220 THE ORIGIN OF HUMAN REASON. 



and thus the whole fortress of my opponents would 

 crumble away." Why, of course, we say there can be 

 concepts without names. We have always strenuously 

 affirmed it, and its affirmation, instead of being destruc- 

 tive to our "fortress," is the very rock on which it is 

 built. Mr. Romanes says * that if his opponents do 

 not "commit argumentative suicide" they must con- 

 cede that the speechless infant is "confined to the 

 receptual sphere of ideation." But instead of conceding 

 this we have strenuously affirmed the very reverse.f 



Having, then, so mistakenly assumed that self-con- 

 sciousness must be reflex, and having attributed to the 

 logical and conceptual gesture-language of children no 

 more value than to the emotional manifestations of 

 brutes, he says J : " The named recepts of a parrot 

 cannot be held by my opponents to be true concepts, 

 any more than the indicative gestures of an infant can 

 be held by them to differ in kind from those of a dog." 



Certainly, we are far indeed from regarding "the 

 named recepts of a parrot" as concepts, but we none 

 the less affirm that "the indicative gestures of an in- 

 fant " are " different in kind from those of a dog " — 

 just as "the indicative gestures" of the arms of a dog 

 are different in kind from those of a telegraph post. 

 External resemblance in action does not prove simi- 

 larity of kind, if there is reason for thinking that 

 the actions are respectively the result of influences 

 which themselves are radically different in kind. The 

 actions as external motions may be similar in appear- 



* p. 225. t See " On Truth," p. 234. 



X p. 226. 



