REASON AND LANGUAGE. 183 



true that we can know by reflection that we have had 

 sense-impressions which did not, when we received 

 them, rise into consciousness ; but such impressions were 

 not and could not be knowledge, but only some of the 

 conditions of knowledge. Consciousness must accom- 

 pany knowledge, but it need only be direct conscious- 

 ness, and need by no means be reflex j^^-consciousness. 



Mr. Romanes fully admits "that no animal can 

 possibly attain to these excellencies of subjective life," 

 but this he assures us we shall find to be due to " the 

 absence in brutes of the needful conditions to the 

 occurrence of these excellencies as they obtain in our- 

 selves. From which," he tells us,* " it follows that the 

 great distinction between the brute and the man really 

 lies behind the faculties both of conception and predica- 

 tion : it resides in the conditions to the occurrence of 

 either." 



These conditions Mr. Romanes thinks to find in 

 external circumstances, while we see clearly they reside 

 in difference of kind or innermost nature. According to 

 him, as we shall see, mere animals may give names, 

 and his Nominalism tells him that whatever creature 

 possesses names, possesses concepts also; since the latter 

 are, for him, nothing but names. 



But if a non-speaking, poorly-gesturing, unintel- 

 lectual creature said " Di " when it saw a bear, how could 

 that utterance, accompanying its plexus of sense-im- 

 pressions, give it a power of " objectifying " that plexus ? 

 But a creature endowed with an intellectual faculty, yet 

 unable to say even " Di," would be able by gesture to 

 * pp. 175) 176. 



