A Persian Cafe, Edward Lord Weeks

Thursday 13 April 2017

Christiano's Renewed Defence of Democracy; or, Rule By The People For The People, Except Without The People

These are interesting times at CEU, but that did not prevent a public lecture by the famous political philosopher Thomas Christiano from going ahead. Christiano is perhaps the world's leading democratic theorist, having put more sustained thought and brainpower than anyone else alive into the defence of this ideal. His talk was specifically responding to a series of critiques made in recent years by Bryan Caplan, Jason Brennan, and Ilya Somin, all of whom argue that democracy is undermined by the poor quality of voters.

Christiano began by briefly setting out a key claim, difficult to dispute, that on a variety of metrics - economic growth, protection of human rights, avoidance of war - democratic nations have tended to enjoy greater success than alternative regimes. This is something that our social scientific theories ought to be able to explain.

He followed this by introducing Caplan's theory of the "rational irrationality" of voters. This theory emerged as a response to the "rational choice" theory of voting behaviour, which held that voters behave in their own interest - that is, voting for the parties and policies which stand to benefit them, as individuals, to the greatest extent. Caplan noted that this assumes voters already know which parties and policies best serve their interests, and pointed out that due to the vanishingly small chance that one's vote could ever change the outcome of a national election, the expected benefit of voting wisely could never exceed or even equal the costs of acquiring such knowledge. Indeed, from a rational choice standpoint, it is difficult to explain why one even takes the ten minutes to walk or drive to and from the polling station. So we have a morass of deeply uninformed voters, who are in no way suited to the task of choosing a government and its priorities. Caplan's argument is borne out by multiple surveys which find the average member of the public to be comically ignorant of fairly basic facts of day-to-day politics. If one cannot name the chancellor of the exchequer, what hope does one have when trying to assess complicated macroeconomic theories which do not even command agreement among experts?

Next, Christiano discussed Brennan's argument for how ideology corrupts our political judgements. Brennan provides three models of individual voters: "Hobbits", "Hooligans", and "Vulcans". Hobbits correspond largely to Caplan's picture of people with no clue of even the most trivial facts of politics; Hooligans are arguably worse, possessing some degree of political knowledge but also being highly ideological and more interested in ensuring the victory of their team than in seeking the best outcome overall. Vulcans, by comparison, consider all available evidence in a relatively impartial way, and contribute honest and valuable information to democratic decisions. A democracy of Vulcans might very well be a good system, Brennan says - but the world in which we live is one in which most people are hobbits, the overwhelming majority of the rest are hooligans, and Vulcans - if they even exist - are a microscopic minority. This is all backed up with social science to demonstrate that most people are as Brennan claims them to be. So while it might be nice to live in a world of Vulcans, Brennan says, that fact is that we do not - and our institutions ought to reflect this, in a way that universal democracy simply does not.

From here, Christiano said, Brennan, Caplan, and Somin - libertarians all - conclude that we ought to adopt alternative systems, with Brennan and Caplan suggesting the rule of experts and Somin suggesting a sharp reduction in the role the government is allowed to play in our lives. In particular, all three advocate a greater role for markets in decision-making.

I don't think this is really a fair characterisation of their positions. In fact, I would say it is an outright misrepresentation of Brennan's position. It is unfair on several counts:
-By "epistocracy" Brennan doesn't mean confining politics to the elites with no-one else able to break in. Rather, he has in mind tests of political knowledge, which one would be required to pass in order to vote.
-More fundamentally, Brennan does not actually advocate epistocracy! Rather, he suggests that it is a potentially-viable alternative to democracy, and that our institutions should not be built on the assumption that all citizens will behave as Vulcans. This is an understandable mistake, given Brennan's other writings; on the other hand, both he and Caplan are avowed anarcho-capitalists, so the only sense in which they can possibly be seen as supporting epistocracy is as an improvement over what we have rather than as an end goal.
-While they indeed believe that markets should play a greater role in our society, and believe (in line with the evidence showing that both social and economic liberalism correlate positively with both intelligence and with being politically informed) that the effects of a higher-quality voting population would be to give markets such a role, this is not (at least for Brennan) a core claim. The argument is that better voters would give a better set of political institutions, without any claims about what those institutions would necessarily be except as illustrations of how our current institutions are ludicrously sub-optimal.

Christiano then boils the debate to the following argument:

(1) Voters are subject to rational irrationality, ideology, and other such biases.
(2) If voters are subject to rational irrationality, ideology, and other such biases, then democracy will fail to work well.
(3) Democracy will not work well.

This, then, is the anti-democratic theorists' view in its simplest form, as a simple modus ponens. But, as he said at the beginning, the conclusion is false! Therefore, since the argument is valid, we know that something must be awry with at least one of the premises. Christiano suggests that Brennan et al overplay the evidence for (1), but does not wish to challenge it too much. The problem, he suggests, is therefore with the assumption that these various biases preclude voters from making good choices about who to vote for.

How can this be? To point towards a solution, Christiano attempts to turn his opponents own arguments against their views, by suggesting that the same problems which they attribute to democratic choice apply in the same way to ordinary decisions made within markets. There is then a dilemma for the anti-democratic theorists: either they admit that markets are just as flawed and so democracy may nevertheless be the best system we can get, or we identify some mechanisms by which individual ignorance can be translated into rational decisions.

There is undoubtedly some small truth to this. I have no idea how to repair a car, but this lack of knowledge on my part does not prevent me from hiring a mechanic - that is to say, from outsourcing the relevant expertise. I do not have the time to form opinions on an especially wide range of books, but I can outsource this to people whose comparative advantage lies in quickly reading and accurately assessing the merits of books.

So, Christiano suggests, such sources of information exist for politics. Moreover, they are often available at little or no cost, and include the following:
-newspapers and television
-political parties.
-friends and colleagues.
-many educated people need to understand political events for their work, and so understanding it for voting purposes comes at no marginal cost
-labour unions
-churches

One worry he admitted to this is that these institutions for informing people need "warning lights" for when they are failing to accurately transmit information. When one goes to a mechanic, it is usually quite clear whether the mechanic genuinely has their claimed skills, due to the success condition in which your car starts moving again (or passes its MOT, or whatever). It is not clear exactly what these are intended to be with regard to politics - The Guardian criticises Theresa May but as a left-wing paper they would say that, wouldn't they? And if one takes the criticism seriously, then without becoming something of an expert oneself, how can one establish whether or not the criticism is accurate?

One possibility, which I'm reading into him though not, I think, unreasonably, is for there to be legal requirements of neutrality or truthfulness applying to political broadcasters, as exist in the UK and Canada but not, infamously, the USA. The big worry with this, as Christiano notes, is that in principle democracy is rule by the people made on their own terms. Is it not contrary to this spirit to compel certain terms of discourse upon them?

OK, so that's Christiano's perspective, presented in what I think is a fairly reasonable and sympathetic way. I have a fair few criticisms, and will work up form smallest to largest.

First, his admittedly-only-a-hypothesis about the role of unions seems highly dubious. He suggested that the decline of unions made working-class populations vulnerable to demagoguery and so is responsible for the current malaise of "the US, the UK and France." But this just seems empirically ridiculous: first, demagoguery was just as potent a force in the days when union bosses would trip into Number Ten for beer and sandwiches. Second, Thatcher gutted the unions in the 80s: why did it then take more than thirty years for demagogues to come along for the working class vote? And finally, France suffering from not enough unions? Are you joking, or are you merely unaware that their transport systems are routinely shut down by disgruntled farmers, taxi-drivers, or whoever else is the angry industry of the day?

Second, I think Christiano overestimates the extent to which even intellectually demanding jobs require one to know about politics, and the extent to which such knowledge represents a very thin and impoverished of the infinitely complex reality. As an example: my dad works in estate and property management for the University of Birmingham, and had a great grievance with the EU that whenever he wished to outsource some work, anti-corruption legislation originating in Brussels required him - as an employee of an organisation recieving significant EU funding - to put it out to tender (including, of course, an expensive advert in a Brussels-based EU-approved journal) and placed certain restrictions upon who he could hire. As a result of this, such outsourcing decisions became vastly slower and vastly more expensive, since in the absence of such regulations he would simply have called up a handful of small local firms, asked for quotes, and gone with the cheapest who he thought could actually deliver at the price they gave. (Of course, the regulations require that he hire the cheapest firm, with the result than from time to time they will not manage to keep to the agreed price, and it is rare that this situation does not end up costing the university further money).

My dad has a deep knowledge of one particular aspect of the way the EU affects Britain. Does this equate to a knowledge, or even a reasonable idea, of what the EU is like as a whole? Of course not. (Incidentally, my dad was turned off by xenophobic messaging of the Leave campaign during the last few days before the referendum, and ended up abstaining; since the referendum, he has been quite enthusiastic about its result).

Third, and moving on to more serious criticisms: Christiano appears to go straight from the uncontroversial claim that democracy correlates with various desirable outcomes to the highly dubious claim that democracy works well, i.e. that it is causally responsible for these outcomes. I've seen a plausible case that democracies can enjoy lower borrowing costs, but otherwise this seems entirely to get the causal direction the wrong way round: countries liberalise economically, this creates a growing middle class, and so a demand for democracy. The economic success of the Asian Tigers is not to be explained in terms of their (anaemic) democracy but in terms of their liberal economic institutions. (And before one tries to argue that they have failed to respect human rights, (1) be careful you're not assuming your conclusion by taking democracy to be a human right, and (2) economic growth is highly underrated as a means to securing people's vital rights to food and shelter).

Fourth, Christiano seems to me to ignore, in an utterly irresponsible way, the quality of information being received. To quote him almost word-for-word: "In the US, I find that I agree with the values of the Democrat party... and this means that they can act as a way to distill complicated information to me." If I had not heard this from his own mouth, I would have assumed anyone attributing these words to him to be creating a strawman. Yet on the grounds that they share his goal of helping the poor more explicitly than their opponents, he is apparently willing on precious little further authority to commit to controversial views on a wide range of topics - the optimal minimum wage, the optimal response to global warming, the optimal level of US involvement in the Middle East...

Fifth, and to the extent that it succeeds most damningly, how different is what Christiano proposes from that which he opposes? If all were to vote the party line, we would have an esoteric epistocracy in which the relevant measure of knowledge would be "Are you a party leader?" It will not be this extreme, of course, with hopefully a range of alternative media sources. But insofar as his vision of democracy is parties telling voters what to think, and the voters consequently choosing parties to implement their policies - why not cut out the middleman, and let the social elites get on with ruling the country untrammelled by the inconvenience of needing to face election? (This, I should note, is the criticism in which I have least confidence).

Perhaps an argument could be made that voters don't really need the knowledge that Brennan thinks they do. More information is not always good, after all.  Apart from this, there are plausible cases for democracy which do not rely at all upon claims about its ability to make decisions. But Christiano's case for this is thoroughly unconvincing, to the point where I was inclined to wonder if his claims to know little of day-to-day politics were not, in fact, just modesty.

No comments:

Post a Comment