Remarks by Mohammed Bhabha, Director of the In Transformation Initiative, delivered in Rome on 26 September 2026 at a conference on education, cohesion and post-conflict reconstruction hosted by the Carter Center — setting the stage for Professor Motala’s presentation.
I. Opening
We are not here to tell you how South Africa rebuilt education after apartheid. Thirty years on, we are still trying to rebuild it. The work is unfinished, and some of it has not yet begun. We would be dishonest to present it otherwise.
Our task is a narrower one. We are here to set the stage for what Professor Motala will deliver, so that her analysis is better understood. What apartheid education actually was. What we did, and did not do, when we had the chance to dismantle it.
Because apartheid education was not neglect. It was not underdevelopment. It was not the incidental by-product of a racist society.
It was policy. A deliberately engineered system designed to produce inequality and to keep it in place. And its architects said so, openly.
We are conscious that many in this room are working through transitions of their own. We do not presume to know what travels and what does not. We offer our experience as one story among many, in the spirit of exchange, not instruction.
II. The design: education for labour, not citizenship
In 1953, the Bantu Education Act placed African education under direct state control. By the 1970s, government spending per pupil on a white child ran roughly ten times higher than on a Black child. That was not a gap that emerged by accident. It was the intended output of the Act’s funding formula.
Its architect, Hendrik Verwoerd, told Parliament exactly what it was for:
“There is no place for him in the European community above the level of certain forms of labour.”
That is the design specification. A deliberate ceiling on Black aspiration, enforced through schooling, and paid for, deliberately, at a tenth of the cost. Policy success, judged by its own stated aims.
The architects understood what was at stake. As early as 1946, J.G. Strijdom warned that racial discrimination would become impossible to sustain if Black education kept improving. Education was a threat to apartheid. So it had to be limited, controlled, and pointed toward labour.
The same year as the Bantu Education Act, a companion labour law stripped Black workers of the right to organise. One law closed off advancement through schooling. The other closed off the leverage organised labour might have won back. Together, by design, they locked a generation into a fixed economic role.
And this was reinforced by spatial law, language law, and a funding formula. Apartheid education was not one policy. It was a system, engineered as a whole, and legislated as a whole.
In hindsight, that is the first thing we underestimated. When a system is engineered as a whole, it cannot be un-engineered in pieces. And our settlement was not designed to un-engineer anything. It was designed to transfer power.
There is one more thing to say about apartheid education, and it is the reason it mattered so much. Education was not simply one of the sites where apartheid was enforced. It was the place where the system was most fiercely defended, because it was the place where the future would be decided. Schools were where children of different races would meet each other, and meet the fears the system had spent decades cultivating. That is what apartheid most feared. Not integration in the workplace. Integration in the classroom, where the next generation was being formed. And it was what Black South Africans sought most, alongside dignity and land and the vote. Not just access to a better school. The right to shape what a shared country would look like. That is why schools became one of the most contested arenas of the whole struggle.
III. Language
Language was not a side issue. It was one of the sharpest instruments of that system.
The imposition of Afrikaans as a medium of instruction in Black schools was the immediate trigger for the 1976 Soweto uprising. Language policy in education is never neutral. It is either an act of inclusion or an act of domination. Afrikaans was the language of authority. English was the language of aspiration. African languages were left to the home and the playground, never to knowledge, power or advancement.
It did not disappear in 1994. It persists today. And in hindsight, we treated language as a technical matter that would resolve itself once access expanded. We assumed access would do the work. It did not.
IV. The spatial architecture: where the classroom was
Apartheid’s deepest effect on education was not only what happened inside the classroom. It was where the classroom was.
The Group Areas Act and the Bantustan system fixed, by law, where people could live according to race. That spatial architecture determined which schools existed, what resource base supported them, and what quality of teaching and infrastructure they could sustain.
This matters today because South Africa’s Gini coefficient remains above 0.63, among the highest in the world. Spatial apartheid did not disappear in 1994. It became the geography democratic South Africa inherited. Where you were born under apartheid still predicts, with striking consistency, the quality of schooling available to you now.
In hindsight, this is the deepest regret. We negotiated a political settlement. We did not negotiate the geography. Property was protected. Existing land use was preserved. The townships and the suburbs stayed where apartheid had put them.
We left the map alone.
The map itself was part of the architecture of oppression. And we did not touch it.
V. Decolonisation of a special type
Now, the transition itself. And this is the point that shapes everything else.
South Africa’s transition was not won through civil war. Apartheid ended through a negotiated settlement. And that meant the institutions apartheid had built did not disappear in 1994. They remained standing. Reformed, repurposed, renamed. But not razed.
This is what we mean by decolonisation of a special type. A negotiated transition that spared us a war, but not the burden of an intact, inherited architecture of exclusion.
Other colonised countries in the region had their own cruel systems of education for African people. Zimbabwe and Eswatini are examples. As cruel as those systems were under colonial rule, they were not as diabolical as Bantu Education. They were segregated, underfunded and unjust. But they did not carry the same explicit, legislated ceiling on Black aspiration. They did not set out, as a matter of deliberate policy, to prevent the emergence of an African professional and administrative class altogether. Zimbabwe, in particular, continued to produce educated people of note—teachers, nurses, civil servants, professionals, academics. Its education system, even under strain, still commands respect in the region. That is not a defence of colonialism. It is a distinction that matters for what came after.
Bantu Education was different in kind. It was a totalising design, announced in Parliament, with an explicit ceiling on Black aspiration written into legislation and enforced by a funding formula. Verwoerd said so himself. There was no place for the African above certain forms of labour.
That statement was the point of the system. And its consequence was this. It was designed not only to exclude Black South Africans, but to prevent the emergence of a Black professional and administrative class altogether. So when 1994 arrived, the state that was supposed to deliver transformation found itself short of the teachers, officials, planners and managers that the work required.
The state was, in an important sense, also a victim of the system it was asked to dismantle. Asked to cure a disease with an immune system that disease had already weakened.
Not because the people were not capable. Because the system had spent generations making sure that capacity would not be there. The failure was not theirs. It was designed into the world they were born into.
The administration that had to deliver the new education was itself a product of the old one. The very instrument of reform carried the legacy of what it was reforming.
This is the legacy that lasts for generations. It is a structure we are still living inside, still governing through, still trying to reform without the luxury of starting over. The harm was not undone by the transition. It was inherited by it.
What made South Africa’s inheritance distinctive was not only that apartheid education was diabolical in design. It was that the negotiated settlement left the institutional architecture of apartheid standing. We did not dismantle it. We inherited it.
We have been honest about the cost of that. The institutions we inherited were not dismantled, and we are still living with the consequences.
But the most important thing the settlement achieved was that we avoided a civil war. That has to be said plainly, because it is easy to lose sight of it when you are listing what went wrong. No scorched earth. No generation lost to fighting. No country torn apart along the lines that apartheid had spent decades deepening.
Whatever the settlement cost us in terms of inherited architecture, it spared us that. And it is why, thirty years on, we are arguing about how to rebuild education, rather than arguing about how to end a war.
VI. Exceptional measures, exceptional expectations
South Africa in 1994 was not a normal policy environment. A near-universal exclusion of the majority from quality education could not be corrected by ordinary incrementalism.
That is why the measures taken were, and had to be, extraordinary. Constitutional entrenchment of the right to education. Redistributive funding formulas. No-fee schooling now covering two-thirds of learners. And they achieved a great deal. That is not in dispute.
But exceptional measures also create an expectation of exceptional results, on a normal timeline. In hindsight, we spoke in the language of a generation, and the work required generations. We should have been more honest, earlier, about that.
VII. Devolution: local voice and its limits
After 1994, South Africa devolved responsibility for schooling to the provinces, and gave School Governing Bodies real power over admissions, language policy and resourcing. This was a democratic choice. It was meant to bring governance closer to communities, and to allow cultural and linguistic expression in schools.
That principle was right. Communities should have a voice in how their children are educated. Language and culture should not be dictated from the centre.
But two problems emerged.
The first was asymmetrical development. Provinces started from radically unequal positions, because apartheid had concentrated resources, universities, hospitals, industry and skills in a handful of places and left the rest underdeveloped. The responsibility arrived before the capacity could be built. And the provinces most damaged were the ones asked to deliver the most. Devolution did not correct that asymmetry. In important respects, it gave it a new engine.
The second was that local control can protect cultural expression, and it can also become a tool for exclusion. A school can use language policy to keep out learners who do not speak the language of the community that controls it. A governing body can use admissions policy to preserve the character of the school rather than to open it.
The two risks pull against each other. Devolve less, and you weaken local voice. Devolve more, and you increase the risk of exclusion and of reproducing the asymmetry you inherited.
We did not resolve that tension at the start. We are still working on it. But one thing we would say to anyone designing a settlement now. These safeguards are easier to build in from the beginning than to retrofit later. We learned that the hard way.
VIII. Cohesion and its limits: the non-racial settlement
This is where the question of cohesion comes closest to the surface.
South Africa’s settlement chose non-racialism as its founding principle. That choice was deliberate, and it was about cohesion. Apartheid had spent decades insisting that South Africans belonged to different nations, with different rights, different futures, different worth. Non-racialism said the opposite. One nation. One law. Equal rights for every citizen from the day the Constitution was adopted.
That was the mechanism of cohesion. It made a single country possible where apartheid had built a hierarchy of separate ones. And it worked in ways that matter. It held the centre when many feared the country would not hold at all.
But there is a tension inside any settlement of this kind. A rights-based, non-racial settlement protects everyone equally from the moment it is adopted. That includes those who already hold an advantage. It guarantees that no one may be treated worse because of race going forward. It does not, on its own, change what race already determined in the past.
Redress asks the state to act with intention and resources. Non-racialism asks it to act, in law, as though everyone is already equal.
And this has a consequence for cohesion that we are still living with. A nation can be united in law and divided in fact. If the conditions of people’s lives do not change, the unity of the law begins to feel like a promise that was not kept. And cohesion becomes fragile precisely because it was real. People believed in it.
We chose non-racialism because we believed in a shared country. We still do. And in the absence of a better answer, the old answers quietly persisted. The inherited architecture supplied one. That is why the architecture of cohesion must be built alongside the law, not after it. It is a design choice with consequences, not a flaw to be engineered away.
IX. Access without identity
And then there is the cultural dimension. Perhaps the most difficult of all.
Even as access expanded, the terms of success remained narrowly defined. Over thirty years, families who could afford it faced a choice the transition had never resolved. The public system still carried the deepest legacy disadvantages. The well-resourced schools offered something closer to what any parent wants for a child. So those who could pay, paid. And that created an interdependence neither side can easily break. The schools depend on the fees. The families depend on the schools.
But look at what those schools ask of the child who enters them. Not only better facilities and better results. A particular cultural paradigm. English-speaking. Eurocentric in its canon. Individualistic in its orientation. The message is rarely spoken but always present. To succeed here, you must become more like us. Your language belongs at home. Your community’s ways of knowing are not academic.
The research tells us what this does. Young adults who went through these schools describe negating their own mother tongue. Studies document racial insensitivity, the privileging of white standards of beauty, the erasure of cultural significance. Not because anyone told them their culture was inferior. Because the environment was arranged in a way that made them feel it.
This is the perverse outcome. The very institutions that offer escape from the legacy of Bantu Education reproduce a different form of it. Not the denial of access. The denial of self.
We did not anticipate this. We assumed that if access expanded, affirmation would follow. It did not.
Access without identity is not full inclusion.
X. Closing and handover
We said at the outset that apartheid education was not neglect. It was policy, sustained with intention over generations. Rebuilding it will have to be the same kind of undertaking. Policy, sustained with intention, over generations. Not a project with an end date.
And it will have to begin with an answer to the question we have carried through this conversation. Not only what to build, but what education is for. Because if that question is left unanswered, the old answers will fill the space.
We do not know whether any of this is useful to anyone else. We offer it only because transitions are lonely, and the experience of others, however partial, can sometimes help. What we have tried to share is not a set of answers. It is an honest account of what we attempted, what we underestimated, and what we would have done differently if we had known then what we know now.
That is the terrain Professor Motala’s work will now measure in detail. She will show you, with rigour, where that story stands today, and what thirty years of reform has, and has not, been able to undo.
The question she leaves us with is not only what to build. It is what to dismantle. Each country will make its own choices. Ours are simply on the record.
Professor Motala. The floor is yours.