"The problem from the beginning was not Moynihan’s publication of what were actually well-established facts, but rather his distorted interpretation of these facts. ... If Moynihan’s critics were unusually vociferous, this was because they understood what was at stake." (Stephen Steinberg)
Brayden King's post at OrgTheory is a great illustration of the constitutive tension between social facts and social problems in social science that I was talking about on Monday. Here the question is whether "cultural factors", which are a subset of the totality of social facts, ought to be invoked in our explanation of poverty, which is one of the classic social problems.
Brayden takes an explicitly pro-science position, going so far as to call Steinberg "anti-science":
The main problem I have with Steinberg’s argument, as a complete outsider to the field, is that he seems to be saying that we should reject all cultural mechanisms that might explain variance in poverty, regardless of their actual explanatory power. This dogmatic stance seems very anti-science and anti-scholarly inquiry. Rejecting explanatory mechanisms without empirical evidence is a sure way to stultify the progress of knowledge about our social world.
I am more inclined to grant Steinberg's point, even in the somewhat sharpened version that Brayden offers. In the case of Moynihan's report, in fact, Steinberg, it seems to me, is challenging precisely the explanatory power of cultural factors: "Moynihan made the fatal error of inverting cause and effect." But even in the case where an "empirical" argument could be made that a group's culture explains why it is poor it will, it seems to me, be difficult to decide whether to stop the analysis before explaining that group's culture.
That, again, is the constitutive issue of social science. Once we have discovered a set of social facts that have sufficient explanatory relevance and that we have sufficient "empirical evidence" for, we must ask whether those facts are also relevant to the solution of an interesting social problem. As Small et al.'s paper and Steinberg's response show, social scientists simply don't confine themselves to Brayden's rules of critique. Sometimes social scientists do disregard the "science" of their work and criticize each other based on the more or less obvious political implications of what they are saying. Sometimes they grant the "empirical basis" for the sake of argument. The reason is that there may be some larger fact that sufficiently explains the "factors" in question, and social scientists might see it as their duty to discover that larger fact before legitimizing well-established myths about the factors.
One way to look at this debate is through the implicit principle that any factual explanation must identify that which must be changed to fix the problem. (This principle can be seen in Steinberg's commentary here, for example: "it is not their culture that needs to be changed, but rather a political economy that fails to provide jobs that pay a living wage to millions of the nation’s poor, along with a system of occupational apartheid that has excluded a whole people from entire job sectors throughout American history.") If it would be inappropriate—or even just unrealistic—to "fix" the culture of the poor, then it is irrelevant to study it. Otherwise, the sociologist is merely explaining why society "has to be" the way it actually is. Some social scientists appear to be content to produce such explanations. Fortunately, the field of sociology as a whole remains critical of them.