Today, on an IRC channel run by a friend of mine, someone mentioned a link: http://paulgraham.com/boss.html.
This has led me to a few randomish thoughts. There is little coherence to them, but they do seem to have in common that they're all about human group size, one way or another.
First is that piece. I think he might be taking it a bit too far, but I do think there is some truth in there. I'm not sure how relevant evolutionarily-based remarks are, though; there are lots of things humans are designed for in an evolutionary sense that we do not consider good, such as dying in (what we think of as) middle age to dying young if you happen to have haemophilia, poor vision, diabetes, or any of many other conditions evolution applies its ruthless pruning to but which we consider worth curing or pallating.
Not that I think we're wrong. (That would be pretty hypocritical, seeing as how I have one of those conditions myself—specifically, nearsightedness.) But I do think that making the leap from seeing a trait as evolutionarily designed in to seeing it as something we should do is, at the very least, in need of a sharply critical look at the trait itself.
That is not to say that I disagree with his conclusions. It just means that I think they should be considered on their own merits; the extent to which they seem reasonable from an evolutionary point of view is pretty much irrelevant.
I think I mostly agree with those conclusions. I've long thought that the legal fiction that we call a corporation is mostly not performing the societal function it was created to, the more so the larger the corporation. I hadn't thought of it from the point of view of human group size, though, and it's a very thought-provoking way of looking at it.
There are many other cases where human grouping size is important. For example, communism is a reasonably good system—for small groups. It works routinely for roommates and families. But I think it's no coincidence that successful communes have generally been fairly small; strain starts to show up in communist systems as the numbers get up into the several tens, falling apart somewhere around 100, maybe a bit higher. Every system much above that, from thousands up to millions or more, has not actually been very communist, no matter how much it may promulgate (and perhaps even believe) the fiction that it is. I've long thought that this happens at about that point because that's approximately the point at which the people involved can no longer know all the other people personally, though I have no evidence for that theory beyond the (weak) evidence of the numerical coincidence.
And, thinking about the various human groupings I participate in (and have participated in), it's very interesting to think about how the group dynamics have varied with group size.
This needs more thought.