It's pretty trivial to fabricate a thought experiment where the genetic record looks one way and the historical record another.
Let's imagine we've got a thousand islands a million years ago, each with a thousand proto-humans living on them.
These islands live in peace and tranquility, with people occasionally swimming from one island to another to trade or visit or start a new life.
Then one year, one island decides it's had enough of peace and tranquility. It cuts off contact with all the other islands. From this point on, everyone born on that island is a descendant of only the thousand people living on that island, instead of intermingling with the neighboring islands over time, but there's still a million people.
Then, one generation, they go to war. They swim to a neighboring island, kill everyone there, and take it over, and then divide themselves between the two islands. The population drops from 1 million to 999,000. Over time, they expand to fill the two islands, and the population rises back to 1 million.
A hundred years later, they decide they need another island, so they swim to a third island, kill everyone there, and repeat.
Bit by bit, over the course of a hundred thousand years, the descendants of the original xenophobic island swim to each of the other thousand islands, kill everyone there, and repopulate.
In the genetic record, you have a bottleneck -- everyone descends from this small starting population. In the historic record, you just have a gradual pattern of migration and genocide. At no point does the population drop much below the original million; at no point does the population of pre-humans "almost go extinct".
Great write-up! I wonder if it would be possible to enumerate all possible scenarios like this and assign probabilities to them.
For example in the scenario described here, I'd expect the probability it actually occured to be pretty low given that the geographical facts (the world is not made of small islands) and the archeological findings (no mass graves that can be assigned to a periodic timeline) don't line up.
Maybe I'm just not creative enough but somehow I feel there can't be that many different scenarios and most of them should be easily falisifieable/verifiable.
Nice, so that's another scenario. Not sure if we should rank it higher in terms of probability though - at some point they would either have had to survive some catastrophe while all others died (basically the bottleneck scenario?) or go out and kill all the others (no archeological evidence).
According to the article, that bottleneck lasted for 100,000 years. So in your scenario, the ancestor island cuts themselves off for 100,000 years before starting the extermination.
Also the time period of the bottleneck corresponds with a hole in the archaeological record.
So while your thought experiment is possible, a bottleneck is more plausible.
I think there's actually a really trivial modification for that: every thousand years or so, one of the colonizer islands decides the other colonizers are impure, cuts itself off, and begins the cycle all over again.
Wow. That is a very powerful thought experiment. Is there any evidence that a series of very slow non-comingling death marauders. Took over a population of one million vs there just being a dangerous few individuals at one point? I feel like we haven't seen successful slow march death marauders more often than low populations.
It's entirely a thought experiment. There's no evidence for it whatsoever, and it doesn't even aim to be an accurate portrayal of even my shallow understanding of the history of human precursors.
It's just an example of a scenario where only a thousand people have living descendants today that isn't "a volcano erupted and almost everyone died" or something.
I didn't mean is there evidence in human history. But is there evidence of it ever happening with any species in any recorded natural history? Because we have lots of examples of near extinction crisis.
Let's imagine we've got a thousand islands a million years ago, each with a thousand proto-humans living on them.
These islands live in peace and tranquility, with people occasionally swimming from one island to another to trade or visit or start a new life.
Then one year, one island decides it's had enough of peace and tranquility. It cuts off contact with all the other islands. From this point on, everyone born on that island is a descendant of only the thousand people living on that island, instead of intermingling with the neighboring islands over time, but there's still a million people.
Then, one generation, they go to war. They swim to a neighboring island, kill everyone there, and take it over, and then divide themselves between the two islands. The population drops from 1 million to 999,000. Over time, they expand to fill the two islands, and the population rises back to 1 million.
A hundred years later, they decide they need another island, so they swim to a third island, kill everyone there, and repeat.
Bit by bit, over the course of a hundred thousand years, the descendants of the original xenophobic island swim to each of the other thousand islands, kill everyone there, and repopulate.
In the genetic record, you have a bottleneck -- everyone descends from this small starting population. In the historic record, you just have a gradual pattern of migration and genocide. At no point does the population drop much below the original million; at no point does the population of pre-humans "almost go extinct".