The problem of studying people in the first villages is not only that their problems don't map directly to today. It's also that it's get's really hard to get concrete data. It's much easier to do good science when you have good reliable data.

With 10,000 people you can solve a lot via tribal bonds and clans. Families stick together. You can also do a bit of religion and everyone follows the wise local priest. Those solutions don't scale well.

It seems mostly irrelevant to me though, since I am aware that rubbish social scientists exist and I just want to try and improve and be a good social scientist.

You are likely becoming like the people that surround you when you go into university. You also build relationships with them.

Going to Cambridge is good. Cambridge draws a lot of intelligent people together and also provides you with very useful contacts for a political career. On the other hand that means that you have to go to those place in Cambridge where the relevant people are. Find out which professors at Cambridge actually do good social science. Then go to their classes.

Just make sure that you don't get lost and go on a career of digging up old stuff and not affecting the real world. A lot of smart people get lost in programs like that. It's like smart people who get lost in theoretical physics.