The focus of this book is necessarily as much on the obstacles to accomplishing health- bestowing change as it is on characterising and accounting for why such change is vital. And the canvas has, again necessarily, been large. As was suggested in the Introduction, my principal grouse against much sociological research on health inequalities, and much of the literature on policy recommendations too, is that they barely mention, or even ignore, social class and other beneath- the- surface or structural causal mechanisms and are, as a result, unrealistic and merely aspirational. The agendas of experts such as Michael Marmot are, of course, epidemiological rather than sociological. In our various personal and friendly exchanges, he has stressed that his concern is to provide the scientific evidence on health inequalities and to push for relevant policy reforms to secure their reduction. This is the commendable ‘realism 1’ of the insider. But health inequalities are currently widening in our fractured society and have been hit hard post- 2010 by political austerity and COVID. An alternative approach is via the sociologist as outsider, and this is the one favoured here: my alternative ‘realism 2’ draws on the theories of Marx, Habermas and Bhaskar. And to reiterate, an immediate and pressing concern is why, despite their best- efforts, Marmot and his colleagues have not inspired effective policy change. Another conspicuous theme has been that health inequalities in the UK can and should no longer: (i) be broached in isolation from global health inequalities, and (ii) be divorced from pressing and ubiquitous risks, typically exported from the high- income core countries of the Global North to the low- income peripheral countries of the Global South, such as climate change and conventional and conceivably nuclear warfare and their sequelae.

Using an implicitly critical realist frame, it has been asked what social structures must exist for empirical studies of material, social and health inequalities to deliver the findings they have. Given the nature of the open society and the unavailability of experimental ‘closures’, sociologists have typically to rely on what Bhaskar calls ‘retroductive inferences’ from the ‘demi- regularities’ deriving from quantitative research, or what he calls ‘abductive inferences’ from qualitative or ethnographic research.