Jan

24

 Daniel Cloud is the author of a masterful new book The Lily; Evolution, Play, and the Power of a Free Society. Here is a short piece he submitted for Daily Spec explaining in brief some of the main ideas contained in his book: 

You sometimes hear people say that things would be better if only America were more like China, because without all this democracy and freedom, they can really get things done over there, can really commit to solar power, or nuclear fission, or budgetary discipline, or whatever the person thinks we need more of. Are they right? Historically, absolutely not. Freedom works. People are always saying that kind of thing - Stalin is the future, Louis XIV is just the sort of powerful monarch we should have here in England, the Spartans aren't soft like we Athenians, etc., etc., on and on. In the last four centuries, however, there are very few cases of an illiberal society permanently defeating or outcompeting a liberal one. But why?

Conventional wisdom assumes that it's competition in the market for explicit, rational ideas and plans of action that gives liberal societies their advantage. We must be free because we always are in a position to know what should be done, and just need the liberty to do it. Watching democracy in action, however, soon reveals that many of the plans actually proposed seem to be useless or even counterproductive, that the system in aggregate displays intransitive, inconsistent preferences, and that the people who lead democracies often seem remarkably unimpressive. It's precisely these features that made many Athenians or Florentines doubt that a free society was really a viable option. In their times and places, they were, as it turned out, right. What is it that makes the modern free society, in the last four hundred years, so much more successful? To answer this question correctly, we have to step back a bit, and look at the problem from thirty thousand feet.

There are only two possible explanations for any system that seems to behave in a way that's somehow optimal or effective. Either that optimal behavior was rationally planned by someone, or else it evolved through trial and error and competition. If the amount and quality of explicit rational planning we see doesn't adequately explain the degree of effectiveness observed in the behavior we see, some process akin to natural selection is the only available explanation. Does the modern free society work better than the unfree one because it's somehow a better arena for the evolutionary optimization of some set of teachable practices, or whole institutions?

In fact, in a human society, we should be able to tell, by inspection, which sort of process is responsible for some particular instance of optimal behavior. Optimal behavior that's the result of rational planning should be based on "knowledge" in the conventional sense of the word, true beliefs that come with some justification, or proof, that include an account of how the belief was arrived at and why it should be presumed to be true. They should be persuasive. On the other hand, highly effective behavior that is the result of some social analog of natural selection should be based on beliefs, probably true but possibly even false ones, or even mere dispositions to behave a certain way, for which the believer can provide no plausible justification, no warrant, that don't come wrapped in any convenient logos, but which nevertheless just happen to be exactly the right thing for the person to believe, from a practical point of view. They should be unpersuasive, at least without the help of a lot of deliberate clarification and anthem-writing, because the person didn't get the belief by being persuaded of it in debate, he got it as a result of it having worked out well, in practice, for him and the people he learned it from.

But this is simply a paraphrase of Plato's description of civic virtue, from Meno, as "mere true belief". The really virtuous citizen, Socrates informs us, often seems to know exactly what he must do, though he generally couldn't quite tell you why, or make his beliefs convincing in debate, which is very puzzling. Among economists, it's a long-standing folk-mystery (which never quite makes it into their formal professional discourse) that firms and households behave in ways that appear undeniably optimal, and yet if you go talk to the people involved, they couldn't explain why that way of doing things is optimal in a million years, and have all sorts of surprising and implausible explanations for their own behavior. There is actually quite a lot of this sort of evidence of a long history of social or cultural evolution, once you know to look for it, quite a few common-sense beliefs or attitudes, even within particular professions, that are probably very useful but not obviously justified.

The reason this all strikes us as paradoxical is that we've collectively failed to make a crucial distinction. Knowledge, in general, comes in two very different flavors, declarative knowledge (knowing that Neil Armstrong was the first man on the moon) and performative knowledge (knowing how to throw a baseball, write a contract, trade bonds, or solve a topology problem.) It's particularly easy to imagine some analog of natural selection happening to privately owned firms (conceived of as balls of money with people attached to them by contracts, which are fit or unfit depending on whether they have baby money the people associated with them can make new firms with.) The sort of knowledge this evolutionary process seems to produce is not, or not exclusively, declarative beliefs about facts that come with persuasive justifications. What the firm needs to prosper and grow is performative knowledge, knowledge of how to get things done, and whatever declarative knowledge is needed to support that. A lot of the "institutional culture", at any given institution, always consists of that sort of thing.

Skills and institutional cultures seem like the sorts of things that could evolve even as our public accounts of them don't. Modernity is, above all, rapid technological change, and perhaps the only efficient way of coping with rapid technological change and the radical Knightian uncertainty it continually creates is by creating a freely co-evolving population of firms and individual skill-sets, a system that isn't rigged in anyone's favor by people foolish enough to think they know what's going to happen next.

Why does any of this matter, who cares precisely how freedom produces optimality, if it does? To understand the difference between rational choice and natural or social selection, as mechanisms, it's useful to think about the difference between a computational simulation of airflows around some airplane design, and the tests we can perform, on the same design, using a model in a wind tunnel.

Simulation is cheap, and easy, and we can change anything we like. The problem with it is that its power is limited by the complexity of the problem we need to solve. If the problem gets very complicated, simulation becomes impossible, because you would have to write too many lines of code. (Often, when you run into a really bad simulation problem, you find that the lifetime of the universe wouldn't be long enough to write it all.) On the other hand, the wind tunnel is expensive, and wasteful, and cumbersome - but it just doesn't care at all how complicated the problem is, it isn't a thought or a simulation, it's part of the real natural universe, so it spits out a correct answer without any delay or hesitation, no matter what. We still don't know why that's the answer, but we can be sure that it is. The wind behaves just exactly the way it would behave, as it went around the model, as it goes around the model, because the real world is actually just like itself in every possible way. The wind tunnel is, effectively, what computer scientists call an 'oracle' for solving what philosophers call 'decision problems' (does the model work as expected, or not?) in no time at all.

 Natural selection is the same sort of thing as the wind tunnel, two vines or two prides of lions or two corporations in a real, un-simulated struggle to the death, and it too, is utterly indifferent to the complexity of the problems it is asked to solve. A contest between two complex modern states and a contest between two relatively simple bacteria or two saplings in a clearing can be resolved in the same amount of time, by the Judge of Battles, with exactly the same amount of work. (None.)

So, in general, there are these two very different sorts of optimization process operating in nature. One of them happens in brains, is cheap, is fast, and can conserve solutions to problems that only come up occasionally or locally. The problem with it is that it's limited in power, and gets less and less useful as the problem that needs dealing with gets more and more complicated. The other sort of optimization process doesn't only or primarily happen in brains - it also optimizes flu viruses, and falcons. It's expensive, it's slow, it's wasteful, and it only can conserve solutions to problems that come up repeatedly - but it isn't limited in power, in anything like the same way, it just doesn't care how complicated the problem it's been asked to solve is.

If it's the second kind of optimization process that is responsible for some of the optimality we see in human societies, as it is for all of the optimality observed in human cells (which we also don't fully understand, even though they're much simpler than a whole society of humans each made of trillions of cells) then there's nothing mysterious at all about the fact that societies built around free and fair arenas of limited evolutionary struggle should outperform ones built around attempts to substitute human judgment for this more wasteful but far more powerful mechanism. Perhaps people are, rationally, only able to accomplish just about as much as the economists themselves can - solving static, equilibrium optimization problems - and everything else only gets sorted out correctly if it's left for Nature to decide.

In planned economies, static problems must routinely get solved in ways that only make dynamic ones worse, and there's no obvious Darwinian corrective mechanism to put things back on track. (No real analogs of bankruptcy, or electoral defeat.) To the extent that the unfree society must rely on punishments to elicit the same sort of effort people would put in spontaneously if they thought of themselves as owners, it also suppresses the sort of variation in behavior any such process of social evolution would need as raw material. Nobody wants to be shot for trying some new way of doing things, some playful modification of an existing skill-set, or institutional culture, that doesn't end up working. Stalin was very successful at eradicating that sort of boldness. The problem is that unless people actually are constantly trying out exactly that sort of thing, in large and small ways, there's no source of variation in the population of skill-sets, and no way at all for the society to spontaneously percolate up to the highest point in its adaptive landscape. Thus you end up with the sort of enormous gap in even simple human skills, like the skill of cooking edible food for large numbers of people, that existed between the Soviet Union and the United States during the cold war, and that still must divide Korea today. (Nobody remembers Soviet cuisine now, the Kvass machines, the gristly, horrible "kutlet" smeared with some poisonous orange sauce, which only the really lucky people got…)

Narcissism trumps experience whenever we imagine that we can solve the sorts problems markets and elections exist to solve, because the reason we actually have markets and elections in the first place is that some of the problems we need to solve, in a modern society, are ones not soluble within the cognitive limitations that afflict us as individual humans. There's a kind of observation bias that constantly tempts us to make this mistake; we can see our own thoughts clearly, but our own customs are mostly invisible to us, so we tend to attribute to our cleverness whatever benefits we get from them. Glibness will only help us make fools of ourselves in these cases, because the mere truths that are most important to know and remember are precisely the ones that aren't readily explicable, or immediately persuasive in debate. In fact, what we all should have learned, from the great Communist experiment, is that there's really nothing worse for people than trying to live in the way that seems most reasonable to them on first hearing it described. That, actually, tends to end very badly, that tends to end with you standing in an endless line for a small piece of rotten meat, and very careful of what you say to the other people standing in line with you. (That's if you're lucky; the really bad outcomes are much worse.)

We need to be free, among other reasons, because we need to accomplish things, to have a cutting-edge modern society at all, that exceed our innate capabilities, in ways that defy our expectations. We do that by letting our institutions and skills and ways of speaking evolve freely, and building our whole society around the sort of fairness and respect for individual autonomy that's needed to make that possible. The leaders don't have to be impressive, for the system to work better than an unfree one, because the people at lower levels are, they're very skilled in an amazingly vast number of different skills, and that's what's crucial, that's what really drives the outperformance, the wild variety and vast depth of constantly evolving skills and institutions.

 It's a testament to human intelligence that we were capable of creating and managing a system that can do things impossible for human intelligence. Attempting to operate the system manually is, in fact, not advised; it routinely results in catastrophe, and in principle it shouldn't be possible. Sometimes you can get away with it for a few decades, when conditions are extremely favorable, for example when you start from a very low level of economic development with a very high level of literacy, but it isn't a good place you're ultimately heading towards, even then.

China only seems like an attractive alternative if you don't really know what's going on there; if you do, you know that what they're building isn't a real thing, it's a mere prelude, a temporary fantasy about beating the free world with one hand tied behind their backs, though that's been tried many times and really never succeeds in the long run. The problem is that the oracle of selection is necessarily cryptic, otherwise it would be redundant, so it requires some resolve to really submit ourselves completely to its judgments, and if you just don't have the right anthems, yet, that's difficult to do. (Even Deng Xiaoping couldn't quite make himself believe that the West is where it is because it's what it is.) In fact, we don't need to be more like them - they need to be even more like us, though everyone is now too polite, or too intimidated, to remind them of that. Not only does freedom work, but to sustain a really competitive form of modernity over the long run, nothing else will.

(Readers who found this interesting might also be interested in the more complete version of the argument contained in the author's new book, The Lily; Evolution, Play, and the Power of a Free Society, available from Amazon.)


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search