The concept of social capital has become a prominent one in educational research. The two most influential early authors on social capital both wrote widely about education. Pierre Bourdieu viewed social capital as one of a number of capitals that determined the objective position of individuals and groups in the social structure, while both shaping and being shaped by agents' dispositions. For Bourdieu, social capital is a capital of connections, which agents then mobilise in order to pursue their interests relative to those of others. In studies of the French bourgeoisie, Bourdieu treated social capital as both complementing educational credentials (and indeed financial and cultural capital), and as an alternative resource which agents might use to secure their positions despite a relative weakness in respect of other resources. James Coleman approached social capital from the perspective of rational choice theory, arguing that networks and the norms that they promoted could explain why individual actors chose to cooperate with others while pursuing their own ends. In a series of studies of schools achievement and race in the USA, Coleman and his collaborators argued that in some schools, the existence of strong ties between parents, teachers, pupils and community leaders could explain higher levels of achievement than might be expected given the absence of other resources. This paper builds on previous studies (Field and Spence 2000; Field and Schuller 2002) which were based largely on qualitative data and the analysis of published data sets. These earlier studies suggested that social capital may take different forms, and that these need to be taken into account when considering its relationship with learning at different stages of the life span. While it appears that dense community ties can promote high levels of attainment by pupils within the formal school system, these same ties may inhibit the acquisition of heterogeneous information and skills in adult life. The paper will explore these insights through an examination of findings from a UK-based social attitudes survey. Based on a sample of 1,800 adults, the survey covered a range of issues including attitudes towards (a) school life; (b) learning in adult life; and (c) civic engagement. The findings suggest that there is some association between positive attitudes towards civic engagement and positive attitudes towards adult learning. However, the findings are extremely complex in respect of school life, which is almost universally viewed by respondents as providing a weak preparation for lifelong learning. These findings supplement earlier published work, and will inform the development of a typology of possible relations between social capital and lifelong learning.