I think that perhaps one of the things that makes us believe that the problem is "bad apples" rather than the situation is that problematic situations don't always produce problematic behaviors. Vernon Smith's team is doing fascinating experiments on the emergence of cooperative behavior and property rights in anarchic systems. It turns out that in most of these systems, there are two possible equilibria: cooperative systems in which everyone gets richer, and systems that stay autarkic, where people stay poor. There's not any particular way of predicting in advance which groups will succeed, no particular awesome people who will obviously make things work. Instead, the outcomes emerge from the interactions between people in the various groups. But even though some groups do develop cooperation, the initial lack of cooperative rules is a "bad" situation--many groups fail to develop cooperative systems.
The fact that some groups succeed and some fail, though, focuses us on the characteristics of the people in the group, rather than the context in which the group operates. Yet in Zimbardo's famous prison experiment, the students were selected for their apparent healthiness on psychological tests, and the researchers decided which would be prisoners, and which guards, by flipping a coin.
Milgram and Zimbardo make the best arguments for libertarianism, not Hayek and Von Mises
Of course they're evil. They're Stanford students.
Go Bears!
"bad apples" rather than the situation is that problematic situations don't always produce problematic behaviors.
Bingo: and he also goes into more detail about to avoid such problems in The Lucifer Effect.
Books like his also explain why "black sites" and the like are a bad idea: no oversight.
"It turns out that in most of these systems, there are two possible equilibria: cooperative systems in which everyone gets richer, and systems that stay autarkic, where people stay poor. There's not any particular way of predicting in advance which groups will succeed, no particular awesome people who will obviously make things work."
Are you really sure about this?
You'd almost think that in the real world, groups of people of a single ethnicity called "nations" would then set up governments (call these things "states"). Maybe you could then look into the success or failure of these "nation states" to develop cooperative systems. I wonder if you would find any differences between, say Germany in 2008 and Gambia in 2008. Hmm, you could even look at differences over time, how about Rhodesia 1962 versus Zimbabwe 2008?
I wonder if that would show you anything about the traits of groups that are more likely to succeed or about the systems and administration of the governance of groups that make them more or less likely to succeed.
Then again, you might find things you'd rather not find. Stick to looking at groups of Stamford students when doing experiments. Much safer for your career that way.
I wonder if you would find any differences between, say Germany in 2008 and Gambia in 2008.
Different things. We are already in 2008, so we know how Germany and Gambia have turned out. The problem is predicting what will happen in the future. How many people back in 1950 predicted that South Korea and Botswana would take off, but India wouldn't?
It's very very easy to construct a logical rational story of how Rhodesia in 1962 inevitably developed into Zimbabwae of 2008 or why Germany in 2008 is richer than Gambia in 2008. But that has nothing to do with whether we can predict *beforehand* how a country will change over the next twenty years, or whether we can predict *beforehand* which group of people will cooperate in a lab. Remember hindsight is 20/20.