This paper examines trust and exclusion. Data were collected from 505 Israeli respondents by telephone. Whereas findings show that the influence of social and political variables on both forms of exclusion varied, that of social capital remained constant and substantial. Interpersonal trust was a better predictor of group exclusion, whereas institutional trust was a better predictor of individual exclusion.
Social science research has revealed public dissatisfaction with immigration (Al-Haj, 2002; Freeman, 1997; Watts & Feldman, 2001), and with increases in immigration levels (Holton & Lanphier, 1994). There has been much research into this and other types of exclusion (Giles & Hertz, 1994; Legge, 1996). This study attempts to trace the link between civic values and exclusion, particularly two concepts: social capital and social exclusion. Three research questions are essential: first, what is the association between social capital and exclusion?; second, what is the association between types of social capital and types of exclusion?; third, what are the determinants of exclusion?
Whereas trust involves "rationality" (Newton, 1999), exclusion is "emotional" (Portes, 1998; Portes & Landolt, 1996; Quillian, 1995). Following Quillian, we distinguish between individual and group-level exclusion. Refusal to interact with individuals, because of differences, is an "individual" type of exclusion (Eliasoph, 1999). Group exclusion is collective, and involves denying group socio-political rights and legitimacy (Mudde, 2000).
TRUST AND EXCLUSION
Social capital has recently become an influential concept. Tocqueville (1945) argued in favor of social organizations for practicing democratic principles. More recently, origins of the construct of social capital lie in the theories of Bourdieu (1986), Coleman (1990), and Portes (2000). Putnam (1993; 1995; 1996; 2000) argued that the quality of democracy depends on social capital - networks, norms, and trust (Skocpol, 1996; Tarrow, 1996). Its critics claim that social capital may also prove dangerous (Foley & Edwards, 1999; Levi, 1996).
Trust is regarded as one of the most complex and multidimensional concepts in the social sciences (Fukuyama, 1995; Tsfati, 2003). In the literature on social capital, trust is considered paramount (Kaase, 1999; Paxton, 1999; Uslaner, 2002), involving long-term relations between two sides - a "trustor" and a "trustee" (Tsfati, 2003). The literature stresses the uncertainty and expectation that interaction with the "trustee" will lead to gains to the "trustor" (Coleman, 1990). Uslaner, however, approached the concept as a moral value, stable over time.
Trust is socially learned and socially confirmed expectations that people have of each other, of the organizations and institutions in which they live (Barber, 1983:165). This definition combines trust in individuals and in political leaders and institutions. Interpersonal trust is based on the expectation that others will not harm us (Hardin, 1992; Newton, 1999). However, political trust is targeted at politicians or public institutions (Smith, 1981). Interpersonal trust and institutional trust are not always related (Kaase, 1999; Rose, Mishler, & Haepfer, 1997).
Trusting people are more tolerant and more willing to support antidiscrimination government programs (Uslaner, 2002). Since exclusion is determined not only by social and political factors but also by moral values and attitudes (Rokeach, 1960), the type of trust may be relevant to the type of exclusion. People who trust are likely to have favorable attitudes towards individual immigrants, and those trusting political institutions are likely to accept groups (Putnam, 2000).
OTHER FACTORS DETERMINING EXCLUSION
In line with sociological explanations, if people interact with the "other", they are less likely to harbor negative feelings (Robinson, Shaver, & Wrightsman, 1999; Yiftachel, Alexander, Hedgcock, & Little, 1999). …