REASON AND LANGUAGE. 155 



can be fully accounted for without the presence of any- 

 real " understanding " or " knowledge " at all. Such 

 associations (cited from remarks made by Dr. Samuel 

 Wilks, F.R.S.) as those between the sight of certain 

 persons and sounds or phrases such a bird has heard 

 them utter, or between the sight of the coachman and 

 the words " half-past two," generally said to him when 

 he comes for orders, or between the sound of drawing 

 a cork with a corkscrew and the sight of a bottle, etc. 

 — all such phenomena of association are most easy to 

 understand and are fully to be accounted for without 

 the presence of any faculty higher than that of con- 

 sentience. 



But after thus admitting the position we contend for, 

 Mr. Romanes pfoceeds to retract his admissions,* with- 

 out saying or appearing to be the least aware that he is 

 so doing. He says, " In designating as 'vocal gestures' 

 the correct use (acquired by direct association) of proper 

 names . . . and short phrases, I do not mean to dis- 

 parage the faculty which is displayed. On the con- 

 trary, I think this faculty is precisely the same [!] as 

 that whereby children first learn to talk. . . . The only 

 difference is that, in a few months after its first com- 

 mencement in the child, this faculty develops into pro- 

 portions far surpassing those which it presents in the 

 bird, so that the vocabulary becomes much larger and 

 more discriminative. But the important thing to attend 

 to is that at first, and for several months after its 

 commencement, the vocabulary of a child is always 

 designative of particular objects, qualities, actions, or 



* p. 133. 



