Commentary: Reconciling the three accounts of social capital

Abstract

The subject matter of social capital tends to arouse passions. After years of debate that often generated more heat than light, Szreter and Woolcock1 have come up with a conceptual framework for examining social capital and health which promises to reconcile the opposing camps. They identify three existing accounts of social capital as it relates to population health, which they refer to respectively as the ‘social support’ perspective, the ‘inequality’ thesis, and the ‘political economy’ approach. As noted by Szreter and Woolcock, an often polarized debate has taken place within public heath between proponents of the ‘inequality’ thesis and the ‘political economy’ approach, with the former group additionally tending to emphasize the ‘psychosocial’ interpretation of the mechanism linking social capital to health,2 as opposed to the ‘neo-material’ interpretation favoured by the latter group.3 We have argued elsewhere4 that the ‘debate’ between the psychosocial and neo-material positions poses an unnecessary distraction. We concur with Szreter and Woolcock's view that in the absence of compelling empirical evidence to distinguish between them, ‘it would seem most sensible to assume that both viewpoints could be valid.’1 The novelty of the proposed framework by Szreter and Woolcock lies in distinguishing and unpacking the concept of social capital in its different forms, namely, ‘bonding’, ‘bridging’, and ‘linking’ social capital.1 The distinction between bonding and bridging social capital has existed for some time,5–7 and is by now widely accepted in the field. Bonding social capital refers to trusting and co-operative relations between members of a network who are similar in terms of social identity (e.g. race/ethnicity), while bridging social capital refers to connections between those who are unlike each other yet are ‘more or less equal in terms of their status and power.’1 The bonding and bridging varieties of social capital could be consistent with either or both the ‘social support’ and ‘inequality’ accounts of social capital and its relationship to population health. But according to the ‘political economy’ perspective on social capital, what is missing from the bonding/bridging distinction is an explicit recognition of vertical power differentials in social relations. It is to address this gap that Szreter and Woolcock introduce a third form of social capital—the ‘linking’ variety—which they define as ‘norms of respect and networks of trusting relationships between people who are interacting across explicit, formal, or institutionalized power or authority gradients in society.’1 By bringing state–society relations and power into the equation, Szreter and Woolcock argue that the concept of linking social capital can reconcile the ‘political economy’ and ‘neo-material’ perspectives with the remaining accounts of social capital found in public health. To what extent have they succeeded? Szreter and Woolcock have produced a framework for social capital that is both sophisticated and more complete than any of the three existing versions. Their proposed approach has the merit of incorporating not only power differentials in society, but also a dynamic dimension describing the antecedents of social capital informed by careful historical analysis. Some gaps still remain to be fleshed out, however. In this commentary we draw attention to two additional issues that are not canvassed by Szreter and Woocock's proposed synthesis. The two issues relate to: (1) the distinction between individual versus collective conceptualizations of social capital; and (2) the methodological and empirical implications of adopting the bonding/bridging/linking distinction. One important disagreement in both the theoretical and empirical literatures on social capital relates to the differences between those who view social capital as an individual attribute versus those who view it as a property of collectives (for example, communities or entire societies). This individual/collective distinction is almost orthogonal to the bonding/bridging/linking distinction identified by Szreter and Woolcock. Whilst noting that the centre of gravity in existing definitions of social capital is tilted toward the individual end, Szreter and Woolcock tend to hedge their bets by adopting a position somewhere in between the micro and macro ends of the spectrum. In other words, they recognize that individuals can draw upon resources through their connections to others (the ‘social support’ mechanism), but at the same time, they argue for the need to include the nature and extent of state–society relations as a necessary part of the theory and definition of social capital. Is it legitimate to pack so much into the concept of social capital? Some have questioned whether social capital can withstand such ‘brand-stretching’ or ‘conceptual creep’. For example, Portes8 warned that so many disparate social phenomena have been heaped upon the concept of ‘socia