they are derived from conceptual roots, or, to put it differently, that our words are concepts. If therefore it is admitted that we cannot think except in concepts, it will be easy enough to understand why we cannot think except in words.

Are concepts possible without words?

But you may say, Cannot a concept exist without a word? Certainly not, though in order to meet every possible objection, we may say that no concept can exist without a sign, whether it be a word or anything else. And if it is asked, whether the concept exists first, and the sign comes afterwards, I should say No; the two are simultaneous: but in strict logic, the sign, being the condition of a concept, may really be said to come first. After a time, words may be dropt, and it is then, when we try to remember the old word that gave birth to our concept, that we are led to imagine that concepts come first, and words afterwards.

Berkeley.

I know from my own experience how difficult it is to see this clearly. We are so accustomed to think without words, that is to say, after having dropt our words, that we can hardly realise the fact that originally no conceptual thought was possible without these or other signs. No strong man, unless he was told, would believe that originally he could not walk without leading-strings. Berkeley seems to have struggled all his life with this problem, and honest as he always is, he gives us the most contradictory conclusions at which he arrived from time to time. It was one of the fundamental principles of his philosophy that

B b

concepts, or what were then called general ideas, are impossible except by attaching a word or sign to a percept, or what he called a particular idea. Hence he knew that concepts were impossible without words, and discursive thought impossible without concepts. But in spite of that he was often very angry with these words, and in the Introduction to his Treatise concerning the Principles of Human Understanding' (1710) he wrote: Since therefore words are so apt to impose on the understanding [I am resolved in my enquiries to make as little use of them as possibly I can]: whatever ideas I consider I shall endeavour to take them bare and naked into my view, keeping out of my thoughts, so far as I am able, those names which long and constant use have so strictly united with them 1.'

1

Again, in his Common-place Book (Works, ed. Fraser, vol. i. p. 152), he says: 'If men would lay aside words in thinking, 'tis impossible they should ever mistake, save only in matters of fact. I mean it seems impossible they should be positive and secure that anything was true which in truth is not so. Certainly I cannot err in matter of simple perception. So far as we can in reasoning go without the help of signs, there we have certain knowledge. Indeed, in long deductions made by signs there may be slips of memory.' Having thus delivered his soul against words-the very signs without which concepts, as he shows, were impossible, or which were at all events strictly united with our thoughts-he breaks forth in another place (vol. iv. p. 455) in the following

1 The Irish bull, enclosed in brackets, was omitted in the second edition.

6

panegyric: Words (by them meaning all sorts of signs) are so necessary, instead of being (when duly used or in their own nature) prejudicial to the advancement of knowledge, or an hindrance to knowledge, that without them there could in mathematiques themselves be no demonstration.'

It seems to me that most modern philosophers are just in the same state of haziness with regard to the relation between thought and language as Berkeley was; only they are not quite so honest towards themselves. The Bishop, for instance, in another passage of his Common-place Book (vol. iv. p. 429), after having satisfied himself that it would be absurd to use words for recording our thoughts to ourselves or in some private meditations,' interpellates himself by adding the following note, 'Is discursive thought, then, independent of language?' He forgot that he had given the answer himself, namely, that it was not and that it could not be.

Process of naming.

Suppose we see the same colour in snow, milk, chalk, and linen. We cannot single it out, take it away or abstract it from the different sensuous objects in which it occurs, unless we have a sign or handle to do it with, and that sign, for all the ordinary purposes of thinking, is a word, such as white. Until that word is there, we may have different sensations, but no concepts, not even percepts, in the true sense of the word. It is the electric spark of the word which changes something common to various sensations into a percept, as, afterwards, it changes something common to various percepts into a concept,

and something common to various concepts into a higher concept.

But whence came that electric spark? Where did men find that sign to signify many things; and did not that sign already, in order to be applicable to different perceptions, require something of a comprehensive or conceptual character?

Origin of concepts.

Yes, it did. And here lies the punctum saliens of the whole philosophy of language. Long before the question was asked, how man came in possession of words, there was the old question, how man came in possession of concepts. Nearly all philosophers drew the line of demarcation between man and beast at concepts. Up to concepts the two seemed alike.

Then the question arose, How did man alone go beyond percepts and arrive at concepts ?

The usual answer was that man possessed some peculiar gift or faculty which enabled him to form concepts, and to comprehend the manifold as one. Even now many philosophers are satisfied with that mythology. But this answer is no answer at all. We might as well say that man began to write because he had the faculty of writing. We want to know what forced man to form concepts, whether he liked it or not. Why should he not have been satisfied with what the senses gave him, with seeing this, with hearing that? Why should he have gone beyond the single images and looked for the general? He might have been very happy in the world of sensations, perceptions, and images. Why should he ever have left it?

What we call the roots of language betray the secret. Almost all of them express, as we saw, the common acts of man. Now before man is conscious as yet of any object, as an object, he cannot help being conscious of his own acts, and as these acts are mostly repeated and continuous acts, he becomes conscious, without any new effort, of his many or repeated acts as one. Here lies the genesis of the most primitive and, I may add, the first inevitable concepts they consist in our consciousness of our own repeated acts as one continuous action. To rub, for instance, was not only to rub once, and then again, and then again, but it was the continuous act of rubbing, afterwards of smoothing, softening, appeasing; and thus the root, meaning originally to rub, came in time to mean to appease the anger of the gods. There is an uninterrupted chain or development between our saying, Oh God, have mercy! and our earliest ancestors' saying, Be rubbed down, be smooth, be softened, ye gods!

Former theories,

It will now perhaps become clear why the three old theories of the origin of language and thought, the Pooh-pooh theory, the Bow-wow theory, and the Yo-heho theory, completely fail to explain what has to be explained, namely, how conceptual words arose. Cuckoo would be an imitation of the sound of the cuckoo, bow-wow of the barking of the dog, poohpooh of our contempt, yo-heho of our labour; but with all this we should never get out of the enchanted circle of mere sensuous knowledge. We want conceptual sounds. How can we get them?