As characteristic examples we may quote the titles of some poems and novels: "To a Skylark" (SHELLEY) ,"On a Distant Prospect of Eton College" (GRAY), "Of Human Bondage" (MAUGHAM), "Under the Greenwood Tree" (TH. HARDY). The syntactical function of the prepositions in cases of this type is a peculiar one. The preposition either expresses a relation between the thing expressed by the noun and something not mentioned in the text (as in "To a Skylark"), or it gives the characteristic of the place where something not specified takes place ("Under the Greenwood Tree").
It is evident that in such cases the preposition has only a onesided connection, namely with the noun following it, but we may ask whether it has not also some reference to something not expressed which may be imagined as standing before the preposition.
Let us, for instance, compare the actual title of W. Somerset Maugham's novel, "Of Human Bondage", with a possible variant "Human Bondage", without the preposition. In this way the meaning and function of the preposition become clear: the preposition of is here used as it is used in the phrases speak of something, think of something, etc. In the title as it stands, the preposition implies that the author is going to speak of human bondage, that is, human bondage is going to be discussed. 1
We shall arrive at a similar conclusion if we compare the actual title of Th. Hardy's novel, "Under the Greenwood Tree", with the possible variant "The Greenwood Tree". The preposition implies that we shall be reading about something happening under the tree, rather than about the tree itself. So it will probably be right to say that something is implied (very vaguely, it must be admitted).
We should especially note some peculiar uses of the preposition about, namely in such sentences as, There were about twenly people in the room, which of course means that the number is given approximately. The preposition here has only a one-sided connection, namely with the numeral, and has no connection at all with the preceding verb. It certainly does not express any relation between were and twenty. Syntactically, it makes an element of the subject group (about twenty people). Indeed we may be inclined to doubt whether the word about is a preposition at all in such a case. It rather approaches the status of a particle.
This is still more confirmed by examples in which the group introduced by about stands after another preposition, as in the
1 The title is actually a translation of Spinoza's title "De servitude humana" (a book of his "Ethics"), but this is irrelevant for our analysis.
Special Uses of Prepositions 153
sentence, This happened at about three o'clock. The group about three o'clock here follows the preposition at in quite the same way as the group three o'clock would follow it in the sentence This happened at three o'clock. The group about three o'clock is a designation of a certain time as much as the group three o'clock, and to establish its relation with the verb happened it also requires the preposition at to be used.
We also find two prepositions close to each other in different contexts. Compare, for instance, the following sentence: He sat until past midnight in the darkness while grief and sorrow overcame him. (E. CALDWELL) Here also belongs the phrase from under in a sentence like The cat stretched its paw from under the table. It seems quite possible to take this in the same way as we took at about in the preceding example, and to say that under the table denotes a certain place and from indicates movement from that place. However, it is also possible to view this case in a somewhat different way, namely to suppose that from under is a phrase equivalent to a preposition, and then we should not have two prepositions following one another here. This problem should be further investigated.
Prepositions can sometimes be followed by adverbs, which apparently become partly substantivised when so used. The groups from there, from where, since then, since when are too widely known to require illustrative examples. Another case in point is the following: She is beautiful with that Indian summer renewal of physical charm which comes to a woman who loves and is loved, particularly to one who has not found that love until comparatively late in life. (O'NEILL)
Prepositions in English are less closely connected with the word or phrase they introduce than, say, in Russian. It would be impossible in English for a preposition to consist of a consonant only, that is, to be non-syllabic, which is the case with the three Russian prepositions в, к, с. This greater independence of English prepositions manifests itself in various ways.
There is the possibility of inserting, between a preposition and the word or phrase it introduces, another phrase, which can, in its turn, be introduced by a preposition. Here is an example of this kind: The first of these, "The Fatal Revenge", appeared in 1807, and was followed by, among other, "The Milesian Chief"... (COUSIN) The two prepositions, by and among, stand one after the other, but there is certainly no syntactic connection between them, and probably there is a pause, corresponding to the comma of the written text. The connection between followed and by appears to be closer than that between by and the phrase which it introduces, namely, "The Milesian Chief". Unless it were so, the preposition by would come after the inserted phrase among others, rather than
154 The Preposition
before it. But that variant, though perhaps not impossible, would certainly be less idiomatic than that in the text.
This way of making one preposition come immediately after another, showing the independence of the first preposition, is also seen in some cases where the status of the second preposition may be doubted, that is, it may be doubted whether the word is really a preposition in that context (compare what has been said on p. 152). The following sentence, which is fairly characteristic of modern usage, will show the essence of the phenomenon: His industry was marvellous, and its results remain embodied in about 40 books, of which about 25 are commentaries on books of Scripture. (COUSIN) Of course all this is made possible by the fact that prepositions in English do not require the word they introduce to have a specified case form.
Sometimes even a parenthetical clause come between the preposition and the noun it introduces, e. g. Some weeks ago Mr Blessington came down to me in, as it seemed to me, a state of considerable agitation. (CONAN DOYLE)
The looseness of the tie between the preposition and the following noun can be offset by a closer tie between the preposition and the preceding word. This may be seen, for instance, in some passive constructions with the phrase "verb + noun + preposition" acting as a kind of transitive unit. Examples of this use are well known. Compare the following sentence: Their conference was put an end to by the anxious young lover himself, who came to breathe his parting sigh before he set off for Wiltshire. (J. AUSTEN) The active construction would have been, The young lover put an end to their conference, where an end would be a non-prepositional, and to their conference a prepositional object. It might be argued, however, that put an end is something of a phraseological unit and should therefore be treated as the predicate. Be that as it may, the fact remains that the noun end is included into the passive form of the verb, and the subject of the passive construction is the noun which, in the active construction, would have been part of the prepositional object.
It should also be noted that a preposition does not necessarily connect the word which immediately precedes it with the one that follows. Cases are frequent enough in which there is no connection at all between the preposition and the preceding word. For instance, in the sentence, This beauty is a trifle dimmed now by traces of recent illness (O'NEILL) there is no connection between the words now and by. The preposition by is of course connected with the passive participle dimmed and the adverb now could be left out without affecting the connections and the functions of the preposition: This beauty is dimmed by traces of recent illness. The same may be said about the sentence I get the same tale of woe from
Functions of Prepositions 155
every one in our part of the country (Idem); the preposition from is not connected with the noun woe which precedes it, it is connected with the verb get, which is separated from it by five other words. Many more examples of this kind might be given. This should warn us against an oversimplified understanding of the syntactical function of a preposition.
Special attention must be given to groups of words whose meaning and functions in the sentence are the same as those of prepositions. Here belong the groups out of, as to, as for, instead of, in spite of, etc. We cannot term these groups prepositions, since a preposition is a word, not a word group, and it is essential to keep up the distinction between words and word groups; neglect of it would bring about a muddle both in grammar and in lexicology. The current haziness in the treatment of such groups and the vague terms "compound preposition" and the like are not conducive to a clear and consistent grammatical theory. Since much the same can be said about phrases equivalent in meaning and function to conjunctions, we will return to this problem after having considered the conjunctions.
Chapter XIX
THE CONJUNCTION
Taking up the definition of a conjunction given above in cur general survey of parts of speech, we must first of all, just as we have done with prepositions, consider the question of the meaning of conjunctions. Many authors, in defining a conjunction, limit themselves to indicating that they serve to connect words (or parts of the sentence) and clauses. 1 This would seem to imply that conjunctions have no meaning of their own, that is, that they do not themselves express any phenomena of the extralinguistic world. This is untenable, as may be very easily shown by the simplest examples. Compare, for instance, the two sentences, He came because it was late, and He came though it was late. The different conjunctions obviously express different real relations between two extralinguistic phenomena: his coming and its being late. The causal connection between them exists outside the language, and so does the concessive relation expressed in the latter of the two sentences. There is no difference whatever in the grammatical structure of the two sentences: the difference lies only in the meanings of the two conjunctions. The same observation can be made on comparing the two sentences, We will come to see you before he comes back, and We will come to see you after he comes back, and also in a number of other cases. All this goes to prove that every conjunction has its own meaning, expressing some connection or other existing between phenomena in extralinguistic reality.
So far our reasoning and our conclusions have been the same as in the case of prepositions. Now, however, comes a point in which conjunctions are different from prepositions. When discussing prepositions, we noted that in a certain number of cases the use of a given preposition is predicted by the preceding word: thus the verb depend can only be followed by the preposition on (or upon), the adjective characteristic only by the preposition of, etc. In such cases the preposition has no meaning of its own. Conjunctions in this respect are entirely different. The use of a conjunction is never predicted by any preceding word. We will no longer inquire into the meanings of conjunctions, as this is a question of lexicology rather than grammar.
In studying the syntactical functions of conjunctions, we have, just as with prepositions, to distinguish between two levels — that of phrases and that of sentences.
On the phrase level it must be said that conjunctions connect words and phrases. It is the so-called co-ordinating conjunctions that are found here, and only very rarely subordinating ones.
1 See, for example, Грамматика русского языка, т. I, стр. 665.
Prepositions and Conjunctions 157
On the sentence level it must be said that conjunctions connect clauses (of different kinds). Here we find both so-called co-ordinating and so-called subordinating conjunctions.
The division of conjunctions into co-ordinating and subordinating is one that can hardly be dealt with outside syntax: co-ordinating conjunctions imply co-ordination of clauses, and subordinating conjunctions imply subordination of clauses. So we shall have to look again into this question when we come to syntax. 1 Here it will be sufficient to say that there is nothing in the conjunction itself to show whether it is co-ordinating or subordinating, and even in the structure of the clauses there is no unmistakable sign of this (as is the case, for instance, with word order in Modern German).
Conjunctions can sometimes lose their connecting function, as is the case with the conjunction if in sentences expressing wish, like the following: If only she might play the question loud enough to reach the ears of this Paul Steitler. (BUECHNER) Probably we shall have to say that if here is no longer a conjunction but a particle. We will consider such cases in Syntax as well.2
PREPOSITIONS AND CONJUNCTIONS
In comparing prepositions with co-ordinating and subordinating conjunctions we cannot fail to notice that while prepositions have nothing in common with co-ordinating conjunctions, some prepositions are very close in meaning to subordinating conjunctions, and in some cases a preposition and a subordinating conjunction sound exactly the same. As examples of similarity in meaning we may give, for instance, such phrases and clauses: during his illness = while he was ill', examples of complete identity in meaning and sound are the words before, after, since.
All this presents us with intricate problems. On the one hand, it seems doubtful whether we are right in uniting subordinating conjunctions (that is, words like when, as, after, before, since) together with co-ordinating conjunctions (that is, words like and, but, or) into one part of speech and separating them from prepositions (that is, words like of, from, after, before, since), with which they obviously have much more in common. On the other hand, it remains doubtful how we should treat the relations between the preposition after and the conjunction after (and similarly, before and since). None of the treatments so far proposed seems satisfactory.
One way is to say, there is the word after, which may function both as a preposition and as a conjunction. But then the question
1 See below, p. 315 ff.
2 See below, Chapter XXXVII, p. 293 ff,
The Conjunctions
arises, what part of speech is after? If it can only function as a preposition and as a conjunction, this would mean that it is neither the one nor the other.
Another way is to say that after the preposition and after the conjunction are homonyms. This will not do either, since homonymy, by definition, supposes complete difference of meaning, as between saw 'instrument for sawing' and saw 'old saying', whereas the meaning of after the preposition and after the conjunction is absolutely the same.
These considerations apply as well to the words before and since, and here the question is further complicated by the fact that they can also be adverbs. 1
The difficulty with the word after would be overcome if we were to unite prepositions and conjunctions into one part of speech (as hinted above, p. 33), which would then have to be given a new name. The difference between what we now call the preposition after and the conjunction would then be reduced to different syntactical uses of one word. But the difficulty with the adverbs and preposition-conjunctions before and since would not be solved by this: it would not do to say that an adverb and a word uniting the qualities of preposition and conjunction are the same word.
A fully convincing solution of this problem has yet to be found.
As to the relation between prepositions, co-ordinating conjunctions, and subordinating conjunctions, it must be said that on the ground of the peculiarities which have been pointed out a completely different treatment of the three types of words is possible. An idea to this effect was put forward by the French scholar L. Tesnière in a book on general principles of syntax. Tesnière classes what are usually called co-ordinating conjunctions as a type for itself: he calls them "jonctifs" (that is, junctives), whereas prepositions and what we call subordinating conjunctions come together under the name of "translatifs" (translatives) and are distinguished from each other as subclasses of this large class: prepositions are called "translatifs, premier degré" (translatives, first degree) and subordinating conjunctions, "translatifs, second degré" (second degree).2 This is quite natural in a book on syntax, in which things are looked at from a syntactical angle and words classified according to their functions in the sentence.
It should also be noted that the difference between prepositions and conjunctions is much less pronounced in Modern English than in Russian, where prepositions are closely connected with cases, while conjunctions have nothing whatever to do with them. In English, with its almost complete absence of cases, this difference be-
1 After is also an adverb in the phrase ever alter.
2 L, Tesnière, Elements de syntaxe structural, 1959, pp. 386—387,
Prepositions and Conjunctions. 159
tween prepositions and conjunctions is very much obliterated. While in Russian the substitution of a conjunction for a preposition makes jt necessary to change the case of the following noun, in English 00 such change is necessary or, indeed, possible. So the distinction between preposition and conjunction is based here only on semantic criteria and, also, on the use of these words in other contexts, where they are not interchangeable.
In discussing prepositions, we noted that there are in English, as well as in Russian and in other languages, certain phrases which cannot be termed prepositions, since they are not words, but which are similar to prepositions in meaning and in syntactical function. The same is true of conjunctions. A certain number of phrases (consisting of two or three words) are similar in meaning and in function to conjunctions. Among them we can quote such phrases as in order that, as soon as, as long as, notwithstanding that, etc. Just as prepositional phrases, these will be analysed in a special chapter in Syntax (see p. 179 ff.).
Chapter XX
THE PARTICLE
To include a word in the class of particles we must find out whether it has the characteristic features of particles which we have described in our general survey of parts of speech, and we should not apply any other criteria. We shall not inquire whether the word has one syllable, or two, or many; this phonetic quality of a word is irrelevant to its grammatical status: just as, for example, a preposition may have one syllable (of, to) or four (notwithstanding), a particle may have one syllable (just) or four (exclusively). Thus the diminutive suffix -icle should not be taken to refer to the length of the word.
In dealing with particles, we will limit ourselves to the grammatical side of the matter. We will not discuss either their meanings, which belong to the sphere of lexicology, nor the morphemes making them up, which should be considered in the theory of word-building.
When speaking of particles in our review of parts of speech we have noted already that they usually refer to the word (or, sometimes, phrase) immediately following and give special prominence to the notion expressed by this word (or phrase), or single it out in some other way, depending on the meaning of the particle.
This usage, which is by far the most common one, can be illustrated by a variety of examples. We will give a few: One just does what is reasonable, and everything is bound to go all right. (R. WEST) She could feel anger stir, even at this late date, as she thought of that night, but she subdued it and tossed her head until the earrings danced. (M. MITCHELL)
Sometimes a particle occupies a different position in the sentence. This question will be dealt with in the chapter on word order.
The question of the place of a particle in sentence structure remains unsolved. It would appear that the following three solutions are possible: (1) a particle is a separate secondary member of the sentence, which should be given a special name; (2) a particle is an element in the part of the sentence which is formed by the word (or phrase) to which the particle refers (thus the particle may be an element of the subject, predicate, object, etc.); (3) a particle neither makes up a special part of the sentence, nor is it an element in any part of the sentence; it stands outside the structure of the sentence and must be neglected when analysis of a sentence is given. 1
1 The latter view is the common one. It is put forward by the authors of the book Грамматика русского языка (т. I, стр. 639).
Function of Particles 161
Each of these three views entails some difficulties and none of them can be proved to be the correct one, so that the decision remains arbitrary.
The view that a particle is a part of the sentence by itself makes it necessary to state what part of the sentence it is. Since it obviously cannot be brought under the headings either of object, or attribute, or adverbial modifier, we should have to introduce a special part of the sentence which ought then to be given a special name.
The second view would be plausible if the particle always stood immediately before (or immediately after) the word or phrase to which it belongs. But the fact that it can occasionally stand at a distance from it (for example, within the predicate, while referring to an adverbial modifier) makes this view impossible of realisation; compare, for instance, I have only met him twice.
The last view, according to which a particle stands, as it were, outside the sentence, seems rather odd. Since it is within the sentence, and is essential to its meaning, so that omission of the particle could involve a material change in the meaning, it is hard to understand how it can be discounted in analysing the structure of the sentence.
Since, then, the second view proves to be impossible and the third unconvincing, we shall have to adhere to the first view and to state that a particle is a separate secondary part of the sentence which ought to be given a special name.
THE PARTICLE Sot
The particle not deserves special attention. It can, as is well known, be used in two different ways. On the one hand, it may stand outside the predicate, as in the following sentence: Not till Magnus had actually landed in Orkney did he consider the many difficulties that confronted him. (LINKLATER) It also stands outside the predicate in a type of so-called short answers, in which the negative is expressed by the particle not, if it is accompanied by a modal word like certainly, perhaps, or a phrase equivalent to a modal word, e. g. of course: Certainly not. Perhaps not. Of course not. 1 Compare also: / am afraid not, I think not, etc. In these cases the particle not appears to be the main part of the sentence.
Another use of the particle not is that within the predicate. In these cases it is customary to treat it as part of the verb itself. The usual way of putting it is this. The negative form of the present indicative, e. g., of the verb be, is: (/) am not, (he) is not, etc., or, the negative form of the present indicative, e. g., of the verb
1 The use of these modal words and phrases with the sentence-word no is impossible.
6 Б. А Ильиш
162 The Particle
sing is, (I) do not sing, (he) docs not sing, etc. The particle not is thus treated as an auxiliary element making part of the verb form. This of course appears to be especially necessary with verbs whose negative form includes the auxiliary verb do, i. e. with the vast majority of Modern English verbs. Here the particle has obviously no syntactic function of its own, and is an auxiliary element within the morphology of the verb. 1
The particle not undergoes further fusion with forms of the verb in the following cases, where indeed it is no longer a word at all but a morpheme within a verb form. The first step in this direction is clearly seen in the form cannot, where it preserves its vowel sound, and the next step in the contracted forms isn't, aren't (also the subliterary ain't), wasn't, weren't, haven't, hasn't, hadn't, shan't, won't, shouldn't, wouldn't, don't, doesn't, didn't, mayn't, mightn't, mustn't, oughtn't, can't, and occasionally also usen't for used not. Here the two elements have quite coalesced into a unit, and some of these forms (e. g. shan't, won't, and don't) cannot now even be divided into morphemes.
DOUBTFUL WORDS
There are some words which may be classed either as particles or as adverbs, since the criteria which we apply to distinguish between these two parts of speech do not appear to yield a clear result here.
Among these we should cite the words almost and nearly, which are close to each other in meaning. Taking a sentence like The boat almost overturned, we can say that it is a matter for discussion whether the word almost does or does not denote the manner in which the action of the verb was conducted. Again, talcing the sentence He is nearly thirty years old now, we can also doubt whether the word nearly does or does not modify the word thirty (or, perhaps, the phrase thirty years). It would rather seem that it does not, but any judgement on this issue is bound to be subjective to a considerable extent, since, as we pointed out above, objective criteria do not yield any clear results. Accordingly, the syntactical function of the words almost and nearly will also remain doubtful and a matter for subjective opinion.
In weighing different considerations that may be put forward in favour of including the word nearly into one or other morphological category, it is essential to bear in mind a phenomenon which quite definitely speaks against including this word in the class of particles. The word nearly may occasionally have the adverb very
1 See above, p. 125 ff.
Doubtful Words 163
standing before it and modifying it, as in the sentence: The time is very nearly seven fifteen. In the sentence The time is nearly seven fifteen we might bring forward certain arguments to prove that nearly is a particle. However, the possibility of its being modified by the adverb very is a powerful argument against that view: a particle cannot be modified by an adverb, or by any other kind of word, for that matter. Since the status of the word nearly was doubtful anyway, the phrase very nearly casts a definite weight against its being a particle and in favour of its being an adverb.
