I've been interested in the relationship between crime and environment for several years now, ever since I saw some compelling data from Martin Day on the strong correlation between income inequality and rates of homicide, across states, provinces and countries. People don't seem to commit murder because they're poor, they commit murder because they're poorer than many of the people around them. India, for instance, is a poor country but has a similar income equality to Switzerland and a lower homicide rate than the United States. This kind of makes sense since poverty is fundamentally a relative term - the nobility of medieval times had a standard of living lower than many Wal-Mart employees today.
We also see that racial and economic segregation increase the rate of violent crimes like aggravated assault and robbery. It's no coincidence that in the US, cities like Detroit, Baltimore, Oakland, New Orleans, St. Louis and even Chicago among the big cities, are known to be among both the most segregated and most crime-ridden. It's also probably not a coincidence that South Africa has both one of the highest levels of segregation and highest levels of violent crime (it also has a high degree of income inequality as well).
Putting on my psychologist hat, it makes me wonder about the mechanisms at the individual level that result in a higher disposition to committing crimes in these environments. It seems to be a complex interaction among trust and trustworthiness, unwritten societal rules, and who you see as "like you."
There's plenty of data to show that we treat people who are "like us," who have been given entrance into our mental representation of the circle of kindred, very differently than we treat people who are unlike us. Philip Zimbardo, in his research, saw that humans are at their cruelest when they can dehumanize the other (Stanford prison experiment) and anonymize themselves (e.g. with masks or facepaint). If what is truly evil are free personal choices to harm others who can feel, for selfish motivations, then most humans shy away from the truly evil. They create a mental framework where it seems like, on some level, their victims are not "like them" and cannot "feel." It's like the mafioso who are family men, yet horrifically cruel to those who belong to economic competitor groups. You can be trustworthy within the circle of trust, but behave completely different to those you consider "out-group." You abide by the unwritten societal rules of your group, which has its own self-governing mechanisms.
The factors that constitute whether you consider someone "in-group" are some combination of shared genetics, in-common relationships, physical proximity, income and educational levels, race, gender, values, and finally, where you live. This final factor of where you live is important because in-group is about who you trust, and trust can be a strange thing. We need some trust to operate in the world, to take calculated risks, and it comes in three flavors: dispositional, situational, and institutional (Dignum 2005, Chervany 1996). Our past experiences in a given environment shapes our dispositional trust, our personal worldview. The context of the environment shapes our approach to situational trust, how we react to the players and situation at hand. And institutional trust, when we can rely on the larger system, institutions and government mechanisms to enforce expected behavior, enables us to expand the group we trust (or when the converse is true, depend on a limited set of kin relationships and local protection rackets). It's easier to see more people as "like you" when the potential cost of that trust is lower. Where you live makes a big difference.
When we have strong governance institutions, lower levels of income inequality, and less segregation (and more in-common relationships), we see low levels of crime. But when we have the opposite, the quicksand trifecta of weaker governance institutions, high levels of income inequality, and high segregation, we have a situation like South Africa.
Paul Collier has spoken and written about the role of weak governance in reinforcing the cycle of persistent poverty. The answer has to start with good governance, which drives institutional trust, which helps suppress organized crime and promotes a private sector more attractive to investment (and more likely to create jobs). To the average citizen, good governance looks like effective and just policing, clean well-paved streets, fast and fair public services, documented and enforced contracts and property rights, and transparent reporting and feedback systems.
Next is skills-building, the sort of education that builds sustained advancements in human capital, so that the population can take on the jobs that are created by the private sector and alter the income dynamics. This is not necessarily university education. It does include capability-building around the productive use of technology, one major factor in the income discrepancy between educational classes.
Finally, and probably most difficult, is promoting physical community spaces, next-generation civil engineering, and public school systems that enable relationships to be built across socio-economic classes. Social groups are generally formed by the time a given individual is in their late-twenties, often in school and work environments. There are relatively few direct opportunities to shape these formations - probably the most significant are primary and secondary schools that break the linkages between neighborhood-income and school attended. Otherwise, a city may be subject to silent segregation, where different social groups may live close to each other but rarely interact (e.g. the Mission of San Francisco).
When I think about someone like John, who refuses to target "mommas" because they make him think of his own mother, it makes me wonder what else we can do to get people to think about others as "like them." What stories can we share, what introductions can we make, what career opportunities can we offer. We all win when we have more trust in our civil society.