My class autobiography got people riled up. More commenters were riled up at Economist's View than here, by the way. There are some things that I wrote that I don't wish to stand by and defend, after thinking about it more. But there are some elements of truth in it. My guess is that any effort to defend the latter is ill-conceived, but here goes.
Remember that I am talking about people living in poverty in the United States, excluding new immigrants. I would also like to exclude people who are temporarily poor, for example someone who was poor in 2003 because he was unemployed most of that year, but who now is making at least a modest living.
The term "mental illness" is misleading and comes across as pejorative. I should not have used it in such generality in talking about poverty. Within my narrow experience of people I know well (including relatives), it is the major explanatory variable. But I will admit that people I know well includes only a small and non-representative sample of people with very low incomes.
However, I would point out that
1. From the perspective of the theory of social class in the United States, differences in economic outcomes across siblings are inconveniently large relative to economic differences across families. It is common to find siblings born to identical parents who end up near the opposite end of the income distribution.
2. I think that many people who work with poor people would argue that poverty includes elements of pathology. One of the commenters on my post put it, "the bulk of the poor are not mentally ill -- the rules of their lives are so radically different from those of the affluent, it just looks that way from the outside in." The commenter recommends two books. I recommend another one, Random Family.
3. Although one can argue that poverty is a factor causing mental illness, I think it would be very rash to dismiss the other direction of causality. Again, look at siblings. There can be very large differences in mental health across siblings, and that to me suggests that a univariate explanation of mental health as being determined by social class will not work.
4. Arguing against social class as a univariate causal force may seem, and perhaps it should seem, like arguing against a straw man. But there are people who focus on social class almost as if it were a great causal absolute. For these class-obsessed academics, it is as if nothing has happened, either in the real world or in social science, since the days of Karl Marx and Max Weber. I think that this class-centric view of the world forces one to become highly angry and rhetorical in debates, because there are so many inconvenient facts with which one has to deal.
5. This not simply an academic debate, however. Those of us who focus on pathology (and in the case of poverty in the underdeveloped world, we look at pathological government and pathological culture) are skeptical of handouts. Handouts encourage people to think of wealth as something that is transferred or redistributed, rather than something that is earned. This in turn may encourage people to focus on redistribution as a way of capturing wealth. Such a focus makes poverty worse, not better. There is an uncomfortable association between welfare and crime and between foreign aid and government corruption.
Thus, the class-centric school of thought believes that handouts are helpful and just. The pathology-focused school of thought believes that handouts ultimately serve to exacerbate the very problem that they are intended to solve. There is probably more passion than evidence on both sides of the debate, but my opinion (perhaps because I am more pathology-focused) is that the class-centric school works hard at tuning out evidence against its viewpoint.