This pattern recurs over and over again in many different forms. e.g. High trust societies work really well, low trust societies work really badly because you have to carefully guard against hostile intent in everything you do.

Teams where everyone cares about quality generally find that almost every intervention they make to improve quality improves quality. Teams where nobody cares about quality reliably produce shitty results no matter how many safeguards you put in place.

This makes it very hard to take advice from successful environments and use it to improve unsuccessful environments.

The other obvious conclusion from this is that if you find yourself in a healthy environment you should do everything in your power to keep it that way, and if you find yourself in an unhealthy environment you should focus all your efforts on fixing that instead of its symptoms.

I guess one way to look at this is that healthy environments are ones that are resistant to Goodhart's law. Healthy environments do roughly what you meant to ask them to, unhealthy ones do exactly what you asked them to and don't much care whether you like the way they do it.

@DRMacIver When I use the term "progress" non-specifically, I find I'm attempting to convey the gradual progression from the latter to the former. It's the process of humans escaping from the prisoner's dilemma.

@DRMacIver As a wise friend once said: "The easiest way to improve, is stop doing stupid shit."