Friday 10 February 2017

Against meritocracy as a political system

A friend outlined a proposal for keeping populism in check: he suggested that only the legislative branch ought to be elected, with the judicial and executive run by appointed experts. His idea is that these appointed experts would have the power to veto legislative action that they would deem potentially catastrophic.

In the wake of Brexit, Trump, Tsipras, Le Pen &c, I suppose many people have similar thoughts, so I decided to explain why I think this is a bad idea (also, it's much easier typing this on a keyboard vs debating on whatsapp).

First, I want to challenge the notion that experts operate differently to populists. My friend said policy should be made by adults who base their decisions on evidence, not emotion; I agree, but I do not think experts are much better than the average person at doing that:

a) Academics, elites and intellectuals have all too often been on the wrong side of history (women weren't admitted at Cambridge until 1869, and were only allowed to become full members of the university in 1948; Yale changed its admission policy in the 1920s to reduce the number of admitted Jews ; plenty prominent scientists have held racist views, such as James Watson; &c).

b) Scientists are just as prone to bias as anyone else. Only 6-8% of social sciences professors identify as conservatives, with Democrats outnumbering Republicans 8 to 1. This would be fine, except that researchers tend to let their political leanings influence their research:
  1. They mistake their values for objective truth (e.g. one research team found (and I am oversimplifying here a bit) that conservatives make more unethical decisions than non-conservatives, but whether these decisions were ethical or not were evaluated from a liberal perspective! In other words, research that says "conservatives are more unethical" was basically "conservatives are not liberal".)
  2. They focus on research that reinforces their narrative and avoid research that contradicts it. E.g. there was (apparently) for a long time a surprising lack of empirical research into the accuracy of stereotypes; and when research was finally performed (by conservative academics), it actually found that stereotype accuracy is one of the strongest effects in sociology. Similarly, there has been far more research in prejudice against right wing-targeted groups (e.g. ethnic minorities) than in prejudice against left wing-targeted groups (e.g. devoted Christians).
  3. Researchers are actually more prone to confirmation bias than non-academics, since it seems that confirmation bias is positively correlated with IQ.
  4. Academics are hostile towards those who hold different political views to the extent of discriminating against research that discredits their views. In one experiment, researchers produced two papers, identical in all respects except their conclusion: one showed that leftist activists were less mentally healthy than average, and the other showed the opposite. They then asked other psychologists to rate the two papers. Liberal psychologists rated the second version as more publishable; less liberal ones showed no such bias. Furthermore, most social psychologists explicitly say they would discriminate against conservative job candidates.

c) If science were purely evidence-based, the source of funding for a particular study shouldn't correlate with the study's result. Yet studies show that source of funding correlates with results that favour the funding body's interests. For example, one study found that 65% of pharma industry-sponsored studies produced results in favour of new drugs, vs 40% of independent studies. Wikipedia has an article on funding bias. Whole research fields have been influenced by sponsorship - such as the sugar industry managing to downplay the role of sugar in heart disease, blaming fat instead.

So, I think this all shows that scientists and academics are far from infallible: they can be prejudiced, biased and corrupt just like everyone else.

Second, it's a bit suspicious that people like the idea of meritocracy only when it leads to outcomes that agree with their own intuitions. For example, I am pretty sure that lots of people I know who dislike populists, and to whom this idea of meritocracy appeals, would still consider the Supreme Court to be in the wrong if they strike down Roe v Wade - even though the Supreme Court Justices are amongst the best legal scholars in the world.

Third, and most important, even if it were the case that experts were perfectly honest, scrupulous and unbiased, even if it were the case that people would respect expert judgement, it would still be a mistake to give absolute power to a small group of people. Maybe this group is good; what about their successors? The good thing about democracy is that it is mutable: if we make a mistake now, we can reverse it later. If you move to a system where a group of people can veto anything they don't like, the only way of changing it is violence and revolution.

Finally, I am constantly shocked at the degree of arrogance people who call for such systems are displaying. When I said this to my friend, he replied, "how am I being arrogant? I didn't say I should be the one to rule". True, but the whole call for meritocracy comes from believing that the crowds have got it objectively wrong; that some of us, the better educated ones, know better. I find this hard to stomach.

No comments:

Post a Comment