The Bactra Review: Occasional and eclectic book reviews by Cosma Shalizi   152

Democracy and Knowledge

Innovation and Learning in Classical Athens

by Josiah Ober

Princeton, New Jersey: Princeton University Press, 2008

Liberty was Born from Endless Meetings

This is an interesting look at how the democratic political institutions of classical Athens worked. Ober wants to refute the idea that participatory democracy is useless on any scale bigger than a kibbutz or a backwoods New England village — that it either becomes hopelessly inefficient, or the democratic forms lapse into mere formalities, with oligarchs ruling behind the scenes. He claims instead that the Athenian political institutions should be seen as devices for bringing together the dispersed relevant knowledge of the citizenry, and for diffusing knowledge and skills among the citizens, in ways that less democratic city-states could not match. "The history of Athenian popular government shows that making good use of dispersed knowledge is the original source of democracy's strength" (p. 2).

I will not go into details here — Ober has a an well-written online precis — but the core idea is this. A more authoritarian government might be able to decide what it wants to do faster, and execute its plans with less fuss, but how are the tiny set of rulers (who after all are no smarter than the average bear, though probably more vicious) to figure out what to do? The decisions of the demos draw on the knowledge and intelligence of more minds than any king or tyrant and his cronies. The decisions of the demos are also ones that persuade many minds of their desirability, rather than matching the whims and prejudices of one man or small clique. (To quote the words Thucydides puts in Pericles's mouth in the famous funeral oration, "instead of looking on discussion as a stumbling-block in the way of action, we think it an indispensable preliminary to any wise action at all.") Athens, of course, had its share of disastrous blunders, but so has every government ever, less democratic ancient states very much included. All in all, I find Ober's argument rather persuasive, though my prejudices incline me to want to think things like this are true. There were certainly places where I wish he had gone into more detail about how, exactly, Athenian institutions worked as mechanisms of collective cognition, but I suspect that our sources just don't let us be that much more specific.

Ober has clearly spent a lot of time reading economics and organization theory. In addition to doing unfortunate things to his prose style, this means he thinks of problems in terms of transaction costs, aligning incentives, and creating common knowledge (in the technical sense of game theory; more below). This in turn means that he tends to imagine the problem of figuring out what the polity should do as just one of bringing together existing pieces of dispersed information. Transaction costs and the like are important, and so is sharing what's already known. But, as some democratic theorists have pointed out, much of genuine problem-solving consists of coming up with new ideas and new information, and devising new courses of action. (I'm not thinking of changing the scheme of representation, just exploring new options within it.) Neoclassical economics, even its neo-institutional branch, tends to assume that the only obstacles to acting optimally are conflicting incentives, and ignorance of specific material facts, and wishes away the sheer difficulty of figuring out what the best or even an adequate course of action.

But for lots of problems, finding the optimum is intrinsically hard. I think this sort of decision-making is especially hard for one person or a small group to do, not just because they are throwing limited processing power against problems of high intrinsic computational complexity, but also because of certain foibles of human beings, like being attached to our ideas. One of the potential strengths of participatory democracy is bringing many minds to bear on improve existing ideas, as people who are not skilled enough to come up with good ideas on their own can still see how to make them better. (The funeral oration again: "we Athenians are able to judge at all events if we cannot originate". Karl Popper translated that sentence as "although only a few may originate a policy, we are all able to judge it" [The Open Society and Its Enemies, vol. I, ch. 10, p. 186 of the 5th edition], which is even more to the point, but as I have no Greek I'm not sure how much of that is Pericles and how much is Popper.) Indeed, one of the key points established by Scott Page's work is that a diversity of perspectives and heuristics is at least as important as expertise in group problem solving, and that a sufficiently diverse group of the relatively unskilled can out-perform even the best single expert, if the group can communicate and cooperate — which, in light of the Athenian fondness for using lotteries, is incredibly suggestive. Ober cites Scott's book, but doesn't really engage with it, and more generally, he doesn't really consider the possibility that the democratic process is itself a cognitive or computational device, one capable of generating new knowledge and options, not just picking from a fixed menu. I am not sure whether this blind-spot comes from thinking economically (really, economistically) and not computationally, or whether it comes from taking as one's guides Hayek and Cass Sunstein, rather than Dewey and Popper*; but a blind-spot it is.

A case where the problem of figuring out what to do is particularly difficult is that of designing new institutions, or modifying existing ones. While people might be motivated to come up with institutions which will advance their interests, selfish or shared, figuring out which institutions will do that is not exactly trivial. Ober claims that Athenian institutions were extremely well suited to promoting the survival and "collective flourishing" of Athens, since all its citizens shared a strong common interest in doing so. But Ober never really explains how the Athenians could figure out how to act on that interest. He gestures at a selectionist explanation, pointing out that it was not all that uncommon for a Greek city to be invaded and sacked, with massive pillaging, death, and enslavement as the consequences. This seems to me to be far too weak. Ober (pp. 80--82) estimates that, between 800 and 300 BC, somewhere between 12 and 33 percent of the Greek cities suffered this level of destruction. (The high figure comes from the most prominent and best-attested cities.) But even the high estimate is simply too little selection, too infrequently, to create very complex adaptations**.

Leaving aside that quantitative difficulty, this is gliding over the fact that people perceive themselves to have different interests — whenever there are gains from cooperation, there is a struggle over how to divide them — and conflicting interests shape institutions. This was surely true even among the free male citizens of classical Athens, something Ober systematically neglects, except for some remarks about how elites came to reconcile themselves to the democracy. (He treats the tyranny of the thirty as an embarrassing aberration.) Again: one of the virtues of democracy is that it provides a way other than violent conflict for making such bargains. Even the messy compromises needed to get majority support mean that many interests must be satisfied by policy, a tendency which only gets stronger as the set of those with a voice in democracy grows more diverse.

A second dubious legacy from economics is an attempt to show that Athenian institutions created "common knowledge" about what Athens had decided to do. This is a technical term in game theory; some fact is common knowledge if all players know it; all players know that all players know it; all players know that all players know that all players know it; and so on ad infinitum. This is a cute notion, which leads to some clever mathematics, and is often invoked by classical game theorists as part of the solution to coordination and cooperation problems. (It also — see previous link — has remarkably paradoxical consequences, e.g., rational agents engage in financial speculation, or other forms of gambling, only if the act of betting is not common knowledge.) But, as Herbert Gintis says (Game Theory Evolving, p. 14), common knowledge is not only "ridiculously implausible", it is not even needed to solve those problems! Ober actually quotes this passage from Gintis (on p. 191), but nonetheless uses the concept of common knowledge a lot. My impression is that most of his arguments would work fine in terms of mere shared expectations — which, unlike "common knowledge", might actually have existed.

Ober doesn't just want to say that Athens managed to work without the dubious benefit of oligarchy. He wants to say that it worked really well, better than alternatives, and that the reason it did so was the political, knowledge-aggregating institutions of the democracy. He's convincing that Athens was one of the most prominent, materially successful and influential city-states, perhaps the most prominent. (Did this really need showing?) And he convinces me that the democratic institutions did in fact work democratically. But he really has nothing that shows these were causally linked. The closest he comes is putting together some very rough indices of democratization and prominence, and arguing that democracy was not a lagging indicator of material flourishing. But these numbers are so rough (and the time-scale of institutional change so slow) that this doesn't even rule out democracy being relatively inefficient, a luxury the Athenians could afford because of their silver mines, their geographic position, and the self-reinforcing nature of trade patterns, Athens being the Dubai of Hellas. The only way I can think of supporting Ober's claims would be comparisons — classical Athens versus other classical Greek city-states, and versus Hellenistic Athens*** — which he does not make.

Let me be clear: Ober's book is a remarkable work of scholarship, showing deep knowledge of the ancient sources, modern classical scholarship, and recent work on institutional economics. It is also, transparently, motivated by a real affection for his subject, and for democratic ideals. There is a tension between these commitments, of course, since the Athenian democracy, like the rest of ancient Greek society, rested on an endless stream of injustices whose depth and magnitude are simply horrifying to contemplate. (A report on the human rights situation in Periclean Attica would give it higher marks than the modern United Arab Emirates on transparency and, for a privileged minority, civil liberties, but vastly worse marks than the UAE on torture, the status of women, human trafficking, and in general the rights of the large majority of the population.) Ober admits this, but wants to separate the positive values of Athenian democracy from the violence inherent in the Athenian social system. For all my criticisms, Ober persuades me that this separation can be made and is worth making; that we should learn from Athens. My hunch is that taking these lessons seriously would lead to radical conclusions****, but that is another matter for another time. In the meanwhile, I earnestly recommend Ober's book to everyone interested in democratic theory and collective cognition.

*: It is very odd that Ober has written a whole book about Athenian democracy, and the power of democracy in general, ending with the suggestion that Athens was one of the first instances of a society becoming "open", and owed its success to that democracy and social openness, which nowhere so much as mentions Popper's The Open Society and Its Enemies. Even if Ober thinks the resemblance is completely superficial, one would expect at least a "please don't confuse me with that famous amateur" footnote. And I think the resemblance is not superficial at all.

**: We can try to calculate how much information this intensity of selection can provide about institutions. (The following calculations are very rough, but should get the right orders of magnitude; cf. Rivoire and Leibler, 2010.) Take the high estimate of a 33 percent chance of destruction over 500 years. This works out to an annual risk of polis death of about 0.2 percent. This is not nothing, but it's not very much either. If we applied selection like this to a population starting with a standard Gaussian distribution, removing the lowest 0.2 percent of the population each year, by the end of 500 years we'd have increased the mean by about a standard deviation. Said another way, the entropy rate of selection is about 0.02 bits/year, so, accumulated over the whole of the archaic and classical periods, we get about 11 bits of information on which constellation of institutions works best. This is the maximum amount of information which could be extracted from the "signal" of city destruction (and yes, that does bring to mind the one good passage in Thomas Pynchon), and it would be reduced by correlations over time, etc.
I think that selection is actually vital to the development of institutions. It's just that the selection which matters is not the rare destruction (or duplication) of whole polities or societies, but rather the continual choice by individuals about whether to conform to institutionalized expectations about their behavior, or to do something else. Renan said that a nation is a daily plebiscite; in a real sense, this is true of every human institution. They survive if, and only if, they are continually reproduced in human practices. This is where selection gets its purchase on institutions.

***: Was Athens more prominent than less democratic cities with comparable natural and economic advantages? Presumably yes; how much of this excess prominence could be explained by chance and self-reinforcing mechanisms? (Cf.) Did Athens become less prominent, compared to other Greek cities, after they all fell under the domination of the Hellenistic dynasties? If so, this would be evidence that Athens owed its prominence to some (possibly endogenous) source other than the excellence of its democratic institutions. (Alternatively, it could have been brought to prominence by democracy, and then some other cause took over later, but then would need to say what it was, and explain why it wasn't working before.)

****: Ober makes some hopeful noises about turning firms or non-profit organizations into Athenian-style participatory democracies; apparently he has co-authored a whole business book about this, A Company of Citizens. The obvious problem, of course, is that the employees of a firm are not its citizens, because profits flow to, and ultimate legal control rests with, the owners, though in practice they usually delegate to a largely self-perpetuating management. To draw an Athenian analogy, employees are metics, foreigners (or descendants of foreigners) who are allowed to live in the city, work there, serve in the military, pay taxes, and generally tie their fortunes to it, but are not even second-class citizens. If a firm wants to fully use the knowledge and abilities of its workers, and so runs itself as a participatory democracy, owners and managers would have to actually give up control. I don't want to assert that the owners must be expropriated — perhaps they could live on in some ceremonial or residual role, like constitutional monarchs in a representative democracy — but clearly we're talking about something very different from actually existing capitalism.

(With many thanks to Henry Farrell: this draws on some of our unpublished joint work on the evolution of institutions, and on discussions of this book)

xviii + 342 pp., black and white diagrams and graphs, bibliography, index (useful)

Ancient History; Politics and Political Thought

Hardback, ISBN 978-0-691-13347-8, and paperback, ISBN 978-0-691-14624-9

17 December 2010; restored missing clause in footnote on selection, 24 December