Kinds of Killing

How bad is genocide, really?

Like ‘fascism’ – with which it is closely connected in the popular imagination – ‘genocide’ is a word carrying such exorbitant emotional charge that it tends to blow the fuses of any attempt at dispassionate analysis. We can thank the political black magic of Adolf Hitler and his Nazi accomplices for that.

Prior to the Third Reich and its systematic, industrialized attempts to eradicate entire ethno-racial populations (Jews, Roma, and perhaps Slavs) along with other numerous other groups (mental and physical ‘defectives’ or ‘useless eaters’, homosexuals, communists, Jehova’s Witnesses …) international law restricted its attention to the actions and grievances of states and individuals, with the latter subdivided into combatants and noncombatants. The National Socialist trauma changed that fundamentally.

On December 9, 1948, the United Nations adopted the Convention on the Prevention and Punishment of the Crime of Genocide (as Resolution 260), defining a new category of internationally recognized crimes as “acts committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group.”

Since 1948, defending genocide has been the surest way to ruin a dinner party. That doesn’t mean, however, that the topic deserves to be immunized from controversy. There is one question in particular that merits intense and prolonged scrutiny: Is genocide really worse than killing a lot of people?

Posed slightly more technically: Is there a crime of genocide that stands above and beyond mass murder (of equivalent scale)? Or (a rough equivalent): Can groups be the specific victims of crime? This is to ask whether groups exist – and have value — as anything more than a nominal or strictly formal set, whose reality is exhausted by its constituent individual members. The existence of genocide as a legal category presumes a (positive) answer to this question, and in doing so it closes down a problem of great and very general importance.

The classical liberal presumption is quite different, as summarized (a little bluntly) by the provocative remark made by British Prime Minister Margaret Thatcher in 1987 “… there is no such thing as society. There are individual men and women, and there are families.” Harshly extrapolating from this position, a certain irony might be found in the fact that a horrified response to National Socialist crimes has taken the form of a legal codification of racial collectivism. At the very least, it is puzzling that suspicions directed at legal references to ‘group rights’ and ‘hate crimes’ among those of a libertarian bent has not been extended to the category of genocide.

In the opposite camp, the most fully articulated defense of collectives as real entities is found, as might be expected, in the foundation of sociology as an academic discipline, and more particularly in Émile Durkheim’s argument for ‘social facts’. Larry May looks back further, to Thomas Hobbes’ Leviathan, or social being, in which human individuals are absorbed as organic parts.

Whilst the distinction of ‘society’ and ‘individual’ has colloquial (and political) meaning, those inclined to the analysis of complex systems are more likely to ask which groups or societies are real individuals, exhibiting functional or behavioral integrity, as self-reproducing wholes. In pursuing this line of investigation, it is far more relevant to discriminate between types of groups than between groups and individuals, or even wholes and parts. It is especially helpful to distinguish feature groups from unit groups.

A feature group is determined by logical classification. This might be expressed as a self-identification or sense of ‘belonging’, an external political or academic categorization, or some combination of these, but the essentials remain the same in each case. Certain features of the individual are isolated and emphasized (such as genitalia, sexual orientation, skin-color, income, or religious belief), and then employed as the leading clue in a process of formal grouping, which conforms theoretically to the mathematics of sets.

A unit group, in contrast, is defined as an assemblage, or functional whole. Its members belong to the group insofar as they work together, even if they are entirely devoid of common identity features. Membership is decided by role, rather than traits, since one becomes part of such a group through functional involvement, rather than classification of characteristics. Social instances of such groups include primitive tribes (determined by functional unities rather than the categories of modern ‘identity politics’), cities, states, and companies. The most obvious instance in socialist theory is the ‘soviet’ or ‘danwei’ work unit (whilst social classes are feature groups).

To take a non-anthropomorphic example, consider a skin cell. Its feature group is that of skin cells in general, as distinguished from nerve cells, liver cells, muscle cells, or others. Any two skin cells share the same feature group, even if they belong to different organisms, or even species, exist on different continents, and never functionally interact. The natural unit group of the same skin cell, in contrast, would be the organism it belongs to. It shares this unit group with all the other cells involved in the reproduction of that organism through time, including those (such as intestinal bacteria) of quite separate genetic lineages. Considered as a unit group member, a skin cell has greater integral connection with the non-biological tools and other ‘environmental’ elements involved in the life of the organism than it does with other skin cells – even perfect clones – with which it is not functionally entangled.

Clearly, both feature groups and unit groups are ‘fuzzy sets’, and the distinction itself – whilst theoretically precise – is empirically hazy. An urban American street gang, for instance, will in most cases be vague in its features and unity, perhaps ‘ethnic’ to some degree of definition, with a determinable age-range, and with ambiguous functional connections to groupings on a larger scale, or to peripheral members whose status of ‘belonging’ is not strictly decidable. Tattoos and other membership markings are likely to involve both identity and integrity aspects – traits and roles. Rituals of belonging (ordeals, oaths, rites of passage) are designed to disambiguate membership.

Despite such haziness, the distinction between these two types of groups strikes directly at the core problematic of genocide (as a legal category). When a unit group is destroyed, a real individual is ‘killed’ above and beyond whatever human losses are incurred. The destruction of a feature group, in contrast, whatever the cultural loss, is not any kind of killing beyond the mass murder of human individuals. If this is worse than murder, we should know why.

This conclusion seems relevant when weighing, for instance, the 1937 Massacre of Nanjing on the scale of historical atrocity. It suggests, at least, that an act of violence directed against a city – or integrated population unit — is no less worthy of specific legal attention than a quantitatively equivalent offense against an ethnicity, or determined population type. It seems to be no more than an accident of history that, in order to appropriate the category of genocide, massive crimes of the former variety need to be recoded as if they more properly belonged to the latter.

Complex systems ontology aside, these matters resolve ultimately into obscure social values. Orthodox conceptions of ‘genocide’ assume that ethnic identity simply and unquestionably means more than active citizenship, or participation in the life of a city. Perhaps this assumption is even arguable. But has it been argued?