The concept of trust is currently on the lips and in the minds of many social scientists. Interest in the concept has
expanded rapidly in the last few years and has generated a literature in the social sciences and philosophy (see, for
example, Gambetta 1988; Misztal 1996; Bianco 1994; Fukuyama 1995; Warren 1996; Hardin 1996; Sztompka 1996).
One reason for the interest is that trust represents a central concept in social capital theory, which is commanding a
good deal of attention. Bothsoc ial capital and trust are close cousins of the writings on civil society and
communitarianism, and all are developing along rather similar lines. Moreover, the concepts of social capital and trust
are, in some ways, modern social science equivalents of the classical idea of fraternity, and in the late 1990s social
scientists have rediscovered the importance of fraternity, or something like it, as a necessary condition of democracy.
For much of the past twenty or thirty years, politics and political science have tended to ignore fraternity, concentrating
instead on liberty and equality. Indeed, in the Reaganite and Thatcherite ethos of the 1980s, political discussion often
focused exclusively on liberty and overlooked or ignored completely both fraternity and equality. More than this,
discussion of liberty tended to circle around a peculiarly narrow view of the concept as if it were related only to
economic and market freedom.
The late 1990s have reacted against this view of the world, recognizing that citizens are more than the sum of their
shopping. Such things as economic growth, social integration, political stability, and government efficiency may well
depend upon underlying values, norms, and attitudes, and social conditions which create the capacity to trust and cooperate
(see, for example, Inglehart 1990: 23–4, 34–8; Inglehart 1997a: 172–4; Inglehart and Abramson 1994).
Without these underlying values and social capacities, society could barely function, and then only with great friction
and inefficiency. In the
51 I would like to thank the Wissenschaftszentrum Berlin fur Sozialforschung where I spent a wonderful sabbatical year, 1996–7, during which this paper was written. I would
especially like