REASON AND DIVERS TONGUES. 279 



is no proof whatever that the bird "names." The bird 

 may, on seeing a dog, be thereby excited to emit the 

 sound the emission of which it had previously associ- 

 ated with the feelings aroused by the dog's presence. 

 Supposing the bird to have a consentient, unconscious 

 craving* for the sight of the dog, the automatic 

 emission of the sound would then be abundantly 

 accounted for by such past association. It would be 

 an unconscious employment of a means to an end 

 sensuously craved after. The subsequent history, or 

 outcome, in the case of the child, gives us reason to 

 suppose that it really named at first, because it indu- 

 bitably " names " afterwards. In the case of the parrot 

 this kind of evidence tells the other way. 



Reversing, then, Mr. Romanes's concluding observa- 

 tions, t we say : brief and imperfect as our criticism of 

 Mr. Romanes's position has been, we are honestly 

 unable to see how the testimony of consciousness and 

 observation combined could have been more uniform, 

 multifarious, consistent, complete, and overwhelming, 

 than we have found it to be. In every single case 

 the witness of philology has agreed with the teaching 

 of psychology. The faculty of language being a power 

 living in us, directly and circumstantially narrates to us 

 the necessary conditions of its own origin and evolution. 

 It has told us that even if we suppose there was once 

 a time when men were altogether speechless, and able 

 to communicate with one another only by means of 

 gesticulation and grimace, that yet bodily and facial 

 expression were the expressions of conceptual thought. 



* See " On Truth," pp. 200, 350. t pp. 357, 359- 



