- Jespersen, Negation, p. 144.
- Cf. Hans Reichenbach, Elements of Symbolic Logic (New York, 1947), p. 93.
- It is not claimed that the statement John's talk was neither intelligible nor unintelligible cannot be interpreted by speakers of English so as to yield some message; it might for instance be understood as an oblique expression of the fact that John did not give a talk.
- For the results of a test designed to measure the agreement among native speakers of English in making the distinction, see Appendix (sec. A).
- We are of course only referring to the logical interpretation of such phrases, and are not concerned here with their rhetorical and stylistic use as affirmations by understatement (litotes).
- C. K. Ogden, Opposition: A Linguistic and Psychological Analysis (London, 1932), pp. 46–47.
- Cf. Bertrand Russell, “On Denoting,” reprinted in Bertrand Russell, Logic and Knowledge, ed. R. C. Marsh (New York, 1956), pp. 52–53.
- For the term “nexal negation,” cf. Otto Jespersen, The Philosophy of Grammar (London, 1924), pp. 329–31.
- Cf. for instance the following remarks by Hilary Putnam: “If someone says, ‘Well, just consider the context of philosophic discussion. Suppose one asserted as an abstract truth, “Ideas are green” he would have made a false statement, wouldn't he?’ My position is that he wouldn't have made any statement that I understand at all.” (Hilary Putnam, “Some Issues in the Theory of Grammar,” Structure of Language and Its Mathematical Aspects, ed. Roman Jakobson [“American Mathematical Society: Proceedings of Symposia in Applied Mathematics,” Vol. XII (1961)], p. 31.) For an opposite view cf. A. N. Prior, “Entities,” Australasian Journal of Philosophy, XXXII (1954), 159–68.
- G. H. von Wright, On the Logic of Negation (“Societas Scientiarum Fennica: Commentationes Physico-Mathematicae,” Vol. XXII, No. 4; Helsinki, 1959), p. 4.
- The etymology is taken from Webster's Third New International Dictionary, hereafter referred to as NID3.
- Of 69 adjectives in a-/an- listed in the 1958 edition of Webster's New Collegiate Dictionary (hereafter referred to as NCD), 23 are cited as being exclusively terms from the vocabulary of biology, botany, medicine, optics, psychopathology, physics, and zoology, 3 others have at least one similarly specialized sub-meaning, and of the remaining number many are hardly in very general use (e.g. acaudal, analgesic, anhydrous, athermanous, atrophic). A number of these adjectives have no positive counterparts in English.
- We shall return to this point in Chapter V.
- The Oxford English Dictionary (hereafter referred to as OED) also gives acaudal as being equivalent to acaudate ‘tailless,’ but notes that the latter is “the more correct form.”
- Cf. the following statement: “Dis- combines [among adjectival bases] only with Romance adjs, chiefly such as have a learned or academic tinge. In productivity it cannot compete with un- which is far more common with words of general currency. Though adjs like discomfortable, dissatisfactory, dissocial exist, the commonly used words are uncomfortable, unsatisfactory, unsocial.” (Hans Marchand, The Categories and Types of Present-Day English Word-Formation [Wiesbaden, 1960], p. 113.)
- Cf. Herbert Koziol, Handbuch der englischen Wortbildungslehre (Heidelberg, 1937), p. 115.
- Note the contrast between disinterested and uninterested.
- The criterion of what might be called “hypothetical synchronic derivability” that was used for determining whether a form in the dictionary should be included in the above and in subsequent lists is not an altogether satisfactory device, but a weighing of diachronic factors would raise practically insurmountable obstacles in an investigation such as ours.
- Jespersen, Negation, p. 140.
- Marchand, p. 120.
- Ibid.
- The texts used were: Kingsley Amis, New Maps of Hell: A Survey of Science Fiction (New York, 1960), about 40–45,000 words; three essays from James Baldwin, Notes of a Native Son (Boston, 1955), totaling about 15,000 words; Shirley Jackson, The Lottery (New York, 1949), a collection of short stories totaling about 80–85,000 words; and three essays from David Riesman, Individualism Reconsidered (Glencoe, Ill., 1954), totaling about 17,000 words.
- cr. above, p. 27.
- Jespersen, Negation, p. 142.
- OED, Un-, prefix1, 7.
- For a more detailed presentation of this interpretation, see below, pp. 38–39.
- The productively derived form uncorrupted is of course also a counterexample to the hypothesis (cf. below, pp. 35, 38–39.)
- An examination of one of the texts mentioned above (p. 29, n. 4), New Maps of Hell, sufficed to turn up the following adjectives, not listed in the NCD, the NID,2 and the NID3: non-addicted, non-gifted, non-malevolent, non-sensationalist. None of them seem very unusual.
- Compare the previously cited contrast between unhuman and inhuman (pp. 29–30), or between nonhuman and inhuman.
- 2
- For a discussion of the notion of “degrees of grammaticalness,” see e.g. Noam Chomsky, “Some Methodological Remarks on Generative Grammar,” Word XVII (1961), 233–39.—Such a model of linguistic processes might, if collapsed from a hierarchy of degrees or levels to one level, be compared to an electrical network where the relevant factors for different paths are not only open or closed switches but also resistors of varying degrees of resistance, so that for two open paths the traversing of one may require a greater amount of electrical energy than that of the other.
- This class of course also contains the fully acceptable derivatives in non-; the point is that full acceptability is not specified as criterial for inclusion in it.
- We would of course hope that this “ideal” classification has some testable behavioral correlates. In asking subjects to rank the four forms just cited—i.e. nongood, nondelicious, nonlong, nonelongated—in order of acceptability, we have obtained complete agreement on nonelongated as the most acceptable item. Either nongood or nonlong was generally ranked last; possibly further evidence of this nature might lead us to revise our schema so that the criterion “evaluatively neutral” applies only to those adjectives which have no obvious antonyms. The ranking would then be the following:
- most acceptable: non- derivatives from stems which have no simplex antonyms and are evaluatively neutral;
- less acceptable: non- derivatives from stems which have no simplex antonyms and are evaluatively positive or negative;
- minimally acceptable: non- derivatives from stems which have simplex antonyms.
- A number of the adjectivally used past participles have the interesting characteristic that they are used attributively only if qualified in some way, un- being one of the possible qualifiers. Cf. an uneaten meal, a half-eaten meal, but not, as a rule, an eaten meal.—There are of course a number of forms in un- which have no commonly used unprefixed counterparts at all, such as unassuming, unprecedented (and the customarily cited uncouth, unkempt). Other forms are used in certain contexts where the unprefixed stem does not occur; cf. an unthinking action, unswerving determination, an unerring aim.
- The Introduction to the NID2 contains the following remarks: “A new feature of this edition is the inclusion, in lists,… of words made from prefixes…. These lists contain many words… attested by citations, but never before appearing in any dictionary. Such entries will show the existence of such words. Words… are not entered in these lists unless the meanings can be inferred from the definitions of the separate parts.” (NID2, p. xvi.) The NID3 has abandoned this method of listing.
- We shall not deal here with forms in -ible, since this derivational morpheme is at best of very marginal productivity in present-day English. In most of the forms where -ible occurs we can, however, assign the same semantic interpretation to it that we have proposed for -able, and such forms in -ible usually have derived antonyms; in the great majority of cases the prefix encountered in these antonyms is of course in- rather than un-. We can thus distinguish between a productive pattern un-x-able and an unproductive pattern in-x-ible. (The distinction between the suffixes -able and -ible is of course relevant to written rather than to most varieties of spoken English.)
- This is true to a much lesser degree of terms of the type un-x-ed.
- It should be observed that the claim that e.g. unintelligent is a contrary opposite of intelligent implies nothing about its equivalence or non-equivalence to e.g. stupid. The latter term may well be “stronger” (or more “negative”) without thereby preventing unintelligent from being interpreted as a contrary opposite of the “positive” term on the dimension intelligent—unintelligent.
- OED, Un-, prefix1, 7.
- Above, p. 30.
- NID2, under the second un- entry.—It is incidentally rather odd that the NID2, which gives a statement of productivity for un-, includes many more individual forms in un- in its listing than the NID3. Of course there may be non-linguistic reasons for this, e.g. reasons of economy.
- Marchand, pp. 151–52.
- Cf. John Locke, An Essay Concerning Human Understanding, Bk. II, chap, viii, sec. 9.
- Marchand does take note of the older forms such as unbroad, undeep, cited by the OED (cf. above, p. 40, for the relevant quotation from the OED), but says no more about them than that “the character of these cbs [=combinations] would…have to be investigated.” (Marchand, p. 152.)
- Cf. the quotation from the OED above, p. 40.
- The following quotation might be interpreted as evidence for the productivity of the pattern of derivation with non-: “‘Adlai wanted a Munich,’ says a nonadmiring official.” (Stewart Alsop and Charles Bartlett, “In Time of Crisis,” Saturday Evening Post, December 8, 1962, p. 20.) In this case nonadmiring is used in spite of what I would consider the relative familiarity of unadmiring, which is for instance listed in the NID3. But the main factor here may well have been a stylistic one; non- is probably better journalese than un-.
- Un- was at one time used as a redundant (perhaps intensifying?) negative with such forms. According to the OED, “it is sometimes redundantly prefixed to adjs. ending in -less…. The type… belongs to the later 16th and 17th cent.; among the instances from that period are unboundless, uncomfortless, undauntless… as late as 1786 unquestionless is found, and unrestless exists in modern dialect.” (OED, Un-, prefix1, 5.a.)
WORD
Volume 20, 1964 - Issue sup1: Monograph No.5