Apartheid’s trauma persists through present conditions and contested memory more than through inheritance

Introduction

Talk of intergenerational trauma has become almost automatic in South Africa. Point to a township with high levels of violence, or to young people who seem angry or stuck, and someone will say that apartheid trauma is being handed down. The idea has real pull. Apartheid meant forced removals, pass laws and state violence, and it only ended in 1994, so plenty of today’s parents and grandparents lived through it themselves (Keet, 2021; Pretorius et al., 2021). Their children are living with what came after, and psychologists in places like Gauteng now routinely talk about a transgenerational legacy in the born free generation (Crankshaw & Dwarika, 2023). Whether that talk rests on solid ground is the thing worth checking.

But the phrase hides at least three claims that get run together. One is biological, that stress in one generation leaves a mark on the next. Another is social, that families and communities carry harm through how they live and through what goes unsaid. The third is political, that the past keeps being reproduced because present conditions have barely shifted. This paper argues that the evidence supports the second and third claims much better than the first, and that apartheid’s trauma is better understood as sustained by current conditions and contested memory than as something inherited in a literal sense (Kim et al., 2023; Pentecost, 2021). I start with what the transmission evidence actually shows, move on to the biologising of trauma and the critique of historical trauma frameworks, then look at ongoing conditions, and finish with memory, silence and the stories families tell, where Gobodo-Madikizela (2024) and Adebayo (2022) offer the most useful ideas.

What the transmission evidence shows

The strongest South African study of transmission is also a fairly cautious one. Kim et al. (2023) drew on the Birth-to-Twenty cohort in Soweto, where pregnant women were enrolled in 1990, during the last violent years of the apartheid state, and their children were assessed again at 17 to 18. Full data were available for 304 mother-child pairs. Maternal prenatal stress was not directly linked to psychiatric morbidity in the adolescents. What the authors found instead was that the link showed up for children of younger mothers and for young people dealing with more household adversity at the time, while social support did not buffer the effect (Kim et al., 2023). The authors read this as stress-sensitising, meaning earlier stress may make later hardship hit harder. That is quite different from saying trauma was passed down, and it matters that the effect appears only when current adversity is also present.

Qualitative work paints a wider and messier picture. Pretorius et al. (2021) asked eight caregivers, three black, two white and three of mixed race, to look back on being children under apartheid and on caring for children now. The socio-ecologies they described ran from pathology and intergenerational trauma through to indifference and reconcilement, and the authors suggest the effects are likely to be intergenerational. Eight people is a very small sample, the study is a short four-page report, and these are adults remembering decades back, so the study can point to a range of outcomes but cannot say how common any of them is. Still, the range itself tells us something. If the harm were handed down in a uniform way, we would expect less variation than this, and Keet (2021), working from documents on Helenvale, Alexandra and Lavender Hill, also describes exposure to trauma as one theme among several rather than the whole story.

Outside South Africa the position is not much firmer. Nagata et al. (2024) review the historical trauma literature for Indigenous, African American and Asian American groups in the United States, and they set out the critiques of the concept before reviewing the evidence, ending with what they call benefits, limitations and complexities. Gone (2023) is blunter on one point, noting that the mechanisms of cross-generational transmission remain unknown, with proposals ranging from the spiritual to the epigenetic. So South African researchers are working in a field where transmission is often asserted more confidently than it has been demonstrated. That does not make it false. It does mean a claim like ‘apartheid trauma is inherited’ needs to be held loosely, and I think the Soweto cohort data, with their conditional pattern, are about as far as the evidence currently lets us go.

Biologising trauma and the critique of historical trauma

The risk of leaning on inheritance is not only empirical. Pentecost (2021) spent time in Khayelitsha, Cape Town, in 2014 and 2015 and watched violence prevention being framed more and more in epigenetic terms, with a focus on maternal investment as a way of stopping cycles of violence passing between generations. She argues against merging trauma with intergenerational epigenetics, because doing so produces a newly biologised view of the bodily marks of apartheid history, and that, as she puts it, is itself a re-inscription of damage and a form of violence (Pentecost, 2021). Her alternative is collective intervention built around accountability and social justice. There is a fair reply available here. Kim et al. (2023) are asking a legitimate developmental question about prenatal stress, and a finding is not responsible for how a programme later uses it. Even so, Pentecost’s worry about what happens when mothers become the site of repair looks reasonable to me, especially where the finding is as conditional as the Soweto one.

The wider critique of historical trauma sharpens the point. Gone (2023) traces the concept to Brave Heart in 1995 and describes it as colonial, collective, cumulative and cross-generational. When he and colleagues reviewed studies that linked it to health outcomes in Indigenous samples in the United States and Canada, they found two main ways of measuring it. Nineteen studies used the Historical Losses Scales and eleven used residential-school ancestry, and in both groups the results were mixed and sometimes contradictory. A pilot trial of group interpersonal psychotherapy with a historical trauma module did no better on depression than the control condition (Gone, 2023). On the conceptual side he lists essentialism, pathologisation, a shift of attention from material causes to individual therapy, and the awkward problem of attribution, since it is hard to tell whether present distress is caused by past trauma or is simply explained by it afterwards. Nagata et al. (2024) also treat critiques as part of the story and not an afterthought.

What South Africa changes is mostly about time and about who is in the frame. Gone is writing about Indigenous communities in North America, where the main dispossession lies further back in time. In South Africa the adults who were pregnant in 1990 are alive and often still raising children, so what looks like ‘historical’ and ‘intergenerational’ trauma blurs into their own direct exposure and into what followed (Kim et al., 2023). The population is also not one colonised minority. Pretorius et al. (2021) included white caregivers as well as black and mixed-race ones, and who counts as a victim of the apartheid past is actually quite contested (Adebayo, 2022). A framework built for a single dispossessed group fits awkwardly here. I would keep the language of legacy, since it names something real, but I would be careful about borrowing the whole historical trauma package without asking what each part is doing.

Present conditions do much of the work

If inheritance is thinly evidenced, the more plausible explanation for continuing harm is that the conditions apartheid built have largely stayed in place (Keet, 2021; Pentecost, 2021). Both studies describe a present in which marginalisation did not simply stop with the 1994 elections. Keet (2021) is the clearest on this. Her document analysis of Helenvale, Alexandra and Lavender Hill identified four themes, forced removals or relocations, social problems, exposure to intergenerational trauma, and ongoing cumulative marginalisation, and she argues that poverty and inequality are key drivers of violence in such communities. That fourth theme is basically the whole argument in miniature. Marginalisation that is cumulative and ongoing is a present fact, and a social worker in one of these places meets it every day regardless of what anyone’s grandparents went through. The limit is that a descriptive study built on documents cannot separate inherited distress from continuing deprivation, so it shows the entanglement and not the proportions.

The Soweto data point the same way. Kim et al. (2023) found that the prenatal stress effect was tied to household adversity in late adolescence, which implies that what looks like carried-over trauma may often be a sensitivity that current hardship switches on. Gone’s (2023) attribution problem fits here too, because a researcher who knows about apartheid and sees a distressed young person in Soweto can always read the distress as inheritance. Pentecost (2021) adds something more useful. In Khayelitsha, she suggests, violence shapes how people anticipate the future and how generations relate to one another now, and it does so through the present arrangement of risk, not only through a past event. That gives intervention a different target. Housing, safety, schooling and services start to look like trauma work in their own right, which is what her call for collective accountability seems to imply.

None of this means the trauma vocabulary should be dropped. Gone (2023) argues that historical trauma is probably better seen as a kind of moral rhetoric than as a generative scientific construct, since the ambiguity about causes is unlikely to be settled by data, and he accepts that softer terms such as postcolonial distress might weaken its political force. I think that trade-off is real in South Africa, where trauma language does support claims for redress, and Keet (2021) uses it to argue for social workers who understand how forced removals shaped the communities they serve. The fairest position is to use it for that purpose while being plain that, as a causal claim about inheritance, it has not been shown. Calling a political and economic injustice a family pathology would be a poor outcome of a good intention.

Memory, silence and who claims injury

The social route is more convincing, and some of the best South African work on it dispenses with biology altogether (Adebayo, 2022; Gobodo-Madikizela, 2024). This part of the literature is mostly conceptual, so it earns its place through the quality of its argument and not through numbers. Gobodo-Madikizela (2024) takes the concept of Nachträglichkeit, the idea that earlier events gain their meaning retrospectively, and uses it to move beyond a straight line from past to present. She applies it to the affective responses of young black students when they interact with white peers, and to the Truth and Reconciliation Commission’s public testimonies, including the scream at the opening of the hearings, arguing that these help us see a present in which the past seems to replay itself. On that reading the past is not stored in bodies and handed on. It is reactivated and given new meaning in encounters, such as a seminar room where a young person is read through a history they did not live. It is a less deterministic way of talking about transmission, although as a conceptual and interpretive piece it cannot tell us how common this kind of reactivation is.

Practitioners seem to be working with something similar. Crankshaw and Dwarika (2023) interviewed South African psychologists about the born free generation in Gauteng and found that they identified transgenerational trauma through stuckness paired with guilt, grief resulting from silence, and certain difficulties with identity and relationships. The study also points to certain approaches being useful at individual, group, family and community level. Silence as a mechanism is the interesting part, because families that do not talk about the past seem to be shaping what children understand without any need for a biological story. The caution is that the markers psychologists use are broad. Guilt, stuckness and identity struggles turn up in many young people for many reasons, and a clinician who already expects apartheid’s legacy may find it, which is Gone’s (2023) attribution worry in another form.

Adebayo (2022) widens the lens from families to the whole society. Reading contemporary South Africa through melancholia, Adebayo argues that structural traumas and moral anguish have not been worked through, that identity groups make different, sometimes disputed, claims to victimhood, and that melancholia takes white, black and intergenerational forms. The position taken is that if we do not rush to pathologise it, this affective structure might help resist hurried attempts to close the door on the past (Adebayo, 2022). That cuts against Gone’s (2023) concern that trauma framing can pathologise whole groups, and it is not clear the two can be reconciled. My own view is that this reads as a description of a shared, contested memory culture and not a clinical condition, which is exactly the level where the South African evidence is strongest.

One more study, from outside the country, is useful for what it complicates. Khanna and Priya (2024) interviewed parent survivors and their children after the 1984 anti-Sikh violence in India and found suffering moving in both directions. Parents’ remorse over being unable to care, and the inadequacy of resources or support, shaped children’s experience, while parents’ healing was helped by their children’s growth and happiness. The findings did not clearly show healing passing from parent to child (Khanna & Priya, 2024). The context is very different, so I would not stretch the result. But it hints that born free South Africans are not just recipients of a legacy, and that how they are doing may in turn shape how older generations live with the past.

Conclusion

Apartheid harmed people in ways that are real and that continue, and nothing in this paper doubts that. The question is what carries the harm forward. The Soweto cohort suggests earlier stress may sensitise young people to later hardship but not that trauma passes down by itself (Kim et al., 2023). The critics of biologised and historical trauma frameworks show that measurement is inconsistent, that attribution is ambiguous and that a medical vocabulary can pull attention away from structure (Gone, 2023; Pentecost, 2021). Studies of communities, practitioners and memory point instead to continuing marginalisation, silence in families and a contested public memory as the more convincing carriers (Keet, 2021; Gobodo-Madikizela, 2024).

There are limits to my argument. Most of the South African qualitative work is small, interpretive and often based on documents or on professionals’ impressions (Crankshaw & Dwarika, 2023; Pretorius et al., 2021), and the one longitudinal study covers a single township cohort (Kim et al., 2023). So absence of strong evidence for inheritance is not evidence of its absence. What follows in practice is fairly clear, though. Research should test conditions and transmission together instead of assuming the second, services should treat housing, safety and schooling as part of healing, and the language of legacy should be kept for what it can plainly claim. If trauma is mostly being re-made by the present, then the present is where most of the work lies.

Reference list

Adebayo, S. (2022). Post-apartheid melancholia: Negotiating loss and (be)longing in South Africa. Social Dynamics, 48(2), 275-293. https://doi.org/10.1080/02533952.2022.2099175

Crankshaw, A. T., & Dwarika, V. M. (2023). Exploring psychologists’ interventions for transgenerational trauma in South Africa’s born free generation. Trauma Care, 3(4), 212-236. https://doi.org/10.3390/traumacare3040020

Gobodo-Madikizela, P. (2024). The time of Nachträglichkeit and the afterlife of apartheid trauma. The International Journal of Psychoanalysis, 105(5), 766-777. https://doi.org/10.1080/00207578.2024.2403235

Gone, J. P. (2023). Indigenous historical trauma: Alter-Native explanations for mental health inequities. Daedalus, 152(4), 130-150. https://doi.org/10.1162/daed_a_02035

Keet, A. (2021). A revision of social work practice: Conceptualising cultural trauma and the role of social workers in South African under-resourced communities. Southern African Journal of Social Work and Social Development, 33(1), Article 6859. https://doi.org/10.25159/2708-9355/6859

Khanna, A., & Priya, K. R. (2024). Rethinking ‘transfer’ in the transgenerational transmission of trauma: A qualitative study of the 1984 anti-Sikh violence. Theory & Psychology, 34(1), 18-39. https://doi.org/10.1177/09593543231204576

Kim, A. W., Said Mohamed, R., Norris, S. A., Richter, L. M., & Kuzawa, C. W. (2023). Psychological legacies of intergenerational trauma under South African apartheid: Prenatal stress predicts greater vulnerability to the psychological impacts of future stress exposure during late adolescence and early adulthood in Soweto, South Africa. Journal of Child Psychology and Psychiatry, 64(1), 110-124. https://doi.org/10.1111/jcpp.13672

Nagata, D. K., Kim, J. H. J., & Gone, J. P. (2024). Intergenerational transmission of ethnoracial historical trauma in the United States. Annual Review of Clinical Psychology, 20, 175-200. https://doi.org/10.1146/annurev-clinpsy-080822-044522

Pentecost, M. (2021). The politics of trauma: Gender, futurity, and violence prevention in South Africa. Medical Anthropology Quarterly, 35(4), 441-457. https://doi.org/10.1111/maq.12678

Pretorius, J., Petersen, C. A. L., & Pedro, A. (2021). The socio-ecology of childcare in apartheid South Africa: Retrospective experiences of a sample of caregivers. Journal of Psychology in Africa, 31(1), 98-101. https://doi.org/10.1080/14330237.2020.1871248