Salvador Giner (1985: 254) wrote of civil society that “the imprecision of the notions used may perhaps be more symptomatic of the object described by them than a reflection of carelessness on the part of its Interpreters. In stark contrast to the clearly defined boundaries of its ‘opposite’ entity, the State, those of civil society must always remain unclear. For the State, demarcation is all, whereas for civil society ambiguity – the ambiguity that stems from a certain kind of freedom – is all.” I disagree. Many contemporary commentators define civil society in determinedly idiosyncratic ways. The boundaries of the State are anything but clear. And neither is it clear why the concept of civil society must remain ambiguous. It is notoriously difficult to define, however.

John Keane has written eloquently of this difficulty. “Civil society has no natural innocence; it has no Single or eternally fixed form,” he observes (1988a: 14; cf. Harbeson 1994: 23), and even when investigations focus on the present alone, “modern civil societies have comprised a constellation of juxtaposed and changing elements that resist reduction to a common denominator, an essential core or generative first principle” (1988b: 19). The very concept of civil society is contested, and the problem has only worsened over time: “The ‘language’ of civil society … increasingly speaks in tongues, in accordance with different rules of grammar and conflicting vocabularies” (1998: 52). This confusion has led some scholars (e.g., Honneth 1993; Kumar 1993) to question the usefulness of the term. The answer, though, is not to discard it, but to clarify it and formulate it in such a way that it illuminates a specific problem at hand.