Democracy's Past and Future: Twenty-Five Years, Fifteen Findings

When Guillermo O'Donnell and I were writing Transitions from Authoritarian Rule: Tentative Conclusion about Uncertain Democracies a quarter of a century ago, we had few cases and almost no literature upon which to draw.¹ Mostly we ransacked the monographs of colleagues who were taking part in the same Woodrow Wilson Center project as we were. We also reached back to the classics of political thought. I personally drew much inspiration from the work of Niccolò Machiavelli who, I discovered, had grappled some time ago with regime change in the opposite direction—that is, from "republican" to "princely" rule.

Neither of us imagined that the fledgling efforts we were then observing in Southern Europe and Latin America would soon be followed by more than fifty other regime transformations all around the world. These "divine surprises," especially the ones in Central and Eastern Europe and the former Soviet Union, brought not only much scientific opportunity and personal normative satisfaction, but also a major intellectual risk. To what extent could the assumptions, concepts, hypotheses, and "tentative conclusions" that we had derived from the early cases be stretched to fit a much larger and highly varied set of countries? The stretching that we were considering seemed even more problematic in light of how opposed our ideas were to most prevailing theories about "really existing democracies."²

We insisted, for example, on a clear distinction between liberalization and democratization. We refused to accept the notion that democracy requires some fixed set of economic or cultural prerequisites. We emphasized the key role of elite interaction and strategic choice during the transition and in most cases ascribed limited importance to mass mobilization from below. We pointed to the demobilizing effect of the electoral process and said that while civil society might have a significant role, it would be a short-lived one. We noted how most transitions began from within the previous autocratic regime, whose collapse or self-transformation by no means guaranteed the eventual success of democracy. Finally, and perhaps most subversively, we argued that it was possible (if not always probable) that one could bring about democracy without having any democrats on hand. In other words, the favorable cultural and normative traits or "civic culture" that comparative survey research had detected and found essential to all stable democracies was better conceived as a product of democracy rather than its producer.

As a comparativist, I welcomed the challenge of "stretching" our original work and applying it to such different cases. I found it gratifying to observe how often, how far away, and even how controversially these "cross-regional" comparisons were attempted, and I am convinced that they contributed to a fuller understanding of democratization. What I found much less gratifying was the tendency of critics and other readers to apply our book to topics that were manifestly not within its purview. It had been no accident that Guillermo and I had given the book a title stressing transitions away from authoritarianism rather than to democracy, yet many treated our tome as if it purported to contain a magic formula for success or even lessons in how to consolidate democracy.

Not only did we refuse to presume a telos that would lead to such a felicitous result, we were obsessed with the likelihood of regression to autocracy. Admittedly, we were concerned all along with the implications that different transitional situations might have for democracy's ability to emerge and persist, but we wrote nothing about what such an outcome might look like. Guillermo and I have since written a good deal on this topic, but nothing in our original joint effort allows one to assume that voluntaristic, structurally underdetermined action would continue to dominate the politics of new democracies once they passed through the highly uncertain transition period, or that strategic machinations among elites would continue to count for more than mass mobilization and popular participation.

Much has happened over the last two decades, including a burgeoning of democratization studies, from which I have learned much. The editors of the Journal of Democracy have asked me to share...