> If your subjects half-suspect that you are deceiving them, what are you really measuring? You are measuring how people react when they suspect that they are

> If your subjects half-suspect that you are deceiving them, what are you really measuring? You are measuring how people react when they suspect that they are being deceived by an authority figure. And the result is that many of them obey you anyway. As people did, and continue to do, in countless dictatorial regimes throughout history. Many North Koreans probably know that the Kims have been lying to them about all sorts of things. Kim also knows that they know. And yet the regime continues to spew lies, and ordinary people continue to pretend to buy those lies. I'm pretty sure that Milgram's experiment failed to account for all sorts of confounding factors. Scientifically, it might or might not tell us anything useful about human nature. But when it comes to the dark side of social psychology, those very factors can help make the conclusion even more relevant to the real world. In a sense, Milgram's team was experimenting not only on their subjects but also on themselves. They showed that a certain combination of authority, deception, and context is highly effective at producing obedience. Hopefully, their result will help us design future political institutions in such a way that it becomes very difficult for any single person or group to obtain that kind of combination.