The hyperbolic discounting hypothesis has pushed us beyond both documented fact and common sense. Yes, the hyperbolas themselves are well-established facts, and yes, people do suffer from persistent motivational conflicts that conventional utility theory can't explain. We do talk sometimes about arguing with ourselves. Nevertheless, we experience ourselves as basically unitary; and if bundling choices into mutually dependent sets is central to the process of intending things, it has gotten amazingly little recognition.

In common speech “personal rules” don't mean the same thing as “willpower.” They sound trivial, like guidelines for deportment – something a given person might not even have. I've applied the term to something much more central in human decision making. However, the very existence of personal rules as I've hypothesized isn't proven. So far the only evidence I've presented for them is that (1) hyperbolic discount curves predict a limited war relationship among interests, and (2) bundling choices together has been observed to increase patience.

Even our one robust experimental finding is counterintuitive: that the valuation process is based on hyperbolic discount curves, and hence is prone to extreme instability because of a tendency for decisions to reverse simply because time has passed. It's not that we don't observe the problems that these curves predict: The human bent for self-defeating behavior has been in the forefront of every culture's awareness. This, after all, is sin, or the weakness of the flesh, or the mistaken “weighings” of options that Socrates complained of.