When Kuhn says that the “most fundamental aspect of … incommensurability” is “the proponents of competing paradigms practice their trades in different worlds” (Structure, 150) or that a scientific revolution is “a displacement of the conceptual network through which scientists view the world (ibid., 102), it is difficult to understand Kuhn as saying anything other than that incommensurability implies incomparability. With the incomparability of two theories, then theory choice is necessarily irrational, commitment to a paradigm or lexicon follows, and so on.
The assumption that theories are incommensurable can lead to, in this most radical version, parroted by post-modernist college kids all across the globe, ‘conversation-stoppers’, modes of speech that deny the very possibility of an assumption’s falsity, and are self-reinforcing enough that the most committed individual can defend the assumption come what may. In short, it’s far too convenient for anyone to claim that they do not understand the meaning of a sentence.
If someone should deny that they understand a sentence because they assume sentences in different lexicons or paradigms are necessarily incommensurable, then they can refuse to understand empirical sentences that would contradict their scientific theories. Criticism that assumes any different lexicon or paradigm from the theory is, assuming this broadest interpretation of the incommensurability thesis, impossible. Yet, the broadest incommensurability thesis is, at most, debatable. In the worst possible cases, these individuals can refuse to understand sentences that would contradict the metaphysical assumption of the incommensurability of theories. Rational theory-preference is, as they understand it, impossible, and the only available option is commitment.
Oh, what a problem for those that think progress in science is possible!
When criticized, Kuhn could retreat to the more reasonable position that the language in scientific theories made numerous implicit assumptions, and that on occasion these assumptions are not translatable. Kuhn says at one point that two theories are incommensurable in the narrow sense just in case concepts cannot be systematically mapped from one language to another with the same extension and meaning. At another point, Kuhn, after borrowing a page from Quine’s Word and Object, adopts the broader sense of incommensurability that two theories are incommensurable just in case there were no neutral “language into which at least the empirical consequences of both can be translated without loss or change” (Criticism and the Growth of Knowledge, 266). While both are different criteria of incommensurability, neither of the two imply that theories are incomparable, which is exactly opposed to what the post-modernist college kids claim.
There can be two incommensurable theories, each with a disparate lexica, when understood in either the broader or narrower sense expressed above, that still target roughly the same domain, as far as the world-in-itself is concerned. While Ptolemaic and Copernican theories have different concepts of a planet, the predictions of planetary position may be immediately comparable. One need only see what deductive consequences of the theories, coupled with identical background assumptions, make about the approximate location of planets, and then measure the approximate location of the planets. While the broader interpretation of incommensurability does not say that the measuring of the planet’s location is in any way manifestly the case, the theoretical language and theory-impregnated tools used to measure location can be used as an arbitrator between the two, though it is in no way final (Duhem-Quine). Therefore, in light of both the narrow and broad interpretations of incommensurability, if two incommensurable theories may still be compared in their empirical predictions, theory choice is, according to these two interpretations of incommensurability, not necessarily irrational.
In sum, don’t let people take incommensurability when understood in its narrow or broad interpretations to immediately apply to its broadest possible interpretation.