August 5th, 2010
I recently stumbled upon an impressive blog: The Crooked Timber. It’s hard to characterize its content, but what makes it so interesting is the high level of intellectual sophistication among all the participants. No flamers, trolls, spammers, or nitwits here. Everybody is highly educated, possessed of intellectual integrity, and impressively courteous. At least, that’s what I’ve seen so far. I got drawn into a fascinating discussion: how many people would you have to send to another planet to establish a self-sustaining colony? That of course depends upon the level of technological sophistication you desire for that colony. But the debate was over a claim that you required a large number of people to maintain the technological expertise necessary to sustain such a colony. Many correspondents rejected this, claiming that a broad education combined with a big library should permit anybody to handle any technological problem.
I was surprised by the positions taken by the two sides, because I developed my own position on this more than twenty years ago with my game Guns & Butter. In that game, you had the technological know-how to build any technology you desired, but economies of scale constrained your ability to economically utilize such technologies. That is, you couldn’t build heavy artillery until you had sufficient steel-making capacity, and you couldn’t make steel until you had sufficient iron-making capacity, which in turn required sufficient mining capacity, sufficient charcoal capacity, and sufficient transportation capacity. These capacities were in turn constrained by your population. You start off with a purely agricultural economy with very little surplus labor to apply to technology. You use that surplus labor to build simple technologies that improve agricultural productivity, creating a larger labor surplus that you use to establish even more sophisticated technologies, and so on. The key point here is that economies of scale only work when you have some scale. You can’t build and operate a big steel mill with a dozen workers -- you need hundreds just to operate the mill.
Here’s the theoretical basis for my hypothesis:
1.Population growth increases the supply of surplus labor. There is one situation in which this would not be the case: when the increment in population must farm on marginal land so poor that one farm laborer can generate only enough food to feed himself.
2. The increment in surplus labor, when applied to existing economic functions, increases the degree of specialization used in handling those functions. For example, if you’ve got ten people making pottery, and suddenly got another ten to help, you’d break the tasks down more finely, attaining greater efficiency from greater specialization.
3. Greater specialization increases the productivity-enhancing character of tools. For example, suppose that one your original team had the task of going out, digging up clay, carrying it back to the workshop, and piling it up. Suppose that you get a second person to help the first. You could break the task down into a) digger and b) transporter. Because the digger is now devoting all of his time to digging, the purchase of a shovel yields greater benefits than it would have earlier. The same thing goes for, say, a wheelbarrow to transport the clay.
4. Greater productivity-enhancing use of tools encourages the search for better tools. For example, when there was just one guy doing all that work, he never bothered to think about getting a shovel, shovels not having been invented yet. However, once the guy starts digging full-time, it’s likely that he’ll realize that he could do a better job if he could get a thin slab of metal with which to pry underneath the clay, with a handle on it. Maybe this guy won’t figure it out, but if you get more and more people digging in the dirt like that, eventually one of them will get the inspiration.
5. The search for better tools yields improved technology.
Thus, population growth yields improved technology.
Thus, human technological advance has not been driven by the felicitous discoveries of rare geniuses -- they are the inevitable result of population growth. If James Watt had not come along to improve the Newcomen steam engine, somebody else would have done it; perhaps they would have come along a few years or even a few decades later, but it’s not as if Mr. Watt had some unique talent. He was obviously a clever guy, but there a lot of clever people in the world.
Thus, the answer to the colonization question is even more pessimistic than the advocates on both sides realize. You need more than the chiefs who know how to run the technology: you also need the indians who’ll actually do the work. For example, if you want a society able to support itself at the technological level that the USA achieved in, say, 1830, then you’ll need a population similar to that of the USA in 1830: 12 million. The USA was a fairly self-sustaining economy by that time: although its economy was boosted by plenty of trade, I think it was isolated enough to give us a decent idea of the minimum level of labor required to achieve its technological level.
Thus, if you wanted to build a colony that could be completely self-sustaining, and you could send only 12 million colonists, then their equilibrium technology level would be that of 1830s America. There is, of course, an ameliorating factor: suppose you equipped them with all sorts of high-tech tools and machinery, gobs and gobs of energy supplies and advanced raw materials, and so forth. Then they’d start at your own level of technological sophistication when they landed, and they’d slowly regress as they consumed the high-tech stuff you sent along with them. However, their population would be growing, and with luck, the upward curve provided by population growth would intersect the downward curve imposed by depreciation before too much technological sophistication is lost. You’d have to plan that one out pretty carefully.