Sorry for the light posting recently. I’ve been at the AAAS Annual Meeting. It was easily the best conference I’ve every been to. It was great to not worry about presenting, worrying about what collaborators/competitors were up to, etc. I only attended the talks that I found interesting. There were actually too many talks I wanted to see, and I often had to pick and choose. Again, this situation was quite different in grad school!
One of the more interesting science policy talks I attended was on “Value and Limits of Scientific Research” (link). Kei Kozumi and David Goldston always have great things to say. It was refreshing to see people wrestle with the tough policy questions around research funding rather than insist that all problems will be solved by more money for science.
A lot of the discussion focused on the effect of the stimulus bill. Because the stimulus bill was designed to have a near-term effect, much attention and energy focuses on the impact of stimulus dollars spent. And so for the first time (?), we’re really trying to undergo a very detailed evaluation of how R&D dollars move through the economy.
We also discussed the potential opportunities and pitfalls of framing science as a job-creating engine. Historically, science was justified as the source of long-term economic growth. Things could get a lot more complicated if we get into the short-term jobs market. There will need to be real evidence that we meet that mandate rather than nebulous statements about basic research leading to technology. And that’s when we can possibly get into trouble.
David Bruggeman recently attacked Chris Mooney yet again for promoting the war on science meme: the concept is meaningless, incoherent, oversimplified, etc. Dan Sarewitz echoed many these arguments in his review of Mooney’s book.
I also found Chris Mooney’s thesis irritating and sloppy. His constant, unadulterated worship of science gets old very quickly. But it’s important to acknowledge that Mooney has a point. George W. Bush’s administration did politicize science a lot more than his predecessors. Since my placement at the EPA started in September, I can’t count how many times I’ve heard complaints about Bush’s interference. Despite some over-generalizations, Mooney collected a troubling body of evidence.
Complaining only about the former problem implies that abstract concerns–how dare Mooney not discuss social construction!–matter more than real world impact. Can we honestly say that distorting EPA reports is no worse than believing in value-free science? Ironically enough, this attitude makes us STS-sympathizers just like those academic scientists we routinely berate.
So yes, everyone does misuse science for their own ends. And yes, Mooney annoyingly promotes a false purity of science. In the end Bush’s actions were different only in degree, not kind, from previous administrations. Agreeing with all this, however, is perfectly compatible with condemning his egregious politicization. It’s possible to be upset at the exaggerations and distortions of both Chris Mooney and George Bush. Bruggeman’s and Sarewitz’s worthy attempts to bring nuance to policy debates unfortunately spends too much time on the former and not enough on the latter.
My recent post on science and race should have had some caveats on the utility of cost-benefit analysis (CBA). I’ll be the first to admit that it can be abused and has its limits, especially with respect to environmental policy. Although I haven’t actually read the book, Frank Ackerman and Lisa Heinzerling make that point in Priceless.
Nevertheless, I think that on some level we have to use CBA. In the end it is a useful tool, and I mostly agree with Sunstein’s critique of Ackerman and Heinzerling. Yes I know it’s unfair to read the criticism and not the original work. Sue me.
So whatever caveats we attach to CBA, my larger thesis is unchanged: CBA plus values*, not science, should be used to analyze social policy like government funded pre-school. And it goes without saying that when I say “my” thesis, I really mean Nobel-prize winning James Heckman’s thesis. But I’ve already admitted that most of my work isn’t original.
*I think the “plus values” part is important because you can make all sorts of principled, theoretical arguments for or against these types of policies. Go read some Nozik or Mansfield for the anti-view, and Rawls or Galston for the pro-view.
fyi, I list all these philosophers to impress my non-existent readers with my erudition. I also like to believe my (imaginary) readers don’t even know what that word means and are looking it up right now. For what it’s worth, I’ve actually read two of Galston’s books. My fictitious readers are now even more impressed, and even more so by the creative ways I complain about my lack of readership. 13 page views over 10 days isn’t bad…right?
My last post discusses the possible harm of automatically placing science at the forefront of decision-making. In some cases it’s simply not true that a scientific lens is the best way to analyze a problem. It seems that we’re begging for a public vocabulary that lets us meaningfully discuss science. We need a way to accept the importance of facts without allowing them to stifle debate. See for all my problems with the current discourse, I’m also sympathetic to scientists who promote it. People shouldn’t be able to ignore facts they don’t like. It’s not okay to cherry pick data you happen to agree with. It does matter that all the relevant experts agree with evolution and anthropogenic global warming.
The solution to this dilemma, however, is not to insist that science is the foundation of policy. As I also discussed in my last post, doing so is scientifically inaccurate. Engaging in this rhetoric makes us, for lack of a better phrase, somewhat hypocritical. How can we speak about the importance of evidence while ignoring the scientific fact that science is only sometimes the foundation of policy?
We need an intellectual framework which articulates there are instances when science is crucially important, instances when it is somewhat important, and instances when it is relatively unimportant. Some decisions heavily rely on science while others do not. Certain disputes are better resolved with politics rather than science and vice versa.
I’d guess that this message will not fly with many scientists. There’s too much nuance there. It doesn’t quite fit our science is God and you’re either with-us-or-against-us rhetoric. My admittedly naive view is that with respect to public communication, science should be neither deified nor demonized. We should instead strive to highlight that it has its uses and can sometimes be very helpful.
To that end, I’ll suggest yet again that science in decision-making should be thought of as team sports. How about the image of science as a key but not superstar running back? The opponent and game plan dictates how much we play. Sometimes winning the game means handing us the ball 20 times a game. Sometimes we have to sit on the bench. How do you know when to stress the running game? By watching game film!
This image, I believe, is the most accurate one we can paint. Before jumping to any conclusions about the role of science, we must first carefully study the situation. Then and only then can we say whether science is the foundation or a cosmetic fixture. Whether we’re the MVP or 6th man of the year. But these analogies are now getting tedious and I think you get the point.
Part of the controversy over The Bell Curve and James Watson was the idea it must mean something if there were a genetic basis for the black-white IQ gap. We couldn’t just ignore this fact like we do much of science. Surely a putative link between race, genes and IQ has more significance than, say, the existence of the radiation belts. (Sorry, I had to toss in some space physics!) Herrnstein, Murray and Watson themselves used these alleged facts as the basis for social policy recommendations.
It is striking that the authors do not discuss the costs and benefits of various interventions. It is in these terms that public policy discussions regarding skill-enhancement programs are usually conducted. The authors seek to short-circuit all of the hard work required to make credible cost-benefit calculations by claiming that there is a genetic basis for skill differences.
But estimates of a genetic component of skills are irrelevant to the requisite cost-benefit analysis unless it can be established that all differences are genetic. No one, including the authors, claims that this is so. [Emphasis added–PK]
So even if we scientifically proved that some portion of the IQ gap can be attributed to genetics, those facts would not help us decide whether the government should fund pre-school. What does help are data showing a 7:1 return on investment and principled reasons on, e.g., the role of government. But however you make the case, genetics really has no role. At least in this case, science narrowly defined is most definitely not the basis of policy.
None of this negates the idea that we should try to dispel the sloppy science. It is important to explain what is and isn’t known about race and IQ. It is important to explain that science may never be able to determine the link (read towards the end of Metcalf). But it is also important to explain that in this case we can better understand the controversy by ignoring the science.
As I’ve argued before, scientists often place science at the center of decision-making. The pattern holds up here. Herrnstein et al. argued that the science of race and IQ implies early childhood education is a waste of time, while others disputed those narrow claims. But it’s crucial to note that scientists did not stress that the issue is not about science and framing it as such is is wrong. Not wrong in an ethical or moral sense. And not wrong in the sense that people shouldn’t exaggerate their importance. It’s wrong for purely empirical reasons: some policies are not decided on the basis of science.
Which finally brings me to the title of this blog post. Since science is in fact irrelevant to some decisions, are there negative consequences for pretending otherwise? Did scientists inadvertently foster negative racial attitudes by opposing The Bell Curve without also pointing out its irrelevance for social policy? I admit that I’m making a very convoluted argument. But bear with me while I try flesh it out.
There are four key points. First, this dispute centered on the science of genetics. Second, this approach is empirically false. The argument should have been about cost-benefit and government’s role in society. Third (and for the umpteenth time!), scientists’ sole response to Herrnstein, Murray and Watson was to attack the scientific basis of their arguments. We made it sound like Herrnstein et al. would have a point if only their science were correct. Fourth, given how technical the issue is, it’s inevitable that some people were not convinced by our rebuttals.
These facts lead me to believe that the inappropriate public framing decreased support for redistributive social policy. That is, some people who initially things like supported universal child care changed their opinion precisely because of the prominence given to Bell Curve type arguments. I’m not sure how to test this idea. But if I’m even partially correct, it appears that how we frame science might have reduced enthusiasm for policies that most help poor black and Hispanic Americans. In short, how we speak about science may unintentionally screw over poor black people.
Finally, there’s a very good chance this very long post that will be read by no one. Nevertheless, I’d be interested in what my (non-existent) readers think. There’s a non-trivial chance I’m spectacularly wrong and it’d be great to hear why and how.
I’m sick of writing about creationism and intelligent-design. So I’m going to switch to another favorite topic of mine: black people. To be more precise, I’ve always been fascinated by race relations in America. For better and for worse, in America this issue is mostly framed in terms of white and black people. And since I’m also fascinated by science, what I meant to say is henceforth the bulk of my literary endeavors will focus on the myriad intersections of science and race in the American social order. Whew! It’s funny how precision can make your writing wordier and less entertaining. I’ll make sure to avoid it in the future.
This is my last post on intelligent design (ID) for a while. But I want to examine the evidentiary claims surrounding the debate. I’m not talking about the evidence for or against evolution. It’s clear that the science overwhelmingly supports the theory.
Consider typical arguments against ID: Teaching ID risks future generations of scientists; students who learn the theory will be unprepared to wrestle with science-related public policy; economic growth will be harmed if citizens have poor scientific literacy. There are at least four statements here amenable to empirical testing:
- If students learn ID, then their scientific literacy (SL) will be harmed.
- If SL is harmed at any point in students’ education, then they will be less capable of becoming scientists.
- If SL is harmed, then citizens will be unable to reflect on public policy.
- If SL is harmed, then economic growth in an increasingly technological society will be affected.
You can phrase this differently or even make additional claims. But I think I’ve accurately summarized how scientists usually argue against ID.
For the sake of argument, I’ll grant that number 3 is self-evidently true . Poor scientific literacy negatively affects deliberation on science issues, thereby undermining democratic governance on some level. But the remaining claims are not so obvious. It’s not at all clear that learning ID will affect overall levels of scientific literacy. It’s especially unclear how public SL relates to economic growth. And as I discussed previously, many scientists believe in ID.
In proper scientific fashion let me offer some testable predictions and a way to test said predictions. I predict that learning ID only affects SL with respect to the theory of evolution . In all other areas of science, learning ID has no impact. I also predict that in college ID-learning students study natural science and engineering at the statistically same rate as non-ID students. As a test, I propose studying children who were home-schooled for religious reasons. Of course, we’d have to try control for household income and wealth, parents’ level of education, etc. It wouldn’t be easy, but social scientists do these types of studies all the time. Finally, I predict that mainstream scientists (including me!) would still oppose ID regardless of any data.
Despite my constant harping on this matter, it’s our tone and attitude I dislike, not the existence of our opposition. I do think it’s important for scientists to draw boundaries . We almost have an obligation to exclude ID from the realm of science. But disagreement, however fierce, should not corrode public discourse with threats and sloppy arguments. It’s important to note that we effectively bully people who support ID: your children’s future will be ruined if they learn intelligent design! In addition to being distasteful, this threat isn’t even very believable. We not only act like jerks, we act like jerks that make bad arguments. The fact that scientists make these unsubstantiated claims is even worse.
 This claim isn’t as straightforward as it appears. Social scientists have shown that even those who appear to be scientifically illiterate can be surprisingly reflective and engaged in some settings. See, for example, Wolff-Michael Roth and Stuart Lee, Scientific literacy as collective praxis, Public Understanding of Science, 11, 2002, pp. 33-56.
 Of course I’d have to define what I mean by scientific literacy. Considering that no one has ever been able to do that, I’m going to cheat by ignoring that problem. See George Deboer, Scientific literacy: another look at its historical and contemporary meanings and its relationship to science education reform, Journal of Research in Science Teaching, 36 (2), 2000, pp. 582-601.
 Check out Thomas Gieryn, Boundary-work and the demarcation of science from non-science, American Sociological Review, 48, December 1983, pp. 781– 795.