542 BUREAU OF AMERICAN ETHNOLOGY [bull. 40 



§ 57* Irreffular Forms {contitf tied) 



The same is true in all cases where the verb stands immediately 

 before its object or instrument. In these cases, when it has no de- 

 monstrative, the objective -x and -s are suffixed to it. 

 la mix'H'dxa hEgwa'nEm he struck the man 

 Jc'.'e'sdx mix'H'd g'd'xEn this one did not strike me 

 lox Le'gadES TH'sEmg'ite this one had the name T!e'sEmg"it 225.18 



The construction is similar to that in sentences in which nouns 

 occur accompanied by qualifying terms. 



mix'H'deda ^wa'lase h Egwd' nEnnxa g'ind'nEm- this large man struck 

 the child (literally, this tall one struck, man the child). If the 

 noun stand§ by itself, the -a preceding the object (§ 50.4) is 

 retained. 

 mix'H'deda hEgwd'nEmaxa g'indnE7n 



Temporal suffixes are treated in the same manner. 



xn'mtslseda ^wd'lasde g-d'x^sa J^Egwd'nEm. the large house of the 

 man was burnt on the ground (xumt- to burn; -eIs on ground 

 [§22, no. 44]; -{x-)de past; g'ok'^ house); (but xu'mtElse 

 g'd'x^ddsa hEgwdriEm the house of the man was burnt on the 

 ground [see § 50.6]) 



There is still another case in which a similar absence of demonstra- 

 tive elements is observed. The verb may be separated from the rest 

 of the sentence, and its place may be taken by auxiliary verbs or by 

 verbalized nominal ideas. Then it is placed at the end of the sen- 

 tence, and has either no ending, or, better, the ending -a. 



la^Em qd's^ida then he went 



la^mdx qd'sHda then he went 



le'da bEgwd'nEm qd'sHda then the man went 



In this position the verb can not take the ending -e, although it may 

 be made a noun by the appropriate prenominal demonstrative. 

 hEgwd'nEmeda qd'sHde the man went 



§ 58. Metnarks rni Irregtilar^ JPortns 



It is impossible to give a satisfactory explanation for all the peculiar 

 usages of these endings, although the rules for their use can be stated 

 quite definitely. The endings -eda and -x, which in Kwa'g-ul invari- 

 ably have the function of determining subject and object, may have 

 originally performed different functions. This is suggested by the 

 following forms: The Dza'wadEenox" forms -setZa &nd -xeda (see § 49), 

 and the analogous forms -xg-ada, -xdxda, -sg'ada, -soxda, of the 



§§ 57, 58 



