Blog/2025-02-24/Lost Science

There are two dominant schools of thought with respect to history and causative events:

• Great man theory
• History From Below

The short version of the difference is that the former is the belief that significant individuals shape history and the latter is that it is the ebb and flow of the masses that shapes history. You can squint the former into the latter if you believe in Multicellular Mankind by believing that the human organism grows individuals as functional tools that progress it.

That's all very well and well-debated so there's no reason rehashing it here. But I saw a tweet recently that asked the question of "what examples are there":

Peli Grietzer @peligrietzer

Sorry to bite the 'great men theory of history' bait, but I'm suddenly curious if there's a major scientific or technological discovery where we think if the person who made it got hit by a bus it might have taken decades or centuries more

Feb 24, 2025

How does one prove that something wouldn't have happened without someone? The classic examples of simultaneity are Alexander Graham Bell and Elisha Gray for inventing the telephone and filing a patent on the same day, and the invention of the radio which is so populated by inventors that it's hard to credit one person.

So there are definitely ideas "whose time has come", so to speak. At least some things arise from the natural transmission of ideas between human beings. Some things are argued one way or the other: "Hitler was uniquely influential" vs. "Hitler arose because of the circumstances of the times; if it wasn't him it would be someone else". But what are some ways to detect whether some Great Men might exist.

Inventions Awaiting Discovery

One idea to do so might be to consider influential inventions or discoveries for which the pre-requisites existed for a really long time. One such example from modern times is in canoe paddles.

Paddles

One example is the bent-shaft paddle, popularized by Eugene Jensen. Well within the technology of millennia but somehow only really adopted since the 1970s. Strictly bent ones with a sharp discontinuity are harder to do in ancient times, and one could argue that joinery and nailing make one less reliable. But curved carvings could be made quite reliably in the past. But there are arguments against:

• Increased complexity
• Lack of need (a 10% improvement is not noticeable in prehistory and only noticeable in racing)

So it was probably just dominated by the fact that paddling technique and physical characteristics of the paddler dominated this greater than the tool for much of human existence.

Stirrups

Another example, more historical, is the invention of the stirrup. South Indians used toe stirrups for ages, perhaps for stability. But the foot stirrup was invented among the Chinese for ages before it made its way to Europe through cultural diffusion. The cataphracts I was so fond of in Age of Empires were actually completely unstirruped and they packed quite a wallop. But you can imagine the increased efficacy of a cavalry that was foot-stirruped must have been quite desired.

The technology for stirrups existed for a long time and they "should have been" invented much earlier. Some argue that the absence of a rigid saddle tree (the base of a saddle) causes too much pressure so it could conceivably be argued that stirrups absent the padded saddle would position the rider strictly in the same position each time causing the horse to be more tired than with the natural more random movement otherwise.

Conclusion

Many other such "inventions awaiting discovery" exist. A few of them are:

• wheeled luggage: casters existed for centuries before someone considered putting them on luggage and there were no materials innovations before they became popular (though there have been ones since)

• spectacles: lenses of required optical quality seem to have existed for centuries before they were thought to be used for people

And while looking around for the paddle thing I actually found this old thread from Dan Luu on Twitter that has some examples in software:

Dan Luu @danluu

Is there anyone who's exploring the question "why wasn't X invented earlier?" for various Xs, especially for modern Xs?

I've heard that I should read Graeber's new book, but that it won't really be what I'm looking for since it mostly focuses on older Xs.

Dec 29, 2021

These days, in LLMs, "very stupid" things have a lot of success including the classic prompt-enhancer "let's think step-by-step" and sometimes they seem to qualify except for the fact that they're discovered quite quickly.

Nonetheless, these examples all have the problem that someone could conceivably make the argument that economic circumstances did not make the ideas worthwhile at the time and only a change in the requirements of the time (canoe racing, more travel) or a difficult unblocking invention (the miniature trigger) made the device valuable. So these could be another example of History From Below.

Lost Science

Another idea might be useful concepts that some human develops which are then lost until another human being develops them later. The greater the duration between the two inventions, the greater the difficulty must be.

Heliocentrism

Aristarchus of Sarmos came up with this model in the 300s BC but it was dismissed in favour of the Ptolemaic model some two millennia until the Copernican Revolution from Copernicus to Galileo. One must therefore conclude that it was a large conceptual leap to make. Without Copernicus coming up with heliocentrism again, we might have gone even longer without discovering it.

Inheritance

Mendelism was 'lost' in the sense that it did not gain popularity until Hugo de Vries and Carl Correns rediscovered it half a century later. Inheritance and natural selection rapidly became well-known and understood over the next half century but this time, the cycle time for the idea was reduced from millennia to a mere forty years.

Conclusion

There are other such ideas or inventions that are lost for a long time before a new inventor or discoverer shows up to recreate them. In a sense, even the famous trapezoidal rule for integral approximations is a rediscovery in Chemistry is an example of this, though the barrier there is not time by field.

Overall

I set myself an hour to research and write this and sadly I've spent two so the time is out before I could get properly into the main concept of whether or not a good test for the Great Man Theory is ideas that are discovered, then lost, then discovered again. I'm somewhat of the opinion that it's a decent pointer to the fact that we occasionally do need innovative brilliance in idea generation to come up with things and that it's not just that the wisdom of the masses gets you to results given prerequisites are solved.

Cultural diffusion seems crucial to humanity learning, though often invention occurs in isolated areas first before going somewhere else. It brings to mind a Psychology paper I read a long time ago about intermittent interactions vs. no interactions vs. constant interactions. As expected, mean(NT) was worse than mean(IT) ~= mean(CT) but optimal(NT)~=optimal(IT) which were both better than optimal(CT).

I think this interaction is self-similar down to single humans (and perhaps further down than that). The mass of life is a swirling vortex of thought and pattern and occasionally it throws off outliers that think in a different way or act in a different way. In Nature, the process is random mutation and the thing that keeps it in line is natural selection through environmental and peer pressures. In Mankind, the process is both the genetic and memetic mutation that results in individuals (the Great Men) who come up with particularly novel and useful things to humanity. So it's less that it's Great Men vs. History From Below and more that the system of interaction of large numbers of independently varying people creates outliers of some kind that, when placed at inflection points in the environment, find themselves particularly able to be adapted to it.