Abstract
Scholars in the field of population health need to be on the constant lookout for the danger that their tacit ideological commitments translate into systematic biases in how they interpret their empirical results. This contribution illustrates this problematic by critically interrogating a set of concepts such as tradition, trust, social capital, community, or gender, that are routinely used in population health research even though they carry a barely acknowledged political and ideological load. Alongside this wider deconstruction of loaded concepts, I engage critically but constructively with Martin Lindström et al.'s paper "Social capital, the miniaturization of community, traditionalism and mortality: A population-based prospective cohort study in southern Sweden" to evaluate the extent to which it fits with other empirical findings in the extant literature. Taking as a point of departure the intriguing finding that social capital predicts cardiovascular and all-cause mortality only for men, but not for women, I argue that future research on the nexus of social capital, health, and mortality needs to frame gender not only as a demographic and statistical variable, but also as an ontological conundrum and as an epistemological sensibility.