When a sociologist as important as Michael Mann publishes three books in just over a year, two of them clearly major works and the third on the big political questions of the day, it is clearly an event for the field. The first two are fruits of a big detour from the third volume of Mann’s masterwork, The Sources of Social Power, although they will also contribute to it. These richly sourced and authoritative works do more than confirm Mann as an unrivalled practitioner of synthetic and comparative methods in historical sociology. They provide – in reality for the first time – a comprehensive sociological explanation of the darkest side of modernity together with the extreme political form that gave rise to its archetypal case. For sociology as a field might almost join Mann in his confession that, until recently, his work ‘had neglected the extremes of human behaviour’ and he ‘had not thought much about good and evil.’ (The Dark Side of Democracy, p. ix) In this sense, the two books will be compulsory reading not only for students of genocide and fascism but for everyone concerned with sociology’s relevance to the modern world. Incoherent Empire is a less substantial work but by the same criterion a valuable contribution.
Here I concentrate on The Dark Side while indicating the significance of Fascists for its argument. Mann begins by outlining eight theses (pp. 2-9, from which all quotations are taken unless otherwise stated) that frame detailed case studies of New World genocides, Armenia, Nazism, Communism, Yugoslavia, and Rwanda. It is beyond my scope to evaluate these studies, or the comparative analysis of fascist movements in the companion volume: suffice it to say that students of each case will find much to enlighten them, while there are no more coherent overviews. My main purpose is to evaluate Mann’s explanations of murderous politics; I turn later to his conceptual framework. His key argument is that ‘murderous ethnic cleansing’ is the dark side not simply (pace Zygmunt Bauman, 1989) of modernity but specifically of democracy. While all his four ‘sources of social power’ – economic, ideological, political and military – are involved, he offers ‘essentially a political explanation’. His main case is that ‘cleansing is a hazard of the age of democracy since amid multiethnicity the ideal of rule by the people began to intertwine the demos with the dominant ethnos, generating organic conceptions of the nation and the state that encouraged the cleansing of minorities.’ Certainly ‘cleansing’ results from less from democracy itself than from its perversion: ‘Regimes that are actually perpetrating murderous cleansing are never democratic, since that would be a contradiction in terms. … Indeed as escalation proceeds, all perpetrating regimes become less and less democratic.’ And ‘regimes newly embarked upon democratization are more likely to commit murderous ethnic cleansing than are stable authoritarian regimes.’
Indeed ‘stabilized institutionalized democracies’ are the least likely to commit ‘cleansing’, although they often have it in their pasts, the more violently where ‘settler democracy’ took hold: ‘The more settlers controlled colonial institutions, the more murderous the cleansing. … It is the most direct relationship I have found between democratic regimes and mass murder.’
Although Mann labels ‘cleansing’ – perhaps misleadingly – a product of ‘inter-group relations’, he is clear that it is no simple product of ethnic differences. To become murderous these need political expression in ethnonationalism, which ‘arises where ethnicity trumps class as the main form of social stratification, in the process capturing and channelling classlike sentiments towards ethnonationalism.’ Indeed it ‘is strongest where is can capture other senses of exploitation. The most serious defect of recent writing on ethnonationalism has been its almost complete neglect of class relations.’ Yet although this is a political account, Mann shows that genocide is not simply statist. There are ‘three main levels of perpetrator’, radical elites running party-states, violent paramilitary bands, and ‘core constituencies providing mass though not majority popular support’. (Fascists contests common views that fascism is a mainly middle class phenomenon.) Ordinary people ‘are brought by normal social structures into committing murderous ethnic cleansing’, and they have many different motives for taking part.
However Mann recognizes that ‘[e]thnic cleansings are in their murderous phases usually directed by states, and this requires some state coherence and capacity.’ Radical party elites are crucial because they homogenize diverse social forces, welding them into more cohesive blocs, and centralize normally fragmented state structures in the pursuit of exceptional goals. This happens through political struggle, and typically through a series of escalations. The danger zone exists ‘when (a) movements claiming to represent two fairly old ethnic groups both lay claim to their own state over all or part of the same territory and (b) this claim seems to them to have substantial legitimacy and some plausible chance of being implemented.’ Escalation to the brink happens ‘when one of two alternative scenarios plays out. [Either] The less powerful side is bolstered to fight rather than to submit … by believing that aid will be forthcoming from outside … . [Or] The stronger side believes it has such overwhelming military power and ideological legitimacy that it can force through its own cleansed state at little physical or moral risk to itself.’ Finally, going ‘over the brink’ into actual perpetration occurs ‘where the state exercising sovereignty over the contested territory has been factionalized and radicalized amid an unstable political environment that usually leads to war.’
Mann’s account challenges the standard absolutist concept of ‘intention’ in genocide. He demonstrates convincingly that an entire historical episode of murderous politics, over many months or years, cannot be explained by singular intentionality: ‘Murderous cleansing is rarely the original intent of perpetrators. … [It] typically emerges as a kind of Plan C, developed only after the first two responses to a perceived ethnic threat fail … To understand the outcome, we must analyze the unintended consequences of a series of interactions yielding escalation.’ Instead of interpreting murder as the direct consequence of longstanding intentions (or for that matter structural conditions), he emphasizes the contingency of violent outcomes: ‘Out of … political and geopolitical crises radicals emerge calling for tougher treatment of perceived ethnic enemies. In fact, where ethnic conflict between rival groups is quite old, it is usually somewhat ritualized, cyclical and manageable. Truly murderous cleansing in contrast, is unexpected, originally unintended, emerging out of unrelated crises like war.’ Because of the difficulties of identifying or proving ‘intention’, some writers have proposed that we abandon the criterion altogether. Mann enables us to retain the idea of genocide as purposeful action, but relate it properly to political relations and changing contexts.
Although in this richness and complexity Mann’s is a highly plausible sociological framework, inevitably problems remain. He disarms criticism by acknowledging that ‘[g]iven the messiness and uniqueness of societies, my theses cannot be scientific laws. They do not even fit perfectly all my case studies.’ Yet the coherence of his detailed explanations too frequently appears to be, at least partially, seriously at odds with his general theses. His ‘ethnic competition’ framework works well for Rwanda and former Yugoslavia, but it does not fit as well the major cases of twentieth century political mass murder. Yet Mann considers genocide as a maximal form of ‘cleansing’, and if his general theses do not fit – without heavy qualification – archetypal cases like the Holocaust (which occupies a great deal of The Dark Side and whose political pre-conditions are a central topic of Fascists) then their general relevance is diminished.
These very darkest episodes appear to have only tenuous connections to democracy. The most murderous are actually products, Mann acknowledges, of ‘a few highly authoritarian regimes’ that ‘deviate’ from the norm of stable authoritarianism, and ‘mobilize majoritarian groups into a mass party-state mobilizing the people against “enemy” minorities.’ Although Hitler had indeed perverted democracy, it is difficult to fit the Holocaust, Mann recognizes, into his pattern of ethnic competition. Germans and Jews were ‘two fairly old ethnic groups’, but it was hardly the case that they ‘both laid claim to their own state over all or part of the same territory.’ As Bauman (1989: 53) suggested, the nature of the Jewish exception to the ethnic rule was what made them particular targets: ‘The world tightly packed with nations and nation-states abhorred the non-national void. Jews were in such a void: they were such a void.’
Likewise the extensive mass murders of Communist states are problematic. Mann plausibly describes many of them as classicide, because their targets were social classes, and explains them as ‘mistaken revolutionary projects’ rather than as ethnic conflict. But he also offers the perversion of socialism as a class variant of the more common ethnic perversion of democracy: ‘socialist ideals of democracy also became perverted as the demos became entwined with the term proletariat, the working class, creating pressures to cleanse other classes.’ Along with the perversion of national democracy, this was then a second ‘general way in which democratic ideals were transmuted into murderous cleansing.’ However we may question whether the idea of the proletariat (working class) was really a moving force in the perversion of Soviet democracy, or in creating ‘pressures to cleanse other classes’. When this idea had most meaning, just after the October Revolution, the Soviet Union was at its most democratic and although the Bolsheviks repressed the peasantry there was no ‘cleansing’. Later, although the term proletariat remained at the centre of Communist ideology, it had come to mean the rule of the party, which substituted itself for the working class. Only after the idea had lost its class meaning did Stalin develop his most murderous policies towards peasants. Mann’s main thesis foregrounds ideology but in this case at least the argument seems overstated.
A more fundamental question is whether Mann correctly identifies the main explanatory locus of murderous policies among his four types of power. He provided, in the first volume of The Sources, the clearest argument that political and military are two separate forms of power: ‘Marx, Weber and their followers do not distinguish the two, because they generally view the state as the repository of physical force in society. To equate physical force with the state often seems to make sense in the case of modern states that monopolize military force. However conceptually they should be regarded as distinct … . (Mann, 1986: 10-11). Clearly by definition, murderous power belongs primarily to this physical, violent ‘military’ category. Mann uses the general distinction effectively to emphasize the importance of paramilitarism to both ‘cleansing’ and fascism: thus the Nazis were ‘a distinctive nation-statist force that promised a “cleansing” paramilitarism’ (Fascists, p. 206).
That murderous power is primarily military does not of course mean that its occurrence should be explained simply by military developments. Yet Mann’s mainly political explanation, plausible in accounting for fascist movements, is more questionable when it comes to genocide. As he acknowledges, his theses connecting ‘cleansing’ to democracy ‘apply beforehand, to the earlier phases of escalation of ethnic conflict.’ It is military power that ‘proves decisive in the later stages of the worst cases of ethnic cleansing. Armies, police forces, and irregular extrastate paramilitaries are the main agencies … .’ Indeed, he provides detailed arguments that should lead to a heavily ‘military’ explanation of murderous politics itself:
Most 20th-century cases of ethnic cleansing occurred during wars or during the chaotic transfer from war to peace. … Ideologically tinged wars reduce shared rules and convert civilians into enemies. … Civil wars and wars of secession with a strong ethnic component are dangerous for ethnic groups trapped behind enemy lines. The lure towards murderous ethnic cleansing increases when it can be accomplished at low military cost, with little fear of retaliation… . Military campaigns may generate tactical lure towards atrocities against civilians that were not originally intended. … Guerrilla warfare lures guerrillas to kill civilians. … These [and others] are all features of military power that may produce murderous cleansing. (p. 32)
Yet at the most general level of Mann’s analysis, in formulating his theses and locating his explanation in political power, this role of war is strangely neglected – he even falls prey to the old sociological mistake of seeing war as an exogenous variable, in that claim that murderous policies emerge out of ‘unrelated crises like war’. But wars are hardly ‘unrelated’: genocidal powers are usually fighting conventional wars against either states or armed movements to which the target populations are linked – or believed to be linked. Perpetrator regimes, even where they are not fighting conventional wars, are often highly militarized – this is part of how they homogenize and centralize power. Even settlers are ubiquitously armed, often warring against indigenous peoples. So he should go further: extensive, systematic murderous ‘cleansing’ or genocide generally (although not always) arises from and is carried out through war. Most genocide is actually part of genocidal war. Murderous politics do not arise primarily from political power relations, but out of political power that is fundamentally conditioned by military power: out of the interactions of political and military power.
The limitations of Mann’s framework are conceptual as well as analytical. Unlike Fascists or the two volumes of The Sources, The Dark Side short-circuits the conceptual debate – although ‘ethnic cleansing’ and ‘genocide’ are highly contested terms. Mann does not fully justify the adoption of ‘cleansing’ as a master-concept. This is unfortunate, since the term reflects ideologies of racial purification: it is widely rejected as a perpetrator euphemism unsuitable for social-scientific use. As even Norman Naimark, a historian who uses the term, remarks (2001: 193): ‘There is nothing “clean” about ethnic cleansing. It is shot through with violence and brutality in the most extreme form.’ Mann defines it as ‘the removal by members of one [ethnic] group of another such group from a locality they define as their own.’ His insistence that ‘murderous cleansing’ is a subtype gives credence to the notion that removal can be non-violent – a manifestation of political but not military power. Yet the wholesale removal of a population group from their homeland is generally involuntary, resisted, and enforced through extreme coercion. The forms and extents of violence vary greatly, but ethnic removal generally falls under Raphael Lemkin’s (1944: 79) original concept of genocide as the ‘the destruction of a nation or of an ethnic group’. The ‘cleansing’ of other groups (classes, political enemies, etc.) falls under the expanded generic concept of genocide as group destruction that others have developed from his starting-point.
Mann has adopted the 1990s reinvention of genocide as ‘ethnic cleansing’ and the corresponding narrowing of genocide itself to the intentional murder of all the members of a group. This definition leads him to the conclusion that the extermination of the Jews was ‘the only really large genocide attempted by the Nazis.’ (The Dark Side, p. 188) This position is not only at odds with the Nuremberg judgements but also, by emphasizing the differences among Nazi policies, may feed resistance to a coherent explanation of the Nazis’ wide-ranging attacks on many different groups in European society (even if Mann himself avoids this trap). The narrow concept of genocide as premeditated mass killing artificially emphasises this method rather than the content of group-destruction. It separates killing from the other violence (expulsion, rape, torture, robbery, etc.) and coercion (legal, administrative, economic and cultural appropriation) with which states and other armed powers characteristically destroy social groups. Paradoxically, Mann seems to define genocide so that it belongs only to military power – and even war is not defined so simply by killing – although he does add in the cultural dimension. We need to return to the broader sociological concept of genocide initiated by Lemkin, and to the intimate linkages with war that were recognized in its first legal formulations. Suitably refined and expanded, genocide can provide a better framework than ‘ethnic cleansing’ for understanding the range of organized violence against civilian populations.
These criticisms should not detract from Mann’s achievements. The Dark Side‘s theses are enormously suggestive, clearly have considerable purchase on the cases and inform rich interpretations that set new standards of analytical complexity in historical sociology. I have hardly done justice to Fascists, a monumental work in its own right, still less to Incoherent Empire, a formidable account of the limits of new American power-projection that makes excellent use of that distinction between military and political types. It is a measure of these works that, unlike many other texts in contemporary sociology, historians will take them extremely seriously. It may take more time for Mann’s own discipline to absorb their contributions, since their subjects seem remote from most sociologists’ research interests. It is the strength of Mann’s detour, however, that it could single-handedly return the topic of political violence to the centre. Genocide is, after all, the supreme crime of social classification: what question could be more important for sociology to address?