Print This Article


JDN 2457271 EDT 15:50.

scifirocketplanet

Starting in the 1900s and continuing as recently as the 1980s, a lot of people thought that by the 21st century we would be colonizing other planets. In my previous post I argued that we are living in the cyberpunk future, but we are actually missing that particular element. Blade Runner was released in 1982 and takes place in 2019. That's four years from now! Roy Batty would have to have been built this year. We're not going to have replicants and offworld colonies within the next four years, I promise you that.

So why did people think we were going to be colonizing other planets by now? And why aren't we?

I think it is because the period from about 1940 to 1990 was the period of the fastest technological advancement in the history of the human race. Up to that point, technological advancement had been getting faster, and faster, and faster—mind-bogglingly fast—and people naturally assumed that this trend would continue and technological advancement would be even faster now than it was then.

But in fact, that really isn't what happened. It's hard to actually quantify technological growth, but intuitively we can see that while the trends toward corporatization and the erosion of privacy happened at about the predicted pace, the trends toward space colonization and flying cars most certainly did not.

I think this is because we were assuming that technological growth is exponential, when in fact it is logistic. In the graph below, the red line is logistic, the blue line is exponential. In the early part of the graph it's actually quite hard to tell which is which; but later on the exponential growth diverges off to incredible heights while the logistic flattens out completely.tech growth annotatedAnimal populations would grow exponentially if they had no predators or other ecological constraints; but in reality they almost always grow logistically. This Nature article explains in further detail.

Almost every technology has grown logistically in the past, so we really shouldn't have been surprised.

If cars had improved exponentially since their invention, by now they'd cost $0.27 and travel at Mach 4. If spacecraft had improved exponentially since their invention, by now they'd cost $50 each and travel at 99% the speed of light—so of course we'd be colonizing space by now. Instead, cars, airplanes, and most other technologies grew slowly before they were really figured out, then saw a burst of development in which they dramatically improved, and then leveled off to remain in basically the same form with gradual improvements over time. That's not exponential growth, it's logistic growth.

I think a major reason why we began to think technology would suddenly start growing exponentially is that one particular technology almost did—I'm speaking of course about computers and Moore's Law. From their invention in the 1940s until today, computer processors doubled in speed every 18 months. That growth was so close to exponential for such a long time that today's computers are literally one trillion times faster than the first computers, which were built in living memory.

Yet even Moore's Law is showing signs of slowing down. It will probably continue for another 40 years, during which time we'll end up with processors a hundred million times faster than today—just about enough to seriously match the total computing power of the human brain—and then I think we'll find it very hard to go any further than that. (If it were easy to make a brain more powerful than ours, I think natural selection would have.) Perhaps the advent of this advanced AI will allow it to make technological progress faster than we can, but at some point we've got to run into fundamental limits just as we did with every other technology.

A standard measure of technology that a lot of economists use is called total factor productivity; it's not an ideal measure (it's basically “the part of economic growth we couldn't account for any other way”), but a historical analysis of total factor productivity growth does give a sense of logistic growth in the US: the growth rate rose from 1% to 3% from the 1900s to the 1930s, and then slowed back down to about 1% in the 1980s. On the other hand, total factor productivity growth seems to be higher now than it was in the 1980s—so it could just be a sign of bad economic policy under Reagan and his ilk rather than a failure of overall technological progress.

I'm not sure how much of our slowing technological development is due to policy and how much is due to fundamental constraints; but if it is due to policy—and I think to some extent it is—then this gives us a great opportunity. If technological growth isn't just exogenous, a given we're stuck with, but is in fact endogenous, a consequence of policy decisions we make, then with better policy we can make better technology.


Some of the changes are fairly obvious: Increased funding to scientific research will allow more research to be done. Improved education will make better scientists and allow people to better fulfill their full potential.

Some are less so: Open immigration allows your population to grow faster (more minds in general means better odds of more geniuses) and also makes other cultural ideas collide with yours and potentially lead to new insights and innovations. Open trade also encourages cultural cross-pollination.

This is vitally important—perhaps the most important thing in the world, in fact—because technological growth is the fundamental cause of all long-term improvements in standard of living. The reason we live today in cities with cars and computers instead of in tribal villages with spears and loincloths is technological growth. Capitalism likes to take credit for that, and to be fair it probably does have something to do with technology, by providing incentives for innovation and maximizing trade; but if an alien were to compare different human societies over places and times throughout our history, it wouldn't notice differences in economic or governmental systems; the only difference that would really be apparent to it would be technology. The Soviet Union in 1950 and the US in 1950 would look pretty much the same compared to the US in 1950 versus 1750, which in turn would look pretty much the same compared to the US in 1750 versus North America in 10,000 BC.

If we can in fact increase the rate of technological growth, or even simply sustain it at this rate for another few generations, we will be colonizing other planets. Maybe not other stars—they're just so damn far away!—but at the very least there will be research stations on Mars, Mercury, and several moons of Jupiter and Saturn. Human beings will finally escape the boundaries of our one little planet and become something greater. You and I may just live long enough to see it.

Or, we could completely screw it up with bad policies and in 2100 have hardly any better technology than we do today. The choice is ours, as researchers, as voters, as citizens—some of us will even be policymakers. Do we invest in our future by spending on space travel, new energy sources? Or do we do the “practical” thing, the “thrifty” thing that lets us have a bit more money now in exchange for being trapped on this rock forever?

[The cover image is Open Clip Art, and seems pretty representative of 1950s pulp science fiction about space colonization. https://openclipart.org/detail/214100/sci-fi-rocket-and-planet]

Patrick Julius

Patrick Julius

This email address is being protected from spambots. You need JavaScript enabled to view it.
Recent Articles
Inequality: Fantasy versus reality
Would gun control save lives?
Who pays the tax?
We live in the cyberpunk future.
Should we vote for nincompoops?

  • No comments found