Paging Vernor Vinge - Input Junkie
Paging Vernor Vinge|About Tainter's The Collapse of Complex Societies
In 1988, Joseph Tainter wrote a chilling book called The Collapse of Complex Societies. Tainter looked at several societies that gradually arrived at a level of remarkable sophistication then suddenly collapsed: the Romans, the Lowlands Maya, the inhabitants of Chaco canyon. Every one of those groups had rich traditions, complex social structures, advanced technology, but despite their sophistication, they collapsed, impoverishing and scattering their citizens and leaving little but future archeological sites as evidence of previous greatness. Tainter asked himself whether there was some explanation common to these sudden dissolutions.
Tainter’s thesis is that when society’s elite members add one layer of bureaucracy or demand one tribute too many, they end up extracting all the value from their environment it is possible to extract and then some.
The ‘and them some’ is what causes the trouble. Complex societies collapse because, when some stress comes, those societies have become too inflexible to respond. In retrospect, this can seem mystifying. Why didn’t these societies just re-tool in less complex ways? The answer Tainter gives is the simplest one: When societies fail to respond to reduced circumstances through orderly downsizing, it isn’t because they don’t want to, it’s because they can’t.
In such systems, there is no way to make things a little bit simpler – the whole edifice becomes a huge, interlocking system not readily amenable to change. Tainter doesn’t regard the sudden decoherence of these societies as either a tragedy or a mistake—”[U]nder a situation of declining marginal returns collapse may be the most appropriate response”, to use his pitiless phrase. Furthermore, even when moderate adjustments could be made, they tend to be resisted, because any simplification discomfits elites.
Link thanks to andrewducker
An example of this and much more: -- General Motors theoretically could have learned how to make reliable cars from Toyota, but.....
I recommend listening to the radio program, but here's a summary. Toyota needed to start making cars in the US and wanted to learn how to work with American workers. GM wanted to learn how to make a fuel-efficient car.
So Toyota re-opened a GM plant at Fremont which had been closed because the work force was the worst in the country-- with the same workers.
The workers were taken to Japan in groups of 30 for two weeks. And it turned out that they were fuck-ups because they'd been working in a system which didn't support anything else. When they were in a system where making high-quality cars was the most important thing, they loved it.
When there was a later effort to import Toyota standards to GM in general, it wasn't possible. You can't just have high quality in an assembly line, you have to have a culture of making high quality parts, and neither the management nor the suppliers at GM could manage to change the culture of sloppiness.
An irony-- part of the disastrously bad culture at Fremont was the assembly line must never stop. You'd get fired for stopping the line, no matter what went wrong. Not only was there sabotage in addition to accidental mistakes, absenteeism was so bad that sometimes the line was stopped anyway.
At Toyota, any worker could stop the line, so defects were caught and corrected before they were expensive to fix or went out to the customers.
From this, I learn that the assembly line is going to stop one way or another-- it's just a question of which way you prefer.
GM did improve quality, but so slowly and incompletely that the company went into bankruptcy.
As has been in the news, Toyota has let things slip a little, but it's still making the most reliable cars.
Another story about what can be accomplished from the top: Frank Schaeffer's Crazy for God: How I Grew Up as One of the Elect, Helped Found the Religious Right, and Lived to Take All (or Almost All) of It Back
is mostly about the genesis of the American religious right (a rather weirder story than one might expect), but it's also a memoir, and there's a bit about him going to a British private school in the fifties which didn't have bullying. It simply (complexly?) wasn't permitted, and the students were proud of not having it.
The reason I mentioned Vinge is that his A Fire Upon the Deep
mentions a couple of ways that advanced societies can fail-- through universal surveillance and through such an extreme search for efficiency that there's no slack if something goes wrong. Overcomplexity is a third.
|Date:||April 3rd, 2010 02:52 pm (UTC)|| |
In the societies we have, there is also rent-seeking, as discussed in Mancur Olson's The Logic of Collective Action: compact groups of people who stand to make large gains from a change in the law, or from the favor of the government, are more effective political actors than diffuse groups of people who suffer small losses from the same causes, with the result that such changes will take place even when the total of the diffuse harm is far greater than the total of the compact good. Though it's not clear that Vinge's envisioned advanced societies suffer from this process; they may have a way around it, and fall because of other causes.
I'm not really convinced by the ideas of societies "collapsing". Take the Roman Empire. Did it collapse and if so when? It considered itself a going entity right down to the fall of Constantinople to the Turks. Even in the parts where Rome's power was extinguished like Italy, Spain and Gaul there was considerable continuity of culture. I suspect the "collapse" hypothesis relies rather heavily on privileging certain kinds of evidence over others and equating "society" with its ruling elite.
It's a good argument for not having a monoculture, but instead building society out of smaller pieces joined together, so that if one part goes horribly wrong the others aren't utterly destroyed as well.
How hard is it to not have a monoculture, especially considering modern communications?
Good question. Considering the effect globalisation has, where countries are (for instance) trying to set up international accounting laws to try and stop a repeat of the meltdown, there's a real risk that we'll end up with identical laws everywhere, and locking our economies together in a way that brings them all down next time there's a crash.
|Date:||April 3rd, 2010 03:55 pm (UTC)|| |
That already happened on a national scale a century ago, with the creation of the Federal Reserve. Instead of individual banks having runs and collapsing, causing local harm to their less alert customers, we have all the banks tied together to a financial monoculture that succeeds or fails as a whole. Jane Jacobs warned about the dangers of such monocultures in Cities and the Wealth of Nations, though her imagination didn't go so far as to having unregulated competing banks, but only to having each city have a separate financial monoculture rather than sharing in One Big Monoculture. Hayek's more radical proposals in Denationalisation of Money might be more effective if they ever got serious attention from politicians.
|Date:||April 3rd, 2010 03:48 pm (UTC)|| |
Actually, contemporary communications . . . that is, not the mass communications of the twentieth century, but the point to point communications of the twenty-first . . . seem to facilitate the emergence of a more fractionated culture, by making it easy for people with unusual interests to find each other regardless of distance. I just coauthored a book for Steve Jackson Games; only one of my three co-authors was in the United States while we worked on it . . . another was in Australia and the third in Japan. I interact much more actively with other members of the Libertarian Futurist Society now that we have the Internet rather than depending on print.
The emergence of this more differentiated culture in fact is alarming some people who are committed to monoculture, whether ideologically (as when columnists and book authors worry about people turning away from the consensus of the mass media to get news from their own ideological or cultural communities) or economically (as in print and television journalists bemoaning the economic failure of the mass media and often asking the government to step in and subsidize them).
Differentiated culture, similar styles.
You can eat in restaurants with cuisines from all over the world, but it's getting harder to find places where people will feed you because you're an exotic and fascinating stranger, or where you can just hunt and gather for yourself.
|Date:||April 3rd, 2010 04:41 pm (UTC)|| |
That's a result of the replacement of local tribal communities with a widely extended market. But when the world was all local tribal communities, the chances of your being able to experience a wide variety of cuisines were pretty minimal; few people could travel to different tribal communities, or learn to hunt and gather in new environments. Look at how many millennia it took Homo sapiens to colonize the entire habitable world to get a sense of how much travel people did.
The ability to hunt and gather as a lifestyle rests on unclaimed land that can easily be accessed; the ability to be treated as a welcome guest of a community rests on small tribes that can easily be visited. But both are purely transitional phenomena that exist because there's a disequilibrium between the Promethean West and the rest of the world. The long-run equilibrium is for everyone to be incorporated into a global market economy . . . or if the Soviet Union hadn't run into the economic calculation problem, it might have been into a global bureaucracy. I prefer the market economy; it's more hospitable to diverse communities of interest. Indeed, the market monocultures of the twentieth century were massively helped along by government promotion of monopoly and cartelization; they would not have been nearly as sustainable without rent-seeking.
My point was that the sort of varied cultures you can get in the modern world is no protection from from a global economic system going sour, or climate engineering gone wrong, or badly designed viral intelligence increase.
It wasn't the economic calculation problem that did for the USSR, any more than you can describe the western system as a truly free market. It was the lack of internal political feedback loops (accountability), leading to endemic corruption ... in combination with Saudi oil policy in the early-to-mid-1980s undercutting the USSR's ability to barter exports of oil and gas for the grain their farmers weren't bothering to produce.
While the calculation problem looks to be a real issue, I suspect no culture on Earth to date has gotten anywhere near the limits of efficiency imposed by a centrally planned economy wrt. to a decentralized market-based one. (If you contemplate what could be done via central planning using modern IT equipment to do near-realtime adjustments of supply requirements to match up against demand, as opposed to the clunky GOSPLAN five year cycle using clerks working on paper, there's a lot of room for improvement in that model.)
Edited at 2010-04-04 01:39 am (UTC)
|Date:||April 4th, 2010 04:00 am (UTC)|| |
I don't look at it that way. My take is that the endemic corruption was the solution to the economic calculation problem: you have an official bureaucratic command economy, but just as in the American armed forces, you have a subterranean layer of wheeling and dealing, bribery and barter and favor trading, that gets around the massive unworkability of the official system. Had the Soviet system actually been purged of corruption, I think it would have been far less than around half as efficient as the American system. But of course it wasn't in the interest of Soviet officials to get rid of corruption, any more than it's in the interest of Congress to get rid of lobbying and rent-seeking.
Chickens and eggs. As it happens, though, we've got a perfect crossover experiment to monitor in the shape of the post-USSR Russia. Which is, if anything, ever more corrupt than the USSR. I cry foul on your citing the economic calculation problem; there seems to be a real problem with corruption in Russian society going back to before the command economy came in -- probably back to the social disruption of the civil war, if not earlier (to the Tsarist tax farmers).
The mess that is the post-Soviet Russia knocked a lot of libertarian naivete out of me.
Still, it was hardly a free market society.
I brought up the calculation problem, not as an explanation for the problems of the Soviet Union, but as a reason for thinking that even the best computerized central planning wasn't going to be all that efficient.
That being said, who's managed to make centralized planning work on a large scale?
In re military black markets-- do you have any information on their prevalence over time? I used to know someone who worked on the computerized supply logistics for the US military and who said that things had gotten a lot more accurate. And then his department was gutted under GWB, and I have no idea whether logistics have deteriorated, or whether there's been an increase of military black markets.
|Date:||April 4th, 2010 03:36 pm (UTC)|| |
I will certainly grant you the point that corruption went back to before the Communist and, for that matter, to before the tsars (Donald Kingsbury traces it to the period of Mongol rule in The Moon Goddess and the Son). But Nietzsche makes a relevant point here: the explanation of how something came into being is not the same as the explanation of why it exists now, or what its current function is. We hear, for example, with the aid of adapted bits of jawbone.
What I am saying about corruption under the Soviets is that in fact it needed to be practiced and tolerated, because it was the only way to keep the Soviet economy going. The Soviet bureaucracy was too clunky and inflexible. Successful plant managers needed to be able to use unofficial channels, or to have employees who could do it for them. This may have represented a new payoff for corruption that already existed in tsarist society, certainly. But I think that the Soviet system was much more productive with its combination of bureaucratic planning and corruption than it could have been with only bureaucratic planning. Do you think this is a misinterpretation, and if so, why?
And presumably corruption in present-day Russia has still different functions and payoffs. I'm not looking for a timeless ahistorical essence of corruption; I'm talking about what role it played in the specific historical circumstances of the Soviet system.
I'm not sure what "corruption" means within the terms of this debate. I'm guessing that in general it means interactions not permitted by the commander of a command economy. If that's the case, and if you propose that a corruption-free bureaucratic system would outperform (criteria?) a chaotic/"free" market system, do you envision an ideal, rational commander that will be capable of managing the economy of the world or a continental power by dictat?
The largest organizations with centralized planning are probably companies. Do you know how much efficiency they gain from it?
If you contemplate what could be done via central planning using modern IT equipment to do near-realtime adjustments of supply requirements to match up against demand
What do you mean by demand? I'm using to it meaning what people are actually willing to pay, as shown by what they do pay.
If the central planners get it wrong (and, afaik, there's no way to get it right-- even people themselves aren't terribly good at knowing what they're going to pay, and I'm willing to be that to the extent they do know, there are preferences they aren't willing to admit to), then you get surpluses and black markets.
The other piece is a measurement problem-- with centralization, it's very tempting to have a goals which are easy to measure, and very easy to get them wrong.
On the market side, you get a couple of examples from GM-- they were measuring number of cars built without caring about the quality, and there was a cargo cult nitwit executive which had the NUMMI (Toyota in the US) factory photographed and duplicated in a GM factory.
Toyota, of course, is an example of a centralized culture (I don't know how much centralized planning they do) which is measuring the right things.
Edited at 2010-04-04 09:14 am (UTC)
|Date:||April 4th, 2010 03:51 pm (UTC)|| |
R. H. Coase's "The Nature of the Firm" opened up this question of central planning in corporations back in the 1930s; the modern theory of the firm (as in Spulber's book on it, which I edited last year) grows out of that. Spulber would be a good starting point, if you don't want to reinvent the wheel. His book's listed on amazon, and your local university library might have a copy.
Demand is one of the three big problems with central planning. There's the complexity of analyzing the information; a systematic analysis of economic planning might well show that it's NP-hard. There's the complexity of getting all the information together in one place, which Hayek emphasized in "The Use of Knowledge in Society." But then there's the fact that the key information about what sort of production is desirable exists in the minds of individuals . . . and in getting it out you have to deal with the ability of individuals to misrepresent their own priorities for strategic purposes.
One of the most basic measures against this is "skin in the game" requirements: If people have to pay the costs of their own assertions of priorities, they have an incentive not to report those priorities as higher than they really are. If you subsidize people, you encourage them to give inaccurate reports up to the measure of the subsidies; if you give them unlimited subsidies, you render their reported priorities meaningless by making it impossible to falsify their claims about how much a desire means to them. Say's law, which says that supply creates demand, because commodities are paid for with commodities, has a corollary: that meaningful demand, "effective demand," is exerted by suppliers of things that other people demand. If you require people to go to the effort of producing something exchangeable to get into the game (or of getting other people who have produced something to sponsor their presence), you give them a solid reason not to demand things idly or manipulatively. In a centrally planned system, where all production is carried on by the state, it's harder to get meaningful input about how much people want things to be produced.
Interesting discussion. I've seen other corporate examples where it's easier to replace the culture than change it.
Vinge's A Deepness In the Sky is probably a better fit for this discussion.
|Date:||April 3rd, 2010 05:57 pm (UTC)|| |
GM also had the chance to learn how to make quality cars from Saturn -- in fact that was kind of the point of Saturn. What happened instead was that the GM divisions complained that Saturn was making them look bad, and Saturn was forced to learn the GM way of making crappy cars.
I think the moral is that fighting entropy is even harder than it sounds.
I'd previous heard that GM's foreign divisions were in much better shape than the American center. The radio program says it's because the foreign divisions weren't as entangled in the dysfunctional system, so it was easier for them to be influenced by the Toyota culture.
Macroecomonics, Space cars, Ziggurats
The linkage between the Mayans and Toyota vs GM in the arrogance of methodology isn't a clear one, even if I liked the idea of ziggurats being built lopsidedly as the Mayan forepeople dozed on the job and the ever more bureaucratic management structure quietly grafted more and more parrots and hearts.
On the first of the points (a point worn blunt from repeated tackling) I'd like an epilogue if you've got one that indulges some comparison - the long dynasty of China, the sighing post-empire Britain, lessened but hardly annihilated in the sense of being eaten right up by the jungle or raped and plundered by the barbarians at the gates.
As for the theory of universal surveillance leading to plenary social collapse - is there a precedent for such an observation (which seems reasonable but also way out in science fiction towne) or should I put it in the same category as flying cars and atmospheric nuclear detonations?
|Date:||April 3rd, 2010 06:15 pm (UTC)|| |
Re: Macroecomonics, Space cars, Ziggurats
My general theme was the strength of culture, with some examples of how resistant culture is to change. I admit I got started with the collapse piece because it's scary and plausible.
There's more than one way of going wrong. I don't know if you followed the first link, but it was about AT&T losing out to cell phone companies which were willing to be sloppier than it could manage.
There's probably some tautology going on. If a culture survives, then by definition, it didn't get over-complexified.
Afaik, universal surveillance leading to collapse is pure science fiction and might or might not be an actual danger. I think that if you've got universal surveillance, the rules, rewards, and punishments will need to be really well-calibrated.
Re: Macroecomonics, Space cars, Ziggurats
The limit of universal surveillance was always one of attentional allocation - the "eyes" could never actually be everywhere at once so the imperfect supervisory regime had to settle for the illusion of the panopticon as a (very effective) deterrant - not the actuality of constant supervision as much as the universal possibility of supervision, already deployed pretty effectively (and with large and horrifying effects) in Latin America. But with computer facial recognition software and etc coming on-line pretty quickly I was curious if you meant the old school fake panopticon or the Brave New World of google-ware and security cameras that can distinguish you from your twin sister?
|Date:||April 3rd, 2010 07:35 pm (UTC)|| |
Re: Macroecomonics, Space cars, Ziggurats
My assumption was the latter.
Re: Macroecomonics, Space cars, Ziggurats
Have Bentham's/Foucault's ideas ever actually been tested, I wonder?
Or rather, since we know that a scientific test is impossible (problems with isolation of causes, repeatability), do we have any data that suggest a marked improvement in inmate conformity with regulations in penitentiaries vs. other kinds of prisons, or surveillance societies vs. "unsurveilled" ones (whatever those might be)? I'm asking because of the reports over the past few years from the UK, which suggest that, once it becomes inconvenient to avoid cameras, people are willing to commit crimes under their gaze: they do not seem to be discouraged from committing the crimes altogether.
...again, of course, isolation of factors: there has to be some follow up/consequences attendant on the surveillance.
|Date:||April 4th, 2010 03:52 pm (UTC)|| |
Re: Macroecomonics, Space cars, Ziggurats
That sounds a lot like the legalist agenda in Ch'in dynasty China.