Steven Lukes Palgrave Macmillan, Basingstoke, 2005, x+192pp. ISBN: 0 333 42092 6.

Lukes's classic, brief and bold analysis of power, originally published in 1974, is here reprinted with two further chapters. Sympathetic readers of the original work will not be disappointed with these additional chapters, which cover critical reflections on his initial discussion, plus an analysis of some of the developments in power analysis over the last 30 years. There is the same taut, lucid style and the same incisive level of discussion. Moreover, although Lukes makes a number of concessions to his critics, he sees little reason to make any major shift of viewpoint, which defends a ‘radical’, non-Marxist view of power.

For those unfamiliar with Lukes’ original argument, his main point was that the concept of power required a non-behavioural third dimension, pointing out the way in which power can prevent people ‘to whatever degree, from having grievances by shaping their perceptions, cognitions and preferences in such a way that they accept their role in the existing order of things’ (p. 11). The key question then became: ‘how do the powerful secure the compliance of those they dominate — and more specifically how do they secure their willing compliance’ (p. 12, author's emphasis). Put simply, narrowly empirical political scientists underestimated the power of ideas. Although Lukes claimed that power was an ‘essentially contested’ concept, he held that his ‘radical’ view, which owed something to C. Wright Mills’ power elite model, was superior to the liberal orthodoxies of American political science in the 1950s and 1960s. It offered a more rounded account of power in the sense that the concept was intrinsically value-laden, and identified more accurately the mechanisms of power. Lukes's ‘radicalism’ stemmed from his deployment of Marxist concepts, namely, ‘real interests’ and ‘false consciousness’, in a non-Marxist way. Thus, he dissociated himself from the structural determinism of Althusser and Poulantzas, or the potentially authoritarian notion of a proletarian ‘mission’ espoused by Lukacs. However, he still affirmed the Marxist (and Aristotelian) view of an ‘objective’ human nature associated with the idea of human flourishing.

In bringing his ‘radical’ story up to date, he offers some wonderfully sharp, but level-headed observations on Jon Elster, James Scott, Pierre Bourdieu, Peter Morriss and Foucault. As well, some ‘golden oldies’ are introduced, namely, John Stuart Mill and Spinoza, along with insights from feminist literature. In response to his critics, he is largely unrepentant. His main concession, apart from not distinguishing between power ‘to’ and power ‘over’, is that he acknowledges that his original definition of power was inadequate, confusing power with its ‘exercise’. Rather power should be seen in a ‘dispositional’ sense, or as a property, or a ‘capacity’ of ‘an agent or agents’ (pp. 63, 70, 109).

As one of the classic statements on power, Lukes’ work evokes admiration, but also the desire to find a flaw to demonstrate that the critical intelligence has not been effaced. Thus, in pursuit of fault-finding, a question could be posed along the lines of whether he is too indebted to C. Wright Mills’ power elite analysis that yoked power and ‘responsibility’ together, the capacity of individuals, or collectivities to do ‘otherwise’. Analytically, Lukes is very keen to distinguish his own position from the determinism of structural Marxism. Thus, power could not be attributed to ‘structures or relations or processes that cannot be characterized as agents’ (p. 72). Accordingly, he regarded social life as involving the ‘interplay’ between power and structure (p. 68). But could not power/structure be seen as a potentially false dichotomy? Cannot structures or ‘systems’ have powers and be ‘responsible’ in a causal sense? For example, a capitalist in sacking a worker might easily claim that s/he could not do ‘otherwise’ (declining markets, new technology, etc.), as an ‘agent’ of the capitalist system. To this can of course be added numerous systemic imperatives that have a coercive quality (for example, the constant pressure to increase the productivity of labour, or the capitalist market regulated division of labour). The ‘willing compliance’ in this process could be interpreted as arising not from the actions of individual, or groups of capitalist(s) but from an impersonal ‘system’ that generates commodity fetishism, whereby social, human relations are seen as relations between things, experienced at the most elementary and elemental level as the ‘power of money’. Indeed, given that Lukes subscribes to the Marxist-informed concept of human nature, could it not be argued that capitalism alienates those powers that make us distinctively human?

All this is not to detract from Lukes's achievement, or to suggest that he is wrong to link power and responsibility. Rather, this comment aims to flag up the possible limits of this linkage, and propose that power might involve a fourth dimension in capitalist societies.