THE SCIENCE OF LANGUAGE. 53 



oldest form d/ta, slumbered tlie different 

 grammatical relations,* 'verbal and nominal, 

 with all tbeir modifications, unsevered as yet 

 and undeveloped, as we may observe in those 

 languages that have remained stationary on 

 this simple stage of development. What we 

 have shown by an illustration selected at 

 random, applies to all Indo-Grermanic words. 

 You, and your fellow naturalists, will best 

 understand my argument, when I charac- 

 terize the radical elements as the celis of 

 speech, not yet containing any particular 

 organs for the functions of nouns, verbs, etc., 

 and in which these functions (the grammatical 

 relations,) are no more separated yet than 



* Ernest Renan is, so far as I know, the only glossologist 

 who holds the opinion that all the so-called parts of speech 

 had their respective functions eked out for them, so to say, at 

 the very dawn of language. Does he imagine that they issued 

 forth from an arsenal of human speech as "the blue-eyed 

 maid " burst forth, speared and shielded, from the head-womb 

 of thundering Jove ? — T. 



