GEXERAL SUMMARY AXD COXCLUDIXG REAfARRS. 419 



generality is so low as for the most part to belong to that 

 which I had previously called " lower concepts," or " named 

 recepts." Next, that they all bear intrinsic testimony to their 

 own comparatively recent origin, and, therefore, are "primi- 

 tive" only in the sense of representing the last result of philo- 

 logical analysis: they certainly are very far from primitive 

 in the sense of being aboriginal. Again, that they are all 

 of the nature of verbs was shown to be easily explicable ; 

 and, lastly, the fact that none of them betray any imitative 

 source is not to be wondered at, even on the supposition 

 that onomatopoeia entered largely into the composition of 

 aboriginal speech. For, on the one hand, we saw that in the 

 struggle for existence among aboriginal and early words, 

 those only could have stood any chance of survival — i.e. of 

 leaving progeny — which had attained to some degree of 

 connotative extension, or "generality;" and, on the other 

 hand, that in order to do this an onomatopoetic word must 

 first have lost its onomatopoetic significance. A large body of 

 evidence was adduced in support of the onomatopoetic theory 

 and certain objections which have been advanced against it 

 were, I think, thoroughly controverted. Later on, however, 

 we saw that the question as to the degree in which onomato- 

 poeia entered in to the construction of aboriginal speech is 

 really a question of secondary interest to the evolutionist. 

 Whether in the first instance words were all purely arbitrary, 

 all imitative, or some arbitrary and some imitative, — in any 

 case the course of their subsequent evolution would have been 

 the same. By connotative extension in divergent lines, 

 meanings would have been progressively multiplied in those 

 lines through all the progeny of ever-multiplying terms — ^just 

 in the same way as we find to be the case in " baby-talk," and 

 as philologists have amply proved to be the case with the 

 growth of languages in genera' 



That speech from the first should have been concerned 

 with the naming of generic ideas, or higher recepts, as well 

 as with particular objects of sense, is what the evolutionist 

 would antecedently expect. It must be remembered that the 



