1 Lg. 39.170-210 (1963). Hereafter diminished, in length but not in credit, to K-F.

2 This is probably the right place to point out that K-F have tried to protect their flank by disavowing any attempt to build a semantic theory. They undertake no more than to sketch an approach to one. If their examples are matched with convincing counterexamples, they will not be abashed, but will look for better ones or modify some detail of the theory. This of course assumes that the foundations remain intact: it is hard to make alterations in a house whose frame has to be torn out. But the truth of the matter is that without putting together at least a fair sampling of the structure, one can never know whether the theory will account for more than the particular lexical sets from which the examples are drawn. If from a handful of specimens the theory appears as likely to fail as to succeed, there is no defense in promising to patch it up. In this sense, one cannot excuse a taste because it is only a taste. Until more is presented, everything stands or falls by it, and a theory is as subject to criticism in the light of its examples as a principal is liable to damages for the actions of his agent.

3 The problem of marker dominance is not treated by K-F. Presumably there are set sequences of markers, e.g. (Human) vs. (Animal) must precede (Male) vs. (Female) which in turn must precede (Old) vs. (Young), etc. Otherwise it will be impossible to generalize markers to the point where one like (Inferior), for instance, will serve both for an inferior grade of nobility (when dominated by Noble) and for an inferior academic degree (when dominated by other markers relating to such degrees). A more specific marker will have to be used, increasing the number of markers.

4 These markers are easily established by following K-F's lead with old. Their trick for pulling (Young) out of the distinguisher was simply to use its antonym in a sentence: The old bachelor finally died. Similarly, we can take a synonym or an antonym of something in the (Animal) distinguisher, disambiguate with it, and make a new marker: the phocine bachelor (establishing Phocine directly), the hairless bachelor (establishing Hirsute indirectly), etc.

5 They also by implication give at least one sense characterization by distinguisher alone. The word colorful has as one of its paths (198) 'Adjective—(Color)—[abounding in contrast or variety of bright colors]'. If colorful is viewed compositionally in the way that K-F advocate (192), then the only part corresponding to the morpheme -ful is 'abounding in', in the distinguisher. The point is not discussed, and there is no way to tell whether it has any significance.

6 Richard Hofstadter, Anti-inlellectualism in American life 90 (New York), 1963.

7 If it is argued that these are ‘different entries’ because of their etymologies, I would simply get another example. But such an argument would be irrelevant to disambiguation.

8 Lg. 40.568 (1964).

In principle, probably no vocable can be replaced by any other without some sort of reverberation in the system. A displacement of just a few rimes would demolish the poetic framework. If the word too is replaced, say, by plethorly, certain restrictions on the use of too, resulting from conflict of homonyms, would disappear (like excessively, plethorly would remove the asterisk in Is he nice?—*Yes, too; also in *a too large group). If tiny is replaced by perminute, teeny will be left high and dry. If all the one-syllable adjectives are replaced with polysyllables so that the -er comparative is lost, certain semantic distinctions that have arisen as a result of the option to use either comparison will be lost along with it. As Householder points out, Journal of linguistics 1.18 (1965), there are a number of grammars. The grammar in which it makes no difference whether tiny is ‘tiny’ or perminute is only one of them.

9 A bit of wishful thinking that is well founded, as theories go: ‘Economy of description’ (small inventory of symbols) is one of I. A. Mel'chuk's ‘measures of excellence’ according to Householder, ibid. 16.