§ 416. The sentence is usually the limit of grammatical analysis. Conrbinations of sentences have never got adequate attention on the part of linguists. Yet the necessity of extending linguistic analysis beyond the bounds of the sentence has of late been frequently emphasized. 2
1 H. Gleason writes: "The difference between The boy ran away
and The boy did run away is not a matter of the presence or the absence
of did. Only of the stress position. Did is there only to provide a meaning
less carrier for that stress in the required position If anything else
were available did would not occur. Compare The boy will run away.
The boy will run away". (Op cit p 174—175).
2 See K. Pike, op. cit., p 30- "We are forced to insist that linguistic
analysis must take as part of its essential domain the treatment of
units larger than the sentence. Without these higher-level units there
are not available adequate matrices for determining sentences themselves."
See also Z. Harris. Structural Linguistics. Preface for the Fourth Impression: "Exact linguistic analysis does not go beyond the limits of a sentence: the stringent demands of its procedures are not satisfied by the relations between the sentence and its neighbors, or between parts of one sentence and parts of its neighbors. There are however structural features which extend over longer stretches of each connected
We should naturally consider the analysis of a wordincomplete without its combinability. But for some reason the combinability of sentencesis not regarded important. One might think that each sentence is an absolutely independent unit, that its forms and meanings do not depend on its neighbours in speech. But it is not so. As H. Kufner has it, "In a very real sense very few groups of words which we would unanimously punctuate as sentences can really be called complete or capable of standing alone ... Most of the-sentences that we speak ... are dependent on what has been said before". 1
It goes without saying that in a book of this kind the uninvestigated problem of the combinability of sentences cannot get adequate treatment. We can only point out some lines of approach.
§ 417. As we have already noted (§ 399), the demarcation line between a sentence and a combination of sentences is very vague. Some part of a simple or composite sentence may become detached from the rest and pronounced after a pause with the intonation of a separate sentence. In writing this is often marked by punctuation. Here are some examples from A Cup of Tea by Mansfield.
She'd only to cross the pavement.
Give me four bunches of those.
Give me those stumpy little tulips.
But still she waited. And that jar of roses. Those red and white ones.
The connection between such sentences is quite evident. The word-combination those red and white ones can make a communication only when combined with some sentence whose predication is understood to refer to the word-combination as well.
But even in case a sentence has its own predication, it may depend on some other sentence, or be coordinated with it, or otherwise connected, so that they form a combination of sentences. In the first of the examples above this connection is expressed by the conjunction but. The following sentences are connected by the pronominal subjects.
piece of writing or talking. These can be investigated by more differentiated tools."
1 H. Kufner, op. cit., p. 1—2.
Rosemary had been married two years. She had a duck of a boy ... They were rich. (Mansfield). l
The sentences below are connected by what we might be tempted to call 'pronominal predicates', and by the implicit repetition of the notional predicate (group) of the first sentence.
Come home to tea with me. Why won't you? Do. (Mansfield).
The second sentence might be extended at the expense of the first into Why won't you come? or even Why won't you come home to tea with me! Similarly, the third sentence is understood by the listener as Do come, or Do come home to tea with me. .
§ 418. We find no predication in the second sentence of the following dialogue.
— How is the little chap feeling?
— Very sorry for himself. (Galsworthy).
But this is not a sentence of the Rain type, with a zero predication. Here we know the subject, it is the chap-oi the first sentence. And we know the structural predicate is. So the person who asked the question perceived the answer as if it had the predication fully expressed: The little chap is very sorry for himself.
Traditionally sentences like very sorry for himself, with some part (or parts) left out are called incomplete or elliptical. But as a matter of fact they are quite complete in their proper places in speech. They would become incomplete only if isolated from the sentences with which they are combined in speech, i. e. when regarded as language units with only paradigmatic relations, without syntagmatic ones. 2
When a speaker combines a sentence with a previous sentence in speech, he often leaves out some redundant parts
1 Speaking of the 'definite restrictions on order' found in sequences
lojiger than sentences, H. Gleason writes: "For example, John came.
He went away, might imply that John did both. But He came. John
went away, certainly could not have that meaning." (Op. cit., p. 57.)
2 J. Hughes writes: "Meaningless expressions like I'm sorry, he
isn't. Yes, completely, become quite easy to classify if we may count as
part of the immediate utterance anything in the utterance just made by
the previous speaker. — Is Mr B. in> I'm sorry, he isn't (in). — Was
the factory destroyed by the fire? Yes, (it was) completely (destroyed by
the fire)." (The Science of Language. N. Y., 1962;.
that are clear from the foregoing sentence, otherwise speech would be cumbersome. A sentence is thus often reduced to one word only.
— Where are you going, old man? — Jericho. (Galsworthy).
— What have you got there, daddiest? — Dynamite. (Shaw).
Theoretically, one and the same sentence may be represented differently in speech, depending on the sentence it is combined with. Suppose, we take the sentence John returned from Moscow yesterday. If this sentence is to be the answer to Who returned from Moscow yesterday? it may be reduced to John. As an answer to When did John return from Moscow? it may be reduced to Yesterday. In answer to Where did John return yesterday from? it may take the form of Moscow. Thus, John. Yesterday. Moscow, may be regarded as positionally conditioned speech variants of a regular two-member sentence. In this they differ from one-member sentences (§ 406).
The sentence on which such a speech variant depends may be called the head-sentence of which it is an adjunct.
§ 419. The sentence-words yes and no are regularly used as adjuncts of some head-sentences.
— "Have you been talking to Hilary?" — "Yes." (Gals
worthy).
— "I've never really got over my first attack." — 'Wo",
said Dinny with compunction. (Ib).
In the same function we find the typically English short predications of the '/ do' type.
— "/'// go, Dinny, if Hallorsen will take me." — "He
shall". (Ib.)
Sometimes the two go together.
— "He wouldn't want me." — "Yes, he would." (Ib.).