Towards a Real Social “Science”?

The always thoughtful Chris Arnade has another thought provoking “rant,” as he calls these thought trains.  To anyone who’d care, I  strongly recommend taking a close look at it.

I think the money quote is this:

Within the GOP the hyper-rationalism comes wrapped in faux libertarianism— which often views empathy as a fault to be overcome, a kink in the armour. With Democrats it comes wrapped in technocrats armed with big books, big math, and huge spreadsheets. People, to them, are data to be manipulated, or costs to be minimized.

Both are armed with a hubris that views humans as simple and knowable — and primary goal of policy to increase economic growth and efficiency. Together they have ceded moral and legal authority to business and their ruthless quest for profits. The assumption being, if most people have more material things we are all good, and empowering business is the most efficient way to get that.

This is the problem in social sciences that I’ve been fighting, it seems, a losing battle against for years, although my fight is, in a sense, quite different from that being waged by Arnade.  To me, the hubris that human behavior is, or rather, should be simple and knowable is not the essence of “science” but a horrible subversion thereof.  Physicists do not assume a frictionless universe, or Hook’s Law, or whatever other simplification of the physical universe as representation of the real world.  They are simplifications, models, crude approximations of the reality.  The real world is, of course, far more complex.  The point of simple models is to establish a baseline, try and see how much explanatory power can be obtained from a simple-minded set up, and see what still needs to be explained, then go back and think through things more about what needs to be added to the simple model.

This is not, sadly, how many social scientists do things.  All too often, they have simple models that they think is the reality itself.  If the reality contradicts the model, it’s the reality that is wrong, not the model.  The humility, awareness that the model is but a crude approximation that is designed to be the starting point and, when things get realistic enough, it will definitely be wrong, is not a common thing in social sciences.  Instead, the attitude is that, since the model says X, we must find ways to leverage this to take advantage.  This, in turn, feeds wonkism, the ideology held by those who like policy and care for the “science” only as long as it helps justify the policy that they like, and are interested in leveraging the science to help them get there.

For example, institutional design and its subspecies, “agenda control,” are particularly blatantly guilty of such short sightedness.  Agenda-control is, to put bluntly, rigging the institutions.  It means troublemakers are kept out and the rules manipulated that whoever that sets up the institution can maximize the probability that the results that they like will prevail, even when it is not very popular.  Or, in other words, this is how “the party decides.”  Of course, the story of 2016 is that, twice over, the party did not decide.  The Republicans were saddled with a strange candidate that they did not like, who, against all expectations, won the election, too.  The Democrats, too, were almost upstaged by an unexpected troublemaker in the person of Bernie Sanders.

Both Trump and Sanders ranted against “rigged” institutions, and quite frankly, they were fundamentally right. The institutions ARE rigged, not so much as to ensure that outsiders can never win, but enough that they would normally face crippling odds.  When those who are being played by the insiders have enough, things can turn ugly, with the incentive to break the rules of the game, informal or formal, rising high.  When the game is rigged by manipulating the rules in favor of the insiders, this is the only chance that the outsiders see that can even their odds.  This is the mood that Sanders and Trump rode on in 2016.  Contra the famous saying, the house must not “always” win.  It can win on “average,” but unless the gamblers see enough odds in their own favor, they will see no point in wasting their money away at something where they have no chance of winning.  The “social scientists” were not even being good scientists who, more than anything, are cognizant of the limits of their models.  They took their models too seriously and outsmarted themselves by wrapping themselves in short-term arrogance, that they can keep cheating the gamblers and that they would never notice, or resort to changing the terms of the game beyond the control of the casino.  When the angry gamblers are coming for the casino owners with the proverbial rope and shotgun, there is no institutional manipulation that can save them.  The only path to salvation they had should be taken before, by paying the gamblers off enough so that they are content, and this requires a deeper perspective.

In a sense, I share Arnade’s expectation:  that there is an age of “irrationality” coming.  What I fear from the Trump wave–not necessarily just Trump himself, but the general sentiment that undergirds his rise, is that they consider the shallow, short-term oriented, and quite frankly irrationally arrogant sense of “science” that prevails among too many social scientists, and far worse, among most wonks as the science itself and throw it all aside.  Trying to convince the folks who fancied themselves as knowing the “science” that they were worshipping at the cargo temple and that they need to repent and do a real science, with due humility, was difficult enough.  I don’t see the Trumpists (broadly defined) as engaging in deep thoughts about what makes humans tick.  In a sense, this is a bit of relief:  good riddance that a lot of bad science (or pseudoscience, even, in many cases) will be swept away.  Alas, though, that a lot of babies will be tossed away with the wastewater.

Advertisements

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