The most well-established environmental determinant of levels of violence is the scale of income differences between rich and poor. More unequal societies tend to be more violent. If this is a relation between institutional violence and personal violence, how does it work and why is most of the violence a matter of the poor attacking the poor rather than the rich? This paper begins by showing that the tendency for rates of violent crime and homicide to be higher where there is more inequality is part of a more general tendency for the quality of social relations to be poorer in more hierarchical societies. Research on the social determinants of health is used to explore these relationships. It is a powerful source of insights because health is also harmed by greater inequality. Because epidemiological research has gone some way towards identifying the nature of our sensitivity to the social environment and to social status differentials in particular, it provides important insights into why violence is related to inequality. The picture that emerges substantiates and explains the common intuition that inequality is socially corrosive. With an evolutionary slant, and informed by work on ranking systems in non-human primates, this paper focuses on the sharp distinction between competitive social strategies appropriate to dominance hierarchies and the more affiliative social strategies associated with more egalitarian social structures. The implications for policy seem to echo the importance to the quality of life of the three inter-related dimensions of the social environment expressed in the demand for "liberty, equality, fraternity."