I think there's a spectrum from thinking with smaller vs larger chunks of other people's thoughts. We never really think from first principles. The universe of thought is more gunky than that.
If we think with smaller chunks of other people's thoughts, we won't hit hard frontiers as often. But just as some physical problems need specialised tools with complex supply chains (e.g. doing heaps of FLOPs really fast), some intellectual problems are the same. If you lack those mental tools, you just won't be able to solve the problem, and you probably can't build the tools all by yourself. The epsilon-delta definition of a limit is an example of such a tool with a complex intellectual supply chain. It's not a hard tool to think with, but it was (as history shows) really hard to create yourself.
Still, if you think with smaller parts of other people's thoughts, you can generally do more and go further than someone thinking with larger chunks of other people's thoughts, since you can build more possible solutions. On one level this is just combinatorics. You can make a lot more things out of 100 individual Lego bricks than 4 walls of glued-together Lego. On another level, being able to think with smaller chunks of thought shows greater competence and generativity in a bunch of ways. There's also the fact that smaller chunks of thought can't fit as many barriers, tricks, dead-ends, and black holes inside them. It's safer to think with a hundred observations than a single ideology.
If you go out and try to change the world, you'll quickly realise power doesn't work the way conspiracy theories claim it does (not through lack of malice, but lack of competence).
This selects for fatalistic conspiracy theorists in two ways:
1. Directly, if you go and try to change the world, you'll soon abandon your conspiratorial beliefs, leaving only fatalistic conspiracy theorists behind.
2. Indirectly, it means successful conspiracy theories have to induce fatalism in their adherents if they are to survive as memes.
This means conspiracy theories are memetic parasites, sapping the vitality of the host to preserve their own existence.
When Napoleon declared Lombardy the Cis-alpine Republic, not the Trans-alpine Republic, he was deciding not to proclaim a new Rome (yet). I wonder if he considered both options.
Meetings and marketplaces both centralise buyers and sellers (of either information or vegetables) to reduce search costs. But marketplaces let transactions take place in parallel (you don't have to wait for me to buy milk for you to sell an apple), while meetings force them to happen in series. In general, shared communication channels are rivalrous, shared marketplaces are not. This means scaling marketplaces increases overall liquidity, while scaling meetings just creates congestion.
But the point of meetings isn't exchanging information (i.e. increasing volume/liquidity), but creating common knowledge. If meetings were non-rivalrous (i.e. everyone could split off into their own breakout rooms), many more transactions could take place, but you couldn't create common knowledge.
The more memetically fit a religion or scripture, the less likely it is to be true.