about the transmission of going to die, but... well...” ’ (M, p.171). tuberculosis and syphilis were generalized to AIDS. Fears about transmission through everyday contact continued to affect Reference was also made to people people’s responses. In Chiang Mai with AIDS who were seen as the province, for example, workers on potential ‘agents’ of transmission. building sites were said to be so con- Among respondents in the lower cerned about infection being trans- impact area in Santo Domingo, for mitted from one person to another example, it was believed that some that they used their own drinking infected people did not develop cups. Among respondents in AIDS. Instead, they became carriers Mumbai, a majority felt that AIDS and it was only the ‘victims’ with was not spread by physical touch, yet whom they had sex who got AIDS. some were still concerned enough to This image of infected people as suggest that mosquito bites, AIDS active ‘agents’ of transmission could microorganisms in the air, talking to also be found in respondents’ infected persons, and gases created accounts in Mumbai where it was felt through the mixing of urine in public that sex workers could pass on HIV urinals could spread infection. While without themselves becoming ill. As it was said that education campaigns one man explained: had helped allay these fears, confu- ‘ ... it does not spread so easily among sions and anxieties still persisted. In women, it stays like that only ... like you Mumbai, for example, respondents said the prostitute is like an agent ... noted a tension between the mes- she remains an agent, she does not sages in mass media campaigns have the trouble, but ... because of her, (such as AIDS is not contagious) and those who go to her get the trouble.’ (I, the behaviour of medical profession- p.51). als who did not want to touch people with AIDS. Such contradictions led Husbands were often implicated in respondents to question the veracity accounts of people with HIV knowing-

77 CHAPTER 4 ly having sex with others. While in become infected by casual contact.’ Mumbai, as in other sites, most hus- (DR, p.12). bands with HIV were reported as being too unwell to have sex with Perhaps in consequence, children at their spouses, one particular husband this same site were usually segregat- was described who ‘knew of his con- ed from adult household members dition’ and who still had unprotected with AIDS. Here, as in Mumbai, sex with his wife. Respondents in beliefs about young people’s special Netzahualcóyotl suggested that vulnerability were more strongly felt women were expected to have sex when the person with AIDS was with their husbands regardless of symptomatic. whether the latter was infected. A good wife had to comply with her hus- ‘He coughs a lot and keeps spitting band’s wishes. here and there. I hate all this. For children things should be OK, right? ‘ ... women find it difficult or impossi- ... people say that you should not ble to resist having sex with their HIV touch him, give him injections or come positive husbands: their subordinate in contact with his blood. condition forces them to keep from And I feel that my problem, (HIV showing their fear of being infected by infection) which is “little” now, would them, and to suppress the suspicion aggravate all of this. That hanky of that they may even try to infect them his, towel, I don’t let the children use willingly (deliberately).’ them, and keep it separate ... then his plate is also kept separate.’ (I, p.35). ‘He used to tell me: “C’mon, I’m not The real and imagined risks facing going to infect you!” But he, I think he young people were also of concern did it, I don’t know, that fact that he to respondents. Interviewees in Kyela took the condom off to ... he maybe did it because he didn’t want me to remar- district observed that parental roles ry or if he died and then I was left by in relation to young people were myself, well, that I couldn’t go with undergoing change. Rules about another man ...’ (M, p.113) young people’s behaviour were said to be more lax owing to changes in HIV transmission: children and schooling and because young peo- young people ple’s potential to earn money was seen as providing them with a In all five countries there were greater degree of freedom. In both strongly held local beliefs about the Kyela and Chiang Mai districts, an ways in which children could get increase in leisure time was said to AIDS. In Santo Domingo, for exam- put young people at special risk ple, where there was little mention of because of associated increases in prenatal and perinatal transmission, alcohol consumption and sexual it was suggested that ‘weaker’ chil- activity. dren might become infected through ‘Related to alcohol use was the problem ‘casual’ or everyday contact. of promiscuity which appeared to be par- ‘No participant referred to mother to ticularly common amongst youth, leaving child transfer of the virus as a mode of them vulnerable to not only premarital transmission. Nevertheless, they pregnancy, but also sexually transmitted believed that “weaker” children could diseases including HIV/AIDS.’ (Ta, p.84).

78 CHAPTER 4

Household responses The way in which test results were provided by medical professionals was Becoming aware of AIDS much criticized. As this was often the first contact with medical services, it The perceived culpability of individu- would seem important that these situ- als was an important determinant of ations are handled well. Gay commu- household and community responses. nity respondents in Mexico City While in all sites a distinction was reported that test results were provid- drawn between those who were felt to ed with a lack of sensitivity and sup- be ‘innocent’ and those who were port. Other Mexican respondents and seen as ‘guilty’ for their illness, stigma some interviewees in Mumbai stated was generally attached to individuals that positive test results were dis- and households with AIDS. Even closed to family members without the when it was recognized that AIDS had individual’s consent. been acquired through no fault of ‘You see, they ordered me to take the people’s own, and faithful wives ELISA and unfortunately in the labora- infected by their husbands were often tory, when they delivered the paper to described in this way, the latter could my wife, it wasn’t sealed, the envelope be blamed for their husband’s unfaith- was just like that, and my wife thought fulness in the first place (through not if it was easy to open it, and since my satisfying them sexually), or by failing sister-in-law is a nurse, she went (to to persuade their husband to use a see her) and asked her, and there it condom during sex. was, they were both crying ... Then she took it to the doctor. My oldest daugh- The news that someone had AIDS, or ter goes in and the doctor, without any had been diagnosed as being HIV preparation or anything tells them: seropositive, generally brought great “He has, he has the disease”. Then my shame. This shame was related to the wife went out crying and (my son) saw stigma associated with sexual wrong- her and asked her what was the matter, doing as well as likely death. Beliefs and my wife lacking judgement told about sexual transgression had an him ... This is how I found out, I who impact on responses to the disclosure am the patient, the day the doctor told of HIV status. me I felt everything sank around me. Now imagine my family.’ (M, p.101). Relatively little was said in reports about why people would seek an HIV Once a test result had been received, antibody test except in Mumbai decisions had to be made about where recurrent illness, mandatory whom if anyone would be told. testing for work in the Gulf countries Respondents from Mumbai indicated and voluntary blood donations were that within households, disclosure cited as reasons, and in Mexico City, tended to be selective. Where certain where gay community respondents household members were seen as suggested that for gay men, the outsiders, as happened with daugh- death of a close friend or ex-lover, ters-in-law in that city, such members persuasion by friends, the influence of might not be informed. How house- activist groups, or a requirement by an hold members felt about the news of employer, were all factors leading an an HIV diagnosis related to how the individual to take an HIV test. infection was believed to have been

79 CHAPTER 4 acquired. High levels of shame and was the care and support provided for embarrassment were said to be felt by and by household members. The respondents in Mumbai where disclo- kinds of support provided included sure of an HIV positive serostatus was physical care, emotional support and tantamount to admitting to having information (such as news about treat- had premarital or extramarital sex. ments), as well as accompanying people with HIV to medical centres where Gay community respondents in these existed, and when medical care Mexico City reported that family could be afforded. responses to HIV were similar to those that took place when it was found out From respondents’ accounts in that a son was gay. If this latter kind of Mumbai, it was clear that households news had triggered a negative responded fairly positively to a family response then it was likely that the for- member with HIV, seeking to provide mer would do so too. Given high lev- physical, emotional and medical care els of stigmatization towards homo- when needed. In Chiang Mai sexuality and AIDS, there often province, families were reported as occurred a joint ‘revelation’ of HIV sta- generally accepting of caring for a tus and of homosexuality. In these sit- family member with HIV, even though uations, family members were said to the quality of the care offered was be particularly distressed. said to vary greatly. Responses to family members with AIDS in In every site there were reported Netzachualcóyotl were part of a more instances of people with HIV trying to general set of responses to ongoing conceal their serostatus. In Mumbai, crises and became stronger when this was said to be easier for unmar- symptoms appeared. ried men. Because their behaviour ‘[At first], there is the difficulty of rec- was less closely monitored than that ognizing that one is HIV positive ... of women, hospital visits could more This is due, partially, to the asympto- easily go undetected. In study villages matic characteristic of HIV infection at in Chiang Mai province, it was report- its early stages; and, on the other hand, ed that people with HIV tried to con- to the tendency that inhabitants of ceal their serostatus until symptoms Netza have of normalizing all the prob- showed. lems they face’. (M, p.101). ‘Some PLWH/A in both the high preva- However, when the disease starts to be lence community and low prevalence visible ... the disease makes itself bru- community concealed their HIV status tally present ... ’. (M, p.100). until the opportunistic diseases deterio- ‘ ... Many homes see that one of their rated their health and were observable members is affected by HIV, and only by family members ... Stigma largely when the disease aggravates does the accounted for such denial and conceal- issue then become of great interest to ment’. (Th, p.40) the family.’ (M, p.213).

Family support In Santo Domingo, data suggested that families in the higher impact Despite widely held negative percep- areas were the most positive in their tions of people with HIV, what was responses. In the Kyela district of the striking across all five study contexts United Republic of Tanzania, care was

80 CHAPTER 4 provided to those with chronic illness, marriage, were reprimanded, or with such provision being more often worse, for doing so. given in recent years by household rather than clan members. This was ‘But women were not as fortunate. The said to be due to households, rather sexual mode of infection in the case of than the clan, having become the a 17-year-old unmarried girl invited focus of consumption and production. the severest criticism from the father who has ever since severed his rela- Women were central to the provision tionship from the family.’ (I, p.29). of care, and husbands and sons generally expected to be cared for by In both sites in Mexico and in Santo their wife or mother. Even among gay Domingo, the quantity and quality of community respondents in Mexico care was influenced by perceptions of City, where support was also provided innocence and guilt. Respondents in by social networks, gay men with HIV Neztahualcóyotl suggested that if HIV not infrequently returned to their was seen as having been acquired by mother’s home for care. Respondents ‘accident’ (the acquisition of HIV by a from the non-gay community sample ‘faithful’ wife from her husband was in Mexico also suggested that when construed in this way), household people were diagnosed as HIV members were likely to respond with seropositive, they tended to return greater solidarity than if the person home. Although care was often pro- with HIV was seen as culpable. Men vided by kinship networks, it was usu- who were seen as blameworthy for ally women within these networks who their illness were nonetheless cared provided the bulk of the support. for, a response which women in this same site felt to be unfair. It was generally perceived as more shameful for a woman to have While women were customarily acquired HIV than for a man. In expected to care, and more often Mumbai, men were often not asked than not did so ‘with love’ for their how they had been infected, it being husbands and sons, there were cir- assumed that transmission had been cumstances where they cared dutiful- through sex, and that sex outside mar- ly while feeling trapped within the riage was ‘natural’ for men. This ‘non- relationship. This could happen, for probing’ helped to sustain the gener- example, when a relationship ally supportive atmosphere within between a wife and husband had which care took place. already broken down, and where the husband had subsequently become ‘In most households the infected male infected. member was protected from shame and embarrassment in answering questions about mode of infection. Generally no Women were reported as receiving questions were asked, or if asked, little support in providing care and answers were not pressed for’. (I, hesitated to ask for help, this being p.28). seen as likely to annoy those they were helping. When women them- For women, things were very differ- selves needed HIV-related care, they ent. They were often questioned and, were generally unable to expect, and after perhaps admitting to sex outside often did not get, the same level of

81 CHAPTER 4 support as men. Moreover, their own to the person with HIV. Nonetheless, psychosocial needs tended not to be other respondents’ accounts suggest- addressed when providing care for ed that care by men was often limited, others. since it went against generally held notions of what men should do. ‘The needs of PLWHIV/AIDS differed along gender lines ... Men’s needs ‘There are indications that perhaps largely centred around the “self” - the male heads of households would wish problems in sharing HIV status with to do more when their partners fall ill others and their health, emotional, and but are curtailed by cultural definitions financial situation and counselling of maleness and the roles defined which needs; women’s needs extended determine masculinity’. (Ta, p.68). beyond the “self” to include the “family” - namely, needs of the spouse, wel- Income and household care fare of children, shelter, economic support and so on. Their own health needs In all five studies, financial resources and emotional needs were low on [the] within the household influenced the priority list.’ (I, p.46). level and quality of care provided. Expenses associated with caring Care for women was often lacking included payment for travel to and within the household of marriage. from hospital, purchase of medicines Unless other women in the communi- and other treatments, special diets, ty were able to provide care, a and time away from paid work (for the woman might return to her parents’ person affected as well as those car- home to receive additional support if ing for him or her). In poorer house- ties with her own biological family holds, adults and children suffered remained. However, women were most. also reported as having to put up with stigmatization and even abuse in In the study villages in Chiang Mai order to access care for themselves. province, for example, distress caused by the lack of money to pay ‘HIV positive women whose husbands for AIDS-related care, and not that have already died because of AIDS, caused by stigma or physical symp- besides having to solve by themselves toms associated with the disease, was financial and emotional problems ... said to have led to the few known also have to struggle with the image of instances of suicide. In Mumbai also, the “single” (unaccompanied) woman, it was recognized that the ‘moods’ of that for many men is equivalent to family members were exacerbated by ‘available’: worries about money. ‘... I don’t like him (my brother-in-law) to fondle me around, because he is not Households hardest hit were those even taking charge of me (financially), which were already poor and those and beats me.’ (M, p.116-117). where the main earner became unable to work due to AIDS-related Although women were the most fre- illnesses. In Mumbai, such house- quent providers of care, men were holds were said to need to borrow also reported to offer some forms of from friends and take out loans in support (often of a less personal kind), order to pay for expenses. In Santo especially if they were closely related Domingo, very poor families in the

82 CHAPTER 4 higher impact area were reported as Illness progression relying on money provided by relatives living abroad. When a wage The progression of illness, and in par- earner was affected, other family ticular the visibility of symptoms, members not only had to provide affected household responses in all care and support, but also had to find sites. Visible signs of illness were also ways to bring in extra income. Both reported as influencing community spouses and children took on this members’ labelling of disease, the responsibility. care provided, and perceptions of the ‘When the ill [person] is the main likelihood of HIV transmission. Where household provider, the spouse and people with HIV had tried to hide their children usually have to do some diagnosis, visible symptoms could menial job to help generate an income force disclosure to other family mem- for regular and extraordinary expens- bers. Respondents in Santo Domingo es. Several affected families had indicated that relatives would be invested all their resources in the dis- asked for help only when symptoms ease.’ (DR, p.23). became impossible to conceal, and that people with HIV were sometimes Care for children was severely com- offered little support until they were in promised by lack of money in affect- the terminal stages of illness. ed households. Children might not Otherwise, people with HIV remained have enough food to eat and their largely ‘hidden’. In Mumbai, it was education might suffer through the only when symptoms became ‘very need to work to earn money for the visible’ that people with HIV and AIDS household. Women, as wives and were encouraged to go to hospital. mothers, could again find themselves providing for others, doing what they Caring for people with HIV who had could to look after both their husband visible symptoms appeared to be more and their children. difficult than caring for those who were asymptomatic. As one respondent in ‘Faced with limited resources and mul- Netzahualcóyotl explained, tiple demands, households exercised choice in spending money. A substan- ‘He was all bones, believe me, it was hor- tial amount of money was spent on the rible bathing someone like that. It was infected person’s special foods and like crying, as if you wanted to scream nutrition and if symptomatic, on med- out because of what you feel, of what you ical care ... The worst affected were the see; then, I used to take a drink or two in children with regard to their nutrition, order to have the courage to take him, education and psychological well- undress him, bathe him and shave him’. being ... In some households, there was (M, p.84-85). sometimes no food and children went hungry. Their nutrition and health People often believed that HIV was were a major source of worry to their more infectious when symptoms were mothers ... In sum, it was the economy visible, and were especially worried if of the lower income, single earner there were also children in the house- households which was most severely hold. In Chiang Mai province, respon- affected and the women bore the major dents suggested that condoms were burden of the decline in economic more likely to be used when one of the standard.’ (I, p.45). partners had visible signs of illness.

83 CHAPTER 4

Community responses like this (do not touch the patient) then what will they do in Gujerat?’ (I, p.28). Being diagnosed as having HIV or AIDS had important consequences for In Kyela district, relatives were some- community responses of support and times said to help out by moving into rejection. Four sets of people were the affected household. This kind of identified as providing care for people support was especially beneficial dur- with HIV and AIDS: visiting relatives, ing periods of lengthy illness when community members known to the assistance from other community household (sometimes talked of as members might be less forthcoming. ‘neighbours’), community members not known to household members, Neighbours and other community and volunteers (also described as members ‘activists’ by gay community respondents in Mexico City). Other than in HIV-related stigma not only affected the case of gay men in Mexico City, it relationships within the home but also was mainly women who provided relationships between households. In community support. all sites, households were said to be under a degree of surveillance by neighbours, and special interest was Relatives taken whenever a person became ill. Such interest might turn to suspicion In Santo Domingo, where households about AIDS when it was believed that prided themselves on their ability to someone in the household might be at cope, relatives tended only to ‘high risk’. In Mumbai also, particular become involved at the later stages of symptoms, the manner in which bodies a person’s illness. In Mumbai, relatives were disposed of, and (in a somewhat were often not as fully involved in all-encompassing manner) long-term AIDS-related care as might have been illness as well as quick deaths, could the case with other illnesses, there cause community members to think being a tendency for family members that AIDS had affected a household. to keep an AIDS diagnosis secret wherever possible. Less positively, and in those areas of ‘Relatives who are normally sources of Santo Domingo where AIDS had had support during crises in India, are greatest impact, there were occasion- deliberately kept out of the scene in the al calls for people with AIDS to be HIV epidemic because of the fear that stamped, marked, or in some way they may tarnish the HIV person’s publicly identified. character, or isolate the family’. (I, p.28). ‘In the higher impact area ... some members of unaffected families pro- As one mother of a person with HIV posed sexual abstinence, public identi- feared: fication, stamping or marking, isolat- ing, or putting HIV positive individuals ‘I don’t go to my sister’s house because in jail ... Strong fear of infection per- there are many doctors in her family. sists in the community.’ (DR, p.16). I’m constantly worried that they may guess (about my son). If the doctors in The quality of care provided by neigh- Mumbai, after being aware, treat you bours and other community members

84 CHAPTER 4 varied across study sites. In Kyela dis- where households successfully man- trict, community members were aged the illness of a person with HIV reported to be the least equivocal so that he or she was not seen by about providing care: visits to affected community members to pose a threat, households were made, material then neighbours were on the whole assistance offered and help provided able to respond positively. when visits to hospital were needed. Volunteers ‘Families with sick members were visited by members of the community not Volunteer responses and the role of only to provide emotional support, and NGOs were highlighted to varying to contribute views on action to be degrees in country reports. In Kyela taken, but also to provide material district and Mumbai, NGOs barely fig- assistance ... When a household could ured in local responses and had little not afford to take the patient to hospi- visibility. The NGO response in Santo tal ... members of the community either physically carried the person to the Domingo was said to be related to the chosen health facility, or contributed impact of AIDS. Where the impact towards both the fare and sometimes had been lower, and where poverty the medical bills’. (Ta, p.35). (and its alleviation) was household members’ main concern, few NGOs In Chiang Mai province on the other were involved in the provision of hand, neighbours and other commu- AIDS-related care. In the higher nity members were said to be unsure impact area, however, a well-orga- about visiting affected households, nized network of NGO and grass roots worried that they might be unwel- support was in place, albeit one which come, or fearful about transmission. In was said not to be responsive to the Netzahualcóyotl, there was said to be needs of younger people. little support from neighbours, although this was linked to the fact Lack of resources was said to affect the that because of their recent arrival in effectiveness of AIDS-related NGOs in this newly developing district, com- Netzahualcóyotl and the Village munity members tended not to know Health Volunteers in Chiang Mai one another. Responses in Santo province. In both cases, volunteers Domingo were said to differ accord- had become unwelcome visitors to ing the impact that AIDS had had. In affected households: since they higher impact areas, and despite brought no money, they were not seen occasional demands that people with as being able to offer useful support. HIV should be publicly identified, ‘The sick household also treats the community members visited and pro- house visits by Village Health vided support to affected households. Volunteers with suspicion. These visits In lower impact areas, people with are seen to be without purpose or AIDS were said to be pitied and treat- desire to alleviate the problems house- ed as pariahs, being offered little holds face. “They are just stopping neighbourly support. Greater local by”, said a member of one affected impact, therefore, was said to have household. However, the health volun- created a more positive overall teers said they were aware of the fami- response. From respondents’ ly’s uncertainty over their intentions. In accounts in Mumbai, it appeared that addition, said the volunteers, not only

85 CHAPTER 4

were they not given any rewards or to Marcelino. She replied “No, what do incentives to make visits, they were you think? It’s one thing to work with expected to bring a gift paid for by this (AIDS), and another to distribute themselves to show their sympathy. For condoms” ... and so she didn’t’. (M, other duties such as checking dogs for p.132). rabies vaccinations, the health volunteers were given a small reward’. (Th, A further barrier to the development p.57). of community-based initiatives was highlighted in Chiang Mai province. Lack of understanding was also said to Where there was an understanding constrain the effectiveness of commu- amongst community members that nity-based organizations in professionals had expertise in relation Netzahualcóyotl and in Chiang Mai to HIV and AIDS, and where there province. NGOs in the Thai sites were were accompanying beliefs that care said to have failed to involve people should be provided in a ‘top down’ with HIV, limiting the impact of their way, there tended to be a lack of com- work and its effectiveness. This lack of munity initiative. This left the provi- involvement was said to be related to sion of care to households alone with- NGO’s requirement that people dis- out significant support from wider close their HIV status. Integrating HIV community members. and AIDS-related care with more gen- Care by friends and community eral kinds of community support (such members as the Village Health Volunteer System, the Health Card and the Aspects of community-related care Funeral Associations) was felt to hold among respondents in Mexico City’s the potential to contribute to more gay community differed from those sustainable forms of community care. already noted. While support by household members (particularly In Netzahualcóyotl, however, where mothers) remained important, gay support was provided by non-AIDS men’s social networks also had an specific organizations (such as volun- important role to play in providing teers from religious organizations and care. This was particularly true for Alcoholics Anonymous), questions those gay men who no longer lived were raised about the quality of care near to their biological families. While provided. Attempts to apply an social networks varied in their size and understanding of other health issues complexity, they usually included part- to HIV and AIDS had acted as a barri- ners, friends and volunteers from er to the provision of effective care. ‘activist organizations’. While commu- Sisters from a Roman Catholic order, nity support for a person with AIDS for example, were reported as having was influenced by broader under- found it inappropriate to provide con- standings of gay solidarity, not all doms. friends provided care and a few ‘ ... what happens is that any Sister will became emotionally and physically say that homosexuality is unnatural more distant. and the Bible [talks about] Sodom and Gomorrah ... they have already worked There was reported to be a ‘dynamic it out for me ... For example, one day I relationship’ between partners, gay asked a sister to take a box of condoms social networks, and family members

86 CHAPTER 4 involved in care. Partners, in particular, linked to the costs of providing food not infrequently found themselves and drink at an increasing number of negotiating over the type of care pro- such events. Despite a strong desire vided when it was offered by both bio- by community members to attend logical family members and them- funerals and to support bereaved selves. The kind of care provided to a households, traditional mourning gay person with HIV depended on the practices involving the slaughter of numbers of people involved in differ- cattle and the consequent feeding of ent social networks, their levels of mourners were now less common due altruism, their access to material to the cost and the belief that they resources and their general awareness might exacerbate poverty. Con- of health service provision. While sequently, a part rather than the friends rather than family members whole of a village tended to be more often provided care at times of involved in funeral and mourning crisis, there were limits to such provi- rituals. sion and both ‘stress’ and ‘burn-out’ ‘...despite the fact that the whole vil- were said to limit people’s ability to lage is required to participate in sup- provide care. port during the mourning period, in reality, much of the material support is The household and the offered by villagers living within the community hamlet, rather then the whole village as was the case in the past. Obviously as the death toll increases, which is Death and mourning likely going to be the case in a few years time, even the duration of stay- In all sites, funerals were seen as ing with the bereaved family and the occasions when community members nature of other support currently pro- could offer support to the members vided would drastically change. In one of bereaved households. Never- village an elder noted: “We had just theless, the presence of large num- come back from burying one person, bers of community members at such when we were told that someone else events could sometimes signal curios- had died ... and the body had just ity about, rather than solidarity with, arrived in the village. It was already households affected by AIDS. dusk, and we were tired. We told the Alternatively, as was noted in Chiang household members not to start Mai province, funerals could be an mourning until the morning as we were opportunity for drinking and gam- too tired to respond.”’ (Ta, p.44) bling. In the two villages in Chiang Mai In Santo Domingo, Kyela district and province, funeral ceremonies had Chiang Mai province, it was reported changed to accommodate local fears that fewer community members of infection. expressed support to the members of bereaved households than before, The economic impact of AIDS on especially in places where the impact households was often greatest on the of AIDS had been greatest. Although death of the head of a household. Due in Santo Domingo this was said to be to the high cost of medical treatment, related to fears of infection at funeral there might be particular worries when ceremonies, in Kyela district it was there were unpaid medical bills.

87 CHAPTER 4

Following death, the economic impact ticularly children. In Kyela district, of AIDS on households could lead both there were reported to be clan rules younger and older household mem- to guide the ways in which bereaved bers (who may not otherwise have children should be looked after. done so) to seek work. While there Respondents, in both higher and were opportunities for carers to re- lower impact areas in Santo enter the labour force when the person Domingo, also noted that where being cared for died, such households immediate family members could not were likely to be comparatively worse look after them, children tended to off than those unaffected by AIDS, as be cared for by extended family newly bereaved households were more members. Children’s care by extend- likely to be headed by women whose ed family members depended to a employment position was less strong. great extent on the ability of households to afford such support. ‘Since the majority of early HIV/AIDS Children from one family, for exam- deceased were male this situation has ple, might be split up and looked increased the number of female-headed after by different relatives. households and single-parent families. This has, in turn, increased the burden In Kyela district, the cost of childcare on the communities’ poor. To a substan- influenced with whom children were tial extent, the AIDS epidemic has placed. With adequate resources and brought about the “feminization of poverty” since these female headed where HIV had not had a high level of households usually have few skills and impact, children were usually looked earn much less than their ex-husbands’. after within extended families. Where (Th, p.58-59). this was not possible due to financial difficulties, clan members would pro- The death of a household member vide care. Although clan members could also lead to physical abuse. In were required to look after children Mumbai, for example, respondents belonging to their clan, and desired suggested that in better-off families, a so to do, lack of resources could bereaved wife might be physically sup- make this problematic. In Kyela dis- ported, but in lower-income families a trict, more general poverty allevia- woman might find her access to shelter tion programmes rather than HIV and threatened. She might be required to AIDS ‘orphan’ projects were believed leave her in-laws’ home which was pre- to be better able to address the care viously shared with her husband, and and social support needs of children might have the added disadvantage of who had lost one or both parents. being unable to return to her own family if they never fully approved of her ‘ ... it may be more difficult for the tradi- marriage. Added to this was the possi- tional coping strategies which allow for household reconstitution, to occur, given bility of sexual abuse when bereaved the sociocultural and economic changes wives were seen as being newly sexu- in the study area. However, in planning ally available. interventions which will assist affected and afflicted households to cope with the Care of bereaved children impact of HIV/AIDS, it is deemed wiser to find ways and means which will aug- When women are bereaved, they still ment traditional coping strategies, have to provide care for others, par- rather than creating new structures

88 CHAPTER 4

which may be difficult to sustain beyond Accessing what was perceived as foreign aid. This is particularly the case ‘good’ medical treatment depended for orphans, where there exists a rich on the ability to pay. Public health ser- cultural system to provide for support vices were often viewed more nega- and care of orphans. Attempts at inter- tively than those provided privately, ventions ... aimed at poverty alleviation, not only in terms of the treatment will be in the long run much more sus- available, but also in terms of accessi- tainable than focused attention on AIDS orphans.’ (Ta, p.79-80). bility. The former might be located at some distance from where they were Treatment and health care needed, drugs might only be available at the beginning of a month, Across all study sites, respondents people were not told what to expect noted the high costs of accessing of treatments, there might be little or medical care, both in relation to the no coordination between services, time needed to travel to services people with HIV were treated badly, (which might otherwise be spent earn- public health professionals had little ing income) and the financial costs of knowledge of HIV treatments or were services and drugs. Respondents in thought of as too ‘authoritarian’, and Kyela district, for example, noted that people with HIV using public health medical care was no longer free, and services were felt to die faster than so could not always be afforded. those using private services. In rela- Respondents in Chiang Mai province tion to this last point, the fact that indicated that the long waits involved people with HIV often accessed ser- in accessing treatment meant that vices at a late stage meant that they people lost income by having to take sometimes entered hospital particu- time off work. In circumstances where larly unwell. Their subsequent death people were paid on a daily basis and not infrequently contributed to local had little or no job security, this was ‘rumours’ that a hospital was no good. especially difficult. There were, however, positive ‘Poor households ... generally ... sub- accounts of public hospitals, particu- sist on day-to-day wage employment larly where drugs were available free (and are) not covered by labour laws or of charge. But, as noted in the report social security regulations. Because (these) people have neither job security from Mexico, respondents generally nor social welfare benefits, they have to had mixed experiences of health often choose between recovery and care, often reinforcing the negative going hungry ... In contrast, (for) mem- experiences they had encountered bers of better-off households ... their when first diagnosed. job benefits not only allow them to take ‘The experience of receiving AIDS sick leave, but also gives them access to related health care at the health clinics proper health care’. (Th, p.35-36). can range from being a comforting experience to a completely negative People with HIV from higher-income one. Some informants reported solidar- households on the other hand were ity behaviour in the institutions and said to be able to spend their time ‘in (with) the doctors, who provided them and out of’ various hospitals and clin- with some drugs for free, or who at ics, an option not available to those least organized events that made them from poorer backgrounds. feel better. But, most of the time, the

89 CHAPTER 4

perspective is rather bleak. If the initial Traditional and complementary impact (HIV notification) ... is not an therapies affirmative one ... this provokes (people with HIV) to distance themselves from The use of traditional and complemen- health institutions’. (M, p.117-118). tary therapies was reported by respondents in Mumbai, Kyela district and Respondents in Mumbai noted that Chiang Mai province. Such treatments hospital visits could support the care were often used in an attempt to exert given by household members by pro- some control over health care by offer- viding the carer with information ing hope amid treatment uncertainty. about appropriate treatments, food Although media campaigns in Thailand and nutrition, as well as affording them had stressed there was no cure for the opportunity to meet with counsel- AIDS, higher and lower income respon- lors. Although not often available, pro- dents nonetheless sought treatment fessional support was said to be help- from traditional healers. While it was ful in coping with the emotional harms known that traditional and complemen- related to HIV such as having to put up tary treatments did not cure AIDS, they with abuse and stigmatization. were seen as being able to alleviate certain symptoms. Although the often limited resources available to affected household mem- Respondents in Mumbai reported that bers were one reason why public complementary medicines were more health services might not be used, likely to be used when people with HIV the stigma associated with HIV also were asymptomatic, allopathic treat- impacted on people’s willingness to ment being preferred when serious access certain forms of health care. symptoms appeared. Even so, preparing special diets and foods could help ‘Most PLWH/A in Thailand do not live household members feel they were long. The delay in seeking care is an more actively involved in the care of a important factor (in this) ... PLWH/A household member with AIDS. ... do not enter the public health care delivery system early due to concerns ’Direct and active involvement of the fam- about the care received as well as their ily in health management could be a key to stigmatization ... Many PLWH/A do promoting positive coping strategies and not go to the hospital to avoid arousing possibly in delaying the onset of AIDS. suspicion of having HIV/AIDS’. (Th, The participation of the members in the p.38-39). treatment process, the special efforts they make in promoting wellbeing, and their Respondents at this same site noted general concern for the person, may help instances of a people with HIV paying in building psychological immunity and a for private treatment rather than will to live’. (I, p.34). using less expensive public health care, for fear of being identified as Where it was problematic to medical- having AIDS. Reports of people with ly diagnose HIV or AIDS, as in the HIV being treated badly in hospital in Kyela district in the United Republic of this country and in others were also Tanzania, and where there was stigma said to be due to medical profession- associated with having AIDS, tradi- als being overly influenced by the tional forms of treatment also offered stigma surrounding HIV disease. a way of redefining the illness itself

90 CHAPTER 4

(especially if the traditional treatment Where a traditional treatment was was seen to work). The failure of tradi- seen not to work, and where the ill- tional treatments, an uncertain diag- ness was accompanied by symptoms nosis, or the non-availability of treat- such as weight loss, itchy bodily rash, ment drugs could, however, cause recurrent fevers and chronic diar- people to ‘shop around for cures’. rhoea, household and community members tended to suspect that the ‘Uncertainty regarding the cause of ill- illness was indeed AIDS. In these cir- ness was evident ... where either a cumstances, an official diagnosis was vague diagnosis or no diagnosis was seen as important. Even though AIDS provided by allopathic health services. might be more stigmatizing than Within these households, the perceived other illnesses, knowing a person had causes of illness ranged from posses- the disease helped household and sion by evil spirits, to being bewitched or cursed. Shopping amongst tradition- community members respond appro- al healers was very prominent in these priately and use what limited households, where divination for cause resources are available to best effect. as well as treatment of the conditions was sought’. (Ta, p.60).

91 CHAPTER 5

CONCLUSIONS stifles supportive responses and should be challenged. And while traditional funeral practices continue to be adhered to in many sites, the num- Although theoretically it is possible to ber of deaths from HIV and AIDS- distinguish household from communi- related illness means that funeral ty responses, and the local enquiries practices are changing so as not to described here were designed on this exhaust a household’s or a communi- assumption, evidence from this multi- ty’s financial and emotional resources. site study suggests that responses to AIDS cannot be so clearly differentiat- While some household and communi- ed in practice. Instead, household and ties have responded to HIV and AIDS community responses are perhaps with courage, determination, support best understood as the outcome of an and solidarity, others have done so with interaction between factors such as fear, panic, despair and stigmatization. poverty and gender and traditional These different kinds of response vary ways of coping with illness and dis- not only between communities but also ease. Key influences on household within them. In Santo Domingo, for and community responses to HIV and example, community members’ views AIDS include: about the epidemic were reported as being more polarized in places where • the existing economic situation of AIDS had had the greatest impact. affected households and communities; Paying your way... • prevailing community and household relationships between men Responses to HIV and AIDS at house- and women; hold and community levels were • local health beliefs and health linked to available financial resources. care practices; and Where the main wage earner (most • local levels of stigmatization in often male) was unable to work, it was relation to those living with and usually incumbent upon his female affected by HIV and AIDS. partner as well as the children to support the household financially. When a Local contexts influence responses in main wage earner died, remaining complex and dynamic ways, and while household members were often left household and community responses with serious debts. For many women to HIV and AIDS are shaped by pre- and their children this could mean existing relationships and beliefs, they having to vacate their home. in turn influence these beliefs and practices. For example, because most At community level, both the number HIV- and AIDS-related care is per- of people requiring care and the num- formed by women, the support needs ber of deaths caused by AIDS posed of women are increasingly being rec- serious challenges for existing health ognized and acted upon. Similarly, care systems and funeral practices. while there is much stigma against Even in Thailand, a country which until people with HIV and AIDS, there is recently has been described as one of growing awareness that such stigma the ‘tiger economies’ of Asia (AIDS

92 CHAPTER 5

Analysis, 1997), difficulties were provided care for husbands, sons, reported in matching available daughters, or themselves. Even resources to health care needs. These among gay community respondents were perhaps linked to the benefits of in Mexico City, sons who were gay economic growth failing to reach regularly returned to their biological those in rural areas. For those women family (and particularly to their moth- needing to earn income, traditional ers) to receive care when ill. Women work was often unavailable, requiring to a much greater extent than men some to enter a rapidly developing had to balance the provision of care service sector that included sex work. to various household members with an increasing responsibility to sup- The care of bereaved children was port their household financially. They also linked to the economic circum- were also reported as being physi- stances in which parents and siblings cally, emotionally and sexually found themselves. Most often, exist- abused: by husbands who might ing household, extended household insist on sex even when they them- or clan relationships established the selves were seropositive; by in-laws structure within which care for who might demand that a daughter- bereaved children was provided. But in-law move out of her deceased even in contexts such as the Kyela dis- husband’s home; and by brothers-in- trict of the United Republic of law or other male community mem- Tanzania, where rules were said to bers who might believe a recently exist for the care of children, poverty bereaved wife was sexually avail- may force childcare to be provided able. beyond preferred social networks.

... in a man’s world Stigma and discrimination

In each study site, there was evidence Each country report offered powerful of important double standards govern- evidence of the stigma, discrimina- ing expectations of men and women. tion and secrecy associated with Whereas men with HIV were little ques- AIDS, as well as the tendency for tioned about how they became infect- people to understand AIDS as a dis- ed and were generally cared for, ease of ‘others’. Respondents made women were often castigated for hav- mention of ways in which the pres- ing had extramarital sex (whether or ence of AIDS might reveal itself not this was the case), and received among household or community lower levels of support. For those wives members. Attention was paid to the who suspected they might be at risk of forms of employment, sexuality, ill- infection through their husbands, and ness and death, which could mark who had been informed about the out the infection, and the infectious- need to use condoms, bargaining ness of others. Sex workers, gay power was dependent on economic men, ‘unfaithful’ husbands, people power. The greater the latter, the more who had lost weight for no apparent force a women could bring to bear in reason, and those who had died sud- her arguments for condom use. denly or who had been ill for lengthy periods of time were often assumed In every study, women most often to have HIV or AIDS.

93 CHAPTER 5

HIV- and AIDS-related discrimination Study reports also noted that services was manifest in a number of forms: offered privately tended to be per- through loss of employment, recrimi- ceived as of better quality than those nation, verbal abuse, being ousted provided through publicly funded sys- from the family home, arguments, tems. However, such services were resentment and segregation. These only available to those able or willing responses often caused HIV to be kept to pay for them — which could leave secret, at least until physical symptoms households destitute following a made the person’s condition obvious. lengthy illness. There were also Despite this, it was rare for care not to accounts of health care providers be provided, for men at least. Men breaking confidentiality, of not being were said to anticipate a more nega- aware of the need for confidentiality, tive reaction than they actually of acting in a discriminatory manner, received, and wives in particular usual- and of failing to address the emotion- ly continued to provide care for their al needs of people with HIV and their husbands — a form of provision that families. often went unreciprocated. Where HIV antibody tests were not AIDS-related discrimination was often readily available as was the case in the justified by real or imagined fears of Kyela district of the United Republic transmission. While the main routes of of Tanzania and in parts of Mumbai, transmission were generally under- household members tended to seek stood, there was continuing anxiety traditional cures for the illnesses they about the possibility of infection believed they had, thereby using lim- through bodily contact, bad gases, ited household resources in an often and the sharing of eating and drinking inappropriate way. Even where med- utensils. In all sites, respondents con- ical care was more widely available, flated the ways in which HIV might be traditional forms of health care such transmitted with the manner in which as paying special attention to dietary illnesses such as tuberculosis were needs were also practised. acquired. Fears were most apparent when children shared the household The importance of voluntary sector with a person with HIV/AIDS. When service provision was highlighted in a there were visible signs of illness, chil- number of studies. In dren, and those deemed particularly Netzahualcóyotl, Mexico City and vulnerable, were often kept at a dis- Chiang Mai province, advantages and tance from the person who was unwell. disadvantages were noted in this form of support. While volunteers could provide an important service to peo- Public, private, voluntary ple with HIV and their families, there and informal sector care were occasions when voluntary workers had little practical to offer by way People with HIV and AIDS received of support, or addressed HIV and care and support from those in the AIDS issues in ways which were con- public, private, voluntary and informal sidered inappropriate. sectors. Most reports drew attention to the severe constraints under which Despite the involvement of health medical professionals operated. care professionals, traditional healers

94 CHAPTER 5 and volunteers, the bulk of care and Recommendations support in all sites was provided through the informal sector, by part- A number of recommendations for ners, relatives, and friends. Without policy and programme development the involvement of these other derive from the studies described in groups, communities would have this report. been unable to sustain any form of meaningful health care for their Policy development members. At a policy level, it is important to recognize that: Bereavement and funerals • Contrary to what is sometimes The death of a person with AIDS had imagined, households and com- profound implications for the recon- munities respond to HIV and stitution of households and for AIDS in a variety of ways. While all funeral preparations. Household are likely to be negatively affect- reformation often took place at a ed by the stress and responsibility time of personal loss and mourning, of caring for seriously ill mem- and against a backcloth of impover- bers, responses of tremendous ishment created by the costs of car- support and solidarity are not ing for an unwell household mem- uncommon. ber. For women, the consequences could be particularly severe. If she • Existing household and communi- had previously been living in the ty structures may require rein- home of an in-law, for example, a forcement if they are to respond woman might be asked to move fol- effectively to the challenges lowing the death of her husband. posed by HIV and AIDS. This is Where children had also been particularly true for resource poor bereaved, their emotional and phys- households where the advent of ical health care requirements need- HIV and AIDS may pose the ed to be addressed. On the whole, greatest challenge. however, children continued to be looked after by household members, • Women most usually shoulder the relatives or clan members. burden of providing HIV- and AIDS-related care, adding to their The high cost of funerals was already heavy responsibilities referred to in both the Tanzanian and within the household. Thai studies. Due to the frequency of Households, families and commu- deaths from AIDS-related illnesses, nities need to be helped to funeral and mourning practices were understand that caring for people in a process of transition. with HIV and AIDS is a responsi- Increasingly, and especially among bility to be shared, and that both those who were poor, efforts were men and women have a role to being made to keep funeral costs play in this respect. low by inviting fewer people, having fewer days of mourning and pur- • The needs of HIV–positive women chasing less costly victuals. in both illness and in health

95 CHAPTER 5

require special attention if women a continuing need for contextual inter- are not to find the burden of car- ventions to reinforce understandings ing for others intolerable in cir- of: cumstances where they themselves may be unwell. • HIV’s major routes of transmission;

• While NGOs and CBOs have an • the ways in which HIV is not trans- important role to play in provid- mitted; ing HIV- and AIDS-related care, their actions should not be seen • the differences between HIV trans- as a substitute for indigenous sys- mission and the transmission of tems of care and support. NGO other diseases such as tuberculosis involvement may sometimes be and malaria; unwelcome in affected communities especially when they are seen • the care and support needs of as contributing little to existing people with HIV and AIDS; and systems of support. • the responsibilities of both women and men in meeting these needs. Programme development A combination of mass media and In relation to programme develop- face-to-face approaches may be use- ment, it is possible to distinguish four ful in consolidating accurate knowl- main fields of intervention: (i) those edge and understanding. Where relating to the overall context within some knowledge about HIV and AIDS which household and communities exists, it may be more appropriate to must respond to HIV and AIDS; (ii) begin by examining (and dispelling) those concerning the strengthening of local myths about transmission and existing household and family the course of disease, than to simply responses; (iii) those contributing to provide more scientific information. the development of stronger and more positive community responses; (ii) Strengthening household and fam- and (iv) those seeking to establish ily responses greater synergy between the household and the community in respond- Within the household, a range of ing to the epidemic. interventions may be useful in strengthening positive responses to (i) Contextual intervention HIV and AIDS, and in seeking to reduce negative ones. These include Local beliefs about HIV and AIDS, interventions to modes of transmission, and means of infection influence the ways in which • promote a realistic appreciation of household and communities respond HIV-related risks including to the epidemic. Where people have responsibility for safer sex and accurate knowledge about HIV and (where necessary) safer drug use; AIDS, responses are likely to be more positive than where erroneous beliefs • reduce stigma, prejudice and hos- exist. In many circumstances, there is tility towards people with HIV and

96 CHAPTER 5

AIDS (e.g. by challenging the division of labour with respect to stereotypes and assumptions that AIDS-related care should be linked to underpin prejudice and stigmati- broader programmes to improve zation); women’s overall economic and educational status. • promote the household’s capacity to provide care and support in There was strong evidence from all ways that minimize the likelihood countries that, except in the most of disease progression (e.g. extreme circumstances, special pro- through attention to diet, rest and jects for ‘AIDS orphans’ are unlikely to stress reduction); be of value. Instead, existing households, families and kinship structures • establish and reinforce a more should be reinforced in their capacity equitable division of labour to offer care and support. This may between women and men in rela- require linking HIV-related intervention to the support and care of tions to broader programmes of affected household members — poverty alleviation and health support this may involve a redefinition of in ways sensitive to local family struc- the concept of ‘family support’ in tures, and in a manner which recog- circumstances where such sup- nizes the potential of extended kin port is understood as being networks to cope with often difficult women’s responsibility; circumstances.

• ensure that bereaved children, or (iii) Strengthening community children whose parents are too responses sick to care for them, are provided for equitably and in line with Communities vary in their capacity to local traditions and arrange- cope with HIV- and AIDS-related ments. adversity. In some settings, systems of community support may be strong A number of different strategies may and able to accommodate the new need to be pursued to achieve the demands triggered by AIDS. In other above goals. It is not reasonable to circumstances, especially where mod- suppose that the provision of factual ernization has caused a weakening of information alone will dispel anxieties community affiliations and ties, the about HIV transmission through every- capacity of communities to respond day contact with an infected person. positively to HIV and AIDS may be Instead, it may be better to provide more limited. Programmes and inter- such information alongside interven- ventions to strengthen community tions in which care providers come responses should therefore be tai- into close contact with people with lored to the contexts in which they are HIV and AIDS and suffer adverse con- implemented, taking into account sequences. Likewise, it is not realistic prevailing community norms. Among to imagine that long established gen- the most needed community level der relations between women and interventions are: men can be transformed by a single AIDS-related intervention. Instead, • initiatives to reduce stigma and efforts to promote a more equitable discrimination towards affected indi-

97 CHAPTER 5 viduals and families, emphasizing how (iv) Enhancing synergy between fear, denial and social exclusion can households and community disorient HIV- and AIDS-related prevention and care; One of the greatest anxieties expressed by people with HIV and • actions to strengthen partnerships AIDS and their family members was between communities and formal apprehension about how the broader health care providers, including the community would react should their provision of counselling and support, diagnosis be known about. People clarifying boundaries of confidentiality feared being rejected, isolated and and responsibility, and acting to meet stigmatized in their neighbourhood the specific support needs of women and by friends and relatives. In conse- providing HIV-related care; quence, there was a tendency to conceal an HIV positive serostatus from • steps to involve traditional healers others and to seek care and support and health practitioners more fully in within the household alone. This is HIV and AIDS-related work; in preven- unfortunate since (i) communities are tion education; in enhancing the phys- not helped to confront the realities of ical and emotional well being of peo- HIV and AIDS and (ii) community ple living with, and affected by HIV resources such as peer group support, and AIDS; and in encouraging client advice from older members and referral to formal health services neighbourly concern remain where serious HIV-related illnesses are untapped. New programmes and suspected; and interventions are therefore needed to:

• efforts to develop and sustain • provide opportunities for commu- appropriate systems of voluntary care nities to face up to AIDS and begin to and support, including those which deal realistically with the challenges it seek to integrate voluntary care and poses; support with that provided more infor- • provide opportunities for people mally within households and at com- living with HIV and AIDS to be more munity level. There is clear evidence open about their serostatus and cir- from these studies that NGO and cumstances; CBO effectiveness is dependent on • dispel inappropriate fears and anx- the relationship between these orga- ieties about HIV transmission; and nizations and affected groups and • enable people with HIV and AIDS communities, and the capacity of such to be visible as responsible, pre- organizations to show they are provid- dictable and valuable members of ing a needed service or source of sup- the community — as part of the port. ‘solution’ to the epidemic and not as a ‘problem’.

98 CHAPTER 5

COUNTRY REPORTS Contact Dr Rafael Garcia, Instituto de Sexualidad Humana, Universidad Autónoma de Santo Domingo, Antiguo Hospital ‘Marión’, Aristides Fiallo Cabral, Santo Domingo, Bharat S (1996) Facing the challenge: Dominican Republic. household and community response to HIV/AIDS in Mumbai, India, Tata ******* Institute of Social Sciences, Mumbai, India. Leshabari M, Kaaya S, Nguma J & Kapiga S (1996) Household and com- Contact Dr Shalini Bharat, Unit for munity responses to HIV/AIDS in Family Studies, Tata Institute of Social Kyela District, Muhimbili University Sciences, PO Box 8313, Deonar, College of Health Sciences, Dar es Mumbai, 400 088, India. Salaam, the United Republic of Tanzania. ******* Contact Dr M.T. Leshabari, Director, Castro R, Eroza E, Hernández J, Institute of Public Health, Muhimbili Manca C & Orozco E (1996) University College of Health Sciences, Household and community responses PO Box 65015, Dar es Salaam, the to HIV/AIDS in Mexico, Centro de United Republic of Tanzania. Investigación en Sistemas de Salud, Instituto Nacional de Salud Pública, ******* Mexico. Singhanetra-Renard A, Chongsatitmun Contact Dr Roberto Castro Perez, C & Wibulswasdi P (1996) Household Centro Regional de Investigaciones and community responses to Multidisciplinarias, Av. Universidad HIV/AIDS in Thailand, Chiang Mai s/s, Circuito No. 2, Col. Chamilpa, c.p. University, Chiang Mai, Thailand. 62210, Cuernavaca, Mor., Mexico. Contact Dr Anchalee Singhanetra- ******* Renard, PO Box 87, Chiang Mai University, Chiang Mai, 50202, Garcia R, de Moya EA, Fadul R, Thailand. Castellanos C, Freites A & Guerrero S (1996) Household and community responses to HIV/AIDS in the Dominican Republic, Instituto de Sexualidad Humana, Universidad Autónoma de Santo Domingo, Santo Domingo, Dominican Republic.

99 CHAPTER 5

REFERENCES Vancouver, July. We.D.3702.

Cohen B & Trussell J eds. (1996) Preventing and mitigating AIDS in sub-Saharan Africa. Washington, AIDS Analysis (1997) HIV/AIDS country National Academy Press. profile: Thailand. AIDS Analysis Asia. 3(1),4-7. d’Cruz-Grote D (1996) Prevention of HIV infection in developing countries. Ainsworth M & Over M (1994) The Lancet, 348, 1071-1074. economic impact of AIDS on Africa. In: Essex M, Mboup S, Kanki P & Dodd R (1996) Thailand’s new AIDS Kalengayi M, eds. AIDS in Africa. New plan influences national economic York, Raven Press. planning. AIDS Analysis Asia. 2(6),6-7.

Ankrah EM (1996) AIDS, Socioeconomic Dowsett G & O’Brien S (1994) decline and health: a double crisis for HIV/AIDS in the Asia-Pacific region: the African woman. In: Sherr L, Hankins report of the international seminar on C & Bennett L, eds. AIDS as a gender social science research priorities. issue: psychosocial perspectives. Sydney, Macquarie University. London, Taylor and Francis. Fernandes MEL, Pimenta MC, Vieira Ankrah EM, Schwartz M & Miller J EM, Fini IG, Kottelos K, & Gitchens W (1997) Care and support systems. In: (1996) Capacity building: a critical Long LD & Ankrah M, eds. Women’s component in the implementation of Experiences with HIV/AIDS. An inter- HIV/AIDS prevention in Brazil. XI national perspective, New York, International Conference on AIDS, Columbia University Press. Vancouver, July. Th.D.4870.

Auer C (1997) Women, children, and Forina FM, Matulessy PF, Sahertian E & HIV/AIDS. In Long LD & Ankrah M, Mamoto B (1996) HIV/AIDS prevention eds. Women’s Experiences with program through family in Indonesia. HIV/AIDS. An international perspec- XI International Conference on AIDS, tive, New York, Columbia University Vancouver, July. Press. Forsyth S & Rau B, eds. (1996) AIDS in Barnett T & Blaikie P (1992) AIDS in Kenya: socioeconomic impact and Africa. London, Belhaven Press. policy implications. Arlinton, VA, AID- SCAP/Family Health International. Bor R (1997) The impact of HIV/AIDS on families. Guest editorial. AIDS Grunseit A & Kippax S (1992) newsletter, 12(7),1-3. Response to HIV/AIDS at the individual, household, community and soci- Chuaprapaisilp A & Parsons CDF etal levels. Report to the Global (1996) A family based care project in Programme on AIDS, Office of Thailand: advances in an HIV treat- Interventions Development, Social ment and prevention programme. XI and Behavioural Studies and Support. International Conference on AIDS, Geneva, WHO/GPA.

100 CHAPTER 5

Jayaseelan J (1996) An HIV/AIDS Mello JH (1996) Strategies for coping organization of communities - empha- with HIV/AIDS in self-help groups. XI sizing community building, consulta- International Conference on AIDS, tion and evaluation. XI International Vancouver, July. Th.C.4388. Conference on AIDS, Vancouver, July. Pub.D.1302. Mercer MA, Mariel CE & Scott SJ (1993) Lessons and legacies. Final Kadhumbula I, Tibatemwa M & report of a grants program for Kalemba J (1996) Developing com- HIV/AIDS prevention in Africa, munity micro-project initiatives to Baltimore, The Johns Hopkins combat AIDS: a pilot for (DIFAP) pro- University. gramme. XI International Conference on AIDS, Vancouver, July. Th.B.4131. Outwater A (1997) The socioeconomic impact of AIDS on women in Leonard A & Khan A, eds. (1994) Tanzania. In: Long LD & Ankrah M, Community based AIDS prevention eds. Women’s experiences with and care in Africa. Building local initia- HIV/AIDS. An international perspectives. New York, The Population tive. New York, Columbia University Council. Press.

Levine C (1996) Ethical issues in devel- Piot P (1997) Bridging the divide — oping policies for orphans of the HIV inequalities and AIDS. Martyn Taylor epidemic. XI International Conference Memorial Lecture, London, 12 on AIDS, Vancouver, July. Tu.D.2929. November.

Machinjili AA (1996) Home based care Ruben E (1996) Developing family for chronically ill people: a community knowledge on HIV/AIDS through the programme. XI International radio, television and newspaper pub- Conference on AIDS, Vancouver, July. lication “AIDS Quiz Program”’. XI We.B.3141. International Conference on AIDS, Vancouver, July. Th.C.4383. Mann J, Taranatola D & Netter T, eds. (1992) AIDS in the world: a global Seidel G (1996) AIDS and gender: a report. Boston, Ma., Harvard forgotten dimension. AIDS Analysis University Press. Africa, 6(5), 12-13.

Mann J & Tarantola D, eds. (1996) Shrestha A (1996) The impact of AIDS in the world II: global dimen- HIV/AIDS on Nepalese women: con- sions, social roots and responses. tributing factors and lessons learned New York, Oxford University Press. from community based interventions. XI International Conference on AIDS, Mawejje D, Patrick G, Kiwanuka R & Vancouver, July. Pub.D.1455. Bosco K (1996) Community’s ability in the provision of home based care: the WHO (1993) General protocol for TASO experience. XI International studies of household and community Conference on AIDS, Vancouver, July. responses to AIDS in developing Th.B.400. countries, Geneva, World Health Organization, Global Programme on

101 CHAPTER 5

AIDS, Social and Behavioural Studies WHO and UNAIDS (1998) The and Support Unit. socio-economic impact of AIDS. Report from a consultation on the WHO/UNAIDS (1995) General socio-economic impact of HIV/AIDS research protocol for studies of the on households, Chiangmai, determinants of HIV and AIDS-related September 1995. Geneva, World discrimination, stigmatization and Health Organization /UNAIDS. denial in developing countries, Geneva, World Health Organization World Bank (1997) Confronting AIDS: /UNAIDS. public priorities in a global epidemic. New York, Oxford University Press.

102 Part 3 Use of the female condom: gender relations and sexual negotiation

Peter Aggleton, Kim Rivers1 and Sue Scott2 (from an original analysis by Sue Scott)

1 Thomas Coram Research Unit, Institute of Education, University of London, UK 2 Department of Applied Social Science, University of Sterling, UK. CHAPTER 1

INTRODUCTION Assessment data were first collected over a 2 — 3-month period on prevailing patterns of gender relations, sexual communication and negotiation. Subsequently, an intervention Several studies have demonstrated took place to strengthen women’s the efficacy of the female condom as capacity for sexual communication a method of birth control and as a and negotiation and to provide them means of protection against STDs with access to the female condom and (e.g. Sai, 1993; Soper, Soupe, knowledge and skills relevant to its Shangold et al., 1993; Fontanet et al., use. Participants’ experiences intro- 1998). Despite this encouraging ducing the female condom into sexu- research, the value of the female con- al negotiations were monitored. In dom for HIV/AIDS prevention remains each of four countries — Costa Rica, unclear, not least because unequal Indonesia, Mexico and Senegal — power relationships between men and two groups of women participated in women impact significantly on the the research. One of these groups ability of women to control the use of consisted of women engaged in sex any prevention technology (du work, the other consisted of women Guerny & Sjorberg, 1993). One possi- from a range of other backgrounds. ble means of reducing HIV transmission is to increase women’s control This comparative analysis of findings over the means of prevention, but this from these studies begins with a demands good communication as selective review of the literature on well as access to resources for protec- gender relations, heterosexual com- tion (Mane, Gupta & Weiss, 1994). munication and negotiation, and the While a number of studies have role of the female condom in safer sex focused on the acceptability of the (Chapter 1). The research methods female condom, little research has used in each of the four sites are then been conducted into the ways in described (Chapter 2) before data are which its introduction might affect analysed to highlight similarities and sexual communication and negotia- differences in findings across the four tion between men and women sites (Chapter 3). The emphasis is on (FHI/AIDSCAP, 1997). The primary the analysis of prevailing gender roles question underpinning these studies, and relations, patterns of sexual com- therefore, was to what extent women’s munication, perceptions of risk, ways capacity to negotiate safer sex with of ‘negotiating’ safer sex, the reac- their partners could be enhanced by tions of study participants and their the introduction of the female con- partners to the female condom, and dom. its impact on gender relations. A final chapter discusses the studies’ implica- Simply making the female condom tions for future policy and practice in available is unlikely to shift the bal- HIV prevention. ance of power between men and women in the absence of other sup- The studies described were portive interventions. A two-phase exploratory qualitative investigations study was therefore designed. and therefore exhibit the strengths

104 CHAPTER 1 and limitations of this kind of enquiry. They provide insight into the nature of prevailing norms, relationships and processes of sexual communication and negotiation, but do not seek to measure or quantify these phenome- na. No tests of statistical significance were performed on the data — such would have been inappropriate given sample sizes, the manner in which respondents were chosen, and the commitments of qualitative enquiry. This multi-site study Use of the female condom: gender relations and sexual negotiation seeks therefore to illumi- nate key issues and concerns which might be further examined, in research of perhaps a different kind.

105 CHAPTER 2

BACKGROUND have high and often rapidly increasing rates of infection (UNAIDS, 1997). While HIV can in theory affect anyone, globally there are marked inequalities of risk dependent on cultural and In most countries women are at par- structural factors that are often ticular risk of HIV infection. Recent beyond the control of individuals estimates suggest that nearly half of (Decosas, 1996). Sweat & Denison those infected with HIV worldwide are (1995) have recently described the women, and of the estimated 2.3 mil- ways in which social, economic and lion people who died from HIV-related political forces such as poverty, migra- diseases in 1997 about a half were tion, urbanization, war and civil distur- female (UNAIDS, 1997). Women con- bance facilitate HIV transmission. stitute more than 50% of all adults Women in developing countries may with HIV infection in sub-Saharan be at particularly high risk of acquiring Africa, more than 40% in the HIV, and the special vulnerability of Caribbean, around 20% in central and young women is now well document- south America, and about 30% in ed. Estimates suggest that in Uganda, south and south-east Asia (UNAIDS & for example, HIV infection among WHO, 1996). There is evidence to young women aged 13-19 years can suggest that women have increased be up to 20 times higher than for men physiological vulnerability to HIV of the same age group (Panos infection, it having been estimated Institute, 1996). In some cities in sub- that the risk of transmission from man Saharan Africa HIV prevalence among to woman per exposure can be up to young pregnant women may be high- 2.5 times higher than that from er than 30% (Mann & Tarantola, 1996). woman to man (European Study Group on Heterosexual Transmission Several factors place women in devel- of HIV, 1992). Stereotypical gender oping countries at higher risk of HIV roles and unequal power relationships infection. Access to education is often between men and women, as well as more limited for girls than for boys; increased economic and social vulner- levels of literacy for females are usual- ability, mean that women are less likely lower than for males (Sivard, 1985); ly to control how, when and where sex and women are more likely to be liv- takes place (Mane, Gupta & Weiss, ing in or vulnerable to poverty (Gupta 1994; Heise & Elias, 1995; & Weiss, 1993; Ankrah, 1996). Women International Center for Research on are also more likely than men to expe- Women, 1996). rience sexual coercion or rape, and are more likely to sell sex in order to Context survive. In Malawi, two thirds of 168 young women recently surveyed Developing countries have been reported having sexual intercourse in among those most affected by HIV exchange for gifts or money, while at and AIDS (WHO, 1994a). Although, the University of Calabar in Nigeria HIV prevalence varies considerably nearly 15% of women studying said between countries, parts of Africa, that they had received payment for south and central America and Asia sex in order to finance their education

106 CHAPTER 2

(International Center for Research on 1994). The high value afforded to vir- Women, 1996). ginity in girls, in conjunction with a very different set of values for boys, Women may be drawn into sex work can also enhance vulnerability to HIV as a result of poverty or in order to infection. Moreover, young people pay off family debts. The Chiang Mai interviewed in several recent studies Hill Tribes Welfare and Development have reported that anal sex is often Center in Thailand, for example, has used as a means to protect a girl’s vir- estimated that one in five girls coming ginity (International Center for into contact with the project provide Research on Women, 1996). sexual services for payment, and some have been sold or given into Recent work in Mexico suggests that prostitution by family members gender stereotypes affect young (WHO, 1994b). In less precarious cir- men’s willingness to use condoms. cumstances women often lack the They also promote the idea that resources and power to negotiate whereas one partner may be appro- effectively with men over safer sex. priate for women, a ‘real man’ must Women participating in focus group have several different partners concur- discussions in Mumbai, India, for rently (Lemus, 1992). Stereotypes and example have stated that, because of expectations also place women at risk economic dependence and fear of in cultures where being too knowl- physical violence, they must submit to edgeable about sex is seen as evi- their husband’s sexual demands with- dence of promiscuity. Recent studies out opposition (George & Jaswal, in Thailand, for example, have shown 1994). that here as in other countries there are pressures on young women to be In many parts of the world, the ten- ignorant about sex for fear they will be dency for women to have sexual part- seen as being too experienced (Cash ners older than themselves also & Anasuchatkal, 1993; Cash et al., increases vulnerability to HIV infection 1997). (Heise & Elias, 1995). Since such partners are often sexually experienced, Women have often been identified as they are more likely to be infected key players in the control of HIV — a (Panos Institute, 1996). Research also focus that has been seen by some as shows that women’s risk of acquiring reinforcing their traditional responsi- HIV increases dramatically as age at bility for the health of the household first intercourse falls (Palloni & Lee, and as guardians of morality (Bland, 1992), and exposure to HIV before` 1985). A major criticism of some early physical maturity greatly increases the HIV/AIDS prevention initiatives was risk of infection (United Nations that they failed to give enough atten- Development Project, 1992). tion to women’s economic and social subordination, and the implications of Social attitudes towards sexual activity this for their ability to negotiate vary dramatically. While sexual activity where, when and how sex took place may be considered acceptable and (FHI/AIDSCAP, 1997). This situation even healthy for boys and young men, has changed, however, and there is this is not the case for young women now an increasing body of literature (see, for example, Bezmalinovic et al. that suggests that gender empower-

107 CHAPTER 2 ment is key to women’s ability to pro- Gender inequalities are reflected in sex- tect themselves from HIV infection ual relationships between women and (Preston-Whyte, 1995). Additionally, men. Pressure and violence may be the there is growing acknowledgment outcomes when male demands for sex that structural and political changes are not met (George, 1996, 1997; du can help both women and men in Guerny & Sjorberg, 1993). Women not developing countries protect them- infrequently have to juggle the twin selves from HIV infection. Baldo expectations of being sexually available (1995), for example, has stated that without appearing to be sexually active. responsibility for the prevention of They may find themselves in circum- HIV infection is not solely a matter for stances where men expect them to pro- the individual, but that social policies vide sexual services either privately in are needed to inhibit the growth of the context of relationships and mar- the epidemic. Policies, programmes riage, or commercially in the context of and interventions to promote greater prostitution, on terms that are not of equality between men and women, their making — circumstances which and provide greater access to educa- deny women the right to sexual auton- tion for girls, are crucial to HIV pre- omy and sexual pleasure. Women in a vention (d’Cruz-Grote, 1996). diverse range of countries have reported being unable to act upon what they Gender, power and know about HIV and AIDS for fear of sexuality implying through requests for condom use that a partner is not loved or not The term gender emphasizes the trusted. Such requests may disturb the social shaping of femininity and mas- intimacy that is central to many rela- culinity, and challenges the idea that tionships and can result in violence, relations between women and men abandonment or rape (Ankrah & Attika, are ordained by nature. Social differ- 1997). ences between men and women profoundly affect women’s sexual rela- In most countries women’s experience tions with men, and gender and sexu- of sex is constructed in relation to male ality intersect with other social divi- sexual needs and wishes in a context of sions such as those of race and social dependence and, in some cases, a dis- class (Holland, Ramazanoglu, Scott et course of romantic love. This is not, al., 1992; Jackson & Scott, 1996). In however, to suggest that women are most societies, gender inequalities simply the passive victims of male sex- cause women to have less access to ual demands. Both individually and col- and control over economic resources. lectively, women develop strategies for This in turn leads to dependence on gaining sexual pleasure and for resist- male partners and relatives for mater- ing the pressures imposed upon them ial survival. Women’s socioeconomic by men, but they do so from an position in turn affects their ability to unequal starting point. enter into sexual relationships with men as equal partners and, even Sexual communication and where there is a measure of equality, ‘negotiation’ powerful norms about sexuality and sexual behaviour construct and con- The focus of many HIV prevention strain women’s behaviour. programmes to date has been the

108 CHAPTER 2 promotion of mutual monogamy and in Zambia have pointed out that while condom use, both of which require wives can (and possibly should) be good communication and negotiation monogamous, they cannot hope to between partners. Verbal communica- control the number of partners had by tion may, however, be one of the most a husband (Mushingeh, Chama & problematic aspects of heterosexual Mulikelela, 1991). While as a group encounters (Holland, Ramazanoglu women are more likely than men to be and Scott, 1990), and difficulty in sex- monogamous, individually they are ual communication appears prevalent also more likely to exchange sex for in countries as culturally disparate as money or goods as part of a survival the USA, Brazil and India (Goldstein, strategy (Heise & Elias, 1995). 1994; Bang & Bang, 1992). Promoting monogamy among women Participants in recent focus group dis- living in precarious economic circum- cussions in Mumbai suggested that stances may therefore be unrealistic unequal power relations between unless alternative income-generating men and women serve as a barrier to opportunities are also in place (Gupta, communication about sex between Mane & Weiss, 1996; Preston-Whyte, husband and wife (George & Jaswal, 1995; Benjamin, 1996). 1994; George, 1997), and an earlier survey of spousal communication in Programmes promoting the use of Asian countries found that almost a condoms in HIV prevention also face third of the women interviewed never difficulties caused by gender inequali- talked to their husbands about sexual ties. Using condoms, for example, matters (McNamara, 1991). Research requires male agreement and women suggests that overt sexual communi- often have to negotiate with an unwill- cation between husbands and wives is ing partner (Heise & Elias, 1995; Elias also rare in many countries in sub- & Coggins, 1996). Communication Saharan Africa (Caldwell, Caldwell & and negotiation about sex may be Quiggin, 1989). particularly difficult in cultural contexts where women are not meant to know Negotiating mutual monogamy most anything about sex (Mane, Gupta & probably entails major changes in Weiss, 1994). Condoms can also carry men’s understandings of themselves negative associations. They may be as sexual beings, and while this seen as suitable for ‘casual sex’ but emphasizes the need to prioritize inappropriate in the context of a work with men, bringing about such longer-term relationship (Cash & changes is probably not a realistic Anasuchatkul, 1993). Condoms may short-term expectation. Women par- be associated with mistrust and some ticipating in research in Thailand have women fear that their partner may commented that they would tend to suspect them of unfaithfulness if they think that a man who does not visit suggest condoms should be used. sex workers is not a ‘real man’ (Cash & Others feel that suggesting the use of Anasuchatkul, 1993), while women in condoms is tantamount to accusing Mumbai have expressed the view that their husbands of infidelity (Heise & a wife must accept her husband hav- Elias, 1995; Ankrah & Attika, 1997). In ing partners outside marriage order to persuade women and men to (George & Jaswal, 1994). Women par- see condoms as a valuable means of ticipating in focus group discussions protection, it is necessary to over-

109 CHAPTER 2 come these negative associations. But The female condom it is also necessary to address inequalities of power as they influence sexual The female condom was developed communication and negotiation for as a barrier method of contraception safer sex. in the late 1980s by Phoenix Health Care (now Chartex International) of While the concept of sexual negotia- Chicago, Illinois. It is a cylindrical tion is problematic, since it implies a polyurethane bag about seven inch- rationalism and autonomy that may es long with an integral outer ring at not exist (Mane & Aggleton, 1996), its wide end and a loose smaller there is evidence to suggest that in inner ring at the closed end. This circumstances where women are able inner ring is used to guide the con- to influence the forms and contexts dom into the vagina and is then used where sex occurs, HIV-related risks to keep the condom in place against may be lowered. This may be true for the cervix. The outer ring then holds some sex workers who are able to the condom in place outside the determine through the price charged vagina covering the woman’s exter- the kind of sex that will take place. It nal genitalia. may also be true for women who are in a position to withhold sex or some Initial responses other service. Research has shown that some women in Nigeria, for Initial responses to the female con- example, are able to refuse sex with- dom were muted and there was out reprisal if their partner is known to sparse attendance at a poster presen- have a STD, although it should be tation on its potential role in HIV pre- noted that the particular women sur- vention given by Leeper at the Vth veyed enjoyed higher than usual International Conference on AIDS in degrees of economic independence Montreal in 1989. Anxieties were (Orubuloye, Caldwell & Caldwell, voiced about the likelihood of sex 1992). workers re-using the condom, a response that was in line with the Central to the long-term success of view that women are ‘risky’ rather HIV prevention are forms of protec- than at risk. For others working in HIV tion over which women have more prevention, anticipation was tem- control and to which men offer less pered by concern. While some were resistance (Heise & Elias, 1995; Elias positive about the prospect of a & Coggins, 1996). Ideally, these device that was under women’s con- might take the form of safe and trol, others were worried that it would undetectable microbicides over place the task of preventing HIV which a woman has control (Elias, transmission even more firmly in 1996; Elias & Heise, 1996), but in the women’s hands, absolving heterosex- absence of these, barrier forms of ual men of responsibility for changing protection such as that offered by the their behaviour and thereby reinforc- female condom may have an impor- ing traditional gender relations tant role to play. (Panos Institute, 1990).

3 The female condom received approval in the UK and some other European countries in 1991, and US Food and Drug Administration acceptance in 1992.

110 CHAPTER 2

Acceptability potential to offer protection against STDs among women sex workers and Since it became available in the early their clients (Fontanet et al., 1998). 1990s,3 the female condom has been fairly widely assessed for its accept- Cost ability and for its effectiveness as an STD prevention measure (George & A major barrier to the use of the Mane, 1996). In Europe and North female condom is cost. This was America it has received a mixed found to be the main obstacle to use response from health professionals, among women, including sex workers, users and potential users (Lehto, 1991; in both urban and rural areas of Hoffman, 1991; Ford & Mathie, 1993; Uganda who were in other respects Gollub, Stein & El-Dadr, 1995; Perry, overwhelmingly positive about its use Sikkema, Wagstaffe et al., 1996). (Dithan et al., 1996). The relatively Studies in a variety of African (Ray et high cost of the female condom com- al., 1995; Ankrah, Kalckmann & Kabira, pared to the male condom derives 1996; Timyan et al., 1996; MacIntyre et from several factors: polyurethane is al., 1996; Musaba, Morrison & Sunkutu, more expensive than latex, more of it 1996; Ankrah & Attika, 1997), Asian is used, and the manufacturing (Tansathit & Cheevakej, 1990; Chan, process is also expensive. Female 1994; Jenkins et al., 1995) and South condoms sell for between US$ 1 and American (Gindin, n.d.; Ankrah & US$ 3 in the USA and Europe, a price Attika, 1997) countries have shown that that is too high for the majority of some groups of women relate positive- potential users in developing coun- ly to its use. A recent review of over 40 tries. Expanding the range of choices studies of female condom acceptabili- is clearly important though, and ty conducted worldwide concluded research suggests that, if provided that the overall balance of view lies in with a broader set of preventive its favour (UNDP/UNFPA/WHO/World options, women will try new and Bank Special Programme of Research sometimes multiple methods, thereby on Human Reproduction, 1997). achieving a higher proportion of protected sexual encounters (Elias & Coggins, 1996; Gollub, 1996). For this Effectiveness reason, a guaranteed purchase price While there is evidence that the of less than US$ 1 for public-sector female condom can be as effective as agencies in developing countries has other barrier contraceptives (FHI/AID- recently been negotiated between SCAP, 1997), its effectiveness as a UNAIDS and the female condom’s means of protection against STDs manufacturers. including HIV has as yet been less studied. What evidence there is sug- Sexual communication gests that it may be as likely as the male condom to protect against STDs What remains to be ascertained is such as trichomoniasis (Soper et al., the impact of the female condom on 1993). A recent field study conducted sexual communication. Of particular with the support of the World Health importance is the extent to which the Organization and UNAIDS in Thailand female condom may empower suggests that the female condom has women in their negotiations with

111 CHAPTER 2 men over the form and context with- pleasurable to women as well as to in which sexual relations occur. men, exploring sexuality indepen- Strong anecdotal evidence exists to dently, and developing an indepen- suggest that the level of this empow- dent sexuality (Holland, Ramazanoglu, erment may vary, since the impact of Scott et al., 1992). Whether and to the female condom may depend on what extent any of these changes are context (Williamson and Joanis, possible are questions that depend 1994). Learning more about the cir- on the circumstances in which women cumstances in which empowerment find themselves. There is clear evi- takes place is central to the success dence to show that women are more of future programmes and interven- able to put their intentions regarding tions involving the female condom as safer sex into practice when they have an HIV prevention technology. some personal experience of empowerment (Holland, Ramazanoglu, Scott Women’s empowerment et al., 1992; Holland, Ramazanoglu, Sharpe et al., 1992). Empowerment Empowerment is a difficult concept therefore offers a starting point for to define and there has been much practical strategies to transform sexu- discussion as to its meaning in rela- al relationships between women and tion to sexual and reproductive men. Possibilities of empowerment health. For the purposes of this are, however, influenced by divisions study, however, it is understood as and differences between women implying an autonomy in sexual com- including those of class, ethnicity, cul- munication and decision making, ture and religion (Cain & Finch, 1981). based on an understanding of options for prevention, in the context An important study in two contrast- of relevant resources for protection. ing communities in South Africa Such a definition shares features with highlighted the contextual differ- the approaches to empowerment ences affecting women’s empower- recently offered by the ‘Women’s ment in relation to the female con- Empowerment Framework’ (Longwe, dom (Preston-Whyte, 1995). n.d.) and by the Pan American Health Findings from this study suggest that Organization through its work on women in one of these communities gender, health and development — Nhlungwane — were able to (Labonte, 1997). begin using the female condom because existing levels of political Envisaging empowerment for women involvement made them confident requires defining the power relations enough to negotiate its use with sex- that need tackling as well as the ual partners. This political involve- means of changing them. In the con- ment also encouraged other women text of sexual encounters, however, in the community to follow their empowering women can mean a vari- example. These women, who were ety of things. These include not aware of the constraints on their lives engaging in sexual activity with oth- and the threats to their health as a ers, not engaging in sexual activity result of economic dependence on without informed consent, getting men, saw the female condom as a men to consent to safer sex practices, potential means of managing risk. negotiating sexual practices which are Their shared experiences thereby

112 CHAPTER 2 became the resources on which to is important that programmes enable draw in communicating about sex women by offering support, informa- with male partners and in negotiat- tion and opportunities to develop ing the use of the female condom. skills that would not otherwise be Other studies also point to the available. This way of working is part of importance of introducing the a wider movement towards building female condom in the context of sexual citizenship and a more expand- group-based interventions in which ed understanding of sexual and repro- women can gain support for change ductive health (Working Group on both from the intervention itself and Sexual Behaviour Research, 1995; from one another (see, for example, Scott & Freeman, 1995). The studies Shervington, 1993). described in this report reflect the growing desire to promote such ideals Concluding comments and to work with women and men in this way. They also derive from the Much attention has been given to belief that it is crucial to conduct issues of gender, power and women’s research in ways relevant to a particu- sexuality in recent work in HIV preven- lar setting, and with the desire to tion. This has highlighted the impor- ensure that findings are acted upon tance of taking account of women’s speedily, sensitively and appropriately own expertise and understandings. It (Gupta & Weiss, 1993).

113 CHAPTER 3

STUDY DESIGN AND • Under what circumstances do established patterns of behaviour METHODOLOGY undergo change? • How does the balance of power differ in the context of different In May 1993 the World Health types of sexual relationship? Organization’s former Global • What is the effect of specific socio- Programme on AIDS (WHO/GPA) cultural contexts on gender rela- published a general research protocol tions? for studies of Sexual negotiation, the • How do gender relations respond empowerment of women and the to increased awareness of female condom (WHO, 1993). Both HIV/AIDS and STDs? principal and site-specific research • How do gender relations respond questions were included to enable to the possibility of protection? intercountry comparisons to be • If there are such responses, are made, and to ensure that data col- they favourable to women? lected would be of value to those engaged in local work. Subsequent to Communication and negotiation the publication of the general research protocol, potential principal • How do men and women commu- investigators from a range of coun- nicate about the terms under tries were invited to prepare local which sex occurs? research proposals. These proposals • How does this communication vary were reviewed by the former in different kinds of relationships? WHO/GPA Steering Committee on • How does this communication vary Social and Behavioural Research, and in different sociocultural contexts? four studies — in Costa Rica, • How do women negotiate the Indonesia, Mexico and Senegal — terms of sexual encounters specifi- were recommended for funding. cally in relation to protective behaviour? Research questions • How do participatory peer-led discussions strengthen women’s incli- Studies supported under this initiative nation and abilities to negotiate were expected to analyse: (i) prevail- protective behaviour? ing gender relations, (ii) patterns and processes of sexual communication, Risk awareness and protective and (iii) the possibility of empowering behaviour women in relation to sexual communication and negotiation through an • What is the general level of intervention involving the female con- women’s awareness about dom. The principal research questions HIV/AIDS and STDs? to be addressed were as follows: • To what extent are women aware of the protective value of con- Gender relations doms? • To what extent are women able to • Who exercises control in relation influence the terms of any sexual to sex and protection? interaction?

114 CHAPTER 3

• To what extent are women able to involved in sex work (group A), and introduce the female condom into women currently working as sex workers a sexual encounter? (group B). Given the preliminary nature • How does risk awareness vary in of the enquiry, respondents were select- different types of relationship? ed so as to be typical but not represen- • How does risk awareness vary tative of important subgroups of women across sociocultural contexts? for whom the female condom might • What kinds of problems do men offer an important prevention option. encounter in relation to awareness The numbers of women recruited to and protection? each group and their principal social • What do men understand to be characteristics varied across sites (Table the effects of the female condom 2). For example, while women in group on sexual practice and negotia- A in Costa Rica represented a relatively tion? affluent section of the population, this • How do men react to any changes same group in Indonesia was made up that occur as a result of women’s of the non-working wives of poorly paid raised awareness? workers. Similarly, group B in Costa Rica included only literate sex workers, while Study design and samples in Senegal this was not the case.

All four projects followed the design Given the exploratory nature of the of the general research protocol out- studies, control groups were not lined above, but some variations were established except in Mexico where included to meet particular local local investigators were confident needs and contingencies. A two-stage that women already knew about the design was used at each site. First, an female condom and its potential assessment of prevailing gender rela- benefits. tions, sexual communication and negotiation, and the nature of sexual The intervention consisted of two and reproductive decision making was principal elements: (i) group activities made. Conducted over a 2—3-month to encourage women to discuss with period, this assessment elicited the one another prevailing gender rela- perspectives of women and men tions, potential barriers to communi- through key informant interviews, the cation and ways of overcoming these; analysis of existing documents and (ii) the distribution of female condoms reports, and observation. along with guidance on their use. Group activities were designed so as Secondly, an intervention took place to be culturally appropriate and in a designed to strengthen women’s capac- format suitable to local expectations ity in sexual communication and negoti- and needs.1 In addition to the female ation. Two groups of women were condom, participants were provided selected in each site: women not with male condoms on request.

1 The intervention in Mexico, for example, had the following components: a first meeting to explore participant’s attitudes towards gender relations and sexual communication/negotiation in general; a second meeting to help participants identify personal assumptions about gender relations and sexual communication and negotiation; a third meeting to introduce the female condom and details about how it might be used. Further details of specific local interventions can be found in individual country reports.

115 CHAPTER 3

TABLE 2 Respondent characteristics

Country Site Initial assessment Respondent numbers Respondent numbers charac- and characteristics and characteristics teristics (non-sex workers) (sex workers)

Costa Rica San José Preliminary forum Total = 32 Professionals Total=32 province involving local (n=10), students (n=4), Hotel workers n=16), women’s groups office workers (n=8) and bar workers (n=8), and potential study housewives (n=10). boarding house work- participants Aged 20-40. ers (n=6), massage parlours (n=2). Aged 20-55.

Indonesia Jakarta Formal and infor- n=54 n=57 area mal interviews with ‘Housewives’; for two Sex workers recruited key informants; par- thirds husbands worked through ‘rehabilitation ticipant observation in construction industry, centre’. among truck drivers for one third husbands Aged 17-32. and sex workers were lorry drivers. Aged 20-41.

Mexico Mexico DF Key informant inter- n=60 (30 in experimen- n=60 (30 in experi- views, observation tal group, 30 in ‘con- mental group, 30 in of places where trol’) ‘control’) young people Mixed social back- Mixed social back- meet, sex work ground grounds contexts, analysis of Aged 23-35 Aged 23-35. secondary sources

Senegal Kaolack Group and individ- n=25 n=20 and ual interviews, Mixed social back- 10 ‘registered’ and 10 Kolda observation with grounds non registered women focus on form and Aged 16-40 selling/ exchanging context of sexual sex negotiation Aged 16-40

116 CHAPTER 3

Study sites than passive in relation to their own health; and Costa Rica • to ascertain to what extent sharing Background — The research began knowledge and information about with a preliminary forum to which STDs, HIV and AIDS might con- women working in the fields of health tribute to women’s success in and sexuality were invited. Thirty negotiating safer sex. women attended this meeting to discuss the main objectives of the study and to identify themes to be explored Research design — Two groups of 32 during data collection. This initial women participated in the study. All consultation aimed to increase confi- the women were resident in San José dence and broaden ownership of the province, with the majority residing project and was considered highly within the city of San José. The first successful in this respect. It also group (group A) consisted of 10 pro- established criteria for the later selec- fessional women, 4 students, 8 office tion of participants. workers and 10 housewives. These women were aged between 20 and In stage one a questionnaire was 40. Twenty-four of them were married, administered to all the women in a further 3 were cohabiting, and all each of two groups. The question- but 1 had partners who were in naire was designed to elicit data on employment. Twenty-three of the 32 sociodemographic characteristics, women in this group were mothers. daily life, health and self care, sense All were considered to have what of identity, relationship(s) with part- locally would be seen as a ‘comfort- ner(s), sexuality, STD and HIV preven- able’ standard of living. tion, the female condom, and experience of sex work (group B only). In A second group (group B) was made stage two of the study, women were up of women working as sex workers offered psychosocial support through in the following settings: 16 in hotels, participation in confidence-building 8 in bars, 6 in boarding houses, and 2 workshops with the following objec- in massage parlours. Members of this tives: group ranged in age from 20 to 55, with 27 members being aged • to prompt individual and group between 20 and 35. Three women reflection on historical and current were cohabiting and 13 had a regular beliefs and attitudes towards partner. Twenty-nine members of the women and in particular to group were mothers. Most of the women’s right to control their own women working as sex workers had bodies and sexuality and to deter- financial responsibilities for children mine their own health needs; and/or other relatives.

• to strengthen participants’ image Research process — In the first stage of themselves in order to strength- of the study, data were collected on en self respect and self care; participant characteristics, prevailing gender norms and patterns of sexual • to affirm women as active rather communication. In stage two, 16

117 CHAPTER 3 women from each of the two groups • to collect data from women on were selected to take part in a seven- how they negotiated female con- session workshop to explore the dom use with partners; themes that had emerged in the first stage of the research. Selection took • to collect data on women’s experi- place on the basis of the women’s ences of using the female condom degree of commitment and potential over a period of time; for group-based work. Each group met for seven two-hour sessions held • to gain an understanding of how weekly with an experienced facilitator. the use of the female condom may Discussions and activities focused on influence women’s willingness to issues raised in stage one of the work, negotiate safer sex in the future; as well as those suggested by the and investigators. • to gain an understanding of differ- Indonesia ences between the two groups of women in the study and their dif- Background — This study included ferent reactions to the intervention preliminary enquiry into patterns and programme. processes of sexual negotiation followed by the evaluation of the effects Research design — Study participants of an intervention that aimed to were drawn from the Jakarta area empower women to negotiate safer because this was the province with sex through the introduction of the the highest incidence of HIV/AIDS. female condom. The key aims of the Group A comprised the partners of study were as follows: drivers and construction workers. These men were seen as being likely • to obtain more information about to have multiple sexual partners. gender relations and sexual com- Group B was made up of sex workers. munication/negotiation in two The main study was preceded by a samples of women: sex workers preliminary investigation using formal and the partners of drivers and and informal interviews with 14 key construction workers; informants and participant observation conducted among truck drivers and sex workers. • to develop materials to provide women with information about The partners of drivers and construc- HIV/AIDS and the need to negoti- tion workers were selected from two ate safer sex and to introduce the areas of the city — Pasar Rebo and female condom as an alternative Depok — where a large proportion of protective strategy; such workers lives. Women in the Pasar Rebo district were contacted • to gain a better understanding of through the wife of a truck driver the effect of increased knowledge already known to the research team. about HIV/AIDS on women’s will- Participants in Depok were contacted ingness to negotiate safer sex and through a driver who was known to on their willingness to try the the research team. In-depth entry- female condom; level interviews were conducted with

118 CHAPTER 3

11 women in Pasar Rebo and 43 ticipants who each attended every women in Depok. All were Moslems session. In the Depok groups, only aged between 20 and 41 who report- one participant failed to attend all ed having no sex outside marriage, the sessions. Group sessions cov- but who were all sexually active with- ered the following key areas: infor- in marriage. All were mothers, and mation about reproduction and con- the majority had less than six years of traception; STDs, HIV and AIDS; formal education. Around two thirds information and training in the use of the women’s husbands worked in of the female condom; training in the construction industry, while the communication and negotiation rest were drivers. relating to safer sex; and empowerment through sharing experience of The sample of sex workers was con- female condom use. Female contacted through a rehabilitation cen- doms continued to be distributed at tre. A meeting was held at the centre the last two sessions and for one attended by about 100 women, at month after the sessions ended. Exit which the aims of the project were interviews were conducted with par- explained. Entry level in-depth inter- ticipants in the group sessions one views were then conducted with 63 month after the final session to women, and 57 women participated enable individual reflection on their in the study. Study participants were experiences of sexual negotiation Moslems aged between 17 and 32, and use of the female condom. and the majority had no more than six years of formal education. For Mexico most, their first experience of sexual intercourse had taken place after Background — The study was marriage and before the age of 17. undertaken in Mexico City through Over half the women had children CONASIDA’s AIDS Information who were economically dependent Centres. It took the form of two par- on them and taken care of by grand- allel case studies of a group of 60 parents, often in their hometown or sex workers and a group of 60 village. In the month before the women not involved in sex work. interview, the frequency of sexual The study used qualitative methods contacts varied from 4 to 90 and, in and included in-depth interviews the majority of cases, individuals had with individual women in each of the had more than one partner in that two groups; observations in places month. where young people meet; observations in massage parlours, bars and Research process — Within each street sites where sex workers meet session, participants were divided their clients; group discussions and into groups of 10-15, each of which focus groups. Additional back- was led by a facilitator and cofacili- ground data were drawn from a tator. In the ‘housewife’ sample, one questionnaire survey of sex workers set of group sessions was held in attending an AIDS Information Pasar Rebo and a further three in Centre to be tested for HIV and from Depok, each of which had a facilita- data gathered in a focus group tor and a cofacilitator. The group study conducted by the same that met in Pasar Rebo had 11 par- research team in 1990.

119 CHAPTER 3

Research design — The study had Senegal the following components: Background — In Senegal, the study • a rapid ethnographic study of the took place in Kaolack and Kolda, cities two groups as a result of which 30 chosen to reflect the ethnic diversity women, aged between 23 and 35, of the country and the local incidence were selected from each group to and prevalence of HIV and AIDS. take part in the intervention; Kaolack and Kolda are both poor cities experiencing rapid urbanization, • detailed interviews with each of but they have different ethnic profiles: the 60 participants; Kaolack has a predominantly Wolof population whereas Kolda has a • three intervention workshops at majority of Fulbe people. At the time monthly intervals. Participants of the study, Kaolack had the highest were provided with supplies of the prevalence of HIV among sex workers female condom at the third work- in Senegal. shop; Research design — An interpretative • focus group discussions three qualitative approach was adopted. A weeks after the distribution of rapid assessment first took place female condoms at which further using group and individual interviews supplies were issued; and observational techniques to collect data on the form and the context • detailed individual interviews two of sexual communication and negotia- months after the second batch of tion. Ten research sites were subse- condoms were distributed; and quently identified within the two cities, and observations carried out in • detailed interviews with the male settings where nonconjugal encoun- partners of study participants. ters were likely to occur. These included hotels, bars, restaurants, dances Research process — Three interven- and bus stations. Semi-structured tion workshops were conducted as interviews were then carried out with follows: workshop 1 aimed to investi- a total of 182 key informants. The tar- gate participants’ attitudes towards get groups for these interviews includ- gender relations and sexual negotia- ed male heads of districts, leaders of tion; workshop 2 attempted to sensi- women’s associations or influential tize participants to their understand- women in the area, ‘housewives’, ings of their own gender and their women in paid work outside the assumptions about gender relations; home, registered and non-registered and workshop 3 offered information sex workers, and young people in about STDs, introduced the female youth associations. In Kaolack these condom and showed participants individual interviews were followed by how to use it. Interviews and focus five group interviews that brought group discussions took place after together women from different dis- the distribution of female condoms to tricts and ethnic groups, some of assess women’s experiences using whom were involved either officially or them, and the responses of male unofficially in sex work. In Kolda 15 partners. group discussions were held with sep-

120 CHAPTER 3 arate groups for adult women, adult nants. The impact of gender inequal- men and young people. Initially, it had ity on processes of sexual communi- been intended to select a subsample cation was then explored with atten- of study participants who would use tion being focused on factors con- the female condom, but such was the straining and enabling communica- enthusiasm that all group members tion between women and men. were given samples to use. Perceptions of risk were identified before and after the intervention, Research process — One person was followed by participants’ experi- selected from each group to act as ences communicating about and liaison between the research team negotiating the use of the female and the participants. With the help of condom with their partners. Finally, this coordinator, five volunteers from the responses of women and men to each district/target community were the female condom were assessed selected to receive supplies of the along with their implications for pre- female condom on a monthly basis. vention. As a result of the different time scales followed by the study in the two Given the exploratory nature of the areas, participants in Kolda were fol- studies described here, the different lowed up over a six month period and samples selected and the qualitative those in Kaolack over three months. nature of the data collected, it is not Fifteen out of the 50 participants appropriate to report the frequencies dropped out during the follow up with which particular statements were period. These were mainly sex work- made — indeed to do so would run ers who left the area. counter to the established principles of qualitative enquiry. Instead, the Data analysis focus is on the recurrent themes and issues that arose at each site, as well In each of the individual studies, as those that existed across the differ- data were analysed thematically in ent sites in the study. Unless otherwise relation to each of the research stated, the views cited are those con- questions specified above. Thematic sidered by the principal investigator analysis commenced by examining concerned to be typical of the overall existing records, documents, inter- pattern of women’s and/or men’s view transcripts and fieldnotes to responses at that site. Minority and identify prevailing gender relations idiosyncratic views are reported as and their cultural and social determi- such.

121 CHAPTER 4

COMPARATIVE prescribing particular behaviours for men and women, whereas in Mexico ANALYSIS and Costa Rica most women perceived gender roles as being shaped by a male-dominated cultural expec- In this section, data from the four tations. research sites are reviewed to identify what the four studies taken together Irrespective of the ways in which the reveal about the potential of the women saw the reasons for differenti- female condom to contribute to sexu- ated roles, or indeed the merits of this al communication and sexual negotia- differentiation, respondents in all sites tion in ways that are empowering for saw women as more constrained, and women. Data from the four individual perceived female roles as linked to studies are analysed in relation to five expectations that limit sexual behav- key themes: (i) prevailing gender roles iour in a way that men’s roles do not. and relations, (ii) patterns and One woman in Mexico commented processes of sexual communication, that: (iii) perceptions of risk, (iv) the negotiation of safer sex, and (v) the reactions ‘In our society principally our sexual freedom is limited. A woman should be of participants and their partners to married in white, she shouldn’t go out the female condom and its impact on with a number of men because then she gender relations. wouldn’t be a nice girl. The more men who pass through your life the more Prevailing gender roles and devalued your image is.’ relations Comments by women in all four sites In each of the four sites, women par- suggested that female sexuality is ticipating in the study pointed to more constrained and controlled than important gender differences in self that of men. In all the settings where understandings, behaviours, roles and the studies took place, women were expectations. In Indonesia, for exam- expected to have sexual relations ple, women study participants consid- exclusively with one man, while this ered themselves inferior to men, and was clearly not the expectation for deferred to them either out of fear or men in these same cultural contexts. in deference to dominant cultural values (Setiadi and Widyantoro, 1993). In Study participants commented that in Costa Rica and Mexico on the other addition to cultural expectations, the hand, where forms of subordination behaviour of women was also con- may be more subtle, there was evi- strained by their economic dependence that women are beginning to dence on men and the threat of with- question traditional roles and have drawal of this support. In Senegal and not so fully internalized a sense of Indonesia, where polygamy may occur, inferiority. Although distinct roles for the threat of a man taking another wife men and women were apparent in all ensured greater female submission to four sites, the reasons given for these male demands. In Indonesia, women differences varied. In Indonesia, for expressed the view that men will be example, the Islamic faith was seen as unfaithful or take another wife if a

122 CHAPTER 4 woman gives poor service in the committed [married or with a regular domestic and sexual arenas, and that girlfriend or lover] is better than with compliance with his wishes is the best someone who is there only for me, and way of preventing this. One partici- who because of that begins to limit my pant in Senegal highlighted women’s freedom’. economic dependence on men when However, the new freedoms that she told researchers that she always women commented on were framed complied with her husband’s demands by traditional sex roles and the privi- for sex ‘ ... or else he won’t give me leges afforded to male sexuality: money to buy food’. While polygamy is illegal in Mexico, women here ‘I have lived alone for many years. In a described how men commonly have way I have made myself, by myself. more than one family, which limits the With many girlfriends who are profes- resources that can be offered to each sionals I have seen it; we are indepen- woman and encourages greater com- dent and they [men] see us in a com- petitive way... I feel many types of bar- pliance with male wishes. riers; you are a professional and you live alone, they [men] have to be In Mexico, Costa Rica and Senegal “careful” with you ...They still don’t women were increasingly active in understand that it isn’t a competition, paid work outside the home. Women but sharing’. currently constitute a third of the economically active population in Women in this kind of situation are Costa Rica (National Centre for often acutely aware of being caught Women and Family Development, up in a set of contradictions not of 1995) and a similar proportion work their making: ‘As time goes by, your outside the home in Mexico (Garcia, being single is dangerous for married 1993). Women participating in the women, and your girlfriends marginal- study in Senegal reported that in ize you’, commented another woman spite of the traditional sexual division from the non-sex worker sample in of labour whereby men are responsi- Mexico. Even with increased econom- ble for the production of cash crops, ic activity, women reported feelings of women are increasingly involved in frustration in relation to their work and work such as food production, crafts, professional ambitions. One woman clothing and domestic employment. in Costa Rica not untypically com- However, men continue to hold land mented that she ‘wanted to be a jour- rights and have the power to autho- nalist or a physical education teacher. rize whether women can work out- Any profession where I could get out side the home. and about and meet people’.1

Increased economic independence In spite of increasing opportunities for offers women a wider range of sexual paid work, a clear division of labour and lifestyle choices. One woman in was still reported at home. Those Mexico commented that: women who were economically active outside the home still took the greater ‘A clandestine relationship is more share of responsibility for domestic enjoyable, going out with a man who is

1 Ambitions of this nature are influenced by social class as well as gender. The sample of women in Costa Rica included more women with formal education than in other sites.

123 CHAPTER 4 tasks and providing care for others. tional expectations of the female role. Indeed the role of caregiver was a val- One sex worker in Costa Rica com- ued role and a source of pride for mented that she ‘dreamed of making most women interviewed. Moreover, a family, having children, (and) having women in Senegal who had become a man who loves me and looks after economically active outside of the me’. Another woman from the same home were not necessarily less eco- group said ‘I dream of being a mar- nomically dependent on men, and ried lady, a housewife, (and) of finding engagement with commerce could a supportive man who loves me and result in women facing additional sex- my children’. The complex nature of ual pressures. Respondents described gender relations is demonstrated how they were expected to exchange here, in that women hope that men sexual services for commercial bene- will play at least part of their pre- fits such as access to transport, suit- scribed gender role (that of provider), able accommodation and the smooth- while it is also clear that economic ing out of customs formalities. The dependence on men can lock women restaurants and bars run by women into a submissive role demanding were often places where men expect- their compliance with male wishes. ed to receive sexual services. Additionally, young women engaged In two of the sites — Mexico and in domestic work or selling food prod- Costa Rica — some considerable ucts were subject to sexual harass- questioning of traditional gender ment from their employers and cus- roles was apparent, and some women tomers. felt that they had broken out of traditional expectations, although often In Senegal, the registration of sex not without a struggle and not without workers commenced in 1966. paying a price. As a respondent in However, registered sex workers are Mexico put it, still much less common than women ‘We have rebelled. We consider our- offering sexual services on an informal selves less, we let them step on us, but paid basis often out of economic humiliate us, use us. In spite of my necessity. As one such respondent beliefs, I have let myself be stepped on, explained: although I haven’t wanted things to be that way. But we carry a cultural bur- ‘Before I got into this activity, I would den from centuries back which we are serve as a maid in town, the maid’s job not going to get rid of in one genera- could not yield a lot of money in relation, or because you have been to the tion to my needs, sexual activity can university, or because you think a cer- generate nightly the equivalent of a tain way; it’s difficult.’ maid’s monthly pay. At the beginning my parents were reluctant but now I am Such women had clear ideas about my family’s main source of income. My their oppression. As one woman in father is now out of work and it’s my Costa Rica described it, mother, who sells vegetables and fish in front of our house, and myself, who ‘That is what women are for, to be provide for the household expenses’. screwed. At the middle level they see you as the secretary; in the home it’s Some women involved in sex work felt obvious. They see you as the secretary, thwarted in their ambitions by tradi- as the one who screws, “I screwed her and now she is screwed”’.

124 CHAPTER 4

sense of their own autonomy and a Findings from all four countries strong desire for a different kind of demonstrated how power is embed- relationship with men, as was the case ded in the gender relations character- for middle class study participants in istic of each site. Women in all four Mexico, they were reliant not only on research sites were systematically having a high degree of economic socially disadvantaged vis-à-vis men, independence, but also on finding albeit to different degrees and with male partners who were committed to differing implications for their sexual change. In order successfully to relationships. Where women’s levels change gender relations, women of labour-market participation were need not only increased autonomy increasing, such as in Costa Rica and and economic independence but also Mexico, this did not necessarily lead a sense of empowerment and the to economic independence, mainly as experience of acting upon it. To be a result of low wages. However, transformatory in the context of gen- increased access to education and to der relations, empowerment has to be the labour market appears to result in integrated both intellectually and increased expectations about experientially (Holland, Ramazanoglu, women’s place in society and a Scott et al., 1991). greater likelihood that inequalities will be explained as cultural in origin and Sexual communication therefore potentially changeable. Thus, while respondents from Costa There were a number of parallels in Rica and Mexico appear to be less sat- the expectations and relationships isfied with their lot, and in particular underpinning sexual communication with their sexual relationships, this between men and women in each of may be explained as much by a the sites. The most important barrier greater desire for independence as by to open communication about sex more oppressive gender relations. was the belief that talking about sex, and to some degree experiencing Women across all four sites continue to sexual desires and needs, is not have the major responsibility for appropriate for women, or certainly domestic work and child care, and not for women who fear that ques- even those who wished to be free of tions will be raised about their charac- ‘traditional’ roles and definitions of ter and morality. women’s work continued to gain a positive sense of self from being seen as In Indonesia, for example, women caring. Where gender differences were expressed the view that it was not most deeply riven into the culture and appropriate for a woman to discuss religion of a society, however, respon- sex or to express sexual desire openly dents were more likely to accept a def- to her husband. Non-sex workers par- inition of themselves as inferior to men ticipating in the study stated that such and therefore to be less critical, or at behaviour was ‘improper’, that they least to have lower expectations about would be ‘ashamed’, and that they their lives in general and their relation- would be considered to be ‘over- ships with male partners in particular. sexed’. Women in the quite different cultural climate of Costa Rica Even where women had both a strong expressed similar views, giving among

125 CHAPTER 4 other the reasons for not talking with tional social norms that differ signifi- partners ‘He may think I’m a bad girl’ cantly from those found in other sites. or ‘Because it’s vulgar, it is not polite. In Senegal, women’s desire and need Women shouldn’t even think about it’ for sex is acknowledged and can be and ‘I feel shy, inhibited’. In a range of expressed through a variety of sym- contexts, therefore, cultural expecta- bols including wearing special loin- tions about female sexuality and its cloths and pearl belts, and using expression inhibit the open discussion incense and perfumes in the house. of sex. Women in Indonesia were, Although open verbal communication however, able to openly discuss con- is not considered appropriate, the traception with their husbands and in symbols used indicate clearly to men general male partners were described that their wives desire sex. As one as supportive of their wives’ participa- woman put it, tion in family planning programmes, ‘No husband can resist his wife if she at least when female controlled meth- desires to have sexual intercourse and ods were used. uses the set of appropriate means of communication without feeling the Women in each country described need to ask for it by word of mouth, how non-verbal communication was which would be an extremely shameless used to manage sexual interaction, act’. especially that involving expressions of need. In Indonesia, women Cultural norms and economic depen- described ways in which they could dence underpin much of the sexual express an interest in sex: ‘If it’s me communication that takes place who wants it, I pretend to show my between men and women. In each thigh unintentionally, then he will site, women expressed feelings that know’. Women in Costa Rica, while demonstrated a lack of power to saying that they made efforts to ver- refuse undesired sexual activity with bally communicate their desires, said their husbands and partners, and in this was difficult and found it easier to some cases the inappropriateness of use actions rather than words. This refusing to have sex with a partner if illustrates the contradiction for he so wishes. ‘A sane mind cannot women between having sexual needs imagine how a spouse could refuse to met and maintaining a sense of being have sexual intercourse with her hus- a ‘good’ woman. Showing too much band’ commented one study partici- sexual knowledge may cause ques- pant in Senegal, while another said tions to be asked about their propri- that ‘Women must not wear pants in ety. ‘As a woman, we are used to the night in order to make it easy for receiving, the man does everything. their husbands when [they] want to You are afraid they will label you, say- have sexual intercourse’. ing “Why do you know so much? Who have you been to bed with? How do In Indonesia, religious and cultural you know?”’ commented one respon- expectations may mean that women dent in Mexico. do not feel it is correct to refuse sex with their husbands: ‘I don’t feel right Sexual relations in Senegal are under- to refuse because I feel it’s sinful. A pinned by adherence to the Islamic woman is her husband’s property, she faith and a set of prescriptive tradi- is already bought’, commented one

126 CHAPTER 4 respondent. While women in the the wives of men who did not fulfil other sites did not appear to have their economic and other obligations internalized ideas about a man’s right might take a lover without social sanc- to sex, they often complied with men’s tion. However, since men in Senegal desires in order to protect themselves control wealth and decision-making, from economic threat or, in some rarer and since polygamy is widespread, cases, the threat of physical violence. women still find themselves in compe- Women in Mexico expressed a com- tition with each other for limited monly held view that having sex, even resources, which constrains their when they did not want it, would power within sexual relationships with ‘avoid problems later’. Women in men. Indonesia also cited the fear that their partner would go elsewhere for sex as Overall, a clear majority of respon- one reason for compliance. dents at all sites described relationships between men and women char- A clear majority of women inter- acterized by inequality, and expressed viewed in each site reported being the view that they had neither the unable to communicate directly about power to communicate openly about sex in such a way as to influence the sex, nor in many instances the ability form and contexts in which it takes to resist sexual demands that they did place, although some variations were not want to meet. found among some of the more highly educated and economically inde- Women involved in sex work pendent participants in Costa Rica described their relationships with reg- and Mexico. However, adherence to ular partners in terms similar to those strong cultural norms in Indonesia and of women who did not engage in sex Senegal assisted women in resisting work. Paid transactions differed sub- sexual practices they did not want to stantially, however, and open commu- comply with. Most women in nication between partners was much Indonesia, for example, were able to more common. The contractual nature resist oral sex because of cultural and of the encounter between a sex work- religious norms that prohibit it. In er and client meant that women were Senegal, some traditional limitations not so concerned with promoting an to male influence over sex exist. For image of themselves as lacking in example, sexual intercourse should knowledge and experience. only take place at night and not on Registered sex workers in Senegal said nights reserved for another wife. that they expected to control the price Intercourse is also prohibited during charged and the context in which sex menstruation and, for some ethnic occurs. These women usually take groups, during lactation. Outside of clients to their own or a friend’s room, these limits though, force is consid- or to a hotel. Women who are not reg- ered an acceptable means of bringing istered appeared to have less power about a wife’s compliance. In all other to negotiate the form and context of sites, women expressed the fear that the encounter. Clients generally women who refuse to have sex with choose where sex will occur and their husbands may be suspected of expect to pay a lower price for it. having extramarital sex. Interestingly, Unregistered women often do not women in Senegal also described how wish to be seen as prostitutes and

127 CHAPTER 4 avoid being defined as such by not fix- find it easier to set the parameters of a ing prices and hoping that men will sexual encounter that involves pay- show generosity after the event. ment, although this is not true in their Often no money changes hands, personal relationships. Where non-ver- instead women bargain for services bal skills are used, they range from related to their other business deal- complex and relatively formal erotic ings. However, even for sex workers in practices such as in Senegal, through Senegal, symbolic and non-verbal subtle seduction techniques, to the communication is more usual than more common pattern of attempts to explicit verbal negotiation. Sex work- determine the form of a sexual ers participating in the study in Costa encounter by accepting or rejecting Rica commented that while they were certain kinds of sexual advances. asked to offer a wide range of sexual services, the contractual nature of the In all four sites, women had little con- relationship enabled some women to trol over the frequency of sexual con- maintain boundaries, at least with tact and gave in to sex they did not some clients. want in order to avoid problems, ensure economic benefits or to pre- The key issue that emerges from all vent their partner seeking gratification four studies was the extent to which elsewhere. Women rarely referred to gender expectations prevent women such pressures as rape or even coer- from being able to express their sexual cion, but saw their responses as prag- needs and desires directly. matic reactions to the inevitable. Respondents expressed a range of Interestingly, the two countries with reactions from lack of confidence to the clearest gender divisions shame when speaking about sex, appeared to have culturally accept- although in some cases, particularly able strategies of resistance — in among the Indonesian women, dis- Indonesia, women felt they could cussing contraception was more invoke cultural and religious norms to straightforward. Sex not infrequently resist certain sexual practices, where- failed to live up to expectations and as in Senegal it was expected that while women had clear ideas about men’s desires could only be met if how they would like it to be, they were women’s sexual desires and the unable to ask for what they wanted. household’s economic needs were Reasons given ranged from powerful first satisfied. While such resistance is cultural prohibitions against women produced in a context that is not of expressing sexual desire, to seeing any the women’s choosing, and in some problems as their responsibility. Some cases with strong limitations placed women indicated that they would find on what is acceptable, there appears it easier to communicate in more to be a stronger basis for negotiation, relaxed circumstances, but a clear pic- if not for open communication, in ture emerges showing the contradic- these contexts than elsewhere. The tion between being sexually knowing culture of individuality that is more and demanding, and being a ‘good’ prevalent in Mexico and Costa Rica is woman. This tension leads to less direct compromised in the sexual arena by attempts at communication which fall men’s power and greater autonomy, some way short of open discussion or and places women in a weaker nego- negotiation. On the whole, sex workers tiating position.

128 CHAPTER 4

Perceptions of risk ‘good wife’ and therefore a good woman rendered them invulnerable. Levels of knowledge about STDs and The sense that they were protected by HIV/AIDS varied between the groups their status appeared to operate on of women involved in the study. two levels: at the cultural level where Women in Costa Rica, for example, marriage is symbolic of a sacred state, were able to clearly identify STD and and at the level of the relationship HIV transmission routes and methods whereby responsibility for the self is of prevention, while women in invested in the partner. Indonesia more often gave inaccurate answers to questions on these same Women selling sex had no such sense topics. The findings in Indonesia are of being protected. Sex workers consistent with contemporary studies involved in the study in Costa Rica in that same country which indicate said that since they are aware that that most women, even those with a they are at risk of infection, they take high level of education, lack knowl- measures to protect themselves by edge about HIV and AIDS and risks using condoms. However, even here relating to heterosexual transmission there is a question of trust when deal- (see, for example, Setiadi and ing with regular and known clients. Widyantoro, 1993). Sex workers in Mexico had a good understanding of HIV and its transmis- However, even where participants had sion, perceived themselves to be at high levels of knowledge about STDs risk and used a range of strategies to and HIV they did not generally per- insist upon condom use. However, ceive themselves to be at risk of infec- they did not us the same strategies in tion. For example, only four of the relationships with husbands and women in the group of non-sex work- boyfriends, instead ‘trusting to love’ in ers in Costa Rica saw any need to pro- the same way as other women in this tect themselves, while the others sim- study. ply said they were monogamous. These women identified the stability Sex workers in Indonesia did have of their relationships and trust in their some knowledge about STDs and HIV partners as the reasons why protec- and said that under certain circum- tion was unnecessary. Although some stances they would attempt to negoti- of them felt that there might have ate condom use. Although they were been cause for concern in the past, fairly knowledgeable about the symp- they stated that their partner had now toms of STDs, the main strategy they ‘changed’ and in some cases had used to avoid infection was partner been tested for HIV: ‘I have confi- selection, refusing foreign clients or dence in him although it is possible sailors who were perceived to present that he could get infected’; ‘He has a greater risk. Respondents consid- changed and I’m more sure of his con- ered that they were protected from duct although I know he could fall into disease by taking antibiotics and tra- temptation’. ditional medicines on a regular basis, and by always washing and douching For women, ‘trusting to love’ was seen after intercourse. They also reported as part of the role of a wife. There was inspecting clients’ genitals and said an in-built assumption that being a that they would ask for a condom to

129 CHAPTER 4 be used or refuse sex if they thought condoms. The most common of these a client had an infection. This group were to explain that it was good for was also much less likely to suggest both parties, telling the client to look condom use with ‘regular’ clients or for another woman if he did not want ‘friends’, especially if it was assumed to use a condom, and explaining that that the client’s only other sexual part- they might get pregnant. If a client still ner was his wife. resisted, about two thirds of the study participants said that they would con- The research illustrates a range of dif- tinue to refuse to have sexual inter- ferent positions in relation to course, while a third would go ahead HIV/AIDS and STDs varying from a but charge more, although the line high level of knowledge and little or between these two positions was not no sense of risk, through a high sense always clear. of risk with little knowledge or even misinformation, to no knowledge and Within the context of sex work, nego- no sense of risk. Only a minority of tiation about safety and risk is often women in the four studies had a high more open because the exchange is level of knowledge and a sense of risk more explicit: women do not need to proportionate to their sexual behav- protect their reputation in the same iour, and only a small proportion of way as women who are not engaged this group were able to ‘negotiate’ in sex work and are therefore more safer sex on this basis. able to speak more freely. However, sex workers participating in the study in Indonesia said that they were often Negotiating safer sex quite reluctant to speak to clients about potential health risks. Even Not surprisingly given the findings when these women did suggest con- already described, negotiation for dom use, if the suggestion was reject- safer sex in the four sites was limited. ed they tended to comply with the Three principal reasons accounted for wishes of the client. this. First, women did not generally perceive themselves as at risk of STDs In Senegal, both men and women use and HIV infection. Second, open com- symbolic gestures to indicate their munication about sex between men willingness or availability for sex, so and women was limited. Third, women even for sex workers verbal negotia- expressed the view that methods of tion about safer sex was limited. Male prevention were both undesirable and condoms were thought to be unerotic unnecessary within the context of a by study participants in Senegal. As committed and loving relationship. one woman put it, ‘With the men’s contraceptive I do not feel anything Sex workers participating in this study and I do not want my partner to use were most able and most likely to it’. Here, as in perhaps other contexts, negotiate safer sex because of the perceptions of sexual pleasure tended commercial nature of the sexual rela- to override any concern for the pro- tionship and a heightened sense of tection of health. their own vulnerability. Women in both Mexico and Costa Rica used a range of Those most likely to negotiate for strategies to persuade clients to use condom use were sex workers, partic-

130 CHAPTER 4 ularly in Costa Rica, and women more Most of these concerns were allayed generally in Mexico. The common through the experience of using the theme that emerges from respon- female condom. In Mexico one dents not engaged in paid sexual women reported that her initial reser- activities is that of love, stability and vations had faded once she had assumed monogamy providing sym- begun to use it: ‘I didn’t think it would bolic protection. As has been sug- be so safe and so comfortable, it isn’t gested elsewhere, ‘trusting to love’ is even hard to put on, it doesn’t do any- a particularly risky strategy to adopt thing to you, and the pleasure is the (Scott & Freeman, 1995). same’. In all countries, women claimed that curiosity and the novelty Responses to the female of the condom had encouraged men condom to try it out. In Senegal one woman observed that ‘... Men like discovering The principal objective of this multi- tricks, having experiences and new site study was to assess the impact of things as is the case with the women’s the female condom on sexual commu- condom’. nication and ‘negotiation’ between women and men. It is important, how- Most respondents and their male ever, to begin by saying something partners drew favourable comparisons about the acceptability of the female between the female condom and its condom in the particular contexts in male counterpart. In Senegal a which it was introduced, since an women working as a sex worker said: unfavourably received technology is likely to impact less strongly on sexual ‘A lot of clients, once they have tried it communication than will one which is ask me to use women’s condom because more positively evaluated. they do not like the men’s condom which in their opinion decreases vitali- While the female condom was ty and may even make them impotent.’ received favourably by the vast majority of participants at each site, Sex workers gave particularly initial concern was expressed when favourable accounts of the female the female condom was first shown. condom. Listed benefits included Some women were worried by its increased lubrication making sex work unaesthetic appearance, whereas more comfortable and encouraging others were concerned about the clients to ejaculate more quickly; potential reactions of partners. preferable to the use of sponges Among the concerns expressed were when menstruating; and easier to per- worries about anatomy and the suade clients to use. As another sex potential for the condom to cause worker in Senegal reported: harm; the size of the condom and fit; ‘I prefer the women’s condom. Before worries about insertion; and anxieties with the men’s condom I suffered dur- about whether the female condom ing sexual intercourse. I have a narrow would heighten or lessen sexual plea- vagina, I felt so bad and I had things sure. For those women who were not like tears especially ... when the man’s comfortable touching their own geni- drunk and he takes time to ejaculate. tals, the female condom presented Now because of the lubrication of the the greatest challenge. female condom my partners ejaculate

131 CHAPTER 4

rapidly, which tires me less. And that I have gone through (and) that instead of pain I feel smoothness.’ those experiences were not exclusive to me.’ Introducing the female condom in the context of a group setting had a clear Communication skills were crucial in impact on sexual communication gaining male agreement for use. The between men and women. Women in wives of drivers and construction all sites, but most especially those in workers in Indonesia were particularly Mexico and Costa Rica, expressed successful in persuading husbands to feelings of empowerment triggered use the female condom. Women’s by having a form of protection over ability and willingness, developed which they had control. ‘I feel better through group sessions, to assert the knowing I have it, that I’m armed with desire to use protection in the first it, that I have the power’; ‘It is like hav- place was key in sustaining the use of ing control, power. Apart from that it the new device. It is important to feels good’ observed two women par- note, however, that although the ticipating in the study in Costa Rica. women in Indonesia were reticent to Another woman in Mexico comment- raise any issues about sexuality with ed enthusiastically that: their husbands that might cast doubts over their being ‘good women’, issues ‘It feels really great ... You feel really of family planning had been discussed free. Oh, it’s the best invention since birth control! I like it better than the previously, and male support for this male condom because even when you had already been forthcoming. are menstruating you can use it, you can control the situation, you are tak- In Senegal, the female condom was ing care of yourself, you don’t depend perceived by respondents as lying on another person ... You can change firmly within the female domain. As positions without any problem and such, it did not challenge male power begin again’. or virility, which was a positive factor in encouraging men to accept its use. However, some men did express con- Women in all study sites commented cern that the female condom would on how important it had been to learn facilitate greater freedom for women, about the female condom in groups including increased opportunities to and in a generally supportive environ- engage in sex outside of marriage. ment.2 This helped them talk about their experiences and value them- In this same country, women were selves. In Costa Rica one woman said, often able to use powers of persua- ‘... it meant a lot because I have sion even when male partners were learned to value myself more as a per- reluctant to use the female condom: son, as a woman’. Another woman at this same site commented ‘It has been ‘Some time after the start of the follow- important for me to be able to share, up my husband began to grow weary of to be able to study my experiences, to the women’s condom and no longer see that I wasn’t alone in the things wanted to have sexual intercourse with

2 This is not to say that in the absence of group activities the female condom does not offer a useful female controlled prevention technology. It is to suggest that when introduced alongside such activity, its capacity to offer women a way of reducing sexual risk is enhanced.

132 CHAPTER 4

it. I appealed to his feelings and gently Women also reported a greater told him that I was much attached to it degree of male involvement in using because it is an excellent way of con- the female condom than might have traception which does not present any been expected given findings about inconvenience and as we had agreed to poor sexual communication. In each space out my pregnancies, he had to site, a small number of women who accept it. I made tasty dishes for him had some difficulties with insertion and did everything to coax him’. asked for and were offered help by Using all their resources and utilizing partners. This involvement may be local cultural norms, women in very important in encouraging the Senegal were sometimes able to work beginning of a sexual dialogue together with their husband’s other between men and women, and also wives to persuade men to continue to helped men feel that they still had use the female condom. One respon- influence over the form of protection dent told the researcher: used: ‘I told my husband that I had difficulty gripping the internal ring ‘When you came to tell us about the and placing it because of the lubrica- female condom my co-wife was away tion. As he could help me he was then on a trip. When she came back she very proud of succeeding in introduc- heard about it in conversations among ing it into me correctly’ reported one women and informed me of her fear of respondent in Senegal. STDs and AIDS as well as her wish to participate in the follow-up group. The promotion of the female condom Then I asked the researcher on the pro- within the context of a respected ject whether it was possible and it was. research programme also enabled Thus my co-wife and I were accom- women to encourage men to continue plices and my husband could not prevent both of us from using the female use. In addition, in Senegal women condom’. were able to use the dual legitimiza- tion of community support, by argu- Women in Senegal also reported that ing that most local women had taken the female condom had been incor- up the female condom, to influence porated into sexual play between the decision of their partners. husband and wife, often becoming in itself a mechanism for increased sex- Sex workers in all the sites reacted ual pleasure: ‘When my husband more positively to, and reported that wanted to have sexual intercourse their clients also favoured, the female with me ... he went to look for the condom over the male condom. The women’s condom himself, tore the former put more control into women’s sachet and placed it in my vagina with domain, and some sex workers did gestures which pleased me as much not even tell their clients prior to sex as they did him’. The impact of the that they were using the female con- female condom on sex play was dom. In Mexico, sex workers largely extremely important in Senegal imposed the condom on clients rather where the major concerns expressed than negotiating use, although men by the women at the beginning of the could be persuaded because of a study were about whether or not the sense of adventure and novelty: ‘You female condom would enhance sexu- can make them feel ignorant for not al pleasure. knowing it existed and take advan-

133 CHAPTER 4 tage of the novelty of it and the desire of the client to be adventurous and experiment’. In Indonesia, sex workers reported feeling more confident about persuading clients to use some form of protection after the intervention, and that the female condom provided new bargaining power with clients who refused to use the male condom.

134 CHAPTER 5

CONCLUSIONS • with couples where men are already supportive of family planning, such as those in Indonesia;

• where a sense of community Findings from this international study involvement is thought important of gender relations, sexual negotia- and where, through this, men can tion and the female condom suggest be reassured that acceptance is that women are constrained in their high among their peers, as in sexual behaviour and choices by two Senegal; major factors: economic dependence on men which makes it more likely • where the male condom is that women will comply with male unpopular, thus rendering the wishes, and gender stereotypes which female condom a preferable inform expectations of female and alternative; and male sexual behaviour and create problems for sexual communication. • where the female condom can be eroticized and introduced into sex The choices women make about play, such as in Senegal. when, how, where and with whom they have sex are more constrained than those of men. Even with husbands and in the contexts of long- The female condom may have less term relationships, women express impact on sexual communication and reticence in discussing sexual matters negotiation where there is little or no with partners: either because they feel tradition of talking about sex, where this is inappropriate and ‘unfeminine’, women believe themselves to be at or because of fears that such discus- low risk of infection, and where part- sion will be ill received by partners. ners ‘trust in love’ as a means of pro- Although women sex workers find tection against STDs and/or HIV. that the contractual nature of relationships with clients facilitates a more The women in this multi-site study open discussion of sexual matters, reported that men, be they regular some find it difficult to openly com- partners or clients, were with only rare municate and negotiate with clients. exceptions unenthusiastic, if not hos- tile, towards the use of male con- The evidence from this study suggests doms. Male condoms were consid- that the female condom will be most ered to decrease pleasure, pinch, rub successful in enhancing sexual com- and cause men to lose their erections. munication and women’s empowerment in the following contexts: Although some women expressed concern about the decrease in sensi- • with sex workers who already have tivity said to be associated with the some experience of negotiating male condom, the majority of the safer sex with clients, such as those women not favouring the use of the women participating in the male condom in these studies focused research in Mexico and Costa Rica; on men’s problems with condoms and

135 CHAPTER 5 the difficulties of overcoming these. (Mane & Aggleton, 1996; Scott and Findings from this study suggest that, Freeman, 1995), were salient to all the when introduced in a particular con- women taking part in this study. It is text, and when accompanied by difficult for women to accept that they group work and training of the kind may be at greatest risk from those described earlier, the female condom closest to them, especially if this does not have the same negative entails openly acknowledging matters associations that inhibit the use of the it is sometimes more convenient to male condom. ignore. Using the female condom other than for contraceptive purposes Interestingly, some of the women who is likely to entail acknowledging that a were most successful in introducing partner is not trustworthy. This is par- the female condom into their sexual ticularly difficult in situations where relationships were operating in con- monogamy is prized and individual texts where there was least likelihood failure is stressed in the context of a of open verbal communication and partner’s infidelity. It may then be eas- negotiation. They include women sex ier to promote female condom use in workers who simply imposed the social contexts where it is openly female condom on their clients and acknowledged that men may have found that the latter either did not multiple partners, but where they notice it, or were happy not to be resist taking responsibility for this in asked to use male condoms. The relation to their own sexual practices other context in which the female con- and condom use. dom seems to have been relatively easier to integrate was amongst the In all sites, study participants were non-sex worker sample in Senegal. willing recruits with relatively high lev- Here, the important factor was the els of motivation to learn more about legitimacy of women’s engagement in sex, sexuality, contraception and erotic play in pursuit of their hus- health matters or simply to participate bands’ pleasure, which meant that the in some special set of activities female condom could be eroticized focused on their needs. It is clear from and read as an indication of the the data that the majority of women woman’s willingness to have sex. found that the process enhanced their confidence and sense of themselves For the majority of the sex workers as competent social and sexual actors participating in these studies, protect- with rights to some autonomy with ed sex with a regular partner was not regard to their bodies and sexuality. It considered appropriate. This distinc- was also clear that where women were tion was often extended to regular able successfully to introduce the clients who counted as friends. This is female condom into their sexual rela- an understandable distinction and is tions, this in turn increased their sense probably one that, while potentially of empowerment. The success of this risky, is emotionally necessary. particular research/intervention design means that the likelihood of The need for trust in intimate relation- successfully introducing the female ships (Giddens, 1990 and 1991), and condom into these, or any other, set- the contradictions that this raises in tings or situations without group work, relation to the practice of safer sex training and other similar develop-

136 CHAPTER 5 mental work, may not enjoy the same ferently, and they feel good with the level of success. It is clear that the condom. That was when the change most successful interventions to pre- happened and I decided that I can have vent further HIV transmission are a voice and a vote, and it is not only his those that respect the needs of the decision. I have a right to feel loved and desired’. local community and are gender aware (Mane, Aggleton, Dowsett et al., 1996). The female condom also offers women a method with which to pro- For those women who have little or no tect themselves where none existed experience touching their own geni- previously. This increases options for tals, the female condom can present women, especially those who want to some problems with insertion, protect themselves against HIV and although it these difficulties appear to other STDs as well as pregnancy, but decrease with time, may be alleviated who face resistance from men who do with some assistance from willing not want to use male condoms. The partners, and may relate more to anx- female condom also offers an alterna- ieties about the body than to female tive to the male condom which is condom use itself. more dependent on male compliance and not under women’s control. Somewhat paradoxically this study of Follow up work needs to be undertak- Gender relations, sexual negotiation en, however, to ascertain the extent to and the female condom suggests that which the use of the female condom the female condom can be introduced can be sustained over time without with success in certain contexts and the need for wider social and attitudi- circumstances, and without major dis- nal change. ruption to the present balance of power in heterosexual sexual relation- Recommendations ships. However, since the female condom may be used as a tool in the The following more specific recom- development of women’s sexual confi- mendations derive from the findings dence and autonomy, this may open of this multi-site study: up the possibility of greater equality in sexual relations, between men and • The female condom provides women. There are indications that the women with an extended choice introduction of the female condom of means of protection. A number can increase women’s sense of self- of recent studies have demonstrat- efficacy and self worth in ways that ed that extending choice and have effects on their lives beyond the enhancing women’s options immediate issue of condom use. This increases the number of protected possibility for seeing both the inter- sex encounters (Elias & Coggins, vention and the condom as having 1996; Gollub, 1996). Increasing some transformatory value is illustrat- the availability of the female con- ed in the words of one Mexican dom therefore has an important woman: role to play in reducing sexual risk.

‘After the workshops, it’s like you find • There is clear evidence that the more words to talk to them [men] dif- female condom is used within, and

137 CHAPTER 5

integrated into, the context of be an inverse relationship broader, local social norms about between the level of open, verbal sexuality and reproduction. This sexual communication before the has implications for how the intervention and the impact of the female condom is best presented female condom on sexual commu- to women and their partners, and nication afterwards. emphasis the importance of interventions that are context aware • It may be particularly helpful to and context specific. introduce the female condom into environments where there is • Introducing the female condom some history of male condom into a given community is unlikely use, but where this use is charac- to be sufficient to promote its terized by inconsistency and extensive use or encourage any unpopularity. The female con- significant change in sexual com- dom is well received when the munication between men and male condom is an unpopular women. The female condom alternative. needs to be introduced within a context of appropriate training • Interestingly, and somewhat para- and group work which helps doxically, the introduction of the women to build their confidence female condom in a planned inter- and skills in sexual communica- vention may have the dual effect tion. of helping women to feel empow- ered, while not making men feel • The female condom, when intro- threatened. This provides fruitful duced in the context of a planned grounds for further exploration of intervention, can impact on sexual issues around negotiation and the communication and add a new power relations between men and dimension to it. There appears to women.

138 CHAPTER 5

COUNTRY REPORTS Setiadi B, Jatiputra I & Santoso G (1997) Sexual negotiation, the empowerment of women and the female condom - a study with com- Hernandez G, de Caso LE & Aguire mercial sex workers and housewives in VO (1996) Sexual negotiation, female Jakarta. Institute of Applied empowerment and the female con- Psychology, University of Indonesia. dom in Mexico. National AIDS Programme (CONASIDA), Mexico DF.

Contact Dr Bernadette N Setiadi, Consultant and Senior Researcher, Contact Dra Griselda Hernandez Lembaga Psikologi Terapan, Fakultas Tepichin, Coordinación de Centros de Psikologi, Universitas Indonesia, Jl. Información, CONASIDA, Calz. de Salemba Raya 4, Jakarta 10430, Tlalpan No. 4585 2o Piso, Col. Toriello Indonesia Guella, Tlalpan DF, CP 14050, Mexico. ******* ******* Zamora A, Quiros E & Fernandez M Niang CI (1995) Sexual negotiations (1995) Voy paso a paso. and the use of the women’s condom Empoderamiento de las mujeres, in Kolda and Kaolack, Senegal. Institut negociación sexual y el condón femi- des Sciences de l’Environnement, nino. Departamento de Control del Université Cheikh Anta Diop, Dakar, Sida/INCIENSA, San José, Costa Rica. Senegal.

Contact Dra Gisela Herrera, Contact Dr Cheikh Ibrahima Niang, Departamento Control del Sida, Institut des Sciences de Ministerio de Salud, Apartado 6535- l’Environnement, Faculté des Sciences 1000, San José, Costa Rica. et Techniques, Université Cheikh Anta Diop, Dakar, Senegal.

139 CHAPTER 5

REFERENCES Bland L (1985) Guardians of the race or vampires on the nations health. In: Whitlegg E et al., eds. The changing experience of women. Ankrah EM (1996) AIDS, socioeco- Oxford: Blackwell. nomic decline and health: a double crisis for the African woman.’ In: Sherr Cain M & Finch J (1981) Towards a L, Hankins C & Bennett L, eds. AIDS rehabilitation of data. In: Abrams P as a gender issue: psychosocial per- et al., eds. Practice and progress: spectives. London, Taylor and Francis. British sociology 1950-1980. London: Allen and Unwin. Ankrah EM, Kalckmann S, Kabira W et al. (1996) The female condom as a Caldwell JC, Caldwell P, & Quiggin woman controlled protective method. P (1989) The social context of AIDS XI International Conference on AIDS, in sub-Saharan Africa. Population Vancouver, July. development review, 15, 185.

Ankrah EM & Attika S (1997) Adopting Cash K, & Anasuchatkul B (1993) the female condom in Kenya and Experimental educational interven- Brazil: perspectives of women and tions for AIDS prevention among men. Arlington, VA: Family Health northern Thai single migratory International/AIDSCAP. female factory workers. Women and AIDS research program report-in- Baldo M (1995) Youth and AIDS: brief. Washington, DC: International adapting to an evolving epidemic. 6th Center for Research on Women. Congress in Adolescent Health/Youth Health. Vancouver. Cash K, Sanguansermsri J, Busayawong W et al. (1997) AIDS Bang R & Bang A (1992) Why women prevention through peer education hide them: rural women’s viewpoints for northern Thai single migratory on reproductive tract infections. factory workers. Women and AIDS Manushi, 507. research program report-in-brief. Washington, DC: International Benjamin JA (1996) AIDS prevention Center for Research on Women. for refugees: the case of Rwandans in Tanzania. AIDScaptions, 3 (2), 4-9. Chan R (1994) User acceptability of the female condom among sex Bezmalinovic B, Skidmore Duflon W, workers. Unpublished report, avail- Hirschmann A, & Lundgren R (1994) able from Chartex International plc, Guatemala City: empowering a vul- UK. nerable group to prevent HIV transmission. Women and AIDS Research d’Cruz-Grote D (1996) Prevention of Program Report-in-Brief. Washington, HIV infection in developing coun- DC: International Center for Research tries. Lancet, 348, 1071-1074. on Women. Decosas J (1996) HIV and development. AIDS, 10(Suppl.3), S69-S74.

140 CHAPTER 5

Dithan K, Lugada E, Siagi M, et al. ‘Reconceiving Sexuality’ Conference, (1996) Acceptability of the female Rio de Janeiro, 14-17 April. condom in Uganda. XI International Conference on AIDS, Vancouver. George A (1997) Sexual behaviour and sexual negotiation among poor Elias C (1996) Women’s preference women and men in Mumbai. Baroda: regarding the formulation of over-the- SAHAJ Society for Health counter vaginal spermicides. XI Alternatives. International Conference on AIDS, Vancouver. George A, & Jaswal S (1994) Understanding sexuality: ethnograph- Elias C & Coggins C (1996) Female- ic study of poor women in Bombay. controlled methods to prevent sexual Women and AIDS program research transmission of HIV. AIDS, 10(Suppl. report series. Washington, DC: 3), S43-S51. International Center for Research on Women. European Study Group on Heterosexual Transmission of HIV. George A & Mane P (1996) (1992) Comparison of female to male Acceptability of the female condom: a and male to female transmission of review of select studies. (Draft) HIV in 563 Stable Couples. British Geneva: WHO Global Programme on medical journal, 304, 809. AIDS, Unit for Social and Behavioural Studies and Support Unit. FHI/AIDSCAP (1997) The female condom: from research to the market- Giddens A (1990) The consequences place. Arlington, VA.: Family Health of modernity. Oxford: Polity. International/AIDSCAP. Giddens A (1991) Modernity and self Fontanet AL, Saba J, Chandeging V et identity. Oxford: Polity al. (1998) Increased protection against sexually transmitted diseases by Gindin LR (n.d.) Acceptability of the granting sex workers in Thailand the female condom (Femy) (unpublished choice of using the male or female study conducted in Argentina; available condom in a randomised trial. AIDS, from Chartex International Plc, UK). 12,1851-1860. Goldstein D (1994) Culture, class and Ford N & Mathie E (1993) The accept- gender politics of modern diseases: ability and experience of the female women and AIDS in Brazil. Women condom, Femidom® among family and AIDS program research report planning clinic attenders. British jour- series, Washington DC: International nal of family planning, 19, 187-192. Center for Research on Women.

Garcia A I (1993) Latin American Gollub E (1996) The women’s safer sex women in numbers. Spain: FLACSO hierarchy: interim data. Initial responses to counseling on women’s methods George A (1996) Gender relations in of STD/HIV prevention at an STD clin- urban households in Bombay: chal- ic. XI International Conference on lenges for HIV/STD prevention. AIDS, Vancouver, Canada.

141 CHAPTER 5

Gollub E, Stein Z & El-Dadr W (1995) women and the diversity of condom Short-term acceptability of the female use. In: Aggleton P, Hart G & Davies P. condom among staff and patients at a AIDS: responses, interventions and New York Hospital. Family planning care. London: Falmer Press. perspectives, 27,155-158. Holland J, Ramazanoglu C, Scott S, du Guerny J & Sjoberg E (1993) Inter- Sharpe S & Thomson R (1992) Pressure, relationship between gender relations resistance, empowerment: young and the HIV/AIDS epidemic: some women and the negotiation of safer possible considerations for policies sex. In: Aggleton P, Davies P & Hart G, and programmes. AIDS, 7, 1027- eds. AIDS: rights, risk and reason. 1034. London: Falmer Press.

Gupta GR & Weiss E (1993) Women Holland J, Ramazanoglu, C, Sharpe S & and AIDS developing a new health Thomson R (1992) Pressured pleasure: strategy. International Center for young women and the negotiation of Research on Women Policy Series. sexual boundaries. London: The Tufnell Washington DC: International Center Press. for Research on Women. International Center for Research on Gupta GR, Mane P & Weiss E (1997) Women. (1996) Vulnerability and Talking about sex: a prerequisite for opportunity: adolescents and AIDS prevention. In: Long L & Ankrah HIV/AIDS in the developing world. EM, eds. Women’s experiences with Washington, DC.: International Center HIV/AIDS: an international perspec- for Research on Women. tive. New York: Columbia University Press. Jackson S & Scott S (1996) Feminism and sexuality: a reader. Edinburgh: Heise LL & Elias C (1995) Transforming Edinburgh University Press. AIDS prevention to meet women’s needs: a focus on developing coun- Jenkins C et al. (1995) A study of the tries. Social science and medicine, acceptability of the female condom in 40(7), 931-943. Urban Papua New Guinea. Unpublished report; available from Hoffman K (1991) Acceptability of Chartex International Plc, UK. Femidom® among family planning clients in Germany. Unpublished Labonte A (1997) Workshop on gen- report; available from Chartex der, health and development. International Plc, UK. Facilitator’s guide. Washington, DC.: Pan American Health Organization Holland J, Ramazanoglu C & Scott S (1990) Sex, risk and danger: AIDS Lehto H (1991) Femidom® User education policy and young women’s acceptability study in Finland 1990-91. sexuality. London: The Tufnell Press. Unpublished summary report; available from Chartex International Plc, UK. Holland J, Ramazanoglu C, Scott S, Sharpe S & Thomson R (1991) Between Lemus B (1992) Mexican agencies embarrassment and trust: young reach teenagers. Network, 13(1), 14-15.

142 CHAPTER 5

Longwe SH (n.d.) Gender equity and Musaba E, Morrison CS, Sunkutu MR women’s empowerment. Harare/New (1996) Long-term use and acceptabil- York: UNICEF. ity of the female condom among couples at high-risk of HIV in Zambia. XI MacIntyre JA et al (1996) The user International Conference on AIDS, acceptability of the female condom in Vancouver, July. South Africa. South Africa, Reproductive Health Research Unit, Mushingeh C, Chama WK, & University of the Witwatersrand. Mulikelela DM (1991) An investigation of high-risk situations and envi- Mane P & Aggleton P (1996) Shifting ronments and their potential role in paradigms in sexuality research: a per- the transmission of HIV in Zambia: the spective gained through cross national case of Copperbelt and Luapula studies of gender/power. Paper pre- provinces. Report prepared for the sented at the meeting on Reconceiving Population Council of Zambia. Sexuality: Interpreting Sexuality Research, Rio de Janeiro. 18-19 April. National Centre for Women and To appear in Parker R Barbosa R & Family Development (1995) National Aggleton P, eds. Framing the sexual report on women’s situation in Costa subject: studies in gender, sexuality and Rica 1985-1994. San José, Costa Rica. power. Berkeley, CA., University of California Press. Orubuloye IO, Caldwell P, & Caldwell JC (1992) African women’s control Mane P, Aggleton PJ, Dowsett G et al. over their sexuality in an era of AIDS. (1996) Summary of track D (XI Health Transition Working Paper International Conference on AIDS, No.12. Canberra: Australian National Vancouver) Social Science: research, University. policy and action. AIDS, 10 (Suppl.3), S123-S132. Palloni A, & Lee YJ (1992) Families, women and the spread of HIV/AIDS Mane P, Gupta GR & Weiss E (1994) in Africa. United Nations population Effective communication between part- bulletin, 33. ners: AIDS and risk reduction for women. AIDS, 8 (Suppl.1): S325-S331. Panos Institute (1996) AIDS and Young People. AIDS briefing, 4. Mann J & Tarantola D, eds. (1996) AIDS in the world II: global dimensions, social Panos Institute (1990) Triple roots and responses. NY.: Oxford Jeopardy: Women and AIDS. Panos University Press. dossier, London: The PANOS Institute. McNamara R (1991) Female genital health and the risk of HIV transmission. Perry MJ, Sikkema K, Wagstaffe D et Center for Population and Family al. (1996) Perceptions and use of the Health, Columbia University (unpub- female condom among inner-city lished manuscript). women. Poster presentation (Th.C. 4504) at the 11th International Conference on AIDS, Vancouver, Canada.

143 CHAPTER 5

Preston-Whyte E (1995) “Bring us the Sweat MD & Denison J (1995). female condom”: HIV intervention, Reducing HIV incidence in developing gender and political commitment in countries with structural and environ- two South African communities. mental interventions. AIDS, 9(Suppl. Paper presented at the John C. A), S251-57. Caldwell Seminar, Australian National University, Canberra August 14-17. Tansathit T & Cheevakj S (1990) Femshield acceptability study among Ray S, Bassett M, Maposhere C, family planning acceptors (a pilot Manangazira P, Nicolette JD, study). Chiang Mai University, Machekano R & Moyo J (1995) Thailand, Faculty of Medicine. Acceptability of the female condom in Zimbabwe: positive but male-cen- Timyan J et al. (1996) Test marketing tered responses. Reproductive of the reality female condom in four Health Matters, 5: 494-503. developing countries: Bolivia, Guinea, Zambia and South Africa. Washington Sai K (1993) The female condom, DC, Population Services International. international agencies explore new option for HIV prevention. UNAIDS (1997) New UNAIDS World AIDScaptions, 1(1), 28-29. AIDS Day report warns that HIV epidemic is far worse than previously Setiadi B & Widyantoro (1993) thought. Press Release, 26 November. Women and AIDS. Jakarta: Institute of Applied Psychology, University of UNAIDS & World Health Organization Indonesia. (1996) HIV/AIDS: The global epidemic, December 1996. Geneva, Joint Scott S & Freeman R (1995) United Nations Programme on Prevention as a problem of moderni- HIV/AIDS. ty: The case of HIV/AIDS. In Gabe J, ed. Medicine, health and risk: socio- United Nations Development logical approaches. Oxford: Programme. (1992) Young women: Blackwell. Oxford: Blackwell. silence, susceptibility and the HIV epidemic. New York: Author. Shervington D (1993) The acceptability of the female condom among low UNDP/UNFPA/WHO/World Bank income African-American women. Special Programme of Research on Journal of the National Medical Human Reproduction (1997) The Association, 85, 5. female condom: a review. Geneva: World Health Organization, Special Sivard R L (1985) Women: a world Programme of Research on Human survey. Washington DC: World Reproduction. Priorities. WHO (1993) General research protocol Soper DE, Shoupe D & Shangold GA for studies of sexual negotiation, the et al. (1993) Prevention of vaginal tri- empowerment of women and the chomoniasis by compliant use of the female condom. Geneva, World Health female condom. Sexually transmitted Organization/Global Programme on diseases, 20, 137-139. AIDS.

144 CHAPTER 5

WHO (1994a) The HIV/AIDS pandemic: An overview. Geneva, World Health Organization.

WHO (1994b) AIDS: images of the epidemic. Geneva, World Health Organization.

WHO (1994c) Women and AIDS: agenda for action. Geneva, World Health Organization.

Williamson NE & Joanis C (1994) Acceptability of barrier methods for prevention of unwanted pregnancy and infection. In: Mauck CK, Cordero M, Gabelnick H, Spieler JM & Rivera R (eds) Barrier contraceptives: current status and future prospects. New York: Wiley-Liss.

Working Group on Sexual Behaviour Research, (1995) Gender, sexuality and health: building a new agenda for sexuality research in response to AIDS and reproductive health. Rio de Janeiro: AIDS and Reproductive Health Network.

145 Notes: The Joint United Nations Programme on HIV/AIDS (UNAIDS) is the leading advocate for global action on HIV/AIDS. It brings together seven UN agencies in a common effort to fight the epidemic: the United Nations Children’s Fund (UNICEF), the United Nations Development Programme (UNDP), the United Nations Population Fund (UNFPA), the United Nations International Drug Control Programme (UNDCP), the United Nations Educational, Scientific and Cultural Organization (UNESCO), the World Health Organization (WHO) and the World Bank.

UNAIDS both mobilizes the responses to the epidemic of its seven cosponsoring organizations and supplements these efforts with special initiatives. Its purpose is to lead and assist an expansion of the international response to HIV on all fronts: medical, public health, social, economic, cultural, political and human rights. UNAIDS works with a broad range of partners – governmental and NGO, business, scientific and lay – to share knowledge, skills and best practice across boundaries.

Produced with environment-friendly materials Joint United Nations Programme on HIV/AIDS

UNICEF • UNDP • UNFPA • UNDCP UNESCO • WHO • WORLD BANK

Joint United Nations Programme on HIV/AIDS (UNAIDS) 20 avenue Appia, 1211 Geneva 27, Switzerland Tel. (+4122) 791 46 51 – Fax (+4122) 791 41 65 e-mail: [email protected] – Internet: http://www.unaids.org