Here’s a piece of news that’s both disconcerting and inspiring. We already knew that electorates in democracies are heavily polarized (although we also know that public opinion polls tend to exaggerate the distance and divisions between political groups). Polarization leaves democratic outcomes at the mercy of the undecided and the often clueless middle, ready to be swayed by irrelevant and irrational considerations. However, polarization is not only bad for democracy, it’s also bad for the polarized: if the distance to another opinion is too large, you’ll never be persuaded, and people who are never persuaded are a sad kind of being.
The news is this: apparently, the polarization that we witness is only skin deep. People often express partisan and biased ideas not because they really believe them but as a means to signal affiliation, trustworthiness and identity. In other words: they are bullshitting. Alex Tabarrok wrote:
I am for betting because I am against bullshit. Bullshit is polluting our discourse and drowning the facts. A bet costs the bullshitter more than the non-bullshitter so the willingness to bet signals honest belief. A bet is a tax on bullshit …
A recent paper provides evidence. It’s well known that Democrats and Republicans give different answers to even basic factual questions when those questions are politically loaded (Did inflation fall under Reagan? Were WMDs found in Iraq? and so forth). But do the respondents really believe their answers or are they simply signalling their affiliations? In other words, are respondents bullshitting? In a new paper, Bullock, Gerber, Huber and Hill provide evidence that the respondents don’t actually believe what they say and the authors do so by making partisans pay for their beliefs. (source)
Those in the control group were asked basic factual questions about politics; those in the treatment group were asked the same questions but were entered into a raffle for an Amazon gift card wherein their chances depended on how many questions they got right.
In the control group, … [t]here are big partisan gaps in the accuracy of responses. … But when there was money on the line, the size of the gaps shrank by 55 percent. (source)
Hence, it seems possible that people are able, given the right conditions, to stop the bullshit and to persuade each other of certain matters of fact. Hence we should be able to reduce polarization and to improve the workings of democracy and of public discourse in general.
The big caveat here is “given the right conditions”. Those conditions are only rarely in place in actual democracies. Normally, when people vote they don’t get paid to vote (and neither do they have to pay to vote). In a certain sense, that’s a good thing: vote-buying is objectionable (as are the property conditions that used to decide who has a right to vote and that were in a sense a price imposed on voting). In another sense, however, it’s a pity that people don’t get paid to vote. The study above shows that if people get some money when their votes are not based on factual mistakes they will tend not to make those mistakes.
If we want democracy to be more than the expression and aggregation of bullshit, we may have to consider paying people to vote.
More posts in this series here.