In 1979 Theda Skocpol, a young Harvard sociologist of notable strength of will and intellectual energy, published what is still the most illuminating single social scientific study of the character and causation of modern revolutions. It has had many fierce critics, then and since. But it was, and very much remains, a work of such force and intuitive good sense that anyone seriously interested in understanding why these massive and disturbing episodes should occur and take the forms which they do would be inane to ignore it.

Skocpol is a combative (though never a gratuitously offensive) scholar, and much has changed in the wider world since she first wrote. Since 1979, accordingly, she has had many occasions to reflect further on the phenomena which she studied and on the, often strikingly different, revolutionary and counterrevolutionary episodes which have since ensued, and ample opportunity to defend herself against a wide variety of critics. This book places her initial work in its original context, reprints some of the working papers in which she first defined her viewpoint, and brings her assessment both of her initial explanatory success and of her subsequent need to extend or modify that viewpoint firmly up to date.

Skocpol is not an elegant stylist and the result does not always make for easy reading. But it certainly provides much material for thought and more than vindicates the continuing good sense of her choice of starting point against all comers. Where it is less successful, perhaps, is in clarifying just what that viewpoint fundamentally implies, and why it has proved so consistently illuminating, or so plainly superior in critical purchase to those of her most confident and ambitious competitors.

The key to her approach is the recognition that revolutions are not merely defined as, but also most decisively consist in, more or less massive crises in the life histories of states, resolved insofar as they ever are resolved, by more or less drastic reconstructions of the state in question. This is as straightforwardly true of the great social revolutions of France, Russia and China which lay at the centre of her first book as it has been of the thirdworld revolutions in Iran or Nicaragua, or the collapse across Eastern Europe, and eventually even in its own heartland, of Soviet imperial domination, over the decade and a half since its publication. It is as true of revolutions characterised by protracted mass struggle in city or countryside as it is of cases (like the revolutions of 1848) in which incumbent governments collapsed virtually without a fight in the teeth of essentially uncoordinated popular animosity and were reconstructed with little effort (and often rather little modification) not long afterwards.

ÁñÁ«ÊÓƵ

As an admiring (if independent) pupil of Barrington Moore, Skocpol's approach to revolution always drew more from Weber than it did from Marx. But she was (and remains) especially averse to seeing postrevolutionary state-building as the reified representative agency of the people (or of some favoured social class), spontaneously emerged and acting with unfailing insight and trustworthy commitment to further the interests of those whom it represents.

Any realist vision of state crisis and its potential resolution is bound to focus on interstate relations and geopolitical struggle as much as on domestic class relations and their political expression. No student of Weber or Barrington Moore could have regarded an essentially social-psychological analysis of popular sentiment towards incumbent regimes as a promising instrument for distinguishing states acutely vulnerable to crisis from their more robust contemporaries, or for interpreting the dynamics of state collapse and reconstruction.

ÁñÁ«ÊÓƵ

The central political intuitions behind Skocpol's work are so simple, and so palpably well-founded, that it is difficult to see how critics then or subsequently could have both identified them clearly and yet retained the nerve to dispute them frankly and with the slightest conviction. In the conclusion to her new book, she struggles to bring into focus the grounds for objection of three types of critic in particular.

The first is a fideist Marxist (personified by Michael Burawoy), who rebukes her for failing to match Trotsky's insight into the dynamics of revolution, while conspicuously failing to mention the many notable lapses in political judgement on the subject of which his hero was notoriously guilty. (Never explain. Never apologise.) The second critic is a rational choice theorist (personified by Michael Hechter), who rebukes her for not casting her work in the canonical format. Here Skocpol very reasonably responds by inquiring just what rational choice analysis has yet added to our understanding of any revolution whatsoever, let alone of modern revolutions as a whole. (Chacun a son gout.) The third, more sportingly, is a historian (or perhaps posthistorian: personified by William Sewell), who rebukes her for the epistemological indecency of seeking to understand causality in historical episodes in terms which ever stray beyond the format of individual narratives of the episode in question. To this, Skocpol cogently responds that she, like most people concerned with politics, necessarily has wider interests, and can hardly abandon these merely because they ruffle the intellectual sensibility of a professional historian. To anyone with such interests (scarcely despicable in themselves), there is no sensible substitute for careful and analytically disciplined comparative historical study.

Where Skocpol is less successful, however, is in clarifying just what is at issue over one major axis of controversy which her book occasioned: the precise relations between the evident good sense of her essentially structural approach and the very palpable presence in revolutionary episodes of dramatic imaginative energy, political will, daring and strategic insight.

It is these elements which critics have picked out as essentially voluntarist in character, irreducibly dependent on beliefs, perceptions, sentiments, and capabilities as agents of particular human beings and groups of human beings. The point of this criticism is not to deny the power or pertinence of structural factors in economy, society or polity, both domestically and internationally. It is simply to underline the speciousness of any attempt to reduce the former elements to the latter: to elide the opportunity for revolution with its historical occurrence.

ÁñÁ«ÊÓƵ

In the present work Skocpol seems less confident than she was that the relation between these two foci of attention genuinely is a binary choice. Or, if she is no better convinced than she used to be of the explanatory compatibility of both elements, she can certainly now be found, at different points, opting firmly for each. Thus, the Shi'ite ulama of Iran are identified by her as both organisationally and culturally indispensable to the revolution which overthrew the Pahlavi dynasty, and credited with giving the menu peuple of Tehran "the moral will to persist in the face of attempts at armed repression". It is hard to see how one could get more voluntarist than that.

The other main fresh conclusion of her present book, the need to distinguish between the vulnerability of contemporary state forms principally in terms of the density and intimacy of their ties with wider elements of their subject populations, fits comfortably with her initial viewpoint. But it seems perverse to ignore the fact that it also fits naturally with a more activist vision of what all revolutions actually consist in: often very intense, and always highly confused struggles between a wide variety of extremely energetic political agents.

Once culture and ideology are acknowledged as causally significant features of the geopolitical and world historical context of particular revolutions, the case for an exclusively external and structural analysis of revolutions collapses. The need to readmit openly idealist elements - the erratic role of human consciousness in determining human action - explicitly into our conception of what is happening in revolutions becomes overwhelming. No one could mistake it for an alternative to structural analysis. But it remains an eminently necessary complement; and no vision of revolution which excludes it can hope to prove politically adequate.

John Dunn is professor of political theory, University of Cambridge.

ÁñÁ«ÊÓƵ