CONCLUSIONS
Scientists who search for a TOE, or philosophers who advocate such a search, offer very general empirical or a priori reasons for thinking that one does or must exist and can be constructed (“in principle”).
They also offer very general methodological reasons for their search, saying that the practice of science presupposes the existence and constructability of some TOE, or that it is or should be an aim of science to produce one, or at least it would be good if one existed, since otherwise the world would not be completely intelligible. These general reasons are the ones I have been criticizing. They are not strong enough to show that the claims made about the existence and desirability of a TOE are anything more than speculations or, to use Weinberg's term, “dreams.” Nor, considered even as speculations or dreams, are the reasons offered for them strong enough to show that these are good speculations—ones worth pursuing—or so I have argued.Some anti-reductionists (e.g., Cartwright and Dupre) emphasize the complexity of the world as a reason, whether empirical or a priori, for rejecting the strong form of reduc- tionism presupposed by TOE enthusiasts. My argument with TOE theorists is not that there is some good empirical or a priori reason for thinking that a TOE does not exist. (I would not accept the idea that complexity is such a reason.) My argument is that TOE theorists have presented no good general reason for thinking that a TOE does or may well exist. Nor have they presented any plausible reason to conclude that if one doesn't exist, the world is not completely intelligible in a way that matters to science.
This does not mean that string theory, or Nagel's panpsychic theory, or any other purported TOE should not be pursued. My claim is that if such theories are pursued, these endeavors should not be based on very general considerations of the sort mentioned here, but on ones specific to the theory.
Those who work on string theory do so, and should, on the grounds that this theory, or some variation of it, may solve various problems in physics of the sort noted by Smolin; or on the grounds that there is some empirical reason to think that strings of the sort postulated do exist. It is up to string theorists to investigate how many different phenomena such a theory might correctly explain. But the only way to determine this is to produce the problems and the explanations the theory might offer. Those who want to take up the challenge of developing Nagel's dream theory should do the same (except, unlike string theorists, they must first tell us a lot more about what the theory says).Nor does my argument show that scientists and philosophers who (in Weinberg's words) “dream of a final theory” should forget their dreams and pursue what is possible. Perhaps they should. But what I am arguing is that the attempt to discover a final theory should not be based on the speculation that some TOE must exist, and that if it doesn't, the world is not completely intelligible. That speculation, typical of TOE enthusiasts, does not deserve high marks, even as a speculation. If you think there is a TOE, then find a potential TOE and give arguments that support the claim that it is true and that it is a TOE, even if those arguments do not provide evidence sufficient for belief. It is one thing to dream of a final theory, but it is another to pursue that dream on the basis of the very dubious, and unnecessary, speculation that one does or has to exist.