Poli Sci is in Trouble–Wonkism Put It There. And It is a Good Thing, Too.

There is an interesting debate of sorts taking in the shadows of the election:  is political science a casualty of the election, or not?  Jason Blakely, writing in the Atlantic, says it is, and Dan Drezner, in the Washington Post, claims that it is not.

I suppose the title of the post gives away which direction I think the developments will take, but not really for the reasons that Blakely attributes.  In fact, precisely for the opposite reason.  There is no such thing as a “predictive science,” at least not the way Blakely describes it.  Every “prediction” is a dare.  The “good” outcome is where the theory, at least some of it, winds up being refuted, so that we have something to think about.  Science operates on the premise that every theory is incomplete and in need of improvement.  Each prediction where the best theories are shown to be wrong shows us where the improvements need to take place.  In other words, science is not good because it can help us predict the future, but because science provides a coherent explanation of how the world works.  “Wrong” predictions, especially from the best theories, are the best thing that can happen to science.  In other words, from a purely scientific perspective, the unexpected Trump win (but not entirely outside the realm of theoretical possibilities–see everything that I have written here for past half a year) is great for political “science,” the real science part that is.

But political science has been rarely appreciated for being a science, but as a raw material with which wonks can justify their policy positions.  The wonks at Vox might be decried for the combination of their ignorance of the real world and their skills at poring through Jstor, but the wonks do not pore through political science theories to find interesting experiments to explore the limits of their explanatory power, but to find support for their pet agendas.  The data collection and analyses are, as such, biased towards confirmation rather than the contradiction–exactly the opposite of what stats textbooks say that we should do–in part because it is frequently misleading.  (p=0.05 is not that there is better than 95% chance that our theory is true, but there is less than 5% chance that, if our assumptions are true, our theory is false and we still got the data we got.  P-values are problematic because “if our assumptions are true” part is rarely applicable–but that’s not usually what many people think when dealing with p-values.)  In other words, political science is not appreciated by those who do for being a “science,” and when the wonks lose credibility–which they surely will–so will their favorite toy, i.e. political science.  The problem is same as that of “evolution” and Richard Dawkins.  The real science of evolution is irrelevant for most people.  It’s complex and not exactly related to questions of morality or righteousness.  But for Dawkins, religion is amoral and unrighteousness and “evolution” is the hammer with which he can attack it, and for the religious, Dawkins is amoral and unrighteousness and his tool cannot be good.

The analogue to Dawkins problem goes further:  many poli sci folk themselves want to play wonks, more than they want to be “scientists.”  They have their opinions, beliefs, and agendas.  It is tempting for them to subordinate their research as means of putting their agenda forward.  This makes them and their craft mere opinions, in the eyes of those who do not share their views and lack the means to evaluate their argument on their scientific, technical merits–i.e. vast majority of the people.  Worse, they need to make their argument with conviction, in order to be trusted by the lay audience at all–so they might as well drink their own theories and become believers.  But once you believe, it’s a religion, not a science.  To repeat, the greatest service that Bill Nye, the “Science Guy,” performed for science in his debate with Ken Hamm, the creationist, was showing that he is not a shill for evolution–he stated the specific empirical evidence that, if unquestionably true, would make a creationist. This is how real science works–not believing in something and advocating for it as a wonk, but abstractly and detachedly evaluating evidence and the causal logic without emotional attachment to one side or the other.

As a science, Trump’s election is great for political science.  It exposed so many flaws in our thinking that we need to account for to develop better theories, and better theories are always a good thing.  BUT, we have become too invested in the theories we had, by selling it as a sort of gospel truth to many semi-enlightened consumers who did not care for the “science,” and this rotten part of the edifice is sure to collapse.  In a sense, this is something to be looked at with much relief:  they WERE rotten after all.  But it also got poli sci folk invited to fancy parties and gave them voice in corridors of power, and that will (and should) go away.


One thought on “Poli Sci is in Trouble–Wonkism Put It There. And It is a Good Thing, Too.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s