This article was downloaded by:[Powell, Christopher] On: 21 November 2007 Access Details: [subscription number 784380422] Publisher:

Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Genocide Research

Publication details, including instructions for authors and subscription information: http://www.informaworld.com/smpp/title~content=t713431069

What do genocides kill? A relational conception of genocide

Christopher Powell Online Publication Date: 01 December 2007 To cite this Article: Powell, Christopher (2007) 'What do genocides kill? A relational conception of genocide', Journal of Genocide Research, 9:4, 527 - 547 To link to this article: DOI: 10.1080/14623520701643285 URL: http://dx.doi.org/10.1080/14623520701643285

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf This article maybe used for research, teaching and private study purposes. Any substantial or systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

Journal of Genocide Research (2007), 9(4), December, 527547

What do genocides kill? A relational conception of genocide

CHRISTOPHER POWELL

An essentially contested concept What does a genocide kill? The word genocide rst appeared in print in 1944, as the title and the subject matter of Chapter IX of Axis Rule in Occupied Europe by Polish jurist and legal activist Raphael Lemkin. Briey, it meant the destruction of a nation or of an ethnic group,1 although Lemkins conception of genocide is much more complex than this. Lemkin formed the word by combining the ancient Greek word genos (race, tribe) and the Latin cide (killing), thus corresponding in its formation to such words as tyrannicide, homicide, infanticide, etc; he also suggested in a note that the word ethnocide would be equivalent. He coined the term to refer not only to what the Nazi regime was doing to Jews, but what it was doing and planned to do to other ethnic groups throughout the Reich, particularly in the territories it had conquered in Eastern Europe. Lemkin stated that he wanted the concept of genocide to protect the right of national groups to exist, just as the concept of homicide protects the right to life of human individuals.2 At rst glance Lemkins logic appears straightforward: like tyrannicide or infanticide, genocide is the killing of a distinct type of subject. Unlike tyrannicide or infanticide, the subject in question is collective, not individual. This subject is designated by the word genos, which Lemkin translated as race or tribe. There are two distinct problems with Lemkins concept. The rst has to do with the problematic nature of the concepts of race, tribe, and nation. Lemkin suggested in a footnote that another term could be used for the same idea, namely, ethnocide, consisting of the Greek word ethnosnationand the Latin word cide, and all throughout Chapter IX he wrote of nations or national groups rather than races or tribes. Lemkin considered races, tribes, and nations to be the same thing. They are not, however, and the assumption that they are belongs to racist thinking that contemporary genocide should abhor.3 I will discuss this issue further in the third section. The second problem is even thornier, and concerns the nature of social collectivities. Is a collectivity something more than the sum of its individual members? If not, then what genocide really kills
ISSN 1462-3528 print; ISSN 1469-9494 online/07/040527-21 # 2007 Research Network in Genocide Studies DOI: 10.1080/14623520701643285

CHRISTOPHER POWELL

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

are individuals, and what distinguishes genocides from other kinds of homicides are the reasons those individuals are killed, i.e. the motives of the perpetrator, and a genocide is homicide where the victims are killed because of their group membership. On the other hand, if a group is something more than or other than the sum of the individuals who belong to it (which is my view), then genocide is the killing or destruction of that something more, and we need to specify just what that something more is to dene genocide rigorously. Scholars who study genocide have proposed a large and diverse ensemble of denitions for the term; readers of this journal will be familiar with the lively and interminable denitional debate, and I do not need to summarize it here. But, as Dirk Moses has argued, many of the positions in this debate can be sorted into one of two kinds: liberal and post-liberal.4 Liberals conceive of genocide as the intended action of a coherent agent. Consequently, events which they count as genocide tend to feature states that demonstrate a will to genocide. These authors warrant being called liberals for two reasons: rst, because of their individualist conception of the social; second, because they take a generally afrmative view of liberal-democratic states and regard genocide as mass murder committed by totalitarian or dictatorial states. Post-liberals conceive of genocide as a structural process that does not require any intending agent. They more readily recognize as genocide those events that cause the destruction of a social collectivity but for which evidence of a coherent exterminatory intent is not available. Given the culpability of white settler societies for the destruction of indigenous societies and cultures, the post-liberal position denies any strong negative correlation between liberal-democratic regimes and propensity to genocide. Mosess discussion implies that liberals and post-liberals also differ on the question of what genocide kills. In the individualistic sociology of liberals, Lemkins genos designates nothing more than a group of individuals who shareeither actually, or in the mind of the genociderssome common attribute or attributes. Genocide is the murder of individuals, targeted because of real or imputed common attributes distinct to them as members of a group. For postliberals, on the other hand, the genos of genocide designates an emergent social structure, irreducible to the individuals who make it up. Genocide is the murder or destruction of this structure. Thus genocide might or might not involve any actual homicide. The strengths and weaknesses of these two positions are those of individualist and holist sociologies in general: if liberal positions are too voluntaristic and take little account of how individual action is constrained by emergent social structures, post-liberal positions are too deterministic and leave little room for agency and, with it, moral responsibility. Liberals emphasis on intent means that discussions of whether genocide has occurred focus more on how perpetrators perceived their own actions than on the experiences of victims, which seems perverse. Post-liberal positions dene genocide in terms of what happened to victims, and make intent secondary; in principle, genocide could even happen by accident. Post-liberals regard as indispensable what liberals could dene away altogether: the collective quality of genocide. But they suffer from a tendency to take the 528

WHAT DO GENOCIDES KILL?

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

genos as given: Sartre,5 for example, speaks of the Vietnamese people as an undifferentiated unity, and Barta6 writes as if the Aboriginal race exists as a natural human group. Liberal positions, on the other hand, can more easily recognize that the victim group as targeted by the genociders might be imaginary, either in part (Jews as dened by the Nazis) or almost entirely (bourgeois as dened by the Khmer Rouge). Essential contestedness Denitions of genocide are like denominations of Protestant Christianity: every attempt to end schism with a new, more catholic proposal only succeeds in producing another schism. It seems useful to accept this situation for the time being and recognize genocide as an essentially contested concept, as dened by W. B. Gallie.7 The term genocide is an evaluative concept. It refers to a complex phenomenon, one that can be described coherently in a variety of ways. It has turned out to be open to modication, in unforeseeable ways, in light of new historical events. The parties to the denitional debates engage aggressively and defensively with other usages to maintain their own. The term genocide is based on a single generally recognized exemplar (the Nazi holocaust), and this exemplar is open to a variety of internally coherent but mutually incompatible interpretations. The fullment of these criteria means two things: rst, that genocide scholars will likely continue to feel that there is something to be gained from denitional debates; and second, that these debates cannot be resolved on purely logical grounds, because no general principle for deciding which usage is best has yet to be found or specied.8 Where does this leave us with respect to genocide? Gallie distinguishes between the logical and the historical senses of what it is to understand a concept. The logical sense involves being able to state and conform to the rules governing a concepts use, while the historical sense involves knowing about the whole gamut of conditions that have led to, and that now sustain, the way we use it.9 For evaluative concepts, a purely logical understanding is insufcient, precisely because the rules of use may be essentially contested. In these cases a logical understanding must be supplemented by a historical understanding:
For, if we want to see just what we are doing, when we apply a given appraisative concept, then one way of learning this is by asking from what vaguer or more confused or more restricted version (or ancestor) our currently accepted version of the concept in question has been derived.10

This does not mean looking for an original historical denition of a term. There is no guarantee that earlier versions of a term are any less problematic than contemporary ones. The point of history is not to nd an essence that has been accidentally lost, but to identify the conditions that produced and sustain a classifying practice. The complete project of explicating the whole gamut of conditions that have led to, and that now sustain, the way we use the term genocide would probably be 529

CHRISTOPHER POWELL

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

best served by a genealogy, in the sense that Foucault practised it.11 In particular, a genealogical emphasis on the effectivity of classifying practices, on how classifying practices participate in relations of force that address human bodies, would bring out the practical political stakes that have fuelled dissention over the meaning of genocide over the past six decades. With the remainder of this paper, however, I would like to address a narrower task within this larger project.

Subjectivism and objectivism As might be already evident, the denitional strategies within genocide scholarship that Moses characterizes as liberal and post-liberal correspond to particular methodological tendencies, in sociological theory more generally, that we could call subjectivist and objectivist. Subjectivist sociologies claim that the social exists in the meanings attached to action by human minds. The social world is subjective in the literal sense of being mind-dependent, a product of human consciousness. This view can be found, for example, in the rst chapter of Webers Economy and Society:
For still other cognitive purposesfor instance, juristic onesor for practical ones, it may on the other hand be convenient or even indispensable to treat social collectivities, such as states, associations, business corporations, foundations, as if they were individual persons. [. . .] But for the subjective interpretation of action in sociological work these collectivities must be treated as solely the resultants and modes of organization of the particular acts of individual persons, since these alone can be treated as agents in a course of subjectively understandable actions.12

By contrast, objectivist sociologies claim that the social exists as structural patterns irreducible to, and at least partly independent of, human subjectivity. The social world is mind-independent; its distinctive properties exist apart from human consciousness of them. This view appears in Durkheims The Rules of Sociological Method, when he says that the rst and most basic rule is to consider social facts as things13 and that:
Social phenomena must therefore be considered in themselves, detached from the conscious being who form their own mental representations of them. They must be studied from the outside, as external things, because it is in this guise that they present themselves to us.14

Different denitions of genocide, different conceptions of what genos could refer to, presume different theories about the nature of social life. Each position faces the problem of how to connect the individual and collective dimensions of social life and, in our case, of genocidal violence. Durkheim, for example, took more account of the meaning individuals attach to their own action than his uncompromising statement in the Rules would imply. Individuals motives for committing suicide and the meanings they attach to the act are part of what distinguishes Durkheims three principal types of suicide.15 The problem of the social genesis of individuals own mental representations was a major theme 530

WHAT DO GENOCIDES KILL?

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

of his The Elementary Forms of Religious Life. Conversely, Weber treated intentionality only as a necessary starting point for sociological investigation, which then proceeds to develop a rigorous nomothetic analysis.16 Webers methodological individualism has an important relational dimension: social action is by denition oriented to the behaviour of others17 and his concept of the social relationship, dened as the behaviour of a plurality of actors insofar as, in its meaningful content, the action of each takes account of that of the others and is oriented in these terms,18 is indispensable to his institutional or structural analyses of bureaucracy, protestant asceticism, and so on. Martin Shaw builds on Webers approach, to understand genocide as one example of a structural phenomenon in the sense that it is a recurring pattern of social conict, characterized by particular kinds of relationship between actors, and with typical connections to other conict structures in society.19 Shaw goes on to redene genocidal action as action in which armed power organizations treat civilian social groups as enemies and aim to destroy their real or putative social power, by means of killing, violence, and coercion against individuals whom they regard as members of the groups.20 In this way, structural concepts can emerge from a subjectivist epistemology. However, the social collectivities that emerge from such analyses always have a provisional candidacy to the status of real phenomena;21 such entities are not autonomous or self-organizing, as they are in objectivist sociologies, but are assumed to be reducible entirely to individual action. Whatever the strengths and advantages of an approach like Shaws may have on its own terms, it has the slight disadvantage of leaving out a crucial feature of Lemkins thinking. For, as I will argue in the next section, Lemkins conception of the social was more like Durkheims, or Spencers, than like Webers. Lemkin treated social groups, or some of them at least, precisely as if they were individual persons. This position has its own problems: it relies on essentialist and even racist assumptions about the nature of human communities. In the nal section of this paper, I will propose that a third general approach to conceptualizing social structure, a relational approach derived from the work of Norbert Elias, can rescue the concept of genocide from these essentialisms without sacricing the notion that the life of a collectivity involves something more than the sum of its individual parts. Lemkins functionalism Lemkins two faces? In the course of any particular episode of organized mass violence directed towards a social collectivity, the relationship between the killing or bodily harm of individuals and the destruction of collective cultural life will vary according to the circumstances. But what kind of relationship between the two is necessary to the concept of genocide? Strongly liberal positions, to use Mosess terminology again, treats killing as the substance of genocide and considers the destruction 531

CHRISTOPHER POWELL

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

of cultural life secondarily or not at all.22 Strongly post-liberal positions that take the destruction of collective life as the substance of genocide might reasonably treat actual killing as not necessary to the process.23 But Lemkin emphasized both the physical and the cultural dimensions of genocide, creating something of a puzzle for most genocide scholars up to this point.24 The concept of genocide was Lemkins second major intervention into international law. The rst was unsuccessful but important nonetheless. In 1933, Lemkin submitted a proposal to the International Conference for the Unication of Criminal Law in Madrid to declare the destruction of racial, religious, or social collectivities a crime under the law of nations.25 This proposal included two crimes: barbarity, dened as action against the life, bodily integrity, liberty, dignity, or economic existence of a person, if taken out of hatred towards a racial, religious, or social collectivity, or with a view to the extermination thereof; and vandalism, dened as destruction of cultural or artistic works for the same reason. Just as Nazi atrocities provided the occasion for Lemkins germinal discussion of genocide in 1944, his 1933 project found its specic inspiration in the atrocities committed against the Armenian national group by the Young Turk government of the Ottoman Empire and, subsequently, by the state of Turkey.26 Writing in 1947, Lemkin characterized his later project as an extension of the former, despite their terminological differences.27 Elements of both barbarism and vandalism, of physical and cultural violence, are incorporated in the 1944 concept of genocide (I will say more about this in a moment). However, in the drafting of the United Nations Convention on the Prevention and Punishment of the Crime of Genocide 28 (in which Lemkin was a key participant), physical and biological aspects predominated. An early draft of the resolution included political groups among the groups whose destruction could constitute genocide, but this was opposed by representatives of the Soviet Union and others.29 An entire article delineating criteria of cultural genocide was removed at the urging of US representatives.30 The wording of the Convention was shaped by the desire of its framers not to criminalize their own behaviour. Important as the Convention is in its own right, it expresses political forces beyond Lemkins control and belongs to the genealogical history of contested versions of genocide. So what of Axis Rule? Up to now, liberals and post-liberals have both been able to cite Lemkin in the defence of their position. Post-liberals can point to Lemkins discussion of German colonial ambitions in Eastern Europe, his emphasis that genocide was to serve as a tool for such colonialism, and his treatment of the genocide of Jews as one among several undertaken or planned by the Nazis. They quote Lemkin as saying that [g]enerally speaking, genocide does not necessarily mean the immediate destruction of a nation, that the objectives of genocide are disintegration of the political and social institutions, of culture, language, national feelings, religion, and the economic existence of national groups, and the destruction of the personal security, liberty, health, dignity, and even the lives of the individuals belonging to such groups and that 532

WHAT DO GENOCIDES KILL?

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

genocide is directed against the national group as an entity, and the actions involved are directed against individuals, not in their individual capacity, but as members of the national group.31 They can also point to his claim that genocide has two phases: one, the destruction of the national pattern of the oppressed group; the other, the imposition of the national pattern of the oppressor and that this imposition, in turn, may be made upon the oppressed population which is allowed to remain, or upon the territory alone, after removal of the population and the colonization of the area by the oppressors own nationals.32 However, liberals can point to Lemkins distinction between genocide and denationalization, which does not connote the destruction of the biological structure.33 Lemkin distinguishes between genocide and terms like Maygarization or Germanization, the latter of which means that the Poles, as human beings, are preserved and that only the national pattern of the Germans is imposed on them. Such a term is much too restricted to apply to a process in which the population is attacked, in a physical sense, and is removed and supplanted by populations of the oppressor nations.34 And, commenting retrospectively on Axis Rule, Lemkin stated that genocide may involve not only murder but the prevention of life through abortions and sterilization, and the deliberate endangering of life such as biological warfare, death by slave labour, and separation of families.35 Although there has been some important work done recently examining Lemkins unpublished writings on colonialism in the Americas and Africa,36 the dilemma still stands. If we assume that genocide is either about the destruction of individuals on account of their membership in a group, or the destruction of the culture per se, by various means of which killing actual people is only one, then Lemkin would seem to have it both ways.

The life of nations How to turn this apparent incoherence into coherence? The answer to that question turns to the (often unacknowledged) theoretical assumptions that inform competing denitional projects. Different ways of dening the same social phenomenon often express different theoretical conceptions of what social phenomena generally consist of. Liberal denitions of genocide express a methodological individualism in which society is made up of individuals and of subjectively meaningful, intentional social action. Post-liberal denitions express a methodological holism in which social structures exist objectively, independently of individuals and their subjective intentions. Now, if we read Lemkin closely, he was unambiguously a holist. In Axis Rule he talks about nations as having a life of their own, analogous to the life of individual human beings. He says that:
nations are essential elements of the world community. The world represents only so much culture and intellectual vigour as are created by its component national groups. [. . .] The destruction of a nation, therefore, results in the loss of its future contributions to the world. Moreover, such a destruction offends our feelings of morality and justice in much

533

CHRISTOPHER POWELL

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

the same way as does the criminal killing of a human being: the crime in the one case as in the other is murder, though on a vastly greater scale.37

And in notes for his unpublished History of Genocide, Lemkin underscored this point, saying:
The philosophy of the Genocide Convention is based on the formula of the human cosmos. This cosmos consists of four basic groups: national, racial, religious and ethnic. The groups are protected not only by reasons of human compassion but also to prevent draining the spiritual resources of mankind.38

Whether politically sovereign or not, the national group for Lemkin had an inherent right to life equivalent to that of the sovereign individual, and such groups provided the essential basis of human culture as a whole. The concept of genocide was designed specically to protect that life. This explains Lemkins insistence that genocide attacks the essential foundations of the life of national groups.39 For Lemkin, a nation was something like a person writ large, having an intellectual and cultural creativity of its own, but having also a biological life on which these higher faculties depend. So even if the point of genocide is the destruction of the nation as a social and cultural fact, this destruction necessarily involves an attack on its physical existence. The physical extermination of members of the nation is not necessary, as long as they are made to be no longer members of one social body. A Spencerian organicism about the social makes sense of Lemkins otherwise contradictory claims about what genocide involves. If individuals are organically integrated into the culture to which they belong, then they are not separable from it (or it from them). The killing of individuals and attacks on language, religious practice, or other cultural institutions are not two different kinds of violence belonging to two different kinds of genocide, but differing aspects of a single multidimensional process. Genocide does not reduce to the killing of individuals because the life of a nation does not reduce to the physical survival of the individuals who make it up, any more than a persons life reduces to the cells that make up their body. However, murderous violence will necessarily be part of the genocidal process, in the same way that killing an individual will necessarily involve the destruction of at least part of their body. If we recognize that Lemkin denes genocide in terms of the violation of a nations right to its collective existence, we can identify an important source of confusion over what he has said. He combines performative statements about what the word genocide means (by genocide we mean the destruction of a nation or of an ethnic group) with descriptive statements about how he thinks genocide works and what it does. We can make a distinction between these two aspects of Lemkins thinking. Genocide is the destruction of a nation. However, it may happen in two ways: immediately, through the mass killings of all members of a nation (but generally speaking it does not necessarily involve this); or through a coordinated plan of different actions aiming at the destruction of essential foundations of the life of national groups.40 Lemkin described seven different elds in which genocide was being carried out41 by 534

WHAT DO GENOCIDES KILL?

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

German occupiers in Eastern Europe at the time he was writing: political, social, economic, biological, physical, religious, and moral. These elds do not seem to form part of the conceptual denition of genocide, but we can read them as part of a theory about what are the essential foundations of the life of national groups. In other words, Chapter IX of Axis Rule can be read as implying sociological claims about the nature or structure of nations and the necessary conditions for their survival. As Moses observes, these claims are of a communitarian nature and so bafe anyone who thinks in exclusively individualist terms.42 Lemkins tacit functionalism This functionalist reading of Lemkin is corroborated by material from his unpublished later writings. Lemkin planned to follow up Axis Rule with a multivolume History of Genocide, covering ancient, medieval, and modern time periods.43 His notes for this project, intended as his magnum opus, include numerous chapters on genocide by European colonizers against indigenous populations in the Americas, Australasia, and Africa.44 To organize his treatment of these genocides, Lemkin prepared a template; this template lists the following types of methods and techniques of genocide:
physicalmassacre and mutilation, deprivation of livelihood (starvation, exposure, etc. often by deportation), slaveryexposure to death; biologicalseparation of families, sterilization, destruction of foetus; culturaldesecration and destruction of cultural symbols (books, objects of art, loot, religious relics, etc.), destruction of cultural leadership, destruction of cultural centres (cities, churches, monasteries, schools, libraries), prohibition of cultural activities or codes of behaviour, forceful conversion, demoralization.45

This treatment, which Lemkin applied in chapter after chapter of this unpublished draft material, implies that physical genocide and cultural genocide were not two distinct phenomena in Lemkins mind, but that genocide was one process that could be accomplished through a variety of means. If a nation is an organic structure, the physical and cultural dimensions of which are interdependent and mutually indispensable, then the murder of a nation might be accomplished by attacking any one part of this structure or several at once. To sum up, three unstated assumptions give coherence to Lemkins claims about genocide: rst, national groups have an existence sui generis, irreducible to the individuals who make them up; second, national groups have natural foundations of life; third, national groups are the essential bases or units of human culture. Beyond functionalism: a relational conception of genocide The irreducibility of social structure Perhaps the boldest and most succinct formulation of strong objectivism about culture is Durkheims The Rules of Sociological Method, which declares that 535

CHRISTOPHER POWELL

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

the rst and most basic rule is to consider social facts as things.46 Durkheim denes a social fact as a way of acting that individuals confront as external to themselves, capable of constraining their actions, and that exist independently of any one of their individual manifestations. Social roles provide everyday examples of this; each individual who becomes a wife, or a mother, or a student, or a welder encounters a set of expectations, obligations, privileges that she has not herself devised and which she can modify only with effort if at all. In a similar fashion, each individual member of society encounters institutionalized forms of interaction such as money, or the local language, which he can ignore only at cost to himself and by which he therefore is constrained to some degree. Of course, all sociology includes some notion of social constraints that do not reduce simply to the intentions of individual actors; Webers analyses of bureaucratic rationality and of protestant asceticism are preoccupied with how social action creates unintended consequences that run counter to the motives of the involved actors. But Durkheims insistence that social facts are explicable only in terms of other social facts,47 and not in terms of the subjective intentions of actors, makes social facts into a class of natural phenomena with a unique type of reality, capable of reproducing themselves over time independently of any specic human agency. A religious community, for example, is at any given moment composed of people who share a commitment to certain doctrines and practices. If this commitment entails inculcating these practices in others (in the children of the community, for example, or in outsiders), then those others are induced to take up the same practices, and also the same task of reproducing them. In this way, social facts reproduce themselves across different actors and populations of actors. Furthermore, just as social facts are greater than the sum of their parts, so whole societies are greater than the sum of their parts; for Durkheim, societies are ensembles of organically interdependent social facts, just as human bodies are ensembles of organically interdependent organs.48 This strong view of society as an organism has little support among sociologists today. But within Durkheims functionalism we can discern a more durable and robust pair of claims: rst, that social structures have properties irreducible to individual action; and second, that these structures connect up with each other to form self-reproducing wholes or systems. This position is central to some of the most important projects in social theory of the last century, cutting across left right political divides, including: the structural functionalism of Talcott Parsons; the ethnomethodology of Harold Garnkel; the structural Marxisms of Nicos Poulantzas or of Louis Althusser; the world systems theory of Immanuel Wallerstein; and the autopoietic systems theory of Niklas Luhmann, to name only a few. So although we may disagree today with Lemkins organicism about nations, it belongs to a trajectory of holist social theory that is still inuential. That said, the claim that social structures exist objectively does have its share of problems. Or rather, the whole opposition between subjective and objective facts in social life is recognized as problematic because it implies a series of irreconcilable oppositions: agency and structure, individual and society, micro and macro, meaningful and meaningless, and so on. A growing tendency in sociology 536

WHAT DO GENOCIDES KILL?

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

seeks to transform these either/or dualisms into both/and monisms, by abolishing the opposition between subjectivity and objectivity altogether and replacing both poles of the opposition with a relational conception of the social.49 The list of theorists who could be said to be involved in this tendency, either directly or indirectly, is long, and some theorists (Marx, for example) can be read either dualistically or relationally. For my purposes I would like to focus on Norbert Eliass version of relational sociology, because of his unique criticism of egocentrism and his emphasis on process. A relational conception of social structure Elias afrms that social life is characterized by structures that are not explicable in purely individualist terms, but he rejects the Durkheimian-Parsonian claim that social structures exist independently of individuals. In Eliass view, the positions that I have been calling subjectivist and those I have called objectivist suffer from the same mistaken premise: that individuals exist prior to social relations.50 Elias calls this premise egocentrism, and refers to the imaginary self-sufcient individual required by egocentric theory as homo clausus. It is pure ction, Elias argues, to suppose that human beings could ever exist autonomously: from before our birth we depend on other human beings for the necessities of life, and for all the possible means by which we could realize our selves in the world. Our very subjectivity is formed out of the practical and taken-for-granted set of skills, attitudes, understandings by which each of us conducts our life, and these develop only through our relationships with each other. The essential human condition is not being, but being with others. And, just as concrete human beings cannot exist independently of society, society cannot exist independently of concrete human beings. Egocentric sociology (and Eliass primary target is Talcott Parsonss structural functionalism) puts the abstracted gure of the individual at the centre of concentric series of rings that represent successively more macroscopic levels of reality, each somehow separate from the levels above and below it. Elias substitutes, for these neat concentric circles, the image of a tangled network of relationships: interactions, interdependencies, balances of power, all in a constant state of ux.51 Order is immanent to this network as the pattern of motion and change within it; there are no transcendent structures outside or above the individual. The propensity of modern Western subjects to perceive themselves as isolated units (individuals) surrounded by a social world external to themselves is itself a product of specic cultural tendencies that can be explained sociologically.52 The other distinctive feature of Eliass relational sociology is its orientation to process. The dominant conceptions of social structure in the twentieth century have generally been synchronic, or static: structure is a pattern that exists at a given moment in time; when it reproduces itself over time, it does so in the same xed form. Elias, on the other hand, insists that change is integral to social structure, which can only be understood if viewed diachronically.53 Just as glaciers, for example, seem static to ordinary human perception but are actually 537

CHRISTOPHER POWELL

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

uid and moving, so all social structures that seem xed and stable are actually in a continuous process of change. Process is an integral dimension of structure, in just the same way that movement is essential to dance, for example.54 Failure to recognize this can lead social scientists to imagine culture as something essentially xed, and so to misperceive the very fact of change, inevitable in itself, as sign of pathological breakdown or decay. It can also lead them to treat local and contingent social structures as universal and timeless. For example, Talcott Parsonss universal theoretical model of the social system is patterned after the modern sovereign nation-state (and an idealized version of that), which is in the long view a very recent and transitory form of social life.55 Indeed, Elias proposed using the term guration rather than structure to refer to the ordered quality of social life, on the grounds that structure has become essentially and irrecoverably associated with stasis.56 The Oxford English Dictionary lists, among the meanings of the word guration, both (a) the action or process of forming into a gure and (b) the resulting form or shape; Eliass use of the word invokes both of these meanings simultaneously. What does this say about genocide? In a relational view, genos must connote a type of social guration. The collective object designated by Lemkins use of genos must, like other social structures, have the general property of being a dynamic relational network formed through practical social interactions in historical time. Several points ow from the fact of treating genos as guration and genocide as the destruction of a guration. First, given that a genos is a network of practical social relations, destruction of a genos means the forcible breaking down of those relationships. This corroborates Lemkins insistence that genocide can be accomplished in a variety of ways. The most obvious way to destroy a network might be to kill all the people in it, but it might be easier and less expensive to kill only enough people to cause the network as a whole to collapse; or to disperse the people in it so that they can no longer be connected to one another; or to prevent them from communicating with other; to suppress their language, religion, law, kinship systems, and other cultural practices through which they maintain the relations among themselves; to break up the central institutions that maintain the cohesiveness of the network their government, economic system, religious organizations, centres of learning, and so on; to impose on them harsh conditions of life that break down social solidarities; and on, and on. These effects could be produced without a coherent intent to destroy; they could result from sporadic and uncoordinated actions whose underlying connection is the production of a new society in which there is simply no room for the genos in question to exist. They might even result from well-meaning attempts to do good: to enlighten, to modernize, to evangelize. Second, given that a genos, like all social institutions, is a process of change and transformation, the effect of genocide is to disrupt that process. Because genocide is itself a form of change, this may seem counterintuitive, until one considers that survival does not mean preservation of a xed, unchanging set of practices. Neither Jewish culture in Europe, for example, nor pre-contact Cree culture on Turtle Island, existed as a timeless and static roster of traditions; though each 538

WHAT DO GENOCIDES KILL?

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

preserved important continuities with its past, each was nonetheless in a process of continual transformation, which had a denite historical form, and this form of change was radically violated by the respective genocides that each experienced. A living, breathing social guration (as it were) decays and grows at the same time, producing new ideas, new institutions, new practices, from which emerge the future contributions to the world that Lemkin wrote of. Genocide violently interrupts this process. We may count among the means by which genocide may be committed the measures that interrupt the reproduction of the guration over time, the passing on of culture to children, the renewal of social institutions, and also the measures that prevent change, through the silencing of innovation in thought, art, technology, everyday practice, or through forcible connement to a fossilized tradition that is not allowed to be transformed. Third, this conception has the merit of directly opposing the grounds on which genocide is often justied. Genociders rely on stereotypes to dehumanize their victims or otherwise remove them from the universe of moral obligation.57 Stereotypes differ from ordinary types or categories in that they work to essentialize, naturalize, and x social identities and the boundaries between identities,58 thus absolving the stereotyper of any involvement in or responsibility towards the stereotyped Other. Understanding genos as relational and processual works directly to undermine stereotyping. More than this, it shows that, because genocidal violence is necessarily enacted through a relational process itself, genocide involves an interconnectedness between perpetrator and victim that the former would prefer to deny. The perpetrator in a sense depends on the victim: identity is always actually a relation of identity-difference, or in process terms, of identication-differentiation. I cannot be my Self without standing in relation to some Other. Thinking of identity in relational terms could aid the construction of what Bauman thinks is essential and always given: a Levinasian sense of fundamental obligation to the other.59 But if genos is a guration, what type of guration is it? Lemkins 1944 denition treats race, tribe, and nation as interchangeable terms. This assumption is itself symptomatic of nationalist thought, and of racist nationalism at that. Of the three terms, race is the most obviously problematic. Races, in the most prominent sense of the word as distinct biological groups within a species, do not exist within the human population.60 If races do not exist, one cannot kill them. Human groups identied on the basis of race do exist, but only as a result of social processes of racialization.61 The cultural characteristics of these groups derive from social dynamics and not from any distinct biology; the idea that biological difference explains cultural difference is the dening feature of racist thought.62 Racism, moreover, is far from a universal feature of human social organization. Bauman points out that the idea of a systematic, and genetically reproduced distribution of such material attributes of human organism as bore responsibility for characterological, moral, aesthetic or political traits is distinct to modern Western science.63 Heterophobia and contestant enmity may be universal human phenomena, but racism is modern,64 and by extension so is the notion of race in anything like the sense that the Nazis used it. So translating an ancient Greek word as race 539

CHRISTOPHER POWELL

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

commits anachronism, and dening genocide in terms of race makes the concept racist, according to the primary but less familiar meaning of the term (see Note 3), even if the purpose of the concept is anti-racist in another sense. The term tribe is more complex but equally unsound. Some anthropologists have questioned whether it, like race, is irredeemably incoherent.65 The term has many denitions, some of which emphasize shared lineage, some of which connote a self-governing socio-political group, and some of which connote both at once. The word is itself derived from the Latin tribus, which was rst used to refer to a (mythical) threefold division of the peoples of Rome, the Latins, Sabines, and Etruscans.66 Tribe has also been used as a translation of the Greek word phyle and also of the Greek word gens, but this is necessarily imprecise given that the phylae (tribes) were frequently composed of phratries which, in turn, were thought to be assemblages of clans (gens).67 In late medieval Europe the word began to be used to refer to the 12 confederations of ancient Israel, and in this sense came to designate a group of persons forming a community and claiming descent from a common ancestor and to be conated with lineage group.68 But at some point in the modern era the word acquired pejorative connotations and came to designate a form of grouping specic to people seen as primitive or barbarous.69 Finally, the word also sometime connotes a socio-political unit, united under a headman or chief or united into a social or political community.70 It is clear that there need be no tension among these various connotations if we subscribe to the racist assumption that lineage groups form the natural basis of sociopolitical organization within the human species. But otherwise, the term generates problems. Lemkin wished to protect the collective existence of national groups: not only Jews, but Armenians, Poles, Aztecs, Herero, and so on. As Weber observed, national or ethnic groups are not generally made up of members of a single lineage group; the claim that they are is often made by political elites hoping to use that belief to consolidate state power.71 Again, dening genocide in terms of tribe panders to the same essentialist mythologizing that has been so useful to genocidaires. Nation is the least problematic term of the three, but it has its share of problems. The most conspicuous of these is narrowness. The United Nations Convention on the Prevention and Punishment of the Crime of Genocide denes genocide as the destruction in whole or in part of a national, ethnic, racial, or religious group. As I mentioned earlier, social and political groups were seriously considered for inclusion in the UNCPPCG, until it was removed in an eleventhhour compromise.72 Many scholars dene genocide in ways that explicitly or implicitly allow political and social groups generally to count as victims of genocide.73 Lemkin himself was satised with the list of victim groups dened by the UNCPPCG, and this makes sense given his view that national, racial, ethnic, and religious groups form the essential basis of human community. But Lemkins primordialism is mistaken; nations are not universal categories of human community. Nations are specically modern phenomena.74 Nationhood is one culturally specic way of giving form to human culture. Both Axis Rule and the United Nations dene genocide in terms that make essentialist assumptions about 540

WHAT DO GENOCIDES KILL?

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

human identity. If we do not share those assumptions, what are we to make of genocide? One possibility is that we understand the concept symptomatically, in terms of the historical situation from which it emerged and the problem-situation it attempts to address. What is important about nationality, ethnicity, race, and religion? They are all forms of identity that have been particularly important to the power of the nation-state. In a sense we can say that Lemkin was preoccupied with nations for the same sorts of reasons that the Young Turks and the Nazis were preoccupied with nations. A Foucaultian genealogy of genocide would be useful to us here. A genealogy would situate the criminalization of genocide and the moral appeal of Lemkins claim that nations have a right to life in the context of that ensemble of practices by which states have sought to govern populations through nationhood practices by which states have sought to cultivate and to take advantage of this particular form of collective identication, in which dispersed practices of language, religion, and folk custom join together to form a solidarity capable of uniting subjects with their sovereign. The Ottoman Empire, for example, identied itself by its royal house, not by the name of the dominant ethnicity within its multi-ethnic polity. The ethnic-cumreligious communities under its sovereignty were called millets, and each millet had its own form of citizenship, was subject to its own laws; the millets had obligations toward the state and the state had moral and political responsibilities to them. The notion of millet connoted just this relationship, which had no place in the nationalist community of the Young Turks.75 In imperial, Wilhelmine Germany also, the Junker aristocrats did not identify themselves with the German Volk; neither did the Freikorps who sought by paramilitary means to topple the Weimar Republic and restore German imperial glory. It was left to the Thule Society and its disciple Adolph Hitler to join German imperial ambitions to volkish populism.76 If nations are imagined communities, in Andersons phrase, then Raphael Lemkins own political consciousness was formed at the very climax of the period during which nations were being imagined into existence. And genocide greatly facilitated this process of construction. It has been generally observed that identication of an enemy Other works to foster solidarity within a group;77 genocide takes this common process to its furthest extreme. This, too, illustrates the relational quality of social identity: Self and Other are not separate or separable, but exist in relation each other, even where that relation is one of opposition. A relational conception of genocide Lemkins conception of genocide emerged in the context of this problemsituation, in which states had turned to nationality, ethnicity, race, and religion as means by which they could foster collective identication of their subjects with the project of sovereignty. Violent annihilation of the identities so constituted was one logical extension of this process. Lemkins primordialism about nations is a symptom of the same nationalist ideology that rationalized the atrocities of the 541

CHRISTOPHER POWELL

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

Young Turks and the Nazis. But it also expresses a more protean concern to oppose the uses of state force and protect the right to existence of forms of collective identity. In a relational sociology, individual and collective identity what Elias called I-identity and we-identity depend on each other. Individuals form their personal subjectivity through relations with others in a denite social context, and identication with others is a crucial part of this process. To violently destroy a collective identity, then, is to violently destroy a crucial part of the individual self. This fact alone might entitle collective identities to legal protection on liberal terms. Only in relation to a collective self is it possible for me to be my individual self. What does genocide kill? I propose that the collective subject we seek to protect through the proscription of genocide should be no less complex and protean a phenomenon than that of collective social identity. In this view, genocide is the process of violent destruction of that type of guration (a self-reproducing dynamic network of practical social relations) whose organizing principle is a collective identication. (This process does not have to be complete, and in fact rarely is; European Jews suffered a genocidal process that, thankfully, was not completed, and the same can be said of surviving indigenous peoples in the Americas.) Or more radically, we could understand genocide as an identity-difference relation of violent obliteration. This formula more easily accommodates cases where genociders channel violence through a ctive identication, as when people are killed for being Jews or Communists or bourgeois when they are not actually Jewish or Communist or bourgeois, either in their own eyes or by previously established standards. This point bears some elaboration. As I have already mentioned, in a relational view identity always operates by virtue of identity-difference relations. The collectively (if heterogeneously) shared identity of members in group always refers to the collectively if heterogeneously shared difference between members and outsiders. Identity, then, always results from ongoing processes of performance or construction or constitution, and never exists apart from these processes. It follows that genocide, as a performance of an identity-difference relation where the gure of the Other is subjected to violent obliteration, always transforms the performance of those relations, and thereby transforms the nature of identity within each of the involved groups (including any involved third parties). In cases of genocide against imaginary groups (such as the enemies of the people that the Khmer Rouge sought to eliminate), the identity of the victim is effectively produced through the process of annihilation. The group is constituted as the network of people interpolated by the perpetrator as occupying an otherwise ctive subject-position; such groups, having no other basis of common identication, may not outlive the genocidal process, but while it lasts they are all too real for those involuntarily enrolled in them. But where ctive identications encounter pre-existing gurational networks, durable new identities can be forged in the crucible of genocide. Before Columbus, there were 75 to 100 million human beings in the Americas,78 living in hundreds of different cultural groups each with their own names for themselves but there were no Indians. 542

WHAT DO GENOCIDES KILL?

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

Jewishness as a pan-global identity became an altogether more formidable movement after the Shoah than it was before it. This is why I refer to ctive and not simply ctional identications; ctive means [o]riginating in ction, created by the imagination, ctitious, but also of a name: assumed, and assumed names become real when they are established as such. At the same time, this conception excludes from the category of genocide forms of violence that have common typological properties but that are not connected into a denite network by any actual social relations. For example: even if one were to allow that the exercise of womens right to abort a foetus she is carrying constitutes the murder of a human being, this practice could not register as genocide because there is no guration of collective identication involved. Human foetuses cannot actually or potentially constitute a relatively integrated self-reproducing social network. In a different vein, the violent persecution of an ethnic minority does not in and of itself constitute genocide. Collective identication can survive or even thrive in contexts where the human beings who bear that identity suffer outrageously. The shift from abuse and persecution to genocide involves a fundamental qualitative transformation, from a relation that assigns the Other an inferior or denigrated position in the wider guration to which both persecutor and persecuted belong, to one that works to deny them any position at all. This relational conception also claries the vexed question of whether the transatlantic slave trade should be considered a case of genocide. For all the enormity of their suffering, the millions killed in the Middle Passage and the atrocities of the New World, slaves qua slaves were not victims of genocide because neither the intent nor the practical consequence of the slave economy was to destroy the social category of slaves; indeed, it was concerned precisely to produce and maintain it. However, the tribes and kingdoms from which human beings were taken to be made into slaves, the many distinct sociocultural formations that were smashed and torn and mutilated by the slave trade, did suffer genocide. Like the colonization of the Americas, the transatlantic slave trade was not one genocide but many. In that context, events occurring outside of Africa the deaths of the Middle Passage, the tearing apart of families, the suppression of African religions were components of those genocides. Slaves were victims of genocide, not in virtue of their identity as slaves, but in virtue of their identity as members of societies destroyed by the slave trade. I recognize that this amounts to adding yet another denition of genocide to the already long list. Not all readers will accept this denition: genocide will remain an essentially contested concept for the foreseeable future. But this denition has at least three features to recommend it. First, if we see Lemkins coining of the term genocide as a symptomatic response to a particular problem-situation, then a relational conception of genocide accommodates the fact that this problem-situation has continued to evolve. Nation-states and empires nd it expedient to ravage people through any number of forms of collective identication: through the imagined communities of nation and ethnicity; through the ctive devices of race which produce real racialized subjectivities; through religion, and also through secular political identication; through sexual identities that 543

CHRISTOPHER POWELL

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

did not exist prior to the late nineteenth century;79 through physical disability; and so on in ways that will continue to evolve as new forms of collective identication emerge and old ones lose their urgency. Second, a relational conception of genos directs our attention to the empirical investigation of the self-organizing properties of social networks. If the denition of genos as a guration organized by collective identication sounds a bit vague, that is because we as genocide scholars (I include myself in this category) lack an informed understanding of the empirical historical dynamics of collective identication. In Canada, for example, much hinges on the question of who can and cannot legitimately claim to be Cree, or Metis, or Ojibway, and so on (and legitimately according to whom). A relational conception of genos tells us we can only get so far answering these question in terms of abstract ethical principles; we need an empirical understanding of how these identities have functioned historically to connect people through practical relations that foster a shared sense of self. Ironically, the study of genocide can contribute to this investigation: because genocide attacks the processes by which gurations exist, they in a sense expose those processes and reveal their importance. This leads to my third and nal point in favour of a relational conception of genocide: it enables us to pose the question of reconstruction, of renaissance. How can the survivors of genocide rebuild the partially but not completely destroyed gurations to which they still belong? Clearly this rebuilding would not mean an identical reproduction of institutions and practices that existed before the genocide. It would mean a re-growth of the network of practical relations that sustain collective identication. It would mean a restoration of the process of change and transformation that gives gurations their vitality and coherence. Given enough space and time for successful healing, even the trauma of genocide might be retroactively integrated into that transformational process. How concrete human subjects accomplish this is as much a part of the scope of genocide studies as the process of destruction that makes it necessary.

Notes and References

1 Raphael Lemkin, Axis Rule in Occupied Europe: Laws of Occupation, Analysis of Government, Proposals for Redress (Washington, DC: Carnegie Endowment for International Peace Division of International Law, 1944), p 79. 2 Lemkin, Axis Rule, pp 79, 91. 3 Many readers, equating racism with assertions of the superiority of one race and the inferiority of others or with hostility towards particular racial groups, may nd this a surprising statement. However, the Oxford English Dictionary denes racism in this way: a. The theory that distinctive human characteristics and abilities are determined by race. b. RACIALISM. Racialism, in turn, it denes as: Belief in the superiority of a particular race leading to prejudice and antagonism towards people of other races, esp. those in close proximity who may be felt as a threat to ones cultural and racial integrity or economic well-being. Oxford English Dictionary, http://dictionary.oed.com (accessed July 8, 2007). Therefore, a concept or idea can be racist in the rst and more rigorous sense of the term even if it is anti-racialist. 4 A. Dirk Moses, Conceptual blockages and denitional dilemmas in the racial century: genocides of indigenous peoples and the Holocaust, Patterns of Prejudice, Vol 36, No 4, 2002, p 19. 5 Jean-Paul Sartre, On Genocide (Boston: Beacon Press, 1968).

544

WHAT DO GENOCIDES KILL?

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

6 Tony Barta, Relations of genocide: land and lives in the colonization of Australia, in Isidor Wallimann and Michael N. Dobkowski, eds, Genocide and the Modern Age: Etiology and Case Studies of Mass Death (New York: Greenwood, 1987), pp 237252. 7 W. B. Gallie, Essentially contested concepts, Proceedings of the Aristotelian Society, Vol 56, 1956. 8 Ibid, p 189. 9 Ibid, pp 196 197. 10 Ibid, p 198. 11 Michel Foucault, Two lectures, in Colin Gordon, ed., Power/Knowledge: Selected Interviews and Other Writings 19721977 (New York: Pantheon Books, 1980), pp 78108; Michel Foucault, Nietzsche, genealogy, history, in Paul Rabinow, ed., The Foucault Reader (New York: Pantheon Books, 1984), pp 76102. 12 Max Weber, Economy and Society (Berkeley: University of California Press, 1978), p 13. He goes on to say that: These concepts of collective entities which are found both in common sense and in juristic and other technical forms of thought, have a meaning in the minds of individual persons, partly as of something actually existing, partly as something with normative authority. [. . .] Actors thus in part orient their action to them, and in this role such ideas have a powerful, often a decisive, causal inuence on the course of action of real individuals. [. . .] Though extremely pedantic and cumbersome, it would be possible, if purposes of sociological terminology alone were concerned, to eliminate such terms entirely, and substitute newly-coined words (p 14). See pp 431 more generally for Webers methodological position. 13 Emile Durkheim, The rules of sociological method, in Steven Lukes, ed., Durkheim: The Rules of Sociological Method and Selected Texts on Sociology and its Method (New York: The Free Press, 1982), p 60. Emphasis in original. 14 Durkheim, The rules of sociological method, p 70. 15 Emile Durkheim, Suicide (New York: The Free Press, 1979), pp 208254. 16 Martin Shaw, What is Genocide? (Cambridge: Polity, 2007), pp 8991. 17 Weber, Economy and Society, p 22. 18 Ibid, p 26. 19 Ibid, p 96. Italics removed. 20 Weber, Economy and Society, p 154. Italics removed. 21 Kyriakos M. Kontopoulos, The Logics of Social Structure (Cambridge: Cambridge University Press, 1993), p 94. 22 For example: Genocide in the generic sense is the mass killing of substantial numbers of human beings, when not in the course of military action against the military forces of an avowed enemy, under conditions of the essential defencelessness and helplessness of the victims, from Israel W. Charny, Toward a generic denition of genocide, in George J. Andreopoulos, ed., Genocide: Conceptual and Historical Dimensions (Philadelphia: University of Pennsylvania Press, 1994), p 75. See also: Frank Robert Chalk and Kurt Jonassohn, The History and Sociology of Genocide: Analyses and Case Studies (New Haven: Yale University Press, 1990), p 10; Helen Fein, Scenarios of genocide: models of genocide and critical responses, in Israel W. Charny, ed., Toward the Understanding and Prevention of Genocide: Proceedings of the International Conference on the Holocaust and Genocide (Boulder: Westview Press, 1984), p 4; Barbara Harff and Ted R. Gurr, Toward empirical theory of genocides and politicides: identication and measurement of cases since 1945, International Studies Quarterly, Vol 37, No 3, 1988, p 360; Irving Louis Horowitz, Taking Lives: Genocide and State Power (New Brunswick, NJ: Transaction, 2002), p 23; Steven T. Katz, The Holocaust in Historical Context (New York: Oxford University Press, 1994), p 131; R. J. Rummel, Death by Government (New Brunswick, NJ: Transaction, 1994), p 31. 23 For example: In Australia very few people are conscious of having any relationship at all with Aborigines. My thesis is that all white people in Australia do have such a relationship; that in the key relation, the appropriation of the land, it is fundamental to the history of the society in which they live; and that implicitly rather than explicitly, in ways which were inevitable rather than intentional, it is a relationship of genocide. [. . .] My conception of a genocidal society as distinct from a genocidal state is one in which the bureaucratic apparatus might ofcially be directed to protect innocent people but in which a whole race is nevertheless subject to remorseless pressures of destruction inherent in the very nature of the society. Tony Barta, Relations of genocide: land and lives in the colonization of Australia, in Isidor Wallimann and Michael N. Dobkowski, eds, Genocide and the Modern Age: Etiology and Case Studies of Mass Death (New York: Greenwood, 1987), pp 239240. 24 Perhaps the strongest statement against allowing culture to matter in the denition of genocide is made by Stephen Holmes (Looking away, London Review of Books, Vol 24, No 22, November 14, 2002), who described Lemkins conviction that killing a hundred thousand people of a single ethnicity was very different from killing a hundred thousand people of mixed ethnicities as strange, archaic-sounding, medieval, and disturbing. Holmes reacts against Lemkins racist essentialisms about culture, but does not consider that a non-essentialist, non-racist account of culture is possible.

545

CHRISTOPHER POWELL

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

25 Raphael Lemkin, Genocide as a crime under international law, American Journal of International Law, Vol 41, 1947, p 146. 26 Lemkin, Axis Rule, p xiii; Samantha Power, A Problem from Hell: America and the Age of Genocide (New York: HarperCollins, 2002), pp 21 23. 27 Lemkin, Genocide as a crime under international law, p 146. 28 Hereafter abbreviated as UNCPPCG. 29 Leo Kuper, Genocide: Its Political Use in the Twentieth Century (New Haven: Yale University Press, 1981), pp 2330. 30 Ward Churchill, A Little Matter of Genocide: Holocaust and Denial in the Americas, 1492 to the Present (San Francisco: City Lights, 1998), p 365. 31 Lemkin, Axis Rule, p 79. 32 Ibid. 33 Ibid, p 80. 34 Ibid. 35 Lemkin, Genocide as a crime under international law, p 147. 36 Michael A. McDonnell and A. Dirk Moses, Raphael Lemkin as historian of genocide in the Americas, Journal of Genocide Research, Vol 7, No 4, 2005; Dominik J. Schaller, Raphael Lemkins view of European colonial rule in Africa: between condemnation and admiration, Journal of Genocide Research, Vol 7, No 4, 2005. 37 Lemkin, Axis Rule, p 91. 38 Helen Fein, Scenarios of genocide: models of genocide and critical responses, in Israel W. Charny, ed., Toward the Understanding and Prevention of Genocide: Proceedings of the International Conference on the Holocaust and Genocide (Boulder: Westview Press, 1984), pp 11 12. 39 Lemkin, Axis Rule, p 82. 40 Ibid, p 79. 41 Ibid, p 82. 42 A. Dirk Moses, The Holocaust and genocide, in Dan Stone, ed., The Historiography of the Holocaust (Hampshire: Palgrave, 2004), p 551n21. 43 Tanya Elder, What you see before your eyes: documenting Raphael Lemkins life by exploring his archival papers, 1900 1959, Journal of Genocide Research, Vol 7, No 4, 2005, pp 471, 489. 44 McDonnell and Moses, Raphael Lemkin, p 502. 45 Ibid, pp 504 505. 46 Durkheim, The rules of sociological method, p 60. 47 Ibid, p 134. 48 Ibid, pp 135 142. 49 Pierre Bourdieu and Loc J. D. Wacquant, An Invitation to Reexive Sociology (Chicago: University of Chicago Press, 1992), p 15; Mustafa Emirbayer, Manifesto for a relational sociology, American Journal of Sociology, Vol 103, No 2, 1997. 50 Norbert Elias, What is Sociology? (New York: Columbia University Press, 1978), pp 13 18, 118128; Norbert Elias, The Civilizing Process: The History of Manners and State Formation and Civilization (Oxford: Blackwell, 2000), pp 473474, 455456, 470471, 481482. 51 Elias, What is Sociology?, p 15. 52 Elias, The Civilizing Process, p 481; Norbert Elias, The Society of Individuals (New York: Continuum, 2001). 53 Elias, What is Sociology?, pp 113 116; Norbert Elias, The retreat of sociologists into the present, Theory, Culture & Society, Vol 4, 1987; Elias, The Civilizing Process, pp 452 455, 551n4. 54 Elias, The Civilizing Process, p 483. 55 Ibid, pp 466 467. 56 Elias, What is Sociology?, pp 15, 129 133. 57 Helen Fein, in Accounting for Genocide: National Responses and Jewish Victimization during the Holocaust (New York: The Free Press, 1979), p 169. 58 Stuart Hall, Heroes or villains? And stereotyping as a signifying practice, in Michelle Webber and Kate Bezanson, eds, Rethinking Society in the 21st Century: Critical Readings in Sociology (Toronto: Canadian Scholars Press Inc., 2004), p 223. 59 Zygmunt Bauman, Modernity and the Holocaust (Ithaca, NY: Cornell University Press, 1991), p 1982. 60 American Association of Physical Anthropologists, AAPA statement on biological aspects of race, American Journal of Physical Anthropology, Vol 101, No 4, 1996; Ashley Montagu, Statement on Race: An Annotated Elaboration and Exposition of the Four Statements on Race Issued by the United Nations Educational, Scientic, and Cultural Organization (Oxford: Oxford University Press, 1972), pp 157 153.

546

WHAT DO GENOCIDES KILL?

Downloaded By: [Powell, Christopher] At: 00:18 21 November 2007

61 62 63 64 65 66 67 68 69 70 71 72 73

74 75 76 77 78 79

Robert Miles, Racism after Race Relations (London: Routledge, 1993). See note 3. Bauman, Modernity and the Holocaust, p 69. Ibid, pp 6273. Morton H. Fried, The Notion of Tribe (Menlo Park, CA: Cummings, 1975); Peter T. Suzuki, Tribe: chimeric or polymorphic?, Studies of Tribes and Tribals, Vol 2, No 2, 2004. Fried, The Notion of Tribe, p 3. Ibid, p 5. Ibid, p 7. Ibid, p 7. Ibid, p 8. Weber, Economy and Society, pp 387393. Adam Jones, Genocide: A Comprehensive Introduction (London: Routledge, 2006), p 14. Chalk and Jonassohn, The History and Sociology of Genocide, p 23; Israel W. Charny, Toward a generic denition of genocide, in George J. Andreopoulos, ed., Genocide: Conceptual and Historical Dimensions (Philadelphia: University of Pennsylvania Press, 1994), p 75; Churchill, A Little Matter of Genocide, pp 432435; Fein, Scenarios of genocide, p 4; Horowitz, Taking Lives, p 23; Kurt Jonassohn and Karin Solveig Bjornson, Genocide and Gross Human Rights Violations: In Comparative Perspective (New Brunswick, NJ: Transaction, 1998), p 10; Katz, The Holocaust in Historical Context, p 131; Lyman H. Legters, The Soviet Gulag: is it genocidal?, in Israel W. Charny, ed., Toward the Understanding and Prevention of Genocide: Proceedings of the International Conference on the Holocaust and Genocide (London: Westview Press, 1984), p 65; Jack N. Porter, Introduction, in Jack N. Porter, ed., Genocide and Human Rights: A Global Anthology (Washington, DC: University Press of America, 1982), p 12; Uriel Tal, On the study of the Holocaust and genocide, Yad Vashem Studies, Vol 13, 1979; John L. Thompson and Gail A. Quets, Genocide and social conict: a partial theory and comparison, in Louis Kriesberg, ed., Research in Social Movements, Conicts and change (Greenwood, CA: JAI Press, 1990), p 248. Benedict Anderson, Imagined Communities (London: Verso, 1991), pp 3746; Ernest Gellner, Nations and Nationalism (Oxford: Basil Blackwell, 1983), pp 4050; Charles Tilly, Coercion, Capital, and European States, AD 9901990 (Oxford: Basil Blackwell, 1990), pp 23, 58, 114117. Robert Melson, Revolution and Genocide: On the Origins of the Armenian Genocide and the Holocaust (Chicago: University of Chicago Press, 1992), pp 54 56, 167 169. Norbert Elias, The Germans (New York: Columbia University Press, 1996), pp 185, 197; Reginald H. Phelps, Before Hitler came: Thule Society and Germanen Orden, The Journal of Modern History, Vol 35, No 3, 1963. Lewis Coser, The Functions of Social Conict (New York: The Free Press, 1956), pp 3338. David Stannard, American Holocaust: The Conquest of the New World (Oxford: Oxford University Press, 1992), p 268. Mary Louise Adams, The Trouble with Normal: Postwar Youth and the Making of Heterosexuality (Toronto: University of Toronto Press, 1999); Michel Foucault, The History of Sexuality, Vol I: An Introduction (New York: Random House, 1990).

547