What do publicly-funded scientists owe the public?

Yesterday, I discussed what scientists supported by federal funds do, and do not, owe the public. However, that discussion was sufficiently oblique and ironic that the point I was trying to make may not have been clear (and, I may have put some of my male readers at greater risk for heart attack).

So, I'm turning off the irony and giving it another try.

The large question I want to examine is just what publicly-funded scientists owe the public. Clearly, they owe the public something, but is it the thing that Dean Esmay is suggesting that the public is owed?

So as not to present a "flash-card" version of Esmay, here's an extended quote from his post:

I'm amused that some people don't like the idea that the President might actually think for himself or question scientific authority. One wonders how unhappy they'd about that if they themselves were dissenters in some area of scientific authority -- say, if they had dissented from the eugenics movement of the early 20th century?

Besides, anyone who claims that science should be immune to politics is peddling rank nonsense. For starters, if you know any real working scientists you know that science is often fraught with petty bickering and squabbles, ego clashes, and ruthless competition for grant money -- even sabotage and character assassination. In short, science is riddled with politics.

Second, and probably more important: because so much science these days is funded by the U.S. government (i.e. the taxpayers) it is outright obscene to suggest that scientists shouldn't answer to our elected leaders. You do not have a right to demand billions of dollars from U.S. taxpayers, then slap a label on your chest and say, "We are scientists! You are not allowed to question us! Just give us your money and accept whatever we tell you!"

If you don't want the dirty, dirty politics in your science, then stop taking the dirty, dirty taxpayer money. Until then, it is entirely appropriate to note that an awful lot of scientists who advocate various theories -- including the theory that humans are causing catastrophic global climate change -- have a perverse incentive: they're being paid to look for evidence that their theories are correct. They're not getting paid to fail to find any such evidence. In the science of economics, this is what is known as a "perverse incentive," and it's hardly inappropriate to note when this happens, and to ask hard questions about it.

So while scientists are still accepting their paychecks from government-funded agencies, or from universities that depend heavily on government funding, and doing studies based on fat grants handed out by the government, let's have a little less hubris, hmm? We didn't elect you. We did elect the President -- for, even if we voted against him, we still had our say in the process.

Our elected officials have not just a right, but a duty to be skeptical on our behalf when dealing with the people whose livelihood depends on our good graces.

(I've added the bold emphasis.)

Let me set out my answer to the large question of what publicly-funded scientists do owe the public. Then, I'll consider how this squares with what I think Esmay is saying in the post quoted above. If I'm not being fair in my interpretation. I trust someone will set me straight.

The public pays for a good bit of scientific research. Why? In part, this is research aimed at solving particular problems that the public has an interest in solving (e.g., developing energy sources that don't require that we import lots of oil, finding safe and effective vaccines to nasty diseases, etc.). Other pieces of publicly-funded research may not be as sharply focused on solving a specific problem, but they may be aimed at answering questions that the public would like answered (e.g., how much longer until climate change hits a tipping point if we keep doing what we're doing? how old are those fossilized noodles?) The public even funds some "basic" research, whose only projected payoff is more knowledge -- not necessarily anything we'll find a practical application for, just a better understanding of some piece of our world.

So, the public funds scientific research that is aimed at making more knowledge, some of it directly applicable to solving problems (or so we hope), some of it just adding to our knowledge base.

What obligations do the scientists supported by the public's money have? Certainly, they have an obligation to do the research that was funded. If you get a grant to study bird flu transmission, you can't use that grant to study cold fusion (nor, say, to spend six months in the lab playing Tetris). Moreover, these scientists have an obligation to do good science with the public's money. They need to design the best experiments they can, and execute them as cleanly as possible. Making up results is right out, as is conducting the research in such a way that people are harmed (whether as participants in the research or as "collateral damage" from improper disposal of toxic waste, etc.). Finally, once the results are in, the scientists have a duty to present these results accurately -- here's what we found, here's what we think it means, and here's our level of confidence in the result.

In other words, the scientists are obligated to do the research the public funded, doing it in such a way that the public is not harmed, and doing it in such a way that high scientific standards are maintained.

Let's talk "deliverables". The public funds research to solve problem X. Does that mean, at the conclusion of the research, scientists need to have found the solution to problem X? Requiring this presupposes that, given serious effort, good scientists doing good scientific work are guaranteed to arrive at the particular piece of knowledge that solves the problem. Sadly, this universe offers no such guarantees. Scientists can solve a lot of problems, but some are tricky. Usually, scientists can turn up further information about the problem they're trying to solve. Sometimes, they find information that would let us solve the problem, but only if we were willing to change some other parameter that, left to our own devices, we'd rather not change.

At the end of a research project, scientists always have new knowledge that bears on the question they were investigating. But, that new knowledge may not be sufficient to solve the problem the public was interested in solving when the research was funded. And, the knowledge the scientists find may not be what we hoped for or expected.

In this way, funding science is more like working with a realtor than a baker. You can tell the baker what kind of cake you want for your money, and you are within your rights to reject a cake that doesn't conform to your expectations. On the other hand, you can tell the realtor what you're looking for given the money you have to spend (this many bedrooms, this many square feet, in this neighborhood), and the realtor can do everything possible to find something that meets these requirements -- but it may turn out that there's no house that meets your requirements. It doesn't mean the realtor didn't really look, just that your expectations didn't match the reality very well.

Now, let's see how the picture I'm advocating compares to Esmay's.

The President should think for himself and question scientific authority. What kind of "questioning of scientific authority" are we after here? Given the mention of eugenics, I take it this should be questioning of what to do with our scientific knowledge rather than questioning of what it is the scientists know. And, that seems fine. There may be the occasional head of state equipped to say, "Look at this result again -- you seemed to have lost a minus sign between pages 26 and 27," that's not part of the President's job description.

Esmay also points out that some of our scientists squabble with each other within the scientific community, and that the scientific community has internal politics. That's undeniable. Regular readers will know that I get cranky about the moments when scientists let personal gain and personal grudges get in the way of doing good science.

As for the "perverse incentive" of which Esmay speaks, it should be noted that part of doing good science is making a good-faith effort to be tough with your own theory. The choice is not, as Esmay suggests, between finding evidence for your theory or finding no evidence for your theory, but between finding evidence for your theory and finding evidence against your theory. This is Karl Popper's picture of the scientist as falsifier, or the permanent skeptical attitude Robert Merton thought the scientific community applied to knowledge claims. Since scientists are humans, operating in a community which is not free from politics or intrigue, it may well be hard for each individual scientists to be sufficiently skeptical of his or her own theory -- but the rest of the scientific community (skeptical and competitive bunch that they are) can pick up the slack here. This, by the way, is why looking for something like scientific consensus on an issue is meaningful: if this bunch of curmudgeons who are trying to knock each other's theories out of the water actually agree on something, there is probably reasonably good evidence behind it.

My big concern -- the point that is suggested by Esmay's post (although perhaps it is not a point he's really trying to make) is that non-scientists, and their elected representative, ought to question scientific research whose findings they disagree with. And I'm not talking about a disagreement about how to proceed with the knowledge, but a disagreement about the knowledge itself. Knowing that a particular species is going to go extinct if present conditions continue is quite different from concluding that we have to do anything about it. The public -- having many interests -- is certainly in a possition to argue about the latter, but they don't have an evidence-based reason to deny the former. And given the pains to which scientists go to create evidence-based knowledge, "skepticism" of this knowledge calls out for similarly evidence-based criticisms.

In other words, to be "skeptical" of a piece of scientific knowledge just because you don't find it congenial to your world view or your policy aims is to misunderstand the nature of scientific knowledge and scientific skepticism. And frankly, it seems like a waste of the public's money to challenge the research the pubic has funded in such a misinformed way.