Against The Popcorn Theory of Cultural Drift

Additional structure within the grab-bag of cultural values keeps a core subset of those values much more protected from maladaptive cultural drift, averting Robin Hanson's doomsday scenario.

Alas, in my experience so far, few who call themselves “rationalist” seem willing to go this far. They mostly want to accept the values of their culture, and then try to be more rational in how to achieve such values, regardless of how adaptive they are. Unless this changes soon, and among enough elites in some society to influence the direction of that society, cultural drift seems unlikely to be fixed soon by rationality.

Robin Hanson thinks our civilization is doomed. Not because of climate or AI, but because of “cultural drift.” He explains this some more on a recent episode of the

Moral Mayhem Podcast

.

Civilization started out thousands of years ago very decentralized, with tons of different clan-like communities scattered around the globe, each one a separate culture. The adaptive ones grew and outcompeted the maladaptive ones, annihilating and/or absorbing them. This selection process continued until we all coalesced into a single global monoculture. 

Now we're dying out because we're not having enough kids. Global fertility rates have been plummeting as cultures get absorbed into the monoculture. Why? Because the values of the monoculture are about becoming a girlboss and having cocktails in the West Village, not staying home and raising kids. Other problems are popping up as well. 

And that's exactly what you'd expect once your big monoculture no longer needs to compete against rival cultures. We're just becoming fatter, slower, duller in general. Our cultural values are, one by one, getting tweaked or discarded and replaced by random new values that happened to go viral amongst all the girlbosses. The memetic fitness landscape in which these new values thrive is decoupling from the old vicissitudes of hard work, discovery, improvement, and fecundity. The price of neglecting those vicissitudes used to be that our men got killed and our women got kidnapped by new husbands. But with the monoculture, there's no price for neglecting the old ways. So we drift.

What happens next? Hanson is pessimistic. He notes various possible ways to save us. But none seem plausible to him. Each of those adaptive values we started with has some half-life. Like popcorn kernels in a microwave. Every few seconds... pop... pop... pop... and once the popping really starts, you know it's about to be over. The odds are vanishingly small that any significant fraction of unpopped kernels remain after a couple half-lives have passed. It's just the arithmetic of probabilities.

What’s going wrong here, I submit, is what often goes wrong with such arguments—the assumption of zero correlation. We don't need to defy probability theory, we only need some kind of correlation pattern among our values. Maybe “our values” is not just a grab-bag of isolated objects that all face cultural selection independently. Maybe there is some kind of hierarchical structure they're all a part of. Not a perfectly rigid structure, but one that connects them and makes them push and pull on each other when outside stresses get applied.

Our values are less like a bag of popcorn and more like bits in an error-correcting code. How do error-correcting codes work? You encode your information bit in a couple of physical bits by setting their values all equal to its value. A cosmic ray can then come along and randomly flip one of the physical bits. But the next error-correction cycle will just notice one of them being out of sync with its neighbors and flip it back. The information—or the cultural value structure—is preserved.

David Deutsch makes a bolder version of this analogy, about how society creates new knowledge. But it’s especially apt for thinking about cultural values, and how society protects the existing means of knowledge creation.

Who exactly is doing this error-correction for our cultural values? It has to be some kind of group or process within the culture itself, governed by some set of core values among the whole set of cultural values. Aren’t those core values vulnerable as well?

Yes, in principle they are. But it's a much easier problem for the culture to protect a much smaller set of core values. And if that core is protected, then all kinds of errors can keep occurring on the surface. They just get identified and corrected periodically by core operations. The core is what exerts ultimate control over the memetic landscape. Also, like certain heavy-duty error correcting codes, there can be multiple layers of cultural values—bits encoded in bits encoded in bits, etc.—with error-correction operating at each layer. Errors might be severe enough to mess up a whole layer of bits, multiple layers even. But failures get exponentially less likely as you go deeper into the layers. It's a really robust system for protecting the core.

There are many historical examples of this kind of error-correction process being carried out over centuries. Modern physics is a striking one. After Newton figured out how to explain the motion of the planets, physics condensed into one strong monoculture. It has had no real external competitors over the following three hundred years. Plenty of internal competition, though. Phlogiston got discarded. Thermodynamics got invented. Statistical mechanics superseded it. Newton's own theories got superseded by relativity and quantum mechanics. Special relativity got superseded by general relativity, and basic quantum mechanics by quantum electrodynamics, which then itself got superseded by electro-weak theory and ultimately the standard model.

The outward differences between Newtonian mechanics and the standard model are stark. On the surface, they might as well be a medieval English countryside and Manhattan. But underneath, at the deepest level, it's the same core values. Values like what? Clearly reason is one. More specifically, things like the necessity to test theories against observation and to avoid fine-tuned, conspiracy-theory type explanations. Those values have been consistent within physics since Newton and Galileo. And they have also formed the core of our larger culture, since the Enlightenment.

Progress in physics over this period mirrors progress in society as a whole. All the obvious metrics—life expectancy, health, wealth, literacy, etc.—have radically improved. And we've faced terrible problems during this period. We conquered slavery, and pandemics, and aspiring global tyrants. Hanson just thinks that was all due to the old values, and now we're running on fumes.

But why? Why isn't fertility just a more recent major problem for us to figure out how to solve? Just like the nature of protons and neutrons was the next big problem to be solved after quantum electrodynamics, when physics had been “running on fumes” for centuries.

The running-on-fumes account of progress is not a great explanation in either of these cases. A better explanation is that, once reason is installed as the deepest value, there can be a long, robust sequence of further developments, of iterative problem-solving and new-evidence gathering. New solutions can be conceived and tried until one works. The popcorn probability is on our side here.

When Enlightenment reason got installed in the core of our culture, it didn't automatically correct all existing problems within the hierarchy of cultural values. Many were still there and still needed to be fixed. Installing it was the first critical step that gave us the ability even to identify those problems and then to solve them. But it was and is an arduous process.

Culture is an intricate, heterogenous thing, like a species. When some new evolutionary jump occurs that fundamentally changes how a species deals with the world, all kinds of other traits have to be dragged along—traits adapted to its previous way of life that don't work the same now. A species of fish emerges on the shore, grows legs, and starts living on the land. You can bet something radical is going to happen to its dorsal fin in the following generations. But it doesn't just all happen at once, the first moment one of them pops its head out of the water. It takes time for everything to reconfigure. Waves and waves of new trial-and-error, with each morphological structure rearranging, getting in sync with other structures, and re-rearranging.

Likewise for culture, albeit on much shorter timescales. Sacralizing reason was like getting out of the water. Slavery, tyranny, the oppression of women, etc.—all of these things were like the dorsal fin. Each structure, while terrible, was also tightly connected to other cultural technology that was functional, and not easy to just get rid off. So wars were fought. Families and communities torn apart. Assumptions questioned. Norms rejiggered. And guess what? We're still in the middle of that process. Dedorsalization is hard. It's ongoing. And it's worthwhile. We've made a lot of progress so far, and we're still heading in a good direction.

Of course nothing is guaranteed. The real risk is that, somehow, reason gets uninstalled from our cultural core. That's not something that happens merely by “drift.” That's a revolution—a core meltdown. Still, there are reasons to take that possibility seriously. The dinosaurs vanished. What could be the asteroid that hits our culture?

Let's go back to physics. Its accomplishments since Newton have been awe-inspiring. But something has gone wrong over the last 50 years. String theory is not merely a degenerating research program. Since about 20 years ago, it has been putting stress on some of the core values within physics that made possible all that previous progress. And the problems didn't start there. They started a hundred years before that with the advent of quantum theory and what was to become the “wavefunction collapse” orthodoxy around how to understand its physical meaning.

There is something similar happening in our broader culture. Indeed, it's not merely an analogy. Some of the same intellectual forces that warped theoretical physics—without totally crippling it—continued to haunt academia for a number of decades. And more recently these forces erupted into the larger culture. As in physics, their impact was neither total nor permanent. Both physics and our culture are currently in a frothy kind of mixed state.

Fertility is an important problem. But it is not fundamental. It's not in the core. There are all kinds of potential solutions—e.g., aggressively deregulating home-production—for this problem, and for similar ones, so long as the core values of the Enlightenment are kept intact. In general, a host of dorsal fins lingering on the body politic have to do with grand schemes that centralized authorities come up with to address either real or imagined failures of the market.

Getting more realistic about what such schemes can achieve, and about the big side effects they tend to produce, is one way that Enlightenment rationality is still slowly reconfiguring our memome. Hanson understands the cultural benefits of people becoming more market-friendly, but he doesn’t seem to connect this with the larger process of intermediate layers being gradually reconfigured under the memetic pressure exerted by our core.

And the maladaptations Hanson worries about have not just come out of thin air. They are knock-on effects created as old, pre-Enlightenment misconceptions interact with new conditions created by all the positive growth and social change that have been unleashed. What’s surfacing these problems is the progress itself. And more progress, driven by the same core Enlightenment values, is the best solution.

How to safe-guard those values themselves is the critical thing that needs to be understood and pursued. Because a new kind of challenge is being posed by the array of anti-Enlightenment forces operating today—postmodern variants of both left and right political ideologies. 

This is the product of many generations of another strong selection pressure working within intermediate layers, but in opposition to our Enlightenment core. What finally emerged from this process was an anti-rational block of memetic code that was able not only to colonize large sections of academia but to jump outside of it entirely and gain currency within normie culture

I'll share my own views about the nature of this problem in subsequent posts, and so what kind of things we need to discover in order to solve it. But Hanson's is one of the few attempts I've seen to formulate the broad question in what I think are the right terms—inter- vs. intra-cultural selection and adaptive vs. maladaptive values on a memetic landscape. So my appeal to the popcorn theorists seems like a natural jumping-off point.