SA CRIME QUARTERLY No 13 SEPTEMBER 2005 7 A third theme is that the demographic change brought about by the epidemic, specifically an overrepresentation of adolescents and young adults in heavily affected populations, will create problems. Schonteich speculates that because young men are most likely to commit crime, a disproportionate number of young men between the age of 15 and 24 in severely affected countries may lead to higher levels of crime – particularly violent crime and group-based aggression.2 Others argue that by straining social institutions like the labour market and educational system, ‘youth bulges’ resulting from either HIV/AIDS or fertility trends may make countries generally more unstable and prone to violence.3 These arguments suggest that HIV/AIDS will exacerbate crime in two main ways: • that children set apart and damaged by AIDS related orphanhood will be disproportionately more likely to engage in criminal and violent behaviour than other children; • that the epidemic will increasingly create an environment conducive to crime. It is the latter that most accurately frames the issues under discussion. T he literature linking growing numbers of orphans to crime suggests three main ways in which AIDS related orphaning may lead to higher levels of victimisation. The first and most common argument is that the illness and death of parents will leave children scarred and marginalised in ways that predispose them to delinquency and criminal behaviour. Proponents believe that growing poverty, together with the emotional trauma associated with multiple AIDS related losses and stigma, reduced levels of parental care, and the loss of positive role models will place children at high risk of developing antisocial tendencies. The second argument is that growing numbers of orphans will provide a recruitment pool for individuals and organisations wishing to violently challenge the existing socio-political order in African countries. According to Cheek, a swell of young people without family care and formal schooling may constitute an “extra national” population group vulnerable to co-optation into socially disruptive activities and ethnic warfare, which if exploited “could effectively destabilise most countries in southern Africa”.1 Robyn Pharoah, Institute for Security Studies rpharoah@gmail.com AIDS, ORPHANS AND CRIME Exploring the linkages In the hardest hit regions of the world, the HIV/AIDS epidemic is increasing poverty and inequality and reversing decades of improvements in health, education, and life-expectancy. It is also leaving millions of children orphaned and living in situations of acute vulnerability. Yet, even as the international community mobilises in support of these young people, some researchers and practitioners are linking orphaning and crime, suggesting that growing numbers of impoverished orphans may pose a threat to individual and communal security in some countries. This connection is generally presented as a neat, linear relationship. But is this the case? SA CRIME QUARTERLY No 13 SEPTEMBER 20058 PHAROAH AIDS orphaning in context On its current trajectory, the epidemic stands to leave millions of children orphaned. It is likely that many will be left in situations of intense hardship. Common consequences of orphaning including growing poverty and its correlates, the loss of parental affection, reduced levels of care, stigma, and the psychosocial implications of repeated personal and material losses such as trauma, stress, depression, and a loss of social connectivity (Figure 1). Such loss and growing vulnerability are obviously undesirable, but it is questionable whether, in the African context, they make children orphaned by AIDS a ‘special case’ among large numbers of other vulnerable children. As shown in Figure 2, war and poverty have already created large numbers of orphans in sub-Saharan Africa. The numbers vary according to how orphans are defined.4 Using an expansive definition of orphanhood (children under the age of 18 who have lost one or both of their parents), the latest estimates by the Joint United Nations Programme on HIV/AIDS (UNAIDS), United Nations Children’s Fund (UNICEF) and the United States Agency for International Development (USAID) suggest that there may be as many as 43 million orphans living in sub-Saharan Africa, of whom only about 12 million are thought to have lost parents to AIDS.5 Numerous studies also show that children in southern Africa have frequently ‘lost’ parents through the physical and social movements associated with migrant labour and fluid marital and partnership arrangements.6 Even when parents are alive, fostering, or the care of non-biological children whose parents live elsewhere, is common. Data from the South African Project for Statistics on Living Standards and Development survey (SALSS) conducted in 1993, for example, shows that roughly 17% of African children between the age of six and 19 were living apart from their biological parents, while 12% of coloured and just under 5% of Indian and white children were fostered.7 National estimates from elsewhere in Africa suggest similarly high levels of fostering, with as many as one in five children living apart from their parents in Namibia, Zimbabwe, Zambia, and Malawi.8 Figure 1: Problems among children and families affected by HIV/AIDS Source: J Williams, presentation to the US Council on Foreign Affairs, April 2005 HIV infection Increasingly serious illness Children may become caregivers Deaths of parents and young children Children without adequate adult care Discrimination Exploitative child labour Sexual exploitation Life on the street Psychosocial distress Children withdraw from school Inadequate food Problems with shelter and material needs Reduced access to health services Increased vulnerability to HIV infecion Problems with inheritance Economic problems SA CRIME QUARTERLY No 13 SEPTEMBER 2005 9 The implications of AIDS related illness and death are seldom confined to households who lose members to the epidemic. Foster argues that children not immediately touched by the virus may feel the effects of HIV/AIDS when families provide money to support sick relatives, their mothers leave home to provide care, or their standard of living deteriorates as their family takes in orphaned children.9 As the epidemic takes hold, they may also be affected as government services and structures tasked with providing for vulnerable children become overstretched,10 or economies affected by the epidemic provide fewer jobs. High levels of illness and death may also have an insidious psychosocial impact. Killian notes that South African children living in severely affected areas are excessively anxious about death and often reflect obsessively about illness and mortality.11 Such dynamics, together with the high levels of poverty that exist in many of the communities worst affected by the epidemic, mean that few of the above problems are confined to children who lose parents to AIDS. As Ramphele notes, the loss and absence of parents, insecurity, and emotional trauma characterise the lives of many poor children in Africa: Whereas the family is supposed to create a safe haven in life’s troubled waters…uncertainty permeates family life in a manner that is difficult for outsiders to comprehend. The family unit can not be taken for granted and the availability of a mother, let alone both parents, is a luxury few children enjoy…the provision of basic needs is beyond many, and trusting and respectful relationships are an exception rather than the rule.12 This is not to say that children immediately affected by HIV/AIDS do not frequently suffer enormous difficulties. Studies show that they often experience deepening poverty and considerable psychosocial stress. Many also show, however, that these effects vary according to factors such as age and gender, whether children are maternal, paternal or double orphans, whether they are taken in by the extended family, and which relatives then provide care. Several studies similarly indicate that children orphaned by AIDS are often no more disadvantaged than poor children in comparable circumstances. This suggests that the implications of orphanhood vary according to context, and that the boundaries between children orphaned by AIDS and other orphans and vulnerable children are frequently blurred. There are only a handful of studies on the implications of AIDS related illness and death for the psychological, emotional, and social adjustment of children and, as noted by Wild, our knowledge is based on “an intermingling of sound data, less reliable data and clinical data and is therefore somewhat less secure than it might appear at a first glance”.13 The few African studies available suggest that children whose parents have AIDS or have died of AIDS tend to experience more anxiety and depression than other children, but are no more prone to delinquency.14 The available evidence therefore suggests that although children orphaned by AIDS are negatively affected by their parent’s death, there is little about these children that should make them disproportionately more likely to turn to crime and violence than other poor children. Rather than focusing on whether children orphaned by the epidemic pose a peculiar threat to stability and security then, we should look at how HIV/AIDS may Figure 2: The estimated number of orphans in sub-Saharan Africa, 1990 – 2010 0 60 40 20 50 70 30 10 1990 M il li o n s 28 1 1995 33 2000 39 2003 43 2010 50 31 9 12 18 Source: Children on the Brink, 2004 PHAROAH Total number of orphans Number of orphans due to AIDS SA CRIME QUARTERLY No 13 SEPTEMBER 200510 PHAROAH create an environment in which the deepening poverty and vulnerability of a larger group of children, together with demographic change, encourage greater levels of criminality. Linking HIV/AIDS and crime Criminologists acknowledge that pinpointing the ‘causes’ of crime is a difficult undertaking. However, the available literature suggests that there are likely to be strong correlations between the dynamics triggered by the epidemic and crime. Factors like material need, social exclusion, unemployment, poor education, and family breakdown, for instance, lie at the heart of many of the prevailing theories of why individuals commit crime. High levels of inequality are also closely associated with victimisation – and may in fact be more consistently correlated with crime than poverty.15 Researchers working in South Africa have found that “inequality is highly correlated with both burglary and vehicle theft”,16 while research in the United States suggests that economic disparities may foster frustration and anger that contributes to violent crime.17 This relationship between inequality and crime has been explained using the concept of relative deprivation, which breeds social tensions so that “the poor seek compensation and satisfaction by all means, including committing crimes against both poor and rich”.18 Less directly, factors such as urbanisation and its correlates – which could be exacerbated by the growing economic hardship associated with the epidemic – have also been linked to higher levels of criminality the world over. At the micro level, there are a number of relevant personal, family, and environmental variables. As noted by Schonteich, biographical factors such as age and gender are closely correlated with criminality, with official arrest and victimisation figures from around the world showing that most crime is committed by young men.19 The relationship between crime and age is particularly strong. As Smith notes, “probably the most important single fact about crime is that it is committed mainly by teenagers and young adults”.20 American data for the years 1980, 1994 and 2000, for example, shows that arrest rates for both violent and property crime increased dramatically amongst adolescents in their early teens, peaked around the age of 18 and then decreased continually after the age of 20.21 There are also a range of purely social variables associated with a greater propensity towards criminal behaviour. The most relevant include: • family variables, such as growing up in a single parent family, poor levels of supervision, having family members who are involved in criminal behaviour and exposure to strife, violence, and abuse; and • schooling variables, including a lack of formal schooling, failing or dropping out of school, as well as exposure to overcrowded and unsupportive school environments.22 Impact of HIV/AIDS on crime rates Given these correlates, it is likely that the demographic change, growing levels of poverty and inequality, and compromised service delivery resulting from the HIV/AIDS epidemic will be associated with higher levels of crime. To date, though, there is scant evidence to support these conclusions. Most countries in southern Africa are in their third decade of the epidemic and their epidemics have matured to the point that large numbers of people are dying.23 The US Census Bureau estimates that average life expectancy may have already halved in Botswana and Zimbabwe from an expected 70 years to 39 and 38 years respectively.24 The population structure of some of the worst affected countries has deviated from the expected pattern and, although expected to worsen in the future, countries like Botswana may be seeing a hollowing out of their population structures as a result of rising death rates among adults in their thirties, forties, and fifties (Figure 3). Available estimates also suggest that orphan numbers are increasing. In the Southern African Development Community (SADC) alone, it is estimated that approximately seven million children have lost either one or both of their parents to AIDS since 1990, and as many as two million have lost a parent to the epidemic since the turn of the millennium. SA CRIME QUARTERLY No 13 SEPTEMBER 2005 11 Other sources, such as the UNODC Survey of Crime Trends and Operations of Criminal Justice Systems (which collects official police data from UN members states) place Botswana’s victimisation rate second to South Africa and comparable to Swaziland (5,207 per 100,000 compared to 7,997 and 4,803 per 100,000), although such findings may be more reflective of better reporting and recording of crime in South Africa and Botswana than actual levels of victimisation.28 Statistics on homicide, which are often considered to be the most accurate measure of crime due to higher recording rates (and a body that must be accounted for), suggest that Botswana has less of a crime problem than many of its neighbours. Recent Interpol statistics give it a murder rate of 13 per 100,000, on a par with Swaziland and behind South Africa (43) and Lesotho (51).29 These statistics have limitations – including issues of comparability, representivity and accuracy30 – but suggest that while there is likely to be an In 2003 the number of children living without one or both of their parents as a result of AIDS ranged from 120,000 in Namibia to 4.2 million in the Democratic Republic of Congo (DRC); the number of children to have lost both parents ranged from 19,000 in Namibia to 420,000 in Zimbabwe.25 Despite this, it is difficult to pinpoint a significant impact on crime – although a paucity of reliable data makes such seemingly simple comparisons more difficult than one might expect.26 The results from the International Crime and Victimisation Surveys (ICVS) conducted by United Nations Interregional Crime and Justice Research Institute (UNICRI) suggest that despite the makings of an ‘orphan crime wave’, countries like Botswana do not show comparably higher levels of victimisation than other African countries for which data are available. The results indicate a total victimisation rate of 32% of the population in Gaborone, compared to rates of 41% and 38% for Maputo and Johannesburg respectively.27 PHAROAH Figure 3: The demographic profile of Botswana with and without AIDS, thousands, 2005 Source: B Epstein, The demographic impact of AIDS, US Census Bureau, undated 0-4 50-54 30-34 10-14 40-44 80+ 70-74 60-64 20-24 140 120 100 80 60 40 20 0 20 40 60 80 100 120 140 A ge ( in y ea rs ) Without AIDS With AIDS Males Females association between the HIV/AIDS epidemic and crime, this relationship is not a simple, linear one. There appear to be a number of reasons for this. The first is that crime is a complex phenomenon. For instance, although poverty is often a motivating factor, it is not true that the poorest societies have the highest rates of crime or that the poorest people necessarily commit the most crime.31 Similarly, although inequality is more consistently correlated with crime than poverty, economic disparities are not always associated with crime.32 Conflict theorists like Davies attribute this to the relativity of deprivation. He suggests that although the prevalence, duration, and degree of deprivation can help to predict the likelihood of conflict, the likelihood of violence is more closely linked to thwarted expectations than prolonged general poverty.33 ‘Relative deprivation’ may therefore be gauged in relation to one’s own past affluence.34 The second is that, even when young people are exposed to particular risks, they will not necessarily turn to crime. Criminologists recognise that even when levels of offending are high, the decision to engage in crime remains an individual choice35 and exposure to micro-level risk factors does not necessarily condemn a child to problems in later life.36 Like criminology, the burgeoning literature on risk and resilience is far from clear-cut, but work by Garbarino and others in the United States shows that the impact of risk factors is highly dependent on children’s environment, and it is only when three or more risk factors combine with an overwhelming and unsupportive environment that children are likely to become delinquent or violent.37 The implications of exposure to deprivation and other sources of risk are mediated by factors such as personality and temperament, coping style, age of exposure, and the availability of caring adults and social supports in children’s environment. The likelihood of long-term maladjustment is therefore dependent on the availability of conditions for recovery as much as the form, number or severity of precipitating stresses.38 Even low levels of support in childhood appear to enable children to overcome severe disadvantages,39 SA CRIME QUARTERLY No 13 SEPTEMBER 200512 PHAROAH and it is estimated that less than one third of children raised in situations of poverty and deprivation are affected negatively by these experiences.40 The third reason why the relationship between the HIV/AIDS epidemic and crime is not a simple, linear one is that crime is linked to both the opportunities for committing crime and the cost and likelihood of being caught. Even when levels of poverty and vulnerability are high, the likelihood of individuals breaking the law is mediated by factors such as social norms concerning the acceptability of crime, the availability of firearms and other weapons, and the strength of a country’s criminal justice institutions. Countries with weak gun or border controls and ineffective criminal justice systems, for example, are likely to experience higher levels of crime than those in which guns are harder to obtain and criminals stand a greater chance of being caught and punished.41 The heterogeneous nature of crime Another factor complicating efforts to predict the effects of HIV/AIDS on crime is what Leggett refers to as “the diversity of human behaviours that fall under the general heading of ‘crime’”.42 Human beings commit crime for a wide range of reasons and different types of crime have different motivations. Someone who steals food in order to survive or to earn a living, for instance, may be motivated by poverty, while public violence may be linked to factors such as relative deprivation, historical rivalries or economic frustration. Greed-motivated or pathological crimes, on the other hand, are more closely associated with individual variables that have nothing to do with either poverty or the inequitable distribution of wealth. Violent crime may also be, at least in part, linked to factors such as the availability of weapons and attitudes to violence. How the epidemic impacts on crime will thus be bound up with how the effects of the epidemic play themselves out in particular settings, the prevailing macro-economic and social environment, as well as SA CRIME QUARTERLY No 13 SEPTEMBER 2005 13 structural factors such as the availability and acceptability of firearms or other weaponry. Future scenarios may thus unfold in a multitude of ways. It could be, for example, that the correlation between crime and the epidemic is strongest in urban areas where economic stress and disparities are felt more keenly and city living loosens traditional social ties. In poor rural communities, where economic disparities are less marked and prevailing norms reject interpersonal violence, the effects of the epidemic might be confined to petty and non-violent property crime. The stresses associated with the epidemic could also result primarily in higher levels of domestic violence or increased levels of rape and sexual assault rather than other forms more publicly oriented violence. While this is undoubtedly an undesirable outcome, it is unlikely to threaten society in the way envisaged in much of the literature. Conclusion There is very little empirical data available to test the links between the HIV/AIDS epidemic and crime; and a great deal more is needed before definitive conclusions about these links can be drawn. A broad reading of the criminological literature, however, suggests that the relationships involved are not simple ones. There is evidence to suggest that demographic change, growing levels of poverty and inequality, and compromised service delivery may contribute to higher levels of crime, but it is difficult to predict the magnitude of these effects. Given appropriate support and viable opportunities for economic and social inclusion, children are unlikely to turn to crime. Labelling children as a security risk without properly understanding the nature of the links between the epidemic and crime stands to increase the stigma and discrimination to which children orphaned by AIDS are already exposed, and may exacerbate rather than ameliorate security concerns. It is thus important to better understand the nature and parameters of this so- called threat. Only then can we design and implement the vital programmes to support the most vulnerable. Acknowledgement This article is based on a recent paper published by the Institute for Security Studies: R Pharoah and T Weiss, AIDS, orphans, crime and instability: Exploring the linkages, ISS Paper 107, June 2005. Endnotes 1 R Cheek, A generation at risk: Security implications of the HIV/AIDS crisis in southern Africa, National Defence University, Institute for National Strategic Studies, Washington DC, 2000, p 5. 2 See for example, M Schonteich, HIV/AIDS, policing and crime in South Africa: Exploring the impact, draft working paper for the CSIS Task Force on HIV/AIDS, CSIS, February 2003, p 23. 3 See, for example, R Cincotta, R Engelman & D Anastasion, The security demographic: Population and civil conflict after the cold war, Population Action International, Washington DC, 2003, pp 42-44. 4 Some estimates use data for children under the age of 15, while others expand their definitions to include young people under the age of 18, with the literature variously reporting on ‘maternal orphans’ whose mothers have died, ‘paternal orphans’ whose fathers have died and ‘double orphans’. 5 Children on the brink 2004: A joint report of new orphan estimates and a framework for action, UNAIDS, UNICEF and USAID, July 2004, p 4. 6 R Bray, Predicting the social consequences of orphanhood in South Africa, CSSR Working Paper, No 29, Centre for Social Science Research, University of Cape Town, February 2003, p 9. 7 Cited by L Richter, the impact of HIV/AIDS on the development of children, in R Pharoah (ed), A generation at risk? HIV/AIDS, vulnerable children and security in southern Africa, Institute for Security Studies monograph series, no. 109, ISS, December 2004, p 18. 8 M Ayad, B Barrere & J Otto, cited in G Foster, Safety nets for children affected by HIV/AIDS in southern Africa, in R Pharoah (ed), A generation at risk? HIV/AIDS, vulnerable children and security in southern Africa, Institute for Security Studies monograph series, no. 109, ISS, December 2004, p 68. 9 Foster, ibid, p 67. 10 J Williamson and G Foster, cited by J Stein, Sorrow makes children of us all: A literature review on the psychosocial impact of HIV/AIDS on children, CSSR working paper, no. 47, Centre for Social Science Research, University of Cape Town, 2003, p 5. 11 B Killian, Risk and resilience, in R Pharoah (ed), A generation at risk? HIV/AIDS, vulnerable children and security in southern Africa, Institute for Security Studies monograph series, no. 109, ISS, December 2004, p 40. 12 M Ramphele, Steering by the stars: Being young in South Africa, Tafelberg publishers, Cape Town, 2002, p 154. PHAROAH SA CRIME QUARTERLY No 13 SEPTEMBER 200514 PHAROAH 13 L Wild, cited in Stein, op cit, p 9. 14 See, for example, T Marcus, Living and Dying with AIDS, Report prepared for the Children in Distress Network (CINDI), July 1999; and Stein, op cit, pp 9–10 for a review of available studies. 15 P Fajnzylber, D Lederman and N Loayza, Inequality and violent crime,The journal of Law and Economics, pp 25–26, 20 February 2004; T Leggett, The relationship between poverty, inequality and crime in South Africa, unpublished paper prepared for the office of the South African presidency, November 2002, p 7. 16 G Demombynes and B Ozler, Crime and local inequality in South Africa, paper presented to the Development Policy Research Unit/Freidrich Ebert Stiftung Labour Markets and Poverty Conference, October 2002, Johannesburg, pp 25–26. 17 R Fowles and M Merva, cited in Leggett, op cit, p 4. 18 Fajnzylber et al, op cit, pp 2–3. 19 T Newburn, cited in A Maree, Criminogenic risk factors for youth offenders, in C Bezuidenhout and S Joubert (eds) Child and youth misbehaviour in South Africa: A holistic view, Van Shaik publishers, Pretoria, 2003, p 68. 20 D Smith, Youth crime and conduct disorders, in M Rutter & D Smith (eds), Psychological disorders in young people: Time trends and their correlates, Wiley publishers, Chichester,1995, p 395. 21 M Ezel and L Cohen, Desisting from crime: Continuity and change in long-term crime patterns of serious chronic offendors, Clarendon Studies in Criminology, Oxford University Press, Oxford, 2005, pp 1–2. 22 See, for example, K Booysens, the relative nature and extent of child and youth misbehaviour in South Africa, in Bezuidenhout and Joubert, op cit; A Maree, op cit, pp 55–59; Youth risk factors, Centre for research on Youth at Risk, 2002, 2 March 2004; P Goldblatt, Comparative effectiveness of different approaches, in P Goldblatt and C Lewis (eds), Reducing offending: an assessment of research evidence on ways of dealing with offending behaviour, Home Office Research Studies, No 187, 1998, p 123. 23 In the absence of treatment there is generally a five to ten year lag between the time that people contract HIV and become ill and die of AIDS. The first cases of HIV and AIDS were diagnosed in most southern African countries in the early to mid-eighties, and HIV/AIDS epidemics appear to have been well- established in the region by the mid-90s. 24 US Census Bureau, Population Reference Bureau, UNAIDS and WHO, 6 March 2004. 25 Children on the brink 2004: A joint report of new orphan estimates and a framework for action, op cit, appendix 1, table 1. 26 National crime statistics are gathered from both official police statistics on levels of reported crime and, in some countries, victimisation surveys, which ask a sample of the population about their experience of crime. Police statistics are hampered by the universal tendency for such statistics to undercount levels of crime. Statistics from either source are only available for about half of all African countries and are difficult to compare due to varying definitions of crime and, in the case of victimisation surveys, different units of analysis, methodologies, and time periods. 27 The latest ICVS data is summarised in the Human development report 2004: Cultural liberty in today’s diverse world, United Nations Development Programme (UNDP), New York, 2004, pp 215–216. 28 United Nations Surveys of Crime Trends and Operations of Criminal Justice Systems covering the period 1990–2000, United Nations Office on Drugs and Crime, Division for Policy Analysis and Public Affairs, available at < http://www.unodc.org/unodc/en/crime_ cicp_surveys.html> 1rst April 2004. 29 Crime and development in Africa, UNODC, June 2005, p 56. 30 The ICVS has been conducted in several African countries, but surveys have often been conducted in different years and have been confined to a single major city in each country. The available police statistics are also hampered by the universal tendency of such statistics to undercount levels of crime. For more on the problems of comparing crime rates internationally, see A Altbeker, Puzzling statistics: Is South Africa really the world’s crime capital?, SA Crime Quarterly No 11, March 2005. 31 See, for example, Leggett, op cit, p 3. 32 Leggett, ibid, p 4. 33 J Davies, cited in Leggett, ibid. 34 T Szayna, cited in Leggett, ibid. 35 Leggett, ibid, p 1. 36 Maree, op cit, p 73; Youth risk factors, op cit, p 2. 37 J Garbarino, cited by M Roper, Kids first: approaching school safety, in E Pelser (ed), Crime prevention partnerships: lessons from practice. Institute for Security Studies, 2002, p 70. 38 Richter, op cit, p 23. 39 K Hundeide, cited in Richter, ibid, p 22. 40 Richter, ibid, p22. 41 See, for example, Crime and development in Africa, op cit, p 17; T Weiss, Guns in the borderlands: Reducing the demand for small arms, ISS Monograph series, no. 95, January 2004, p 107. 42 Leggett, op cit, p 13.