I am trying to work out some thoughts on Marxism and ethnicity, with respect to Donald Horowitz’s reading of ethnicity. For some part, and this is a massive book, I think Horowitz proceeds properly. He sees ethnicity as something that is neither set in stone nor putty, it is malleable, within limits — to paraphrase Horowtiz (66). Ethnicity is a myth of common ancestry, or common descent (52). It isn’t “natural,” in the sense that it doesn’t proceed logically that, from differences of race, language, kinship groups, etc., there will be differentiated groups. Rather, differentiation of groups picks up on — or even constructs, that is, intensifies or gives new meaning to — language, race, etc. As he puts it, it’s not the attribute that makes the group or group difference, it’s group difference that makes the attribute (50). Culture provides the content, it is not the prerequisite to ethnic differentiation (69). If ethnicity, in and of itself, isn’t “modern” — i.e., identitarian differences existed long before “modernization” (incorporation of the Third World into the world capitalist system) — it certainly has been modulated, moulded, shaped by this incorporation of societies into modernity. Particularly, he notes how colonization gave rise to the significance of identity by incorporating diffuse groups, that is, previously politically diffuse groups, into a singular polity (76). All of a sudden, you have a different dynamic to deal with. He does mention somewhere about how colonialism reinforced ethnic identities, but he doesn’t focus on this too much (as, for instance, Benedict Anderson does in Chapter 10 of Imagined Communities, or how Leroy Vail approaches the question in The Creation of Tribalism in Southern Africa).

Horowitz distinguishes between “ranked” and “unranked” ethnic groups. Ranked is where, for instance, ethnicity maps on, more or less, to class. So slavery, or colonialism, is a clear example of how ethnicity (race) is ranked. The question of a caste system, to the extent that a caste can be seen as an ethnicity, is also a ranked system. (However, here we have to be considerate of the fact that castes exist within ethnic identities. Horowitz doesn’t seem to theorize this.) However, an unranked system is more complicated, in the sense that within an ethnic group you will have different classes (or castes, is it is). The question then is why is ethnic conflict — or ethnicity, at any rate — so salient in these unranked systems?

I think that much of the Marxist study I have done, including Marxist studies of race, have focused on the question of ranked systems of ethnicity. In highlighting the origins of race, the ways that race separates itself from class, the ways in which race articulates with class, etc. In fact, my previous post on Benedict Anderson’s book is an example of this — although I focus more on class than I do on race, the actual fact is that I am focusing on cases where class is determined by race.

Unranked systems, however, are more complicated. How do we talk about them without going back to questions of “false consciousness”? Elites, fractions among the capitalist class, that is, using ethnicity as a tool against other fractions of the capitalist classes — or, if not quite capitalist, then state bourgeoisies, or elites of whatever kind. The question ultimately, as Horowtiz notes, is why on earth the subalterns go along with this? They’re not stupid.

Horowitz takes a flattened and kind of ridiculous view of Marx and Marxism, i.e., that Marx assumes that classes are “closed” social groups, there is no social mobility, and they will fight each other. Rather, Horowitz says, its’ ethnic groups that are the closed social groups. Ethnicity forms the axis here, not class. But this view that class is somehow closed is something I’ve never gotten out of Marxism. It is true that many Marxists will point out that the news of social mobility in capitalist countries is somewhat exaggerated (and it is). But aside from that, the truth is that class in Marxism need not be (as it is in vulgar Marxist views) a static thing, but rather something dynamic. Classes, like all other social phenomena, continuously come into being. That is, society is not just produced, but reproduced. Classes change according to changing historical conditions. And historical conditions change as a result of different articulations of class fractions, and so on.

I think the essential problem of many Marxist formulations, or at least, of critiques of Marxist formulations, is their inability to distinguish between class-in-itself, and class-for-itself. The conceptual difference is important. Something may exist “objectively” as a class, in the sense that a group of people who are alienated from the means of production may constitute, objectively, a proletariat. However, that doesn’t mean that they will, of necessity, all be alienated in the same way, in the same cultural context, in the same proportions, etc., and that they will, therefore and inevitably, develop a uniform consciousness. That is to say, uneven alienation, or differentiated positions within classes can lead to differentiated subjective articulations of identity, and differentiated articulations of agency. This isn’t merely a question of false consciousness, that people are too stupid to see that they are all proletarians and should unite to overthrow capitalism, or what have you. It is more complicated than that, it’s a basic part of how groups come into being, and how group identities come into being.

What does this have to do with ethnicity?

Generally, from what I know, practical Marxist movements haven’t really spent too much time trying to theorize ethnicity. It’s there, you can see it, so you work with it. The principle is national self-determination. The ethnic group or nation is taken for granted. Marxist theorists have also not spent much time theorizing ethnicity — as opposed to race.

What would a historical materialist theory of ethnicity look like?

The problem with ethnicity is that it’s really a mixed bag of things. If we exclude race (in its western articulations) and caste from ethnicity, we are still dealing with a really vast and mixed bag of group identities, with varying levels of salience, all of which depend on the context. That is, a historical materialist theory of ethnicity has to specify its historic character in the first instance. In my studying Mozambique, for the most part ethnicity has not played a salient role in the macro-politics of the country. It has played a role, that’s undeniable — for instance the revolutionary forces of Frelimo were largely recruited from the Makonde tribe in the northwest, who were all up in arms against the Macua. The Portuguese pit the Macua against the Makonde, but they were exploiting already-existing faultlines there. The Macua saw the Makonde as a greater threat than the Portuguese. But generally, politics in Mozambique hasn’t taken on an ethnic character, as it has in other parts of sub-Saharan Africa.

So what? So plenty! I am not convinced by generalized theories of ethnicity for precisely this reason. Somalia, which is largely ethnically homogeneous, is riven with conflict — which probably has something to do with clan politics, but I have no idea what’s going on in Somalia. Mozambique, which is ethnically pluralist, is not split on ethnic lines, politically. There is not necessarily a causal relationship between ethnic plurality and conflict on ethnic lines, or ethnic homogeneity and lack of conflict. I may be talking about outliers on some graph, but these outliers say something — that we specify the historic character of ethnicity in a given social context.

But we still have to explain why it is that ethnicity, where it is “constructed,” say, as a result of colonial encounter, nevertheless maintains a salience that may appear to overshadow class compositions or other forms of identity. Horowitz tries to explain this in terms of group psychology, and to some extent this is plausible (Fanon, after all, dealt in some respects with group psychology). But it seems to me that, even if he explains how a certain psychology is produced, Horowitz still leaves out how that psychology itself is reproduced. How, and why, is it that ethnicity reproduces itself?  What are the material processes by which ethnicity reproduces itself? State structures, power-sharing agreements (or disagreements), the repeated use of ethnic categories to differentiate people and populations, the division of labour — the social organization of production and distribution — has to have something to do with this. When we talk about social organization of production and distribution, we are going beyond just class, but we are still talking Marxism. This is the fundamental task of Marxism. I’m drawing here on Amilcar Cabral, who spoke of the distinction between straightforward class analysis, and a more historical and materialist analysis of mode of production (i.e., as Hamza Alavi calls it, the social organization of production), in all its richness. This richness includes an analysis of culture and its myriad differentiations, such that one can speak of different cultures not only across ethnic groups but also across class groupings. We also clearly need to look at Louis Althusser in greater detail, and return to Gramsci. I also know that John S. Saul has written a piece on the dialectic of tribe and class, that might be useful.

Ultimately, ethnicity needs to be better conceptualized. Is it possible that a Marxist understanding, if undertaken properly, can help in this?