The years around ad 520 were catastrophic for the Eastern Roman Empire. Constantinople and other cities were devastated by a series of earthquakes, the Euphrates brought repeated floods, and the empire had to endure military conflicts with the Persians, the Bulgars and the Saracens. It also had to contend with internal revolts, and to cap it all the appearance of Halley’s Comet spread open panic. Mischa Meier, the German historian of antiquity, lists all these cataclysms and notes something very strange: contemporary sources paint a detailed and dramatic picture of the disasters, both local and wider in scale, but do not give the impression that they came out of the blue or were seen as particularly threatening.1
Twenty years later, around ad 540, a whole cascade of disasters poured over the same area. Again a comet of ill omen appeared in the skies, again Bulgar raids and natural devastation took place in conjunction with each other. The Ostrogoths reconquered large parts of the empire in the prelude to a gruelling war that caused numerous civilian casualties. The capital was once more hit by earthquakes, and plague caused ‘death on a scale never seen before. Trade and crafts came to a standstill in Constantinople and other cities. The infrastructure of the empire broke down, whole villages were emptied of people.’2
Here the sources give eloquent testimony of the panic, fear and dramatic sense of threat. What, Meier asks, explains this striking difference in perceptions over such a short space of time? His answer is astounding but illuminating. People expected the world to end sometime around the year 500; it was an idea based on the calculations of Christian chroniclers, on the experience of the fall of the Western Roman Empire, and on the reign of Emperor Anastasius (whose very name meant ‘resurrection’). This was the frame of reference in which the disasters of 520 to 530 were interpreted; people were mentally prepared for the Apocalypse and explained events as harbingers of it. The patterns of interpretation and orientation, Meier notes trenchantly, probably ended in dissonance only when ‘the charted disasters had not come to pass, when external conditions had not changed’.3
Of course, the fact that the world did not come to an end ensured that the reference framework was different twenty years later. Impending apocalypse was no longer seen as the reason for the increased number of disasters: ‘the perception of events’, Meier writes, ‘could no longer be matched with the prevailing patterns of orientation.’4 This led to considerable criticism of the emperor, who was held responsible for the threatening events.
This is an interesting example, which convincingly shows that disasters are not events predefined in certain ways; it is only in the perceptual horizon of those affected by them that they are interpreted, or not interpreted, as full of menace. Erving Goffman’s classic Frame Analysis5 presents in detail this argument that certain socially moulded patterns or ‘frames’ are available for the understanding of events and their emotional significance – that it is not the sheer objectivity of an event, but these principles organizing its perception, which shape how people respond to it.
Psychology, social psychology and the cognitive brain sciences are in agreement that people take decisions on the basis of complex assumptions, only a small number of which reach the level of conscious reflection. Their reference framework, then, includes not only conscious and unconscious perceptions and interpretations but also background assumptions (‘this is so’, ‘this is how it is done’, etc.), socialized habits and forms of behaviour, situational requirements, the actions, demands and orders of others, and so on.
The interpretation of perceived threats and the motivation for conclusions and decisions are therefore shaped by cognitive appraisals that may be confirmed and strengthened by interactions and group processes, and vice versa. Relevant in this respect are the situations in which the players find themselves, their socially acquired patterns of perception and interpretation, and the changes that various threats, wars or other disasters bring about in these patterns. More abstract conceptions and reality models also come into play, concerning the end of the world, what is and is not to be expected, war and peace, justice and injustice, responsibility, retribution, and so on. Such factors, at different levels of specification, form the reference framework for the situational perceptions, conclusions and action decisions of the individuals concerned. Hence, one and the same situation may be perceived and interpreted completely differently by different persons, or at different points in time. Since the chain of disasters in ad 520 was generally thought to confirm what people expected at the time, they did not give rise to interpretive endeavours; those around 540, however, did not conform to expectations and triggered widespread panic. Only when events, experiences or developments can no longer be understood within the prevailing frame of reference do orientation problems arise and create a need to explain what is going on. In what is felt to be a climate of disorder, a desire naturally grows for overviews and transparency – and, of course, for order.
‘I do not remember the date. But people said the president was dead, and he was our parent. Then the Tutsis immediately fled. We saw houses burned here and there. We were angry after the death of our parent. The war began. The Tutsis were killed.’6 These words refer to one of the shortest and most terrible genocides of the twentieth century, when 500,000 to 800,000 people were killed in Rwanda between April and July 1994. Most of the victims belonged to the Tutsi group, which was reduced by three-quarters in the space of just thirteen weeks. But the mass murder did not strike only at Tutsis; Hutus who were critical of it, or who were considered traitors because they were married to Tutsis or for some other reason, suffered the same fate.
It has now been well documented that the sharp differentiation between the two groups was largely a product of the colonial age, and that in particular the Tutsi minority owed its better social position to the higher value that the colonial authorities attached to it. In the years before the genocide, a widespread sense of social discrimination among the Hutus turned into a perception of threats and finally culminated in a total enemy-image of the Tutsis. At some point the Hutu majority came to perceive a genocidal threat from the Tutsis and, feeling that they had to do all in their power to resist it, actually turned it into its opposite. When President Habyarimana was assassinated on his aircraft on 6 April 1994, the mass killing began.
‘After the crash, people said our parent was dead…. Since he loved us and we loved him, when he was dead, everyone was affected, and we thought we were finished. People said the enemy had attacked us and that we had to defend ourselves.’7 As we see, both this participant and the previous one were able to attach a meaning to the mass murder: there was a deadly attack that had to be resisted. The genocide of the Tutsi was perpetrated by ordinary members of the Hutu majority, although it was mostly army officers, senior state officials and administrators who gave the orders and compiled the lists of those to be killed. The number of murderers was in the six figures; they usually killed with machetes handed out to them beforehand.
A number of violent conflicts and massacres between Hutus and Tutsis had already taken place since the 1960s. It is remarkable that the two groups were not at all segregated from each other – on the contrary, their daily coexistence was largely unproblematic, and they worked together and made friends with each other. So why this outbreak of mass murder along ethnic lines? The quotations have already suggested that Hutus felt there to be a problem, but it may well be that these two killers were not aware of the deeper reasons for their feeling.
Both explain the mass murder by the same personal event: the assassination of their president. They saw him as one of theirs, like a family member, a father-protector who had responsibility for them; they therefore logically saw the assassination as a life-and-death crisis for themselves. However unjustified this sense of threat may seem from outside, it plays the most important role in the subjective motivation of perpetrators of genocide, mass killings and massacres.8 Even if the real majority–minority relationship shows this to have been a grotesque reversal of any actual threat – 90 per cent of the Rwandan population were Hutus – people reacted as if they had to attack and kill in order to save their own and their family’s lives. In this way a sense of fear was bent around and itself became a deadly reality. The Hutu believed there was a lethal threat from the Tutsis, as anti-Semites in Germany in the 1930s and 1940s believed in a Jewish world conspiracy,9 or the followers of Slobodan Milošević believed that the Serbs were in mortal danger. Even when the threat perception was totally irrational, it resulted in the real deaths of countless numbers of people. Irrationality of motives has no influence on the rationality of action. The Holocaust is the most disturbing proof of William Thomas’s theorem: ‘When people define situations as real they become real in their consequences.’
During the Vietnam War, American soldiers engaged in several massacres of the civilian population. The best known and most horrific of these, at the village of My Lai, was the subject of several judicial investigations. The interrogation records express in grotesque terms the soldiers’ perception that they were killing enemies. The following extract gives us a flavour:
From the outside such statements appear completely absurd, or deranged. But a reconstruction of how US soldiers saw things in Vietnam shows an extreme underlying loss of orientation and control, which resulted from the fact that they were unprepared for jungle warfare and unable to cope with the Vietcong’s guerrilla techniques – and therefore that they experienced the whole operational space as threatening. The fantasy of the grenade-wielding baby, which occurs in a number of statements, evidently had its roots in this perception of a diffuse, all-embracing threat from an invisible enemy. It testifies to extreme disorientation and a sense that the surrounding threat was total and incalculable. Anyone not part of the ‘we group’ was a disguised enemy ready to do anything.
This disorientation occurred against the background of a military strategy that released soldiers from the rules of conventional warfare, making the killing of civilians a perhaps regrettable but by no means forbidden accompaniment of battle. It was the strategy of ‘search and destroy’, ‘free fire zones’ and ‘body counts’,11 which measured military success by the number of dead bodies. This deadly mixture makes the fantasy of Vietcong babies less unfathomable. To treat the casualties as an undifferentiated mass can be functional, since it increases the sharpness of vision.
Nor were frontline soldiers the only ones affected by such fantasies. In the US political and military leadership too, the unexpectedly disastrous course of the war spread an irrational perception of reality, so that it was assumed, for example, that enemy fighters would sooner or later desert the Vietcong and hand final military superiority to the Americans. As one contemporary observer wrote, the chiefs of staff and the president’s advisers were ‘like men in a dream’, incapable of making a realistic assessment of the consequences of their action.12
‘Body count’ statistics were supposed to make it possible for the Pentagon to predict the moment of victory (originally set for the end of 1965), when the enemy’s fighting capacity would be exhausted. But, in practice, reliance on such figures led to the indiscriminate slaughter of men, women and children; only the number of casualties mattered, in a kind of logic that led one officer to tell his men that pregnant women should be counted twice.
Historians always look in retrospect for causal sequences and for logical connections between action A and outcome B. But things like the above show that outcome B may be quite different from what was intended for action A. US soldiers in Vietnamese villages reasoned differently from those who gave the orders in the Pentagon, and they did so because they faced different problems. The statistical hubris of a ‘body count’ exit from the war led on the ground to escalations in which the body count became an end in itself: such are the sequences that lie hidden beneath abstract concepts such as ‘escalation dynamic’. In extreme situations of perceived threat, certain kinds of reasoning have an autocatalytic effect and generate certain kinds of action and outcomes that subsequently seem even to participants as curiously alien and unfathomable.13
Another example from the same war concerns the idea that the North Vietnamese army had a high command in the jungle (COSVN – Central Office of South Vietnam) that had to be located and destroyed. This fantasy fed partly on the inability of the US Army to crush the numerically and technologically inferior forces of the Vietcong, and partly on the assumption that the enemy operated in accordance with the same logic as the Americans. This led to a senseless bombing campaign to burn areas of jungle and uncover the (non-existent) ‘nerve centre’. Here too the aim was to allow a clearer overview of the situation. But the result was that world public opinion turned against the American conduct of the war, at the latest after photos were published of a little girl, Kim Phuc, naked and in tears, her hair burnt, fleeing from her bombed village. This too, of course, was a consequence that no one had intended, but it would prove decisive for the future course of the war.
The Vietnam War was also an assault on the ecological conditions under which the enemy lived. False perceptions led to the dropping of ‘800,000 tons’ of bombs on the country, ‘more than in all the theatres of the Second World War combined’,14 and to the widespread use of toxic defoliants to expose the jungle to view, which would have long-term consequences for future generations of Vietnamese.
Of all the wars that punctuated the so-called Cold War, the one in Vietnam was certainly the most absurd, costly and long-lasting. Its consequences run deep in both Vietnamese and US society. Official America suffered there its first major defeat, moral, military and economic, as well as in terms of public confidence in the policies of its president.
The underlying causes of the conflict were in many respects psychological: fantasies of superiority jostled panic fears of losing face. Both Lyndon B. Johnson and Richard Nixon publicly declared that they did not want to be the first American president to lose a war, and since it is hard to go back on such a statement they threw everything into winning the war long after that had become an impossibility. Their advisers (including the highly intelligent Henry Kissinger) and most of the top military brass created an unreal space around themselves, analysing the problems as they saw them and developing various solutions that carried them further and further towards the final debacle.
As Barbara Tuchman points out, those responsible for the conduct of the war typically refused to take on board any information that did not match their expectations. First they considered it out of the question that they could ever be defeated by a ‘fourth-rate’ country like Vietnam; then they went on systematically to overestimate the strength of their own forces and those of the Saigon regime; and, finally, the more they clung to their illusions, the more clearly disaster loomed on the horizon. The familiar mechanism of dissonance reduction was obviously at work here – and, given the obvious parallels with the Iraq War, it is a good example of how psychological phenomena such as hubris and ‘groupthink’15 can be more powerful than the lessons of history.
In view of all this, it is not surprising that radically changed realities often bring violent problem-solving in their wake. The unexpected collapse and system change of the Eastern bloc in 1989 proved too much not only for social and political theorists, whose business it was to predict such things, but also for citizens and politicians in the countries concerned. Post-colonial societies are especially prone to violent conflict, and it can take decades before a relatively stable civil society develops in them. ‘Thus, after decolonization, only 19 out of 44 African countries were capable of developing a stable state.’16 Violent conflicts followed independence in most cases, often persisting to this day in varying degrees of intensity (Sudan, Congo, Sierra Leone, Guinea-Bissau, etc.). In Ethiopia, a war that lasted from 1976 to 1991 caused as many as 2 million casualties (more than 90 per cent of them civilians); half a million to a million died as a result of fighting in Mozambique between 1976 and 1992; and war in the Congo has claimed 4 million victims since 1998.
As to the former Eastern bloc, system change and the road to democracy have by no means always been peaceful. In 2007, nearly twenty years after the collapse of the Soviet Union, the list of the sixty most unstable (and therefore violence-prone) countries in the world included Uzbekistan (23rd place), Kyrgyzstan (28th), Bosnia-Herzegovina (35th), Tajikistan (42nd), Russia (43th), Turkmenistan (45th), Belarus (50th), Serbia and Montenegro (55th), Moldavia (58th) and Georgia (60th).17 In not a few of these countries there were sharp conflicts among different ethnic groups.
Studies of mass violence and genocide have shown little or no understanding of what happens when people begin to attack or kill their neighbours – as if it were normal for them to love, or to have a close relationship with, their neighbours under peaceful circumstances. In fact, as everyone who lives in a shared house knows, and as Jan Philipp Reemtsma has powerfully reminded us,18 close proximity may well be a source of violence rather than an obstacle to it; you can end up hating the people you have to live beside.
It is the ambiguity or ‘viscosity’ of group boundaries that can lead to extreme acts of violence in a crisis. The function of such acts is to establish once and for all who ‘we’ are and who ‘they’ are, who should be seen as a friend and who as an enemy. The violence itself draws the boundary line, since ‘we’ and ‘they’ are unmistakably clear after an attack or mass killing. One man from the former Yugoslavia explains how he distinguished friend from enemy: ‘Civilians are different, civilians don’t walk the streets when there’s shooting going on [laughs]. It’s really quite simple: civilians don’t walk the streets when there’s shooting going on.’19
People develop various techniques for identifying and allocating group membership; ID papers are one, racial theories another, mass murder yet another. Thus, perceived threats to a ‘we group’ from a ‘they group’ make it more and more necessary to establish ways of identifying others, and in cases of extreme violence the outcome itself defines who belongs to which. The result of such self-referential systems is only superficially chaos. For the perpetrators, violence actually creates order.
The most spectacular escalation of violence during the post-1989 systemic breakdown, and the one with the most protracted consequences, took place in Yugoslavia, a country where various ethnic groups had once lived unproblematically side by side in a federation led by the charismatic autocrat Josip Broz Tito. Here too conflicts now became sharper and sharper as these groups felt it necessary to assert their difference through violence. It was happening not in sub-Saharan Africa or Kashmir but in the middle of Europe, at a time when the Cold War seemed to be over and no one thought a hot war was on the cards.
It is also remarkable that no one had foreseen the explosive force of nationalism in the unipolar world that followed the collapse of the Soviet Union, or the disorientation in the republics that had been part of it. The rival forces in Yugoslavia, as well as actors on the international political stage, seemed bereft of ideas amid the turmoil. The disastrous mistakes of the then German foreign minister, Hans-Dietrich Genscher, accentuated the conflict among the republics of Yugoslavia. His decision to recognize the secession of Slovenia and Croatia as sovereign states, which Belgrade saw as sabotage of its attempt to reestablish the federal republic under Serb domination, provided further fuel for the radicalization of Serb nationalism.
This case shows that foreign policy sticks to familiar bearings after a systemic change; it was as if the collapse in the East had changed nothing in the general configuration. As Henry Kissinger famously remarked: ‘It is an illusion to believe that leaders gain in profundity while they gain experience … [T]he convictions that leaders have formed before reaching high office are the intellectual capital they will consume as long as they continue in office.’20 Not only politicians but also managers, academics or doctors keep to models and prescriptions that have brought them success in the past, even when the conditions for their application have totally changed. This often has disastrous consequences.
But such immobility is not the only dramatic aspect of politics visible in escalations of violence after systemic change. Defective models and concepts are also typical of politicians in the newly formed states themselves, since they tend to be people who lack experience of democratic negotiating processes, market economics and liberal constitutionalism, but are well versed in power politics, corruption, propaganda, personality cults and cronyism. This combination of unimaginative and autocratic ways has the most baneful effect where the polity itself is in turmoil. Moreover, whereas the West still favours and promotes the model of the bourgeois-national state, murky situations in which the state has collapsed and there is a pressing need to act often mean that a new state can be conceived only along ethnic lines of divide, since other elements that might generate a national community are nowhere to be seen.21
So, for political actors who take the stage after systemic change, nationalism regularly offers the best promise of success. This in turn intensifies radical tendencies, as we saw in the case of Slobodan Milošević. His efforts to preserve his influence and to keep ultra-nationalist rivals at bay turned him more and more into a nationalist himself.
Changed realities also change those who have shaped them – a process well illustrated by the radical evolution of the Nazi leadership elite. At the level of social psychology, however, one of the reasons why this is such an uncertain process is that people often do not notice how their own mental maps of what is right and wrong, normal and abnormal, predictable and unexpected, undergo change amid the changing realities around them. In other words, as members of a society in the throes of normative change, the fact that they remain constantly in tune with their surroundings means they often do not notice that their own norms are changing. This may be termed the phenomenon of ‘shifting baselines’ (see pp. 139–143 below).
The deadly violence to which this gave rise in Yugoslavia and elsewhere is well known. Less familiar is the idea that our democratic postwar society stands on the foundations of this process of identification through extreme violence. From time to time it resonates in the somewhat equivocal lament that Germany’s non-Jewish majority robbed itself of an essential part of its culture by killing the Jews – as if the main damage was suffered by the society that succeeded the National Socialist Volksgemeinschaft. Extreme violence crosses generations, probably persisting even through several systemic changes, and today’s Federal Republic of Germany is a society that still displays the consequences of violence (as the retired general Klaus Naumann, former head of NATO, put it). But in this respect it differs only in scale, not in principle, from other postwar societies.
Phenomena such as the Holocaust or the violent break-up of Yugoslavia (the latest European instance of state-building) illustrate the truly horrifying point made recently by Michael Mann: that most of Europe’s ethnically homogeneous states are the outcome of processes of ethnic cleansing and mass killing. These murderous options are not simply occupational accidents but the dark side of the democracy that rests upon them. The path to ethnic cleansing and genocide does not follow any master plans and is often strewn with unintended consequences. War and violence intrinsically tend to trigger developments that no one foresaw at the beginning of the state-building process; resettlement can suddenly turn into expulsions, and expulsions into genocide.22 It is important to realize that such dynamics are not historically random. But escalations of extreme violence are aspects of modernization processes that are subject to cultural amnesia after the successful constitution of a new state. One reason why this is possible is that the victims of homogenization have run away or are dead.
If the ethnic cleansings and genocides of the nineteenth and twentieth centuries are understood as dynamos of modernization – and there is much to suggest that they should be – then the social transformations that come in the wake of globalization will produce a further rise in deadly violence. And, if changes in habitat, system change or the resource needs of other countries lead to increasing instability in various societies, attempts to find solutions through violence will become more and more likely.
Established polities also show a failure of imagination with regard to forms of community beyond the nation-state, although these are precisely what the globalization process will require in the medium term.