Normative sociology - the study of what the causes of problems ought to be

Excellent explicative post, On the problem of normative sociology by Joseph Heath.

The whole “normative sociology” concept has its origins in a joke that Robert Nozick made, in Anarchy, State and Utopia, where he claimed, in an offhand way, that “Normative sociology, the study of what the causes of problems ought to be, greatly fascinates us all”(247). Despite the casual manner in which he made the remark, the observation is an astute one. Often when we study social problems, there is an almost irresistible temptation to study what we would like the cause of those problems to be (for whatever reason), to the neglect of the actual causes. When this goes uncorrected, you can get the phenomenon of “politically correct” explanations for various social problems – where there’s no hard evidence that A actually causes B, but where people, for one reason or another, think that A ought to be the explanation for B. This can lead to a situation in which denying that A is the cause of B becomes morally stigmatized, and so people affirm the connection primarily because they feel obliged to, not because they’ve been persuaded by any evidence.

Normative sociology. I've never heard of it but I see it all the time. Assuming the root causes of a problem without actually empirically confirming what the root causes might actually be. The example I see most often relates to reading. Reading is good. Therefore more reading ought to be better. But when you ask for evidence regarding how reading actually measurably causes good outcomes, you are quickly met with stunned but disapproving glares. Of course everyone knows that reading is good. Yes, but show me.

I am an enthusiastic reader and instinctively believe that there is some as yet unexplained causal relationship between good life outcomes and enthusiastic reading. I believe it but cannot meaningfully prove it. Nor can anyone else. The challenge is that if you do not know the nature of the causative relationship, then you have no empirical grounds on which to make a claim for resources. I would love it if schools spent more time on elective reading and spent money building up libraries. But that time and that money has many claims including band and music, sports, science, history, arts, etc. If I cannot show a causal relationship between my objective (more reading and more books) and desired life outcomes (such as aptitude test scores, grades, graduation rates, college acceptances, etc.) then I am making a belief-based argument, which is very weak and ultimately political (can I get enough other parents who believe what I do to prioritize spending on books and reading as opposed to all the other good demands).

I actually think this sort of confusion between the moral and the causal order happens a lot. Furthermore, despite having a lot of sympathy for “qualitative” social science, I think the problem is much worse in these areas. Indeed, one of the major advantages of quantitative approaches to social science is that it makes it pretty much impossible to get away with doing normative sociology.

Incidentally, “normative sociology” doesn’t necessarily have a left-wing bias. There are lots of examples of conservatives doing it as well (e.g. rising divorce rates must be due to tolerance of homosexuality, out-of-wedlock births must be caused by the welfare system etc.) The difference is that people on the left are often more keen on solving various social problems, and so they have a set of pragmatic interests at play that can strongly bias judgement. The latter case is particularly frustrating, because if the plan is to solve some social problem by attacking its causal antecedents, then it is really important to get the causal connections right – otherwise your intervention is going to prove useless, and quite possibly counterproductive.

To solve social problems you have to address the real causal antecedents. The challenge is that often it is exceptionally hard to determine what the real causal antecedents might be. Poverty is not simply a matter of lack of money. It is the product of a horrible stew of historical circumstance, path dependency, behaviors, values, beliefs, experiences, psychology of individuals and groups, institutions, etc. But which of these are most contributive in what fashion and which of those are most amenable to action are major issues.

I recall marvelling at how seldom I had heard this idea expressed: that the left consistently gets it right when it comes to identifying problems, but then gets the explanations wrong (and often clings to those explanations long after they have proven problematic), and so is practically ineffective.

So many social policies not only fail but fail in a fashion that exacerbates the original problem it was seeking to ameliorate. Is this simply a matter of incompetent government? There is plenty of evidence to suggest that that is the case. but I wonder if there is not a subtler issue in play. Perhaps the welter of policy failure is a product of normative sociology. I.e. perhaps government execution is not the biggest issue but the underlying failure to establish causal relationships because of the normative sociology.

Further interesting observations (edited.)

1. Wanting a policy lever. Many of our outstanding social problems remain outstanding because they occur in areas that are outside the immediate jurisdiction of the state: either because they occur in the private sphere (e.g. the gendered division of labour within the family), or because they involve an exercise of individual autonomy, (e.g. students dropping out of high school). As a result, there is no obvious “policy lever” than can be pulled to solve the problem, because the state simply lacks the authority (and sometimes even the power) to intervene directly in these areas.

Think anything to do with poverty and education.

2. Worrying about “blaming the victim.” The most common confusion between the moral and the causal order occurs when people start thinking about responsibility. There is an enormous tendency to think that if person X caused A to occur, then X is responsible for A. As a result, when people don’t want to hold X responsible for A, they feel a powerful impulse to resist any suggestion that X’s choices or actions might have caused A. This is, of course, a confusion, since whether or not X caused A is just a factual question, which doesn’t really decide the question of responsibility. And yet I’ve often heard academics being challenged, after having made an entirely empirical claim about the source of a particular social problem, by people saying “aren’t you just blaming the victim?” One can see here a moral concern intruding where it does not belong. If we follow this line of reasoning, we wind up talking about what we would like the cause of problems to be, rather than what they actually are.

So common. If I cannot make a counterargument, I cast an aspersion on your motivations instead.

3. Picking one side of a correlation. This is a more subtle one. Statistical analysis often reveals a correlation between two things, but as we all know, correlation does not imply causation. If A tends to go hand-in-hand with B, it could be that 1) A causes B, or 2) B causes A, or 3) A and B are mutually reinforcing, or 4) there is some third thing, C, that causes both A and B. It is, however, very very common for statistical correlations to be reported as causal ones.

Excessively common overall but de rigueur in journalism.

4. Metaphysical views. I mentioned this above, but often there is a sense that the moral awfulness of some action or episode requires that it have enormous consequences. This can easily lead to the view that anyone who denies the causal effects is in some way minimizing or downplaying the moral awfulness.

Frequently seen in arguments about poverty, mental illness, crime and gun control. If you dispute the presumed causal flow you are accused of being a denier of something heinous. Again, a symptom of the debater not having an empirical grasp of the issue but a hot flame of moral outrage instead.