Is it, however, true? Almost infallibly one is assured upon the first page of every textbook on English grammar that "language is a means of communication." And from this indubitable, though somewhat shadowy declaration, we should naturally expect to proceed by observing certain cases in which an idea is conveyed from one mind to another and analyzing the process as reflected in the language used.

The office of various elements involved in this communication would then presumably be noted, and the elements defined on this basis. That is, one would analyze a sentence as the unit of language, to discover the parts of speech. But instead of this the accredited procedure up to a very recent date, both for textbook and for teacher, has been first to define the parts of speech in turn and then proceed to join certain of them together in such fashion as to make what was called a sentence. A noun, that is a word, representing a person or thing was prefixed to a verb, that is, a word standing for an action or a state of being, and behold a sentence! Thirty years ago pupils were not infrequently required to manufacture sentences after this method, of which a reductio ad absurdum appears in the following "direction" taken from Reed and Kellogg's Higher Lessons in English.⁴ "Unite the words in columns 2 and 3 below [auxiliaries appear in column 2, past participles in column 3] and append the verbs thus formed to the nouns and pronouns in column 1 so as to make good sentences." And the implication that this is the typical sentence-structure, that language is a mechanical aggregation of separate elements, appears continually in the definitions and rules current during this period.

In his Essentials of English Grammar⁵, Whitney assures us that the parts of speech must be "joined" together, "in order to make a whole, in order to be speech." "For a sentence," he declares further, "there must be not only words of more than one kind, but words of certain kinds, fitted together in certain ways." (The italics are, of course, mine.) Nor can we "make a complete sentence without joining together a subject and a predicate."

There could be no more unequivocal statement of the conception of language as a mechanical aggregation of separate words. And I might quote interminably from Whitney's contemporaries, even, I regret to say, from some modern textbooks also, equally direct implications of this conception.

It is doubtless true that to the grammarian of an older generation there was no apparent inconsistency between this e pluribus unum conception of sentence-structure and the statement that language communicates the speaker's thought, since to the crude psychology which he had inherited from a still earlier time thought itself was "a thing of shreds and patches." One was, indeed, supposed to think first "house" and then "burning" and then put these two thoughts together before he could think--or say--"the house is burning." Granting this as a true account of the structure of thought, language might with entire consistency be described as a similar adding of word to word.

But we all know now, that whatever else this splicing of "percept" to "concept" may be, it is not in the genuine sense of the term "thought," any more than a leaf, a stem and a root tied together are a plant. The leaf, the stem and the root are found in the plant as the ideas of house and burning are found in the thought that the house is on fire; but as the plant is a living growth, which has put out root, stem, and leaf, so the thought is an organic structure out of which its constituent ideas have developed. From a confused sense of something wrong perhaps as one suddenly wakes out of sleep, grows the single thought of the whole situation, namely, the house's being on fire, in which neither the house as such nor the act of burning as such have any separate existence. The thought is, in truth, one before it is many. The growing plant or animal is its fair analogy, not the mosaic or the stone wall.

This organic conception of thought the present generation of English grammar teachers have gained from psychology and from real logic. And further, all that we know of the structure of language from modern philologists and students of literature goes to show that it, too, is a living, growing thing, not in any sentimental or remotely analogical sense, but as sober, scientific fact. The sentence which is spliced together out of the "parts of speech" is, in truth, no sentence at all. It is not language any more than a company drill is fighting, or a scarecrow a man. Thought which is living, growing, organic in structure, cannot be conveyed or represented by a lifeless, static, artificial construction. Nor are we studying language by studying such a construction. The sentences which grammar presents to us have in very truth ceased to be language, once they have been cut off from all reference to the various acts of thought-communication which gave rise to them, so that they seem to exist in and for themselves, mere mechanical congeries of words, brought together only to fulfil certain arbitrary requirements of the sentence form as such.

That an artificial conception of the sentence similar to this, and directly at variance with our best knowledge of its nature and structure at the present time, has conditioned much teaching of English grammar in the past, seems to be indubitable. And that this false conception has actually been conveyed to pupils through their study of English grammar I also believe. A priori we should, indeed, expect it to be so. The mind of the child is extraordinarily sensitive to the images latent in our phrases. Professor Scott's paper on the "Figurative Element in Grammatical Terminology"⁶ discloses some quite unforeseen conclusions drawn by the young pupil from the uses in grammar of such supposably abstract and wholly technical terms as "case," "agree," "govern," "decline." And it is hardly conceivable that he should be insensitive to the suggestions of mechanical aggregation offered by such words as "joined with," "fitted together," "added to," "put with," "put together with," or "put along with," which in the older textbook are continually applied to the relations of words with one another in the sentence.

⁴ Lesson II.
⁵ Chap. ii.
⁶Leaflet No. 36, published by the New England Association of Teachers of English.

Continued on page three