groups which do not bridge social cleavages may create trust internally among their own members, but distrust
externally with other groups and wider society. Organizations which cut across social cleavages may help to teach
tolerance and understanding, and thereby create the ‘habits of the heart’ (Bellah et al. 1985) associated withtr ust,
reciprocity and co-operation.
Two observations can be made about the de Tocqueville/Putnam model of social capital as rooted in community
groups, voluntary associations, and intermediary organizations. First, there is, indeed, a huge number and great
diversity of suchorg anizations in all Western societies, and they cater for almost every conceivable sort of interest and
activity known to humankind (Newton 1976: 31–88). At the same time they account for a relatively small proportion
of the time of most citizens. A large minority of people in most Western societies do not belong to any voluntary
organization, and most of the rest do not spend much time with them. Only a highly active but small stage-army of
community leaders and organizers (the ‘joiners’) give a significant proportion of their adult life and emotional
commitment to voluntary associations.
By most measures the Netherlands has one of the highest levels of voluntary membership and activity in the Western
world, but members and volunteers spend an average of between four and five hours a week with their organizations
(Dekker and de Hart 1996: Table 12). This amounts to around 8 per cent of their leisure time. In the Republic of
Ireland two-thirds of volunteers give less than ten hours a month to the activity (Ruddle and O'Connor 1995: 93). In
ten Western and Eastern European countries the average is around ten hours a week (Gaskin and Smith 1995: 31).
For the great majority who participate in associational life, other institutions are likely to have a much more important
role in generating sociability: for the young it is school and the family; for adults, work and neighbourhood. Coleman
(1988: 109–16) emphasizes school and the family; Putnam (1995b: 667) suggests that education has a strong
connection withboth trust and association membership; Verba et al. (1995: 320, 514) find the workplace and education
important. School, family, work, and neighbourhood are likely to have a far greater significance in the origins of trust,
reciprocity, and co-operation than the limited and sporadic involvement of most people in voluntary organizations (see
also Levi 1996: 48). It is not surprising, therefore, that recent American research finds much higher levels of social
participation than trust (Broder 1997).
Nor is it a surprise that World Values data show no statistically significant relationships between voluntary activity and
social trust in four countries, and only a weak relationship in three others (Table 8.1). Only in Italy is the statistical
association strong, and even here it is weaker than four other measures dealing with satisfaction with life, morality,
national pride, and social and economic status. The voluntary organization variable in these regressions is the strongest
available, dealing not merely with members but with the most deeply involved activists who do unpaid voluntary work.
If activists do not