574 GRAMMAR OF THE KLAMATH LANGUAGE. 



On the other side, the Khimath verb differs from the true predicative 

 verb, and rangesiitself among the noun-verbs of agghxtinative languages by 

 the following characteristic features: 



a. The transitive verb is controlled and 'modified by \ts object (espe- 

 cially its direct object), and not by its subject. This becomes chiefly appar- 

 ent by the way in which the distributive form of the verb is applied. In 

 many intransitive verbs, this form connects itself with subjects standing in 

 the plural number; but, from the study of Morphology, it becomes evident 

 that the true cause of the reduplicative process in this instance lies in the 

 repetition or severalty of an act or state, and not in the grammatic number 

 of the subject. 



b. The verb possesses no personal inflection, if we except the rudiment- 

 arv agglutination to it of some personal pronouns. It has no real personal 

 pronoun of the third person. It has a grammatic form for two tenses only, 

 and the modal inflection is rudimentary also. As to number, a sort of 

 pi-efix-inflection is perceptible in a long series of verbs, which tends to 

 prove their nominal nature. That part of the verbal inflection, which is 

 developed more extensively than all the others, is made up by the verbals, 

 which, by themselves, are nominal forms. 



c. Several suffixes, inflectional and derivational, serve for the inflection 

 and derivation of the noun, as well as for that of the verb. The fact that 

 certain nouns can become preterital by inserting -u-, shows better than 

 anything else can, the imperfect diff'erentiation between the noun and 

 the verb. 



d. For the passive voice, the same form is used as for the active voice; 

 shlca is to see and to be seen. 



c. Some verbs are used as nouns without change — that is, without as- 

 suniino' the derivational suffix -sh, -s of substantives. But the existence of 

 the binary and ternary case-inflection shows that the inflectional, polysyn- 

 thelic power of the noun, theoretically, almost equals the power of affixation 

 in the verb. The mere possibility of a binary and ternary case-inflection 

 proves that some of the Klamath case-signs are of the material kind of 

 affixes, and not of the relational kind, which are not susceptible of any fur- 

 ther affixation to themselves, l^he inflective languages have relational case- 



