My thesis is that it's a mixture of a few things: a) Underappreciation of what things require technological advancements. Imagine being a [1] first person to d

My thesis is that it's a mixture of a few things: a) Underappreciation of what things require technological advancements. Imagine being a [1] first person to discover that you can turn certain grasses into dough that can be cooked into bread... and you can't turn other grasses into that dough. That kind of person could well have been the Lavoisier of their day. Yet today, that's the kind of skill passed down from parent to child as the most pedestrian thing ever. b) Bootstrapping effect. A lot of technology requires itself to be built, whether it's a compiler to compile your compiler, or a kiln to bake the bricks to make your kiln. In the inverse of this effect, until you build up enough infrastructure to use your technology to perpetuate itself, it might not be appreciated that it's worthwhile to do so--I'm reminded of the anecdote that Dijkstra once chastised a student for wasting precious computer time running an assembler program. c) Societal effects of smaller, nomadic societies. If you're small, you run a greater risk of knowledge simply dying out; information loss rates were probably much higher 20,000 years ago than today. Similarly, a nomadic society is probably less willing to invest in infrastructure (like, say, kilns) that isn't readily able to move with them, and this will retard certain technological development. It's just a thesis, though; I don't really have any hard evidence for anything, just thought experiments. [1] As I mentioned elsewhere, this is probably an example of something that was independently invented several times.