In Chapter 7 of his book The Fabric of Reality, David Deutsch has a dialogue in which he discusses critical rationalism with a crypto-inductivist, a philosopher who thinks there is an induction shaped hole in his worldview although he agrees that “inductive inferences” don’t work. Deutsch points out (pp. 150-153) that theories like “general relativity holds except for Gerry” who will not fall toward the Earth when unsupported, which are often said to pose a problem for critical rationalism aren’t a problem because they can be ruled out as bad explanations since they add unexplained complications to theories that solve problems. As part of this argument, Deutsch points out that changing our language to replace fall with x-fall which means “fall when unsupported, unless you’re Gerry, in which case you can float” doesn’t change this. Somebody who thought in terms of x-falling would still wonder why he will die when he x-falls although Gerry won’t and Deutsch states that English has fall rather than x-fall because languages implicitly contain explanations and in this respect English contains good explanations about falling.
In his pro-induction book Hume’s Problem on pp. 98-99, Colin Howson writes:
Deutsch considers a rather special grue variant to gravitational theory, in which a single exception is made to the rule that all bodies fall when unsupported (1997 : 151), but his argument applies to the more general form we have been considering as well. The argument is that such ‘theories’ are not explanatory because they postulate unexplained exceptions to a rule. To the objection that they can be made syntactically universal by introducing appropriate predicates, like ‘grue’, Deutsch replies that these merely conceal the fact that unexplained anomalies are being postulated, a fact which ordinary English, which evolved to express faithfully what is genuinely problematic and what is not, makes clear (p.153). Thus there is, claims Deutsch, a relevant asymmetry: currently accepted theory is explanatory in a way that the grue variants to it are not. It is strange to find the authority of Popper of all people, a thinker vehemently opposed to ‘ordinary-language’ arguments, being enlisted in such an enterprise. But Deutsch is really doing no more than restate the Goodman criterion of projectability based on what is entrenched in common concepts, and the same objection applies that we brought against that: science introduces unconventional concepts, a fact that Deutsch, himself something of a scientific revolutionary, should be the first to appreciate.
Here, Howson is missing the point. The current theory of gravity is conjectured to be true and it is a good explanation. It states that gravitational attraction is due to curvature of the gravitational field and works out the consequences of this conjecture in detail. This theory doesn’t refer to a particular place and time. The reason for framing the theory this way is to make it easier to criticise. Also, if the theory is true then it applies everywhere and explains all gravitational phenomena, so it’s a better explanation. None of this has anything to do with justification. The point of Deutsch’s argument about English is that he is explaining a feature of English using the standard theory, not using that feature of English to prove the standard theory. The fact that Howson sees none of this serves as an interesting illustration of Popper’s point that misinterpretation is unavoidable no matter how clearly you write.
