The resurgence of democracy in Latin America in the last decade or so came as a surprise to many who saw the continent, if not the whole of the Third World, as producing conditions which favoured only the exercise of tyranny. Latin American democracy will indeed remain surprising to those who think of democracy as a single, fixed ideal which nations at one time or another more or less attain. If we ask questions about what kind of democracy has developed, and in whose interests, about the constraints on democracy in the nation-state of globalizing capitalism, then it may be possible to see what it really amounts to, that periodic exercise of the vote. Perhaps the masses are permitted democracy only when the alternatives for the elite seem worse, or when prospects for change are remote.

In the early 1970s, many observers were predicting a poor future for Latin American democracy.footnote1 Guillermo O’Donnell argued that under contemporary Third World circumstances, capitalist development would have corrosive effects on the democratic gains of the recent past. He contended that the specific patterns of industrialization emerging in the more economically developed countries of South America had by the 1950s created a coalition of industrialists, technocrats and foreign bankers committed to capital-intensive producer goods, a course of action diametrically opposed to the interests of labour as well as other beneficiaries of the ‘populist’ policies that in previous decades had emerged around strategies of consumer-oriented industrialization through import substitution. The new industrial coalition, stymied by labour’s earlier gains, now sought to eject labour from the political arena by closing down democracy. O’Donnell’s initial study appeared in Spanish in 1972 and in English the following year.footnote2 It drew on the experiences of Argentina and Brazil in the 1960s. The military seizures of power in those two countries seemed to go beyond the traditional political roles of Latin American militaries; they seemed harbingers of a new form of anti-democratic rule. The immediately subsequent experiences of Uruguay and Chile appeared as striking confirmations of O’Donnell’s argument: the two South American states which seemed to have the most secure claims to some measure of democracy were occupied by their own armed forces.

And even more striking was the Argentine drama. The military regime that had provided the main empirical material for the book came to an end with the book’s appearance, hardly an endorsement of the stark thesis. But the anomaly vanished with the new coup of 1976 and the far more brutal regime that followed. It would have been hard to imagine a more dramatic reinforcement of the claim that a harsh authoritarian order was the normal political concomitant of economic advance in industrializing South America than this swift failure of civilian rule. It is probably unusual for a work in social science to seem so clearly confirmed by unanticipated events in several countries. O’Donnell’s work came to have enormous influence and legions of political scientists developed increasingly complex accounts of comparative economic structures and economic policies as they elaborated, refined, and corrected the central notions.footnote3 Alas for the thesis, the confirmatory Chilean, Uruguayan and Argentine events were at the tail-end of a global wave of anti-democracy that can be dated from the late 1950s.footnote4

O’Donnell had hardly been alone. Many who rejected his economistic explanation had their own theses that extrapolated from an anti-democratic present to an anti-democratic future. Some North American social scientists found a cultural strain towards authoritarianism in the Latin soul, a standpoint from which the earlier optimism of a universal democratic future now seemed a naive form of ethnocentric blindness.footnote5 Samuel Huntington had a different but equally pessimistic thesis. He saw a multi-continental failure of political institutions to contain social forces mobilized by ‘modernization’; this disorderly state of affairs, ‘praetorianism’ in his terminology, was far more likely to provide opportunities for authoritarian forces than democratic ones as Leninist parties—or an occasional local formation like Mexico’s Institutional Revolutionary Party (pri)—managed to make order out of chaos.footnote6 This thesis, too, predicted a dim future for Third World democracy. Huntington lectured to the cia in 1976, and compellingly explained the powerful forces causing ‘The Global Decline of Democracy’, not having noticed that democracy was already on the upswing.footnote7 Thus both those with left sympathies and antipathies had concurred on the swelling dominance of anti-democratic forces to the south of the United States.

Such mishaps of prediction occur easily when we take the way the tide flows at any moment for the long-run trend of history, rather in the manner of those on the Left who used to take downturns in the business cycle for the final crisis of capitalism. But we can also step back from the current trend to search for a deeper historical context. In addition to the distinctive national histories of particular political systems, there is a world history of democratization within which we can situate the present moment.