Sunday, July 27, 2014

Science vs. Politics: A pragmatic argument for why this distinction doesn't work


[X-posted on the HASTS blog.]

Recently, I talked to a doctor and public health professional about the relationship between science and policy; he told me, in a vivid metaphor, of how things work, and should work, in the regulatory process. The science produces the facts, which then get funneled through our values through the process of politics.  What comes out of this machine, he said, are policies.

It was quite a beguiling vision, but as an STS person, I couldn't help asking: did he really believe in it? Yes, he said.  I pressed on.  How, I asked, would he explain the controversy over global warming? Why was it difficult to implement policy when the scientists had a decent agreement over the facts?  His answer was that it was Fox News, fed by the big bad industry, which had fooled certain people into not believing the scientists.  I asked if it might not be more useful to wonder whether this disagreement over what to do about climate change (or about whether anthropogenic climate change even exists) might be an indication of something deeper: perhaps a reflection on the particular ways in which American society is now polarized rather than about Fox News brainwashing susceptible viewers.  He didn't think so, he said.  (He objected strenuously to my use of the word "brainwashing"; I took it back, but I maintain that it was an accurate descriptor of what he was saying.) I asked at the end what he thought should be done about all of this.  He said it was a long-term project; but it began with education; scientific literacy had to begin at a very early age.  Only then would people stop listening to Fox News. At that point, I gave up.


I admit that there is something really alluring about this picture of a science that produces facts which are then funneled through our values by the process of politics, all of which combines to produce rational public policy.  Even if we admit that this isn't really how it works in practice, perhaps this is how it should work.

But even holding on to this vision as a normative ideal may not be in our best interests.  As Sheila Jasanoff and Bryan Wynne have shown, this is because the process of science is shot through and through with values. Wynne suggests that scientific models to measure risk (e.g. risk analysis, cost-benefit calculations) often contain hidden assumptions and prescriptions: about what it means to be social and human, and what an ideal social order should be.  These visions of the human and the social are often found wanting by different publics.  E.g., the language of risk analysis comes coded with what a risk is or is not, and what things humans should worry about, points about which different publics disagreed but a) did not have the tools to express their disagreement, and b) were not taken seriously by experts and understood as only lacking an understanding of the science.  One of Jasanoff's suggestions is that rather than trying to cure science of its values, or create a politics that is based on "facts," we accept the value-riddenness of science and use that to think about how expert advice fits into the political process.  (Needless to say, I agree.)

All of which brings me the real reason why I'm writing this: this Scientific American blog-post which the worst combination, in my mind, of two overlapping tendencies: the plague-on-both-houses bipartisan strategy of journalism (something that journalist James Fallows calls "false equivalence"), and the dichotomous conception of "science" and "politics" as two mutually opposing entities.

The post details the ways in which the EPA's efforts to establish a new regulatory standard for drinking water, with an even smaller permitted amount of arsenic in it, were stymied by a Republican Congress.  The contours of the story itself will not surprise anyone.  Surveying some of the research that had been conducted, the EPA was on the verge of making official its stance that arsenic was a more dangerous carcinogen than it had originally thought.  This would be a prelude to a tougher drinking water standard. Naturally, this meant that corporations that produced arsenic or used arsenic in their products lobbied hard to make sure this didn't happen.  In these polarized days of American politics, it made sense to turn to the Republican party.  And the Republicans delivered by delaying the process.  Essentially, they got the the National Academy of Sciences (NAS) do an independent review.  Read the whole piece; it's detailed and precise to the point where it can exhaust the reader.

And here my problems begin.  Take the headline:
Politics Derail Science on Arsenic, Endangering Public Health
Why "Politics" and "Science"?  Why not say "Republicans Derail Science on Arsenic"?  Or even better and my personal preference: "Republicans derail EPA on Arsenic"?

Then take the leading line after the headline:
A ban on arsenic-containing pesticides was lifted after a lawmaker disrupted a scientific assessment by the EPA.
Again, why this coyness about the identity of this "lawmaker"?  Why not mention upfront that that this is a Republican congressman?  Why does it take until well into half-way into the article to identify the offending Congressman: Mike Simpson of Idaho?

Why, for example, is this sentence worded in this particular way when we know we're talking about the Bush White House?
The White House at that point had become a nemesis of EPA scientists, requiring them to clear their science through OMB starting in 2004.
The piece, for all its commendable whistle-blower reporting, contains the worst tendencies of what journalist James Fallows has called "false equivalence" in journalism, which is the plague-on-both-houses stance (see Fallows' copious collection of examples).  Essentially, newspaper reporting has a tendency to blame both political parties, or politics in the abstract, when things reach a bad state.  Here the newspaper is seen as above politics, which is what grubby politicians do. And therefore the contrast between the policy that the newspaper is advocating (which is not politics but merely good moral sensible stuff), and that what the politicians are doing, which is bad, i.e. politics.  E.g., the tendency to see the US Congress itself as dysfunctional, rather than the threats of the Republican Congressmen to filibuster pretty much any legislation.

The same forces are at work in the Scientific American piece.  Notice that the piece is not explicitly portrayed as a Republicans vs. the EPA piece but rather as a Politics vs. Science piece.  If I had to caricature it, the main point is: science good, politics bad.  The problem is that this often serves to paint politics itself as grubby and well, dishonest.  

This also leads to a manifest lack of curiosity about certain topics.  You might wonder why the makers of the arsenic-containing herbicide choose to work through the Republican Party and not the Democratic Party.  There's no way you could answer this question without looking at the broader trends in American politics over the last 50 years.  The two parties now occupy non-overlapping spaces on the political spectrum: the Democrats are a hodge-podge of interest groups: minorities, relatively affluent social liberals, unions, etc; the Republicans, on the other hand, have only two constituencies: evangelicals and big business.  Perhaps, 50 years ago, a business that wanted to fight a piece of regulation, would have had to think harder before deciding which political channel to use; today, it doesn't take more than a minute to decide what to do.

I understand that this is perhaps unfair criticism.  The piece is long enough, and talking about the realignment of American politics will only make it longer.  But that's exactly the point: if you black-box both science and politics, and paint the regulatory battle in question as a contest between them, then you don't need to think deeply about either.  Framing the article as the Republicans' battle against the EPA would have required the writers to ask the question of why these two actors are arrayed against each other. Editorial choices matter.


But the worst thing about the article is what's NOT even in it.  What, one would wonder, is a citizen to do after reading it?  The article doesn't say but I have an answer: call or write to your Congressman (especially the Republican ones but it doesn't really matter).  Tell him or her that you think the gutting of the EPA's power is something you don't agree with.  That you believe in a robust regulatory structure with teeth.  That perhaps you believe in a more take-precautions-first European style of regulation rather than a do-it-first-deal-with-consequences-later American style of regulation.  Why couldn't the SciAm article include a link for us to call or email our Congressmen?  Because that would have been too political, that's why.  And why should we bother with grubby politics when the science is in our favor?

One of the recent revelations for me has been how easy the Web can make it for us to call or email our legislators to inform them about our opinions on particular issues.  The techies did it really effectively with their "blackout" in protest of SOPA and PIPA.  Recently, in protest against the FCC's proposal to gut net neutrality, we were able to flood the FCC's comment-solicitation notice board with some good arguments for net neutrality.  At heart, this is just good old-fashioned politics, trying to convince our fellow-citizens about the rightness and wrongness of certain causes, sometimes celebrating victory, at other times, accepting defeat and vowing to fight another day [1].

Now I understand that explicitly political action might not be feasible for certain organizations responsible for the article, a collaboration between the Center for Public Integrity, and the Center for Investigative Reporting, both of which may have explicit prohibitions (because of their funding model, for e.g.) against participating explicitly in politics.  But that's part of what's got to change because that's the most important shortcoming of the science-vs.-politics narrative.  It precludes avenues of action for citizens.  What do you do? Trust science, which is what the SciAm investigative piece seems to suggest?  Despair that your representatives are morally and politically corrupt [2]?   Or do the hard work of politics and convince your fellow-citizens that they're better off having a robust EPA?  I vote for the latter.

---------------

Notes
[1] Certainly, citizens are starting to participate in science-politics in other ways, most importantly, through the practices of citizen science. Citizen science is perhaps the most interesting way of making science "impure." But making phone-calls to your legislators, voting, giving money to causes you deem fit, are also equally good ways of participating in the political process.

[2] And that, perhaps, explains why the show of our times is Netflix's House of Cards.  More on that another time.  

Sunday, July 13, 2014

Recent blog-posts around the Web

In the past few months, I've been blogging at multiple places and as a result, have completely neglected this blog.  In the future, when I post somewhere else, I will cross-post it here, or at least, post a link.  In the meantime, though, here are some of the posts I wrote recently:

For the CASTAC blog, a post on the history of artificial intelligence and the new field of machine learning.

Also, for the CASTAC blog, a revised post on the phenomenon called "data science" where I speculate that the proliferation of claims about "big data" is more about a crisis in professional identities (who has the expertise to work on particular problems: those with domain knowledge or those with data manipulation skills?) rather than an epistemological crisis (can we analyze phenomena without pre-existing theory?).

Finally, a post on the HASTS blog about how one might use the game of tennis as a way of understanding what the history of technology is all about.

I've also started posting interesting articles I see to my Tumblr.