Gary Becker is often known for his penchant for the hyper-rational. The large body of his work attempts to tease individualistic rationality out of seemingly irrational behavior. In this pursuit, he sought to include all sorts of moral values into the "utility function" to explain things like monogamy, addiction, and suicide. Thus, it might come as a surprise to most to learn that one of his earlier papers made the case that one didn't need any sort of utility function at all in order to derive normal economic behavior.
According to one of my mentors, this paper is the best thing Becker ever wrote. In the 1962 article in the Journal of Political Economy, Becker shows how it is possible to derive normal aggregate demand behavior - all that a market needs to function properly - from any sort of budget maximization. Becker shows that even with completely irrational and inconsistent actors, overall behavior will be consistent enough for firms to react to market demand. What this shows, among other things, is that it's entirely possible (and perhaps likely) that what determines economic behavior is not preference at all, but rather structural constraints on behavior. Of course, for a Chicago economist in the heart of the microfoundations (counter)revolution, such conclusions are unacceptable. So instead, we got Superfreakonomics.