The Russian/Soviet experience raises complex general questions concerning orientalism, conceptual hegemony, and the politics of (post-)colonial knowledge. Russia was not an empire in Said's sense, and drew much of its orientalist categories from non-imperialist German sources; the Soviet Union was explicitly anti-imperialist, and was dedicated to the emancipation of subaltern classes and nationalities. Yet Soviet orientalism in part reproduced hegemonic categories of “bourgeois” knowledge, notably concerning language and national identity. This becomes especially clear in the case of Soviet studies of Kurdish, a language subaltern with respect to Persian, Arabic, and, increasingly, Turkish. In the 1920s and early 1930s, native scholars like Erebê Shemo, Qanatê Kurdo, and Heciyê Cindî pioneered the creation of both an alphabet and a literature in Kurdish and of scholarly linguistic studies. Their work was shaped (and encouraged) by Nikolaj Marr's rejection of the idea of genetic links between Indo-Persian languages, and of the reification of “national characters.” Marr's “japhetic” linguistics dovetailed with Stalin's nationality policies in the 1920s and 1930s; it is rightly rejected as unscientific, but it did have positive emancipatory effects. It criticized ethnocentric and racist assumptions in contemporary Indo-European linguistics, and emphasized the value of spoken subaltern vernaculars like Ossetian and Kurdish against hegemonic written languages like Sanskrit and Persian. It also had the paradoxical effect of both countering bourgeois nationalism and encouraging national consciousness. The article concludes with a discussion of how the Soviet experience may affect our view of the Gramscian concept of hegemony and of the linguistic turn in later postcolonial studies.