THE WITXESS OF nilLOLOGY. 295 



have had a genesis, — we have next to ascertain whether our 

 deduction admits of corroboration by any inductive evidence 

 supplied by the science of language, as to what this genesis 

 actually icas. 



And here I had better say at once that the results of 

 philological science will be found to carry us back to an even 

 more primitive state of matters than any which I have 

 hitherto contemplated. For, so long as I was restricted to 

 psychological anal)'sis, I was obliged to follow my opponents 

 where they take language as it now exists. In order to argue 

 with them at all upon these grounds, it was necessary for me 

 to consider what they had said on the philosophy of predica- 

 tion ; and, in order to do this, it was further necessary that 

 I should postpone for independent treatment those results 

 of philological inquiry which they have everywhere ignored. 

 But now we have come to the place where we can afford to 

 abandon psychological anal)'sis altogether, and take our stand 

 upon the still surer ground of what I have already termed 

 the palsrontological record of mental evolution as this has 

 actually been preserved in the stratified deposits of language. 

 Now, wlicn we do this, we shall find that hitherto we have not 

 gone so far back in tracing the genesis of conceptual out of 

 receptual ideation as in point of fact we are able to go on 

 grounds of the most satisfactory evidence. 



Up to this time, then, I have been meeting my opponents 

 on their own asstimptions, and one of these assumptions has 

 been that language must always have existed as we now 

 know it — at least to the extent of comprising words which 

 admit of being built up into propositions to express the 

 semiotic intention of the speaker. But this assumption is well 

 known by philologists to be false. As a matter of fact, 

 language did not begin with any of our later-day distinctions 

 between nouns, verbs, adjectives, prepositions, and the rest : 

 it began as the undifferentiated protoplasm of speech, out of 

 which all these "parts of speech" had afterwards to be 

 developed by a prolonged course of gradual evolution. " Die 

 Sprache ist nicht stuckweis odcr atomistisch ; sie ist glcich 

 20 



