Unlike so many areas of Latin America with large indigenous populations (such as Guatemala and Peru), Ecuador did not experience widespread civil war in the 1970s and 1980s, characterized by guerrilla movements and severe military repression. Instead, those years saw the emergence in Ecuador of a strong indigenous movement within the bounds of civil society, which since the 1990s has had a growing role in articulating the demands of subordinate groups before the state. In fact, among all Latin American countries it is, strikingly, only in Ecuador that a national indigenous federation has been established; since 1986, the Confederation de Nacionalidades Indigenas del Ecuador (CONAIE) has become the hegemonic2 institution representing indigenous projects, proposals and problems before the state. The Ecuadorian Indian movement has received a great deal of scholarly (and political) attention recently, particularly since the broadly based indigenous uprising of 1990. Important analyses have been made of the organizational basis and forms of the Indian movement and the challenges it faces in Ecuador today. The argument presented in this article is rather different: I suggest that to understand current processes of organization and mobilization, it is essential to examine ongoing changes in class relations in Ecuador over the last century, situating Indians in relation to dominant groups.The Ecuadorian Indian movement appears to fit into the category of a new social movement, eschewing class as its organizational basis, and demonstrating instead the importance of other forms of identity, in this case, ethnicity. However, I argue that the gains of the Indian movement today can best be understood: through an historical analysis of a century of changing class relations and projects3; as the result of a series of responses to a modernist state project of inclusion imposed from above (with unexpected consequences); and in the context of changing international political-economic processes, reshaping the terrain on which local struggles would be carried out. To some extent, in Ecuador space has been generated for the partial incorporation of subaltern projects, in a way that has not occurred in many other Latin American countries. The generation of these spaces was often inadvertent, as elites pursued policies in their own interests, which then opened up unexpected possibilities for indigenous Ecuadorians.This analysis draws on Marxist resources for an anthropological political economy in two main ways. First, I argue that whfle class has not proven to be the focal point for identities that inform political movements at the beginning of the 21st century, this does not mean that we should reject class analysis as a tool for understanding social processes (for a similar argument applied to Mexico, see Nugent, 1997; and more generally, Thompson, 1978:148149). Indeed as Michael Kearney (1996) has pointed out in regard to indigenous movements elsewhere in Latin America, the defence of indigenous identity (an ethnic issue) is closely linked to the protection of access to land and other essential subsistence resources (a class issue). Thus I argue that an analysis of the changing relations among classes over time (including political struggles to impose projects in a particular class's economic interests) is essential for understanding current mobilizations, even if those mobilizations do not themselves emphasize class. second, this article proposes that we cannot understand subaltern projects in isolation. The concept of the subaltern comes from Gramsci's work (see especially 1971:44120), and he situates subaltern or subordinate groups always in relation to dominant groups, within a broad social field of analysis. Often when Latin Americanists have drawn on the school of subaltern studies, they have lost sight of this broader social field as they focus only on subordinate groups (cf. Mallon 1994). I would argue in contrast that neither dominant nor subordinate groups can be fully understood except in relation to each other. …