Of course this controversy has generated commentary on how far the conservative Heritage Foundation will go in its sponsored "scholarship," or how the "PC police" are shouting down legitimate science, or how such an illustrious institution as Harvard could have let slide such a dissertation, or how the swift movement from empirical findings about IQ to racist immigration policy that Richwine's dissertation appears to substantiate will have white supremacists salivating (and even The Heritage Foundation rapidly distancing itself from Richwine). But there's another side to this story that isn't getting any media attention, though I think it's one of the most important aspects of this controversy.
Let's consider what the academics said about Richwine's work. When The New Republic interviewed controversial scholar Charles Murray, who did groundbreaking (if problematic) work on intelligence and class in The Bell Curve (1994), Murray said that Richwine's work was not inaccurate nor intellectually suspect, but simply "taboo." One of Richwine's dissertation advisers at Harvard, Richard Zeckhauser, saw fit to note that "...none of his advisers would have accepted his thesis had he thought that his empirical work was tilted or in error. However, Richwine was too eager to extrapolate his empirical results to inferences for policy."
Zeckhauser goes on to note that in his opinion,
...our School gives too much emphasis on moving from findings to policy implications in scholarly work...in many cases, merely presenting the facts would be a preferable way to go. That makes it much harder for one's opponents to dismiss what you say, or to accuse you of manipulating facts to reach policy conclusions. Moreover, I believe that policy conclusions usually rest on one's underlying values. If one complements one's empirical assessments with values issues, those assessments get questioned, particularly if one addresses a controversial realm of policy...In many contexts, one's work will have a long run greater influence on policy if the facts are left to speak for themselves.
I want to point out something absolutely essential to understand about what these academics, particularly Zeckhauser, are doing with these comments. They're arguing that there are two extricable components to scholarly work on policy: one, the empirical data, two, the "values issues." And as Murray suggests, if your career is ruined because the empirical question you sought to answer verges into the "taboo" territory of the "values issues" sphere, what this amounts to is people putting their weird, idiosyncratic prejudices and preferences above the value of your empirical work because you've touched on something that people have strong personal response too, i.e. that which is taboo. Zeckhauser, too, wants us to understand that so long as Richwine's empirical work held up, we shouldn't be mad at Harvard for accepting his dissertation; and the real problem that Richwine ran into wasn't how he structured his study, nor how he approached the question, nor even how he answered the question, but rather how he moved so indelicately from that answer to a "values issues" policy recommendation.
For anyone familiar with the Kennedy School at Harvard, you'll know that the full name of the Kennedy School is the Kennedy School of Government, not the Kennedy School of Doing Statistics in a Vacuum. For people who place the entirety of their trust and support behind empirical studies and empirical studies only as ways of producing knowledge, this little shuffle that Zeckhauser is doing in the quotes above is a basic survival tactic. The same academics who win major grants and gain status in the policy world by claiming that empirical scholarship is so important because it's so useful in the policy realm will inevitably find themselves backpedaling from that position anytime an empirical study suggests something horrific that we don't want to see or talk about, or that's self-evidently without use-value because it's so ridiculous or untenable for reasons that go beyond methodology or empirical competence. Once that happened to Richwine's Harvard dissertation advisers, they turned about and said, effectively, whoa, look, our School of Government dissertations shouldn't make policy recommendations! Oh no no no. We should just let the FACTS speak for THEMSELVES!
The trouble, of course, is that facts don't speak for themselves. Facts have no value content, no moral position, and no sense of strategy when it comes to policymaking. And if the worth of empirical data in policy studies amounts to not bolstering policy decisions, but, as Zeckhauser puts it, making it "much harder for one's opponents to dismiss what you say," then something very sneaky, and very important to understand, is going on here:
Facts and empirical data are being marshaled as rhetorical devices to support preconceived, subjective values.
To someone like me, this is neither new nor especially problematic. But for someone like Zeckhauser, who appears to believe that facts can "speak for themselves" without containing or speaking with or from some subjective values position--in other words, that we can completely extricate the empirical from the "values issues"--this is just another way of doing bad science.
In the real world (as opposed to Empirical Study Vacuum Land), however, the answers we get depend largely on not just what questions we subjectively choose to ask (whether they're "taboo" or not), but also on how we ask those questions, and how we interpret the results. It's true that today we can (and too often do) hold up data as a rhetorical device; but we should all be savvy enough to call this out when it happens, and to recognize that data don't speak for themselves, but for the subjective position they're marshaled to speak for. Data should be part of many (though not all) arguments; but data are not an argument in and of themselves. Data cannot speak this way.
The real problem with Richwine's study, then, is that he used, as adviser George Borjas noted, a proxy for "intelligence" or "efficacy" or "success"--IQ--that is only weakly predictive of any of these things. The very basic language problem--that IQ doesn't really mean any of these other descriptors--is by itself enough to demonstrate that there's a subjective slippage from data (IQ) to subjective meaning (whatever it is that you want IQ to stand in as a predictor of). If every study that uses something like IQ--easy to measure, as so many easy-to-measure things are, precisely because of how meaningless it is in the real world--simply explained that IQ is a self-sealing proxy for things that IQ doesn't really mean (only more elegantly)--then these studies would be more intellectually honest and more worthy of passing at the Harvard Kennedy School of Government. But the combination of the ongoing obsession with finding simple, authoritative, empirical answers to messy, non-empirical questions (like policy and race), along with the deep-seated fear of confronting and admitting this reality for those who make a living by passing off "data rhetoric" for truth or even meaningful information, mean that the Richwine controversy will continue to develop as just another instance of "underlying values" corrupting what might have been meaningful empirical work.
More importantly, however: knowledge is always already "corrupted" by subjectivity. That's part of the definition of knowledge, which takes on meaning in human vessels, and nowhere else. The facts can "speak for themselves" no more than can a rock or a piece of petrified dung. "The facts" are born into a preexisting discourse, after which point someone will always be speaking for them. In the case of Richwine, the person speaking for the facts had a clear, partisan agenda (one likely doesn't take a Harvard PhD to The Heritage Foundation without at least sympathy for that agenda), a clear interest in quantifying and giving empirical authority to characteristics of racial difference, and a clear interest, as one should have in policy work, in applying his research findings to policy. Rather than fretting over how the data might have been "corrupted" by subjective values, we should be looking at the bigger picture: people who do studies, interpret studies, and apply studies have personal values; and the studies themselves are a rhetorical mode.
