What makes technology evolve? Not population size

Population size was thought to be a key factor driving technological evolution. A review of the data doesn’t support this conclusion.

Around 100,000 years ago our ancestors began developing a set of new technology called the “Upper Palaeolithic”. It took Neanderthals millennia to catch up. By the time they did they were doomed to extinction. Many think differences in population size explain this difference.

Well, not just population size. Many other social factors may also influence the course of technological evolution. Like how frequently individuals interact with each other. But population size has a big influence these other things. So it’s the one everyone focuses on.

Or at least, that’s what many anthropologists argued. And there were several computer models and case studies that seemed to back them up. Like the peculiar case of the Tasmanians, who “lost” technology when their population size was reduced.

But some computer simulations aren’t as good as real world studies. So whilst this hypothesis for why technology evolved had been gaining steam in the lab, a growing body of anthropological evidence has put an iceberg squarely in its path.

Fishermen, Neanderthals, and other oddities.

Explaining the evolution of technology in the human family is harder than you might think. This is because it doesn’t really follow the pattern you might expect. For most of our evolution stasis, not evolution, was the norm for our tools.

Lissoirs, one of the few tools invented by Neanderthals

Lissoirs, one of the few tools invented by Neanderthals

There were the Neanderthals, who didn’t change their technology (much) until they started stealing it from humans. Before them, Homo erectus in Asia seems to have not bothered adopting inventions made by their Western cousins. Even the first tool makers didn’t really build on their discovery. The first toolkits remain the same for almost a million years. Clearly inventing stone tools was enough innovation for these chaps.

It’s clear there has to be some external factor influencing technological evolution. Otherwise all these groups would be consistently progressing at a similar rate. This external factor could be population size.

After all, this can vary considerably between groups, explaining why one might prosper (technologically speaking) whilst another stalls. Neanderthal groups, living in harsh northern environments, might not have been able to reach the sizes of human groups closer to the equator. Thus we were able to innovate, whilst the Neanderthals were limited by their population.

Why would population size have this impact? Computer models have come up with various ways the two are linked.

  • One factor is the sheer rate of innovation would be larger. More people means more possible innovators.
  • Learning is imperfect. Only a small percentage of a population might learn a skill well enough to teach. So in a smaller population there will be fewer teachers.
  • Larger populations also mean you will interact with others more often. Thus any innovation might spread faster. Particularly between groups.

Crucially, they all also vindicate this link between population size and technological evolution. Anthropological data has also provided some support for this conclusion.

The last indigenous Tasmanians. Their population size at this point was very small: four.

Perhaps the most notable piece of evidence in favour of this are the indigenous Tasmanians. They were isolated from mainland Australia ~14,000 years ago; effectively slashing their population size. After this they seem to have lost a great deal of their toolkit complexity. For instances, fishing used to be a staple food source. Yet within a few thousand years it seems they “forgot” how to fish. It even got to the point that they were disgusted by the idea. Europeans were treated with disdain for eating out the sea.

Other studies on indigenous populations have also found a link between population size and toolkit complexity in Native Americans, African groups, and a sampling from around the world. Some of these studies have also vindicate some of the explanations for why population size matters. It turns out that the Hadza hunter-gatherers often wind up learning from each other, whenever two groups interact.

But as I teased you all with earlier, the evidence for this isn’t as strong as it seems.

Evidence against population size

A large chunk of the evidence for the link between population size and technological change comes from computer models. Being models, they aren’t the real world. They miss out some key things. Like learning.

The person modelled by these simulations

Many of these models . . . model how information is passed on from one generation to the next (and how its transmitted between peers too). However, after this initial period of learning they don’t model any further development. Once the virtual human leaves hunter-gatherer school they learn nothing new. They don’t improve on any of their skills (except via innovation). If they were a sub-par pupil they’ll be a sub-par adult.

This is one of the key reasons these models produce a lack (or even reversal) of innovation in small groups. Only a small percentage will be excellent learners, meaning there will only be a few who can be excellent teachers. If the population is small then – in real terms – there simply aren’t enough teachers to go around.

Which is of course, how the real works. No new skills are developed after school. Nothing new is ever learnt. Oh wait, my mistake. That’s absolute bullshit.  Obviously, people can adapt, learn, and change after school. And this fact has some profound implications for these models. Tinkering with this single premise drastically changes the outcome; sometimes eliminating the link between population size and cultural complexity.

The link between cultural complexity and population size under different models of learning. The top one, with the clear link between the two, is the one produced when post-school learning doesn't happen.

The link between cultural complexity and population size under different models of learning. The top one, with the clear link between the two, is the one produced when post-school learning doesn’t happen.

But if the models are suspect, what about the anthropological data?

The research in support of the link does seem right. There is indeed a correlation between toolkit complexity and population size in non-industrial populations. But this vanishes when you factor in environmental risk. There’s a much, much stronger link between toolkit complexity and how risky your environment is. Groups in a location where resources are hard to come by (and variable) compensate with more complex tools. So when they do encounter resources, they can be sure they’ll nab ’em.

This trend even holds true when you start examining the past. Genetics can tell us a lot about the population size of our ancestors. So surely there should be an increase in size when we started advancing past the Neanderthals? Nope.

Population size inferred from different genetic sources (the different coloured lines) versus when "modern" technology appeared

Population size inferred from different genetic sources (the different coloured lines) versus when “modern” technology appeared

Even the Tasmanians don’t seem to follow this trend. Although it is true that they lost some technology once they became isolated, this is more than offset by increasing complexity and innovation in other aspects of the toolkit.

This hypothesis could have helped understand why the Neanderthals went extinct. It turns out that this idea needs to join them in the graveyard. Mostly. There is still a minor link between population size and technological evolution, after all, but this is dwarfed by other factors.


Population size was thought to be a key factor driving technological evolution. A review of the data doesn’t support this conclusion.


Collard, M., Buchanan, B., O’Brien, M.J. and Scholnick, J., 2013. Risk, mobility or population size? Drivers of technological richness among contact-period western North American hunter–gatherers. Philosophical Transactions of the Royal Society of London B: Biological Sciences,368(1630), p.20120412.

Collard, M., Ruttle, A., Buchanan, B. and O’Brien, M.J., 2013. Population size and cultural evolution in nonindustrial food-producing societies. PloS one, 8(9), p.e72628.

Collard, M., Buchanan, B. and O’Brien, M.J., 2013. Population size as an explanation for patterns in the Paleolithic archaeological record.Current Anthropology, 54(S8), pp.S388-S396.

Hill, K.R., Wood, B.M., Baggio, J., Hurtado, A.M. and Boyd, R.T., 2014. Hunter-gatherer inter-band interaction rates: Implications for cumulative culture. PloS one, 9(7), p.e102806.

Vaesen, K., Collard, M., Cosgrove, R. and Roebroeks, W., 2016. Population size does not explain past changes in cultural complexity.Proceedings of the National Academy of Sciences, 113(16), pp.E2241-E2247.

Related posts

5 thoughts on “What makes technology evolve? Not population size”

  1. Marcel F. Williams says:

    I thought necessity was the mother of invention:-)


    1. Adam Benton says:

      As the link between risk and technology demonstratws, it is

  2. Steve Staloff says:

    Since the article began by measuring prehistorical technological development through stone tools, it ought to have continued to do so. The only conclusion to be drawn about today’s world by using that metric is that technological advance has ended and the technology of our ancestors was lost — apart from the means of crafting known a very few experts, who lack the know-how to use the tools very well or safely.

    One could imagine instead that stone tools were adjuncts to other, less readily preserved technologies which may have been in continuous development for millions of years. And that stones, being difficult to work and heavy to carry, were only placed in use when other technologies were found limiting by one or more nerdish ancestors with spare time to spend while coincidentally being in a good location for prototyping with stone chips.

    1. Adam Benton says:

      Despite the name, the Palaeolithic contains more than stone tools. Indeed, the Upper Palaeolithic referenced at the start of this article is partially defined by the increasing prevalence of bone and ivory tools. It’s not as narrowly defined as you seem to imply; making this sort of research more applicable than you indicate.

      That said, there is a focus on stone tools. Part of this stems from taphonomic bias, but it’s also worth noting that stone tools were infact widely used. For instance, examining butchery marks on faunal remains reveals people from this period preferentially (by a large margin) were using stone tools to cut their meat.

  3. Pingback: Staying mobile was the secret to our evolutionary success
  4. Trackback: Staying mobile was the secret to our evolutionary success

Leave your filthy monkey comments here.

More in Ancient technology
Neanderthals stole human technology

Human-like tools were found at a Neanderthal site. But humans didn't make them. This reveals our designs were stolen by those meddling hominins.