Groupthink vs whips

I started my blogging career with what I still consider one of my best posts, where I said that human history is very easily explained if you take into account the fact that (most)  humans are just plain dumb. Learning is hard, really hard. And it should, animals don't learn if they can avoid to. It takes domestication and industrial amounts of drilling to make an animal learn some behavior. It follows that it takes domestication and quite large amounts of drilling to make people learn some behavior.

While this sinking ship called neoreaction is, if only etymologically, an anti-modernist, declinist crowd, the very fact that I'm here writing a blog instead of just copying and pasting quotes from old books, means I regard myself as having some new insight that wasn't available to my forebearers. As I said earlier, cognitive science is full of true and powerful insights on how people think and why they do so. We now know not only of cognitive "biases", often constructed as surmountable errors, but the very architecture of cognition. Reading through Jonathan Haidt's The Righteous Mind should be a good replacement for years and years of degrees on ethics and moral philosophy, which I now visualize as an industrial sweatshop of rationalization hamsters run by an evil medieval bishop.

While I very much admire the aesthetic sense of traditional societies, the fact is pre-moderns weren't strangers to evil, crime, and sheer ideological nonsense. While comparisons are often made on the degree of evil and nonsense of past and present, I think a reasonable position is to say that it's pretty much the same. And that is necessarily must be the same. People are full of shit, and leaving aside technological differences which may amplify or expand it, the underlying process is the same. And it must be the same, as cognitive science teach us that brains work like that they do, and there's nothing we can do about it. Back to Haidt, you gotta give it to the guy that he proves in himself the truth of his book. He writes how people's morality is based in fact-free, a priori intuitions reached by emotion and sloppy thinking, and reason is a language based coating applied after the fact to rationalize one's decision with one's peers. He makes the point here in this EconTalk podcast. But all this awareness of how moral intuitions work doesn't stop him from saying out loud how he agrees with Paul Krugman's economics even though he knows nothing of the subject! I just couldn't believe he's so lacking of self-awareness. Perhaps he's just advertising his cognitive dissonance as promotion for his book. Think my thinking is fucked up? Buy my book and you'll know why!

Anyway it seems clear that the default mode of thinking for humans is groupthink, and the content of that groupthink is computed with a complex algorithm taking into account one's peer group, the loyalties one owes to whom, who has more status and what utterances fit better with the religion that one has been painstakingly drilled into since infancy. Jim Donald linked to an eerie example of unified TV propaganda here, although that was engineered. Uncouth Reflections links to this piece on the global warming inquisition, and this looks spontaneous. Plain horizontal transmission. Why bother thinking when you can parrot the line? Thomas Schelling of all people has enough brains to think for his own. But he won't.

Now we could read that and lament how far down we have come, how science has been corrupted, people are more evil than they used to, independent thinking has disappeared, etc. But it wasn't that much better in the past. Most big ideological debates in history can hardly be described as being fair and thoughtful. Groupthink and witch-hunts on spurious grounds are hardly new. You may say that mass media means that the scale of groupthink has changed; people used to parrot whatever was popular in the local parish; now they parrot the central propaganda organs of the government. But that's been known for almost 100 years. The Rebellion of the Masses touches on the subject, and Goebbels made an art of propaganda almost 80 years ago. Nihil sub sole novum. The content might change, but the underlying mechanism hasn't changed. It can't change.

So what do we do about it? My previous post argued that groupthink might be universal, but the degree of what Handle eloquently termed "conversational homogeneity" is culture-specific. Groupthink is a coordination mechanism, but it's not the only one. You can also coordinate people from above by using a long whip with spikes on it. People can also coordinate because of loyalty, carefully cultivated over years or generations, producing a sense of duty that doesn't necessarily involve parroting the same opinions as anyone else. Both of these alternatives seemed to have a common defect; they scale badly. Whips with spikes (or machine guns) can only control so many people, and there's the issue of quis custodiet ipsos custodies. Loyalty production is also a very slow and intimate process, which can be accelerated by technology or specific triggers (the threat external aggression mostly), but is generally very hard to sustain.

Groupthink on the other hand might be more efficient as a coordination mechanism. It seems to be part of the fundamental mechanism of the brain that deals with socialization, which is a big part of being human. One of the great triumphs of recent science has been the gradual overturn of Chomsky's theory of language. Chomsky famously wrote that children can't possibly learn language and its grammar from the little input they hear or eavesdrop from their surroundings. Ergo, there must be a module on the brain that specifically deals with language. He then stretched the theory, saying that if there's a specific language module in the brain, i.e. in all human brains, that module works the same for everyone, so there's a single universal grammar in all human brains, and all actually existing human languages are just a surface representation of that. Rivers and oceans of ink ensued in order to devise models for translating all human grammars into that mystical universal grammar we all share.

Cognitive science has been slowly proving that there is no particular language module in the human brain. Instead, language is processed in several different areas around the brain, located in mostly the same areas in most people but not necessarily. The corollary is that there is no separate language module which behaves differently from the other modules; but that brains only have one single way of functioning, learning in all different cognitive areas is all based on the same learning processes, and the structure of language just mirrors the way brains work in general. Meaning people actually think of time as if it were an object moving towards oneself (the time comes, then is past), and categories are remembered as prototypes full of detail (say, cats have 4 legs and meow, or black people are big and dangerous), from which certain specimens might differ in some characteristics without contradicting the category itself (a mute cat with the 4 legs cut off is still a cat, a mellow black midget is still a black person).

Coordination mechanisms must also work the same way. Pain avoidance makes you obey the guy with the whip; friend-foe identification makes you gang up with your tribe. And basic learning processes produce the "beemind". Everybody's conservative about what they know best, but what about everything else? Then you just use general, off-the-shelf learning processes, i.e. drilling. Drilling needn't be always forcible; hear the same thing a thousand times, and you'll assimilate it without even paying attention. The general consensus has a way of making itself known. Japanese people call it kuuki, i.e. air. You're supposed to "read" the air, and avoid saying or doing anything which doesn't fit the groupthink of the moment. A related, if wider concept is the zeitgeist, which is just a description of the general patterns of groupthink of a society.

Asking people to use rational inquiry to make up their own values is not only implausible given the cognitive ability of most people; it's plain impossible because human brains just don't work like that. People's opinions aren't "found" through the use of reason, and they aren't purely a function of hereditary disposition either. Values are picked up from society, and different societies allow a bigger or smaller Overton window, and within it more or less heterogeneity of opinion, depending on their particular mix of coordination mechanisms. Then again the preferred coordination mechanisms might be genetically determined, but I'm skeptical. The Chinese used to be quite a tolerant bunch. Then Mao happened.

Common sense would lead to an Aristotelian arrangement, where a society uses a balanced mix of all coordination mechanisms, whips, loyalty and groupthink, so no single one predominates. This mirrors traditional Chinese statecraft, which was a varied mix of Confucianism and Legalism. Confucianism argues for rule by example; the rulers must be virtuous, promote loyalty and obedience and kindness of all that. Of course that doesn't work, and that's where Legalism comes in, where actual whips with spikes, and many other torture mechanisms were used to control the people. Legalism also had a groupthink aspect in that non approved books were to be burned and their authors be killed along their families. That that only happened twice in all their history is a testament to the strength of Chinese statecraft, that it seldom felt the need to enforce groupthink.

But I'm rambling already. Thoughts?

28 comments