interchange, e.g. to sing — song; to feed — food; full — to fill,etc. which lie outside the scope of word-formation in Modern English.
The conception of conversion as derivation with a zero-morpheme, however, merits attention. The propounders of this interpretation of conversion rightly refer to some points of analogy between affixation and conversion. Among them is similarity of semantic relations between a derived word and its underlying base, on the one hand, and between words within a conversion pair,
e.g. 1. action — doer of the action: to walk — a walker(affixation) to tramp — a tramp(conversion);
2. action — result of the action: to agree — agreement(affixation), to find — a find(conversion), etc.
They also argue that as the derivational complexity of a derived word involves a more complex semantic structure as compared with that of the base, it is but logical to assume that the semantic complexity of a converted word should manifest itself in its derivational structure, even though in the form of a zero derivational affix.
There are also some other arguments in favour of this interpretation of conversion, which for lack of space cannot be considered here.
If one accepts this conception of conversion, then one will have to distinguish between two types of derivation in Modern English: one effected by employing suffixes and prefixes, the other by using a zero derivational affix.
There is also a point of view on conversion as a morphological-syntactic word-building means,1 for it involves, as the linguists sharing this conception maintain, both a change of the paradigm and a change of the syntactic function of the word, e.g. I need some good paper for my roomsand He is papering his room.It may be argued, however, that as the creation of a word through conversion necessarily involves the formation of a new word-stem, a purely morphological unit, the syntactic factor is irrelevant to the processes of word-formation proper, including conversion.
Besides, there is also a purely syntactic approach commonly known as a functional approach to conversion. Certain linguists and lexicographers especially those in Great Britain and the USA are inclined to regard conversion in Modern English as a kind of functional change. They define conversion as a shift from one part of speech to another contending that in Modern English a word may function as two different parts of speech at the same time. If we accept this point of view, we should logically arrive at the conclusion that in Modern English we no longer distinguish between parts of speech, i.e. between noun and verb, noun and adjective, etc., for one and the same word cannot simultaneously belong to different parts of speech. It is common knowledge, however, that the English word-stock is subdivided into big word classes each having its own
1 See, for instance, I. V. Arnold. The English Word. L. — M., 1973.
semantic and formal features. The distinct difference between nouns andverbs, for instance, as in the case of doctor — to doctordiscussed above, consists in the number and character of the categories reflected in their paradigms. Thus, the functional approach to conversion cannot be justified and should be rejected as inadequate.
§ 17. Synchronic Approach
Conversion pairs are distinguished by the structural identity of the root and phonetic identity of the stem of each of the two words. Synchronically we deal with pairs of words related through conversion that coexist in contemporary English. The two words, e.g. to breakand a break,being phonetically identical, the question arises whether they have the same or identical stems, as some linguists are inclined to believe.1 It will be recalled that the stem carries quite a definite part-of-speech meaning; for instance, within the word-cluster to dress — dress — dresser — dressing — dressy,the stem dresser — carries not only the lexical meaning of the root-morpheme dress-,but also the meaning of substantivity, the stem dressy-the meaning of quality, etc. These two ingredients — the lexical meaning of the root-morpheme and the part-of-speech meaning of the stem — form part of the meaning of the whole word. It is the stem that requires a definite paradigm; for instance, the word dresseris a noun primarily because it has a noun-stem and not only because of the noun paradigm; likewise, the word materialiseis a verb, because first and foremost it has a verbal stem possessing the lexico-grammatical meaning of process or action and requiring a verb paradigm.
What is true of words whose root and stem do not coincide is also true of words with roots and stems that coincide: for instance, the word atomis a noun because of the substantival character of the stem requiring the noun paradigm. The word sellis a verb because of the verbal character of its stem requiring the verb paradigm, etc. It logically follows that the stems of two words making up a conversion pair cannot be regarded as being the same or identical: the stem hand-of the noun hand,for instance, carries a substantival meaning together with the system of its meanings, such as: 1) the end of the arm beyond the wrist; 2) pointer on a watch or clock; 3) worker in a factory; 4) source of information, etc.; the stem hand-of the verb handhas a different part-of-speech meaning, namely that of the verb, and a different system of meanings: 1) give or help with the hand, 2) pass, etc. Thus, the stems of word-pairs related through conversion have different part-of-speech and denotational meanings. Being phonetically identical they can be regarded as homonymous stems.
A careful examination of the relationship between the lexical meaning of the root-morpheme and the part-of-speech meaning of the stem within a conversion pair reveals that in one of the two words the former does not correspond to the latter. For instance, the lexical meaning of the root-morpheme of the noun handcorresponds to the part-of-speech meaning of
1 See, for instance, А. И. Смирницкий. Лексикология английского языка. М., 1956, с. 71 — 72, also О. С. Ахманова. Некоторые вопросы семантического анализа слов. — Вестн. МГУ, 1957, № 2, с. 70.
its stem: they are both of a substantival character; the lexical meaning of the root-morpheme of the verb hand,however, does not correspond to the part-of-speech meaning of the stem: the root-morpheme denotes an object, whereas the part-of-speech meaning of the stem is that of a process. The same is true of the noun fallwhose stem is of a substantival character (which is proved by the noun paradigm fall — falls — fall’s — falls’,whereas the root-morpheme denotes a certain process.
It will be recalled that the same kind of non-correspondence is typical of the derived word in general. To give but two examples, the part-of-speech meaning of the stem blackness— is that of substantivity, whereas the root-morpheme black-denotes a quality; the part-of-speech meaning of the stem eatable-(that of qualitativeness) does not correspond to the lexical meaning of the root-morpheme denoting a process. It should also be pointed out here that in simple words the lexical meaning of the root corresponds to the part-of-speech meaning of the stem, cf. the two types of meaning of simple words like black a, eat v, chairn, etc. Thus, by analogy with the derivational character of the stem of a derived word it is natural to regard the stem of one of the two words making up a conversion pair as being of a derivational character as well. The essential difference between affixation and conversion is that affixation is characterised by both semantic and structural derivation (e.g. friend — friendless, dark — darkness,etc.), whereas conversion displays only semantic derivation, i.e. hand — to hand, fall — to fall, taxi — to taxi,etc.; the difference between the two classes of words in affixation is marked both by a special derivational affix and a paradigm, whereas in conversion it is marked only by paradigmatic forms.
§ 18. Typical Semantic Relations
As one of the two words within a conversion pair is semantically derived from the other, it is of great theoretical and practical importance to determine the semantic relations between words related through conversion. Summing up the findings of the linguists who have done research in this field we can enumerate the following typical semantic relations.