42 Wisconsin Academy of Sciences, ArtSj and Letters. 



CHAPTER III.— RELATIVE WORDS. 



These are described by Grammar as words which relate to 

 an antecedent, that is, to a principal. But, even if such be the 

 fact, it does not sensibly differentiate them from the already 

 examined vicarious words. Like these the relati 'cs are also a 

 linguistic luxury; they assume both retrospective and prospec- 

 tive attitudes ; they vary in nearness and fidelity to their prin- 

 cipals and in the certainty of their service; they enter the sen- 

 tence now as one and again as another part of speech. Their 

 principals also vary in number, complexity and thought-factor- 

 ship, being sometimes single and sometimes multiple, present- 

 ing sometimes a single idea and sometimes a plexus, and being 

 at no time confined to appearance as any particular part of 

 speech. 



Turning from characteristics shared by relatives with other 

 words, I wish to find a characteristic, if such there be, which 

 distinguishes relatives only. Such, being found, may show 

 whether Grammar is right in ranking relatives as a special pro- 

 nominal class ; whether, to question more deeply, the relatives 

 after all are pronouns ; whether indeed they are properly rated 

 even as words. 



The following inspection, necessarily somewhat microscopic, 

 I wish to appear as part of a broader survey of mental phe- 

 nomena, for the sake of making sure, if possible, that nothing 

 indispensable is overlooked. Conceding then to mental activ- 

 ity the greatest fullness conceivable, I offer as my initial topic, 



I. MULTIPLE THINKING, 



supposing that a mind may work on several lines of thought at 

 once; that two such lines of thought may be parallel, each fail- 

 ing to coincide in any part with the other ; that they may be con- 

 vergent, passing from* parallelism to coincidence; that con- 

 versely they may be divergent; that two lines of thought, one 



