Haus der Kulturen der Welt

Beyond Nuremberg
The Historical Significance of the Post-Apartheid Transition in South Africa

Mahmood Mamdani is the Herbert Lehman Professor of Government at Columbia University and director of the Markerere Institute for Social Research (MISR), Kampala. His work explores the intersection between politics and culture, the modern state and the colonial subject, and the theory, history, and practice of human rights.

Mahmood Mamdani

I recall a conference ten years ago with Edward Said here at Columbia [University]. It was a conference we organized at the Institute of African Studies and it was called “An Anti-Apartheid Perspective on Israel and Palestine.” Edward was to be the last speaker in the conference, he called me up that morning and said he wouldn’t be there for the first session, because he was going to the hospital for chemotherapy. But after chemotherapy he would come and talk. So I asked him, “Are you sure you’d have the energy to talk after chemotherapy?” He said, “Yes, for Palestine.”

I think it was the last conference that he participated in.

As it was then so today South Africa is a way of talking about Palestine. They are different and yet not all that different.

The contemporary human rights movement holds up Nuremberg as a template with which to define responsibility for mass violence. I will try and understand how Nuremberg was ideologized as part of human rights activism beginning in the late 1970s. Mine is also an argument that a critical appreciation of the post-apartheid transition in South Africa suggests that we should consider a move beyond the logic of Nuremberg.

The human rights movement that gathered steam in the late 1970s anchored itself ideologically in the lessons of the Holocaust and presented itself as a post-Nuremberg movement. [Professor of European Legal History] Sam Moyn has recently suggested that human rights were born as an alternative to grand political mission, as a moral criticism of politics. I will try to connect the moral and the political, the ethical and the historical, through a discussion of two responses to two crimes against humanity: first the criminal trials known as Nuremberg and then the political discussions that took place in South Africa and paved the transition from apartheid. The discussions were known as the Convention for a Democratic South Africa (CODESA).

The first thing about the contemporary human rights movement is that it anchors itself ideologically in the lessons of defeat, not of revolution—as was the case with the human rights movement that followed the French Revolution. Whereas the movement organized around the revolutionary banner—Rights of Man—was highly political, the contemporary human rights movement is consciously anti-political, which is the meaning it gives to the notion of human and humanitarian. Nuremberg redefines the problem and the solution. The problem is extreme violence—radical evil—and the question it poses is that of responsibility for the violence. The solution which we know as Lessons of Nuremberg, is to think of violence as criminal, and of responsibility as individual. State orders, we are told, cannot absolve officials of individual responsibility. Above all, this responsibility is said to be ethical, not political.

Could one argue that the lesson of the transition from apartheid is the opposite? Should extreme violence be thought of more as political than criminal? I was part of an audience one grey morning in Cape Town when the Truth and Reconciliation Commission (TRC) questioned F. W. de Clerk. De Clerk had just read out a statement enumerating the wrongs of apartheid and concluded by taking responsibility for apartheid. But the TRC was not interested. Its interest was narrowly focused: on specific human rights violations such as murder, torture, and kidnapping: did de Clerk know of these? Had he authorized any of these? It struck me how different this was from what I had read of Nuremberg. At Nuremberg, the greatest responsibility lay with those in positions of power, those who had planned and strategized, not those with boots on the ground. At the TRC, the responsibility lay with the one who pulled the trigger. The greatest responsibility seemed to lie with the one closest to the scene of the crime. We may ask: “Why was the leadership of apartheid not held responsible for it?” The answer is political, not ethical.

The negotiations that ended apartheid provide us with the raw material for a critique of the universal claims of the current human rights paradigm. I will suggest that the present rush for courtroom solutions advocated by the human rights community is the result of a double failure: analytical and political. Analytically, it confuses political violence with criminal violence. Politically, the focus on perpetrators is at the expense of a focus on issues that drive the violence. As such it is likely to magnify rather than mitigate violence in the public sphere.

What distinguishes political from criminal violence? The key difference is this. Political violence requires more than just criminal agency; it needs a political constituency. More than just perpetrators, it needs supporters. That constituency, in turn, is held together and mobilized around an issue. More than criminal violence, political violence is issue-driven.

Nuremberg and CODESA need to be thought of as two responses to political violence. Each was a response to a crime against humanity. Each has an implication for how we think of human wrongs and thus of human rights. Whereas Nuremberg shaped a notion of justice as criminal justice, CODESA calls on us to think of justice primarily as political justice. Whereas Nuremberg has become the basis of a notion of victims’ justice—as a complement rather than an alternative to victors’ justice—CODESA provides the basis for an alternative notion of justice, which I call survivors’ justice.


Nuremberg was one of two trials at the conclusion of World War II. The second was The Tokyo [War Crimes] Trial. Nuremberg was an innovation for at least three reasons. The judges at Nuremberg rejected the claim that individual officials were not responsible for an act of state. Nuremberg established the principle of individual responsibility for the violation of human rights. The judges at Nuremberg also established criminal responsibility for these crimes. Finally, Nuremberg stood for a universalism whereby the international community would be able to reach back through the boundaries of state sovereignty to protect individuals or impose norms, thereby holding these individuals directly accountable to that same international community. The international community here was a euphemism for a group of civilized nations, to which otherwise sovereign polities were ultimately answerable.

Nuremberg was born of a debate among victorious powers on how they should deal with defeated Nazis. Winston Churchill argued that Hitler and his gang had forfeited any right to legal procedure and so should be summarily shot. Henry Morgenthau, Jr., the US Secretary of the Treasury and a close friend of Franklin Roosevelt, agreed with Churchill. Morgenthau went further and called for a destruction of German industry so Germany would never again rise as a power. The opposition was led by Henry Stimson, Roosevelt’s Secretary of War. Stimson wanted a trial, not just a show trial, but a trial with due process. His conviction was shared by the Chief Prosecutor at Nuremberg, Robert Jackson.

Even if based on due process, Nuremberg needs to be understood as symbolic and performative. For a start, only the losers were put on trial. The victors appointed not only the prosecutor but the judges too. On their part, the accused preferred to be tried by the US than by any one else. They expected a fairer trial from the Americans who, unlike the victims—Jews, Russians, French, British—had had the privilege of pavilion seats during the war. They also expected softer treatment from the Americans who were likely to be German allies in the brewing Cold War. For official America, Nuremberg was an excellent opportunity to inaugurate the new world order by showcasing a performance of how a civilized liberal state conducts itself. At a time when the air was full of cries for revenge, Robert Jackson told the audience at Church House in London: “A fair trial for every defendant. A competent attorney for every defendant.”

Nuremberg combined elements of both victors’ justice and victims’ justice. Victors’ justice followed from the outcome of the war: victorious powers established a rule of law under which alleged perpetrators were tried. The notion that justice would follow victory was not new. It followed a long-established tradition of how we think of justice in the aftermath of victory, be that victory the result of war between states or revolution between classes or a civil war of a different type. In every case, the assumption is that once the conflict has ended, there is a clear victor under whose power justice can be administered. This overall frame marks Nuremberg as a model for victors’ justice.

The accused at Nuremberg were charged with four crimes: conspiracy to wage aggressive war; that was the first. Waging aggressive war, was the second. War crimes, meaning violations of the rules and customs of war, such as mistreatment of prisoners of war and abuse of enemy civilians; that was the third; and crimes against humanity, which included the torture and slaughter of millions on racial grounds, fourth. Now the first thing striking about this list is that the conspiracy to wage war and its actual waging were defined as the principal crimes. They were labeled “crimes against peace.” The Holocaust came last in this series of four crimes.

The Allies were divided on this order. The French disagreed that waging war was a crime in law; it is, they said, what states did—and do. At the Tokyo trial, which took twice as long, partly because of long and substantial dissenting opinions, Justice Radhabinod Pal of India argued that the charge of the crime against peace was a case of ex post facto legislation which he said, “served only to protect an unjust international order, if there were no other workable provisions for peaceful adjustment of the status quo.” Much later, in 1992, Telford Taylor, who had replaced Jackson as the Chief Prosecutor in the twelve remaining US-conducted trials in Germany, and who then had a distinguished career as professor of law at Columbia Law School, conceded that the court’s judgment on counts 1 and 2 did indeed rely on ex post facto law.

An even more serious problem arose from the fact that the victors’ court was not likely to put the victors on trial. Would not Truman’s order to firebomb Tokyo and drop atom bombs on Hiroshima and Nagasaki, leading to untold civilian deaths at a time when the war was already ending, qualify as “gratuitous human suffering” and “a crime against humanity,” to use the language of the court? Had not Winston Churchill committed a crime against humanity when he ordered the bombing of residential, working class, sections of German cities, particularly Dresden, in the last months of the war? Most agreed that the British policy of terror bombing of civilian areas killed some 300,000 and seriously injured another 780,000 German civilians.

The second dimension of Nuremberg is victims’ justice, what is often thought of as an alternative to victors’ justice, but was in fact a complement to it. One of the charges against the accused was that they had committed crimes against humanity. The charge—crimes against humanity—was first formulated in 1890 by George Washington Williams, a historian, a Baptist minister, a lawyer, and the first black member of the Ohio state legislature in the US. Washington wrote a letter to the US Secretary of State in which he documented atrocities committed by King Leopold’s colonial regime in Congo, concluding that this was “a crime against humanity.” I have already pointed out that crime against humanity was the last of four charges against the accused at Nuremberg. As the trial proceeded, the emphasis on victims’ justice began to diminish. The reason was political: as the Cold War developed, US policy toward Germany moved from a demand for justice to a call for accenting accommodation over punishment. The effect was most evident in the trial of Alfried Krupp, the leading German industrial magnet. The Krupp family had been manufacturers of steel since the early nineteenth century and Europe’s leading manufacturers and suppliers of guns and munitions by World War I. They had armed Germany in three major wars. During World War II the Krupps managed 138 concentration camps. Ranged throughout Europe, all camps were privately owned by the Krupps. Alfried Krupp used slave labor from the camps and prisoners of war to build his factories and provided Hitler’s wars with money and weapons, as combination of investment and commitment. One of those charged at Nuremberg, Krupp was released in 1951, his fortune restored. There was little justice for victims at Nuremberg. For Krupp it was time to serve in another regime in another war, the Cold War. When it came, the justice for victims was political; more importantly, it was obtained outside the court.

To understand the particular form that victims’ justice took, we need to appreciate the political context that framed Nuremberg. Nuremberg functioned as part of a larger political logic shared by the victorious Allied powers. This was that winners and losers, victims and perpetrators, must be physically separated into different political communities. As they redrew boundaries and transferred millions across borders, the Allied powers carried out or sanctioned the most extreme ethnic cleansing in modern history. The possibility of victims’ justice flowed from the assumption that there would be no need for winners and losers to live together after victory. The perpetrators would remain in Germany and the victims would depart for another homeland. Yesterday’s perpetrators and victims will not have to live together, for there will be a separate state—Israel—for survivors. The State of Israel was created after the principal trials at Nuremberg, but it was the culmination of a process that defined the logic of Nuremberg. Indeed, post-Holocaust language reserves the identity “survivors” only for yesterday’s victims. As in Israel, this is also the case in contemporary Rwanda. In both cases, the state governs in the name of victims.

Post-apartheid transition in South Africa

The post-apartheid transition in South Africa is popularly identified with the work of the TRC. This work is presumed to have been guided by the dictum that those in power be forgiven past crimes in return for acknowledging the past—which is called truth. It is said that the TRC created a new precedent: immunity from prosecution—some may say, impunity—in return for acknowledging the truth, forgiveness in return for an honest confession. As the saying goes: Forgive, but do not forget. This claim is central to the contemporary ideologization of the TRC.

The problem with this widely accepted notion is that it is not quite true. Key to the post-apartheid transition was not an exchange of amnesty for truth. The fact is that the amnesty was granted in exchange for the willingness to reform. That reform was the dismantling of juridical and political apartheid. The real breakthrough represented by the South African case is not contained in the TRC but in the talks that preceded it and were known as the Convention for a Democratic South Africa—CODESA. These talks have so far been dismissed as nothing but so much hard-nosed pragmatism. I intend to look at these talks as a source of lessons for us here.

The ground for CODESA was prepared by a double acknowledgement by both sides to the conflict. To begin with, both sides recognized that there was little prospect of ending the conflict in the short run. For those far-sighted in the leadership, this was equivalent to a recognition that their preferred option was no longer within reach: neither revolution for liberation movements nor military victory for the apartheid regime was on the cards. If South Africa is a model for solving intractable conflicts, it is an argument for moving from the best to the second-best alternative. That second-best alternative was political reform. The quest for reform, for an alternative short of victory, led to the realization that if you threaten to put the leadership on either side in the dock they will have no interest in reform. This change in perspective led to a shift, away from criminalizing or demonizing the other side to treating it as a political adversary. Its consequence was to displace the paradigm of criminal justice identified with Nuremberg.

I suggest that we think of CODESA less as an alternative to Nuremberg than as a response to a different set of circumstances. As such, it is also a statement that Nuremberg cannot be turned into a universally applicable formula. CODESA was born of the realization that the conditions that obtained in apartheid South Africa were different from those that led to Nuremberg. The difference was two-fold. First, whereas Nuremberg followed a military victory, the conflict in South Africa had not ended. How do you stop a conflict that has not ended? How do you convince adversaries that it is in their interest to stop an ongoing conflict? Surely, this could not be done by prioritizing criminal justice and threatening to take the political leadership on either side—the apartheid state or the anti-apartheid movement—to court, because the people you would want to take to court are the very people you would need to stop the conflict. Second, whereas Nuremberg was informed by a logic of ethnic cleansing, one that called for a physical separation of yesterday’s victims and yesterday’s perpetrators into separate political communities, in South Africa there was no question of creating an Israel for victims of apartheid. Instead, it was clear that victims and perpetrators, blacks and whites, would have to live in the same country.

Rather than put justice in the back seat, CODESA presents us with a radically new way of thinking about justice. It presents us with a double breakthrough. To begin with, CODESA distinguished between different forms of justice—criminal, political, and social. It prioritized political justice, the reform of the political system, over the other two. The difference between political and criminal justice is two-fold. One, political justice affects groups whereas criminal justice targets individuals. Two, the object of criminal justice is punishment, that of political justice is political reform. A shift of logic from the criminal to the political led to decriminalizing and legitimizing both sides to the conflict. The liberation movements—the [African National Congress] ANC, the [Pan-Africanist Congress] PAC, and the [South African] Communist Party [SACP]—were all unbanned. The apartheid regime, the Nationalist Party, and the highly secretive underground network known as the Broederbond, also ceased to be treated as pariah by anti-apartheid activists. In decriminalizing and legitimizing opponents, CODESA turned enemies into political adversaries. In the process, CODESA also moved the goalpost. The goal was no longer the internment and punishment of individuals charged with so many crimes, but a change of rules that would include them and their constituencies into a reformed political community. CODESA’s achievement was to bring adversaries to agree on a political reform which dismantled legal and political apartheid and redefined an inclusive citizenship. The point was not to avenge the dead, but to give the living a second chance.

The full impact of this change in perspective was no less than a shift of paradigmatic significance. Whereas Nuremberg was backward-looking, preoccupied with justice as punishment, CODESA sought a balance between the past and the future, between redress for the past and reconciliation for the future. The paradigm shifted from one of victims’ justice to that of survivors’ justice—where the meaning of survivors changed to include all those who had survived apartheid: yesterday’s victims, yesterday’s perpetrators, and yesterday’s beneficiaries—some would say bystanders—all were treated as survivors.


The negotiators at CODESA faced a central challenge: how to forge a transition from a white minority regime to a government elected by an enfranchised majority. As an interim measure, the parties to the negotiations agreed to lay down a set of constitutional principles that would define the parameters of the interim constitution. The declarations notwithstanding, the negotiations at CODESA were testimony to so much horse-trading, with each side trying to strengthen its hand through political pressure, including violence, outside the negotiating chambers.

This went on from December 1991 for almost another year, as each side mobilized; outside, sporadic violence sort of punctuated this process and the violence in turn pushed the negotiations forward. Eventually the ANC cleared the ground for agreement with historic concessions, famously known as the sunset clause, announced by Joe Slovo, the Secretary General of the Communist Party, in an article in the party journal, The African Communist. The article, clearly with an imprint of both the Communist Party and the ANC, called for power-sharing between the two sides, retention of the old bureaucracy—and presumably other organs of the state: police, military, and the intelligence services—and, finally, a general amnesty in return for full disclosure. Joe Slovo did not need to state what was clear to one and all: that the real quid pro quo for these concessions was the dismantling of juridical and political apartheid and the introduction of electoral reforms that would enfranchise the majority and pave the way for majority rule. An acceptance of the sunset clause would mean that South Africa would not have its own version of Nuremberg.

The multi-Party negotiating process began on March 5 at Kempton Park and then continued into a second phase. […] Basically the main point here is that the process was not entirely democratic. It was moved forward by something known as sufficient consensus, so long as the National Party and the ANC came together, that was understood as sufficient consensus. Anything agreed upon could be vetoed by the South African Supreme Court. Normally a constitutional principal agreed upon by a constituent assembly would not be vetoed by any external agency, but in this case there were vetoes. My focus here are the concessions that the ANC made. A Bill of Rights was included as part of the constitutional checks and balances. The Bill of Rights included protection of private property as a fundamental human right. At the same time, and without a stated rationale, the clause providing the restoration of land to the majority population was placed outside the Bill of Rights. Where property rights were in clash, as that of white settlers and black natives, the former received constitutional protection, the latter only a formal acknowledgement in law.

More important concessions were made at the local level. These were a response to two political forces that came together in the course of the negotiations: white settlers and Native Authorities. There was constitutional recognition of the Zulu monarchy and of indigenous and customary law. More significant was the passage of the Local Government Transition Act of 1993 which entrenched consociational government at the local level which meant the following: local government elections were structured in such a way that they precluded black voters from obtaining two-thirds majority on a local government council. And then there was a ward limitation system, which guaranteed non-blacks 30 per cent of the seats. And another provision in the same law required a local authority to muster a two-thirds majority to pass its budget. In other words the non-black population had a veto in every local authority. There were other provisions, which forbade the transfer of wealth from one part of a local authority to another part of the same local authority through a taxation mechanism. Taxes had to be equal.


There are two debates in South Africa today. The first focuses on the perpetrator, and thus on criminal justice. The second focus is on the beneficiary, and thus on social justice. Whereas there is hardly a popular demand in contemporary South Africa for perpetrators of apartheid to be tried and punished, it is the debate around social justice that more and more drives the critique of the post-apartheid transition.

The current debate on the call for social justice needs to be taken as a double critique, of both CODESA’s success in putting a constitutional closure on the search for social justice and of the TRC’s failure to reopen that debate. Unlike CODESA, the process guided by the TRC was designed as a civic educational process. Its recommendations were not to have the force of law, but were rather to be considered as advisory in the policy-making process. Its advisory nature gave the TRC greater latitude than it would otherwise have had, for it gave the TRC freedom to think a step beyond CODESA and thereby rethink the future. Such a rethink had the potential to connect the quest for political justice with that for social justice.

But the TRC functioned as a surrogate Nuremberg. It displaced the logic of crime and punishment with that of crime, confession, and amnesty. The TRC attempted to join the metaphor of political injustice with that of crime, but the attempt was not successful. The TRC sought to hold individual state officials criminally responsible for those actions that would have been defined as crimes under apartheid law. Put differently, it did not hold them accountable for the violence undergirded by the rule of law under apartheid, but only for infringements of apartheid law.

The legislation that set up the TRC gave it freedom to define the victim. In interpreting the legislation, the TRC made three key decisions. First, the TRC individualized the victim. To do so was to ignore precisely what was distinctive about apartheid, that it was a system based on group oppression. Secondly, the TRC defined a human rights violation narrowly, as violating the bodily integrity of an individual. This distinction too proved problematic in a context where most violence was extra-economic. The violence of apartheid did not target the bodily integrity of a population group defined as Bantu but it targeted their means of livelihood, land, and labor. Finally, this extra-economic coercion—whether the forced removal of millions from land gazetted as white areas or pass laws that tracked the movement of all black people—was the work of the apartheid authorities, and not of individual operatives. Just as victims were identified as racialized groups and not individuals, or just as victims would have to be identified, but were not as racialized groups, and not individuals, the perpetrators too were part of a racialized power and not just individuals.

In 1959, the apartheid government passed the Promotion of Bantu Self-Government Act. The Act was to provide the legal umbrella for a far-reaching ethnic and racial cleansing of 87 per cent of the land that was defined as white South Africa. A widely distributed and cited investigation by The Surplus People Project [SPP] documented that 3.5 million persons had been forcibly moved by South African authorities between 1960 and 1982 as part of the project to create ethnic homelands. The commission accepted the estimate and acknowledged that the process involved collective expulsions, forced migration, bulldozing, gutting, or seizure of homes. After noting that forced removals were an assault on the rights and dignity of millions of South Africans, the commission claimed it could not acknowledge them since these violations may not have been gross as defined by the Act.

Similarly, the commission also sidestepped the laws by which apartheid was known, the pass laws which targeted every single black African. And the commission also ignored the question of prisoners detained without trial—the most famous of these was Steve Biko—because it argued that these were common law criminals and not those who had been in prison with a political motif.

At the end, the commission came up with three truly bizarre conclusions. First it concluded that the victims of apartheid numbered a little more than 20,000 and the TRC recommended only those 20,000-plus should receive reparations from the post-Apartheid state—and not the millions who were victims of the pass laws, forced removals, and forced labor. Second, the commission compiled a time series of violations over its mandate, which began with the Sharpeville massacre in 1960 and closed with the first democratic elections in 1994. The commission concluded: most violations took place in the period after the unbanning of political parties (1990–4) and that it was the result of conflict between anti-apartheid groups, especially the ANC and the [Inkatha Freedom Party] IFP in Natal. From this followed its most scandalous conclusion based on a list the commission compiled of perpetrator organizations. The IFP was identified as the top perpetrator-organization and the ANC as the third perpetrator-organization. In this list of perpetrators, the state security services came as runner-ups: the South African Police (SAP) was second and the South African Defense Forces (SADF) trailed as the fourth.

In sum, the TRC set aside the distinctive violence of apartheid, the violence that targeted entire groups and that was central to realizing its political agenda. This is because the TRC understood violence as criminal, not political; as targeting identifiable, individual victims, not entire groups. It focused on violence as excess, not as norm. It thus limited the criminal responsibility of individual operatives to actions that exceeded political orders—actions that would have been defined as criminal under apartheid law. In doing so, the TRC distinguished between the violence of apartheid—pass laws, forced removals, and so on—and the excess violence of its operatives. Because it did so, it was unable to achieve even that which Nuremberg did: to compile a comprehensive record of the atrocities committed by the apartheid regime. This is why the TRC should be seen as a special court within the framework of apartheid law.

As interpreted by the human rights movement, the lesson of Nuremberg is that criminal justice is the only politically viable and morally acceptable response to mass violence. Dehistoricized, Nuremberg has been turned into the founding moment of the new human rights movement. Nuremberg is today the model for the International Criminal Court (ICC), held as the fitting anti-dote to every incident of mass violence.

To de-ideologize Nuremberg would be to recognize that the logic of Nuremberg flowed from the context of interstate war, one that ended in victory for one side, which then put the losers on trial. The logic of a court trial is zero sum: you are either innocent or guilty. This kind of logic ill fits the context of a civil war. Victims and perpetrators in civil wars often trade places in ongoing cycles of violence. No one is wholly innocent and none wholly guilty. Each side has a narrative of victimhood. Victims’ justice is the flip side of victors’ justice: both demonize the other side, and exclude it from participation in the new political order. A civil war can end up as a renegotiated union or as a separation between states. The logic of Nuremberg drives parties in the civil war to the latter conclusion: military victory and the separation of yesterday’s perpetrators and victims into two separate political communities. That the founding moment of the South African transition is not a criminal trial, but political negotiations, is mainly the result of a radically different context: apartheid was not a war between states, it was a civil war.

The contemporary human rights movement is permeated with the logic of Nuremberg. Human rights groups focus on atrocities for which they seek individual criminal responsibility. Their method of work has a formalized name: Naming and Shaming. The methodology involves a succession of clearly defined steps: catalog atrocities, identify victims and perpetrators, name and shame the perpetrators, and demand that they be held criminally accountable. The underside of the focus on perpetrators is to downplay issues. Read the field reports of Human Rights Watch or International Crisis Group and you will find that, except for a pro-forma [one] page or two Introduction on history and context the focus is on naming and shaming. Indeed, context is considered a distraction from establishing the universality of human rights.

This is problematic, if one recognizes that political violence is often not a stand-alone incident but part of a cycle of violence—a fact obscured by the absence of a historical context. In a previous book on the Rwandan genocide, I set about constructing a historical account of the violence: the more I did so, the more I realized that victims and perpetrators tended to trade places. Each side has a narrative of victimhood. The more you downplay context, the more you tend to locate the motivation for violence in either the individual psychology of the perpetrator or the culture of a group of perpetrators. The tendency to portray the perpetrator as the driving force behind the violence leads to the assumption that the perpetrator is always the perpetrator and the victim is always the victim. The result is to demonize the agency of the perpetrator and diminish the agency of the victim. Demonizing goes along with branding, and reinforces the assumption that you can easily and eternally separate the bad from the good.

The more depoliticized our notion of violence, the more the temptation to think of violence as its own explanation. Indeed, the tendency is to seek the explanation for violence in the person of the perpetrator. From being a problem, violence also becomes the solution. The temptation is to think that eliminating the perpetrator will solve the problem. But instead of showing a way out of the dilemma, it feeds the logic of revenge and introduces us to a quagmire. It feeds the cycle of violence.

But violence is not its own explanation. To break out of the cycle of violence we need to displace the victim narrative with that of the survivor. A survivor narrative is less perpetrator-driven, more issue-driven. Atrocities become part of a historical narrative. They no longer stand as so many stand-alone acts but parts of an ongoing cycle of violence. To acknowledge that victim and perpetrator have traded places is to accept that neither can be marked as a permanent identity. The consequence is to de-demonize and thus to humanize the perpetrator.

If Nuremberg has been ideologized as a paradigm, the end of apartheid has been exceptionalized as an improbable outcome produced by the exceptional personality of one individual, Nelson Mandela. It is thus said that Africa’s problems—the violence of civil wars—require a different solution, one approximating Nuremberg-type criminal violence, calling for punishment rather than political reform. But African conflicts are in the main civil wars, they are not wars between states.

The lesson of South Africa is to look for the solution within the problem and not outside it. The point is to strive for internal reform, and not external intervention. CODESA has a double significance. CODESA focused on the cycle of violence as threatening the very foundation of a political community. It presents us with a challenge. It dares us to reimagine the political community by recognizing in the aftereffects of violence an opportunity to re-found the political community. In doing so, it underlines the need to return to an older tradition in political theory, one that recognizes a possibility in political violence—civil war in particular—that it is potentially foundational to the creation of an inclusive political order.

This essay is based on the talk with the same title given by Mahmood Mamdani at the symposium “A Journey of Ideas Across–In Dialog with Edward Said”, at Haus der Kulturen der Welt, Berlin, 31.10 2013–02.11 2013.