This is a blog post about how stupid some article is. Those are generally kinda lame, so let me explain why I’m writing this. First of all, the stupid article in question is The open society and its enemies revisited, by Karl Popper. It’s pretty popular within the community of critical rationalist, where it is sometimes used to argue for real world policy—see for instance this tweet. A friend of mine is a member of this community and he asked me to explain why I didn’t like it. Apart from that, it’s also a good opportunity to make a point that I find important: political science is real and genuinely useful if you want to make sense of the political world around us. Discarding the work that political scientist have done in favour of philosophizing, which I use in the worst way possible, is bad and results in stupid opinions. We will see this in a minute.
First, let’s take a look at what Sir Karl is actually saying. This is pretty hard, because he does a couple of things simultaneously and mixes all kinds of arguments. The boundary between a theory and a definition of democracy is often very blurry. To make it even more complicated, the boundary between a theory of actual, real-live democracies and democracies as how Popper would like to see them is also blurred. It is disappointing to see a famous philosopher make these descriptive/prescriptive errors and even more sad to see them gobbled up without question.
Anyway, the classical idea of a democracy, according to the big P, is that the people rule. It answers the who-should-rule-question by an unequivocal “the people”. The people give a government the legitimacy to rule over the people, in the same way that God makes the pope a legitimate ruler of the catholic church. To focus on legitimacy can lead to wrong outcomes, because what if the people choose to be ruled by a tyrant? Let me stress the point from the previous graf, this does not invalidate “rule of the people” as a descriptive theory of government, or even as a theory of the British government. But it does happen that countries with perfectly fine democratic constitutions elect dictators, Venezuela being a recent example. It is true that were you to write a constitution with rule of the people as the only guiding principle you wouldn’t get the constitution of any country, but maybe that is because you need more ingredients than just one principle.
Big Poppa’ goes further than saying that popular rule is a simplistic theory of democracy: he proposes his theory of democracy as a replacement of the ‘classical’ one. Democracy is — -or should be? again, this is not always clear — -all about avoiding tyrants and dictatorships. The fundamental feature is a clear rule of law that allows for the dismissal of bad rulers without violence. The people, or more precisely: a majority, is to be the judge of a ruler and should be able to dismiss them. If you think about it, you can rephrase Popper’s idea of a democracy to be an answer to the question he wants to replace. Q: who should rule? A: someone who’s willing to go if a majority demands so.
This is a bit facetious because this answer is not motivated by concerns of legitimacy and so on, that is what is supposed to be replaced by this new definition/theory. That has not really happened, of course, we are still left with questions like : why is a majority of the populace the force behind removal of bad rulers? There are other options available: the catholic church has a way of preventing tyrants that works pretty well, you could imagine giving a supreme court the power to remove bad rulers, or maybe you instate term limits, or maybe only women over the age of 55 have the power to remove a bad ruler. If you want to prescribe a democracy, you still have to make a moral choice between these options. (If, of course, you’re only interested in describing modern democracies you have to look real evidence of which option they actually use — you can’t guess.)
Even if you let go of the annoying complications of the real world and concentrate on the prescriptive part, Karly P’s idea is still too narrow. Legitimacy does actually matter, as is illustrated by Chapman, Cleese et al., and not just on a theoretical level. A functioning government begs some sort justification of why some people have certain privileges and others don’t in order to feel just, or even to be acceptable. Taxation isn’t theft, because the government has been made legitimated by being democratic.
What’s going on in the first half of the article is that both ideas of democracy presented by Mr P. are incredibly narrow and capture only a bit of what a real democracy is and what it should be. If you were to continue down Popper’s path and think of other things that make up a sustainable democracy you might invent the trias politica and other democratic institutions. A modern, comprehensive look at what a democracy is and how it works is not simple. Democratic norms go beyond who’s in power and who can vote people out of power — see for instance this paper, or a discussion of it. The actual dynamics of voting are surprisingly messy and non-rational, and a good descriptive theory of democracy takes into account that political leaders can actually influence the opinions of voters and other real-world, what-is-actually-going-on stuff.
Popper uses the latter half of his article to bring his theory into practice and compares proportional representation to representation by way of a first-past-the-post system. Here, the article gets a lot easier to refute because it gets slightly less confusing and a lot more wrong.
Take for instance this claim:
“This argument collapses if the old theory is given up, so that we can look, more dispassionately, and perhaps without much prejudice, at the inescapable (and possibly unintended) practical consequences of proportional representation. And these are devastating.”
Now, I happen to live in a country that has a very proportional voting system and I have not experienced any devastation. The more specific claims about multi-party proportional parlements are just as easily refuted by looking at an actual country. The funniest one is the claim that: “In practice, then, a two-party system is likely to be more flexible than a multi-party system, contrary to first impressions.”, even though the Dutch system is one of the most flexible political systems in the world. Populist anti-immigrant parties popped up in the early 2000’s and had real influence on our government, way before they had a (disastrous) influence on the British government.
The reason why he gets so much wrong has nothing to do with sophisticated philosophical reasons: instead, it is easily seen that K.P.’s view of political parties is overly cynical and very simplistic. Add to that some confused reasoning and you get the above b.s.. (About the confused reasoning, some of it is basic logic: small parties do not have outsized influence in coalitions because there are other small parties, making the small party not absolutely necessary to govern. Other mistakes come about because he tends to use ‘logic’ to think about proportionally represented parlements while arguing from an empirical p.o.v. when it’s in favour of FPTP parlements.)
Let me take a break from dunking on the article and its stupid opinions for a moment. Look, I’m not really arguing for or against FPTP here and I do think that the notorious K.R.P. has a point when he says that a FPTP representative has a more direct relation to his constituents. That doesn’t really matter, though, if you surround a good point with lots of things that just don’t describe the real world.
I don’t want to blame Popper too much for this article. It is clear that the man can write a clear sentence, especially for a philosopher. He seems way out of his element with this stuff, and I have 30 more years of Dutch and British parliamentary history to look at, so it might have been hard to see how wrong his statements about parties are. But maybe don’t use his article to analyse actual politics, okay.
Some extra reading: for if you think the title is overly rude and not an The Economist-style pun, if you’re interested in the precise voting dynamics of the Dutch (proportionally represented) parliamentary election (Warning, pdf and in Dutch). The links in this article are also worth checking out.
