The plurals in my title carry two implications, neither of which I take to be controversial, if they ever were: first, that there is more than one kind of “original” state; second, that there is more than one way in which states originate. There is, admittedly, continuing controversy over the definition of “state.” But for the purposes of this article, I assume that there are four necessary and jointly sufficient conditions for the emergence of a state from nonstate or stateless forms of social organization: specialization of governmental roles; centralization of enforceable authority; permanence, or at least more than ephemeral stability, of structure; and emancipation from real or fictive kinship as the basis of relations between the occupants of governmental roles and those whom they govern. All four admit of differences of degree. But they furnish an adequate framework within which the different processes by which different states have come into being can be analyzed and compared.

The process whereby political independence came to the Spanish and Portuguese colonies in America has been the object of a rich body of scholarly interpretation. Although most of this literature concentrates on the causes of independence, several authors, particularly those concerned with the Brazilian case, have tried to explain also the reasons for the differences in the political evolution of the two colonial empires. Without denying the value of some of these explanations, this essay argues that they are not entirely satisfactory and that an alternative, or at least supplementary, explanation can be found in the nature of the political elites that emerged in the two colonies as a consequence of differing colonial policies.

The domestic and global forces that shaped Mexico's revolutionary struggle of 1910–17 continued to shape its state-building conflicts during the 1920s. Recent revisionist historiography underscores this point by emphasizing the bitter irony of the Mexican revolution. This literature contends that, despite the revolution's populist and nationalist aura, by the late 1920s it had done little more than subordinate the masses to an even more centralized state and deepen Mexico's dependence on United States governmental and capital interests. The revisionists argue, then, that the revolution had basically reinforced the country's old regime legacies and created a more institutionalized version of the Porfiriato's authoritarian state.

In a letter addressed to the medievalist Ferdinand Lot and dated June 1941, Charles Seignobos, hereditary enemy of the Annales, declared, “I have the impression that, for approximately the last quarter-century, the effort to think about historical method, which was vigorous in the 1880s and especially so in the 1890s, has reached a stalemate,” and noted that, as a sign of the times, “the Revue de Synthese Historique … has changed its name.” Seignobos, then only a year before his death, was writing a book on “the principles of the historical method.” His letter alluded to American and German output (“a mediocre American, Barnes, published a fat book in 1925 in which he summarized a large number of works….”), but made no mention of Lucien Febvre, Marc Bloch, or of the Annales, then in its twelfth year. To choose to ignore the Annales while discoursing on historical method is of course unjust and absurd. But aside from this omission, Charles Seignobos's remarks are not without pertinence. It is true that France at the turn of the last century and particularly during the first decade of the twentieth century, had been the center of a passionate and fascinating debate on the nature of historical knowledge, on the legitimacy of its pretensions to be a science, and so forth, and that by the 1940s this debate had ceased.

Dominica is the northernmost of the Windward Islands in the Lesser Antilles. Its capital is Roseau. The official language is English, but for most of the island's predominantly Negro population of about 80,000 the language of everyday life is a French-based Creole called patois. Dominica was granted full independence from Britain on 3 November 1978. From November 1977 to December 1978, I lived there in a fishing village called Casse. Casse has about 800 inhabitants.

In the period 1894–96, when the Ottoman Empire was ruled by Sultan Abdul Hamid II, tens of thousands of Armenians were massacred. Nineteen years later, when the empire was weakened by disintegration and war, some one million persons—half of the Armenian population—were killed with the active participation of the Committee of Union and Progress, the ruling party of the day. The massacres and the genocide—for that is what the second act of violence has come to be called—must rank among the most terrible catastrophes of our era. Two questions come to mind: why did these things happen, and what is there to be learned from the Armenian case? We shall attempt to answer these two questions, keeping in mind that for historical events of such complexity and magnitude there are no final answers, merely more or less credible, more or less convincing, formulations. And given the obvious limitations, this article will focus on the massacres alone.

Historical research has supplied extensive information about the stark facts of the Holocaust. It includes efforts both to document the full extent of the horror and to maintain a degree of objectivity and avoid undue sentimentality (Bauer 1978). The historical work includes, and goes beyond, chronicling the details of the murderous events. It points up unresolved-and possibly unresolvable-questions, such as the nature of the involvement and responsibility of European Christians. That issue involves, at one end, the accusation that Pope Pius XII was, at the very least, inactive in the face of a supreme moral challenge (Falconi 1970). At another end, it involves acknowledgement of extensive efforts by Christians to protect Jews, at considerable risk to themselves (Friedman 1980; Flender 1963).