42 BUREAU OF AMERICAN ETHNOLOGY [bull. 40 



of the pronoun is as weak as in the Siouan languages, to which I have 

 referred before, the definiteness of the pronominal forms of the verb, 

 to which we are accustomed, is entirely lost. Tluis it happens that 

 in the Sioux the verb alone may be used as well for the more or less 

 abstract idea of verbal action as for the third person of the indicative. 



Much more fundamental are the existing differences in regard to 

 the occurrence of tenses and modes. We are accustomed to verbal 

 forms in which the tense is alw^ays expressed with perfect definite- 

 ness. In the sentence The man is sick we really express the idea, 

 The single definite man is sick at the 'present time. This strict expres- 

 sion of the time relation of the occurrence is missing in many 

 languages. The Eskimo, for instance, in expressing the same idea, 

 will simply say, single man sick, leaving the question entirely open 

 whether the man was sick at a previous time, is sick at the present 

 time, or is going to be sick in the future. The condition here is 

 similar to the one described before in relation to plurality. The 

 Eskimo can, of course, express whether the man is sick at the present 

 time, was sick, or is going to be sick, but the grammatical form of 

 his sentences does not require the expression of the tense relation. 

 In other cases the temporal ideas may be expressed with much greater 

 nicety than we find in our familiar grammars. Generally, languages 

 in which a multiplicity of tenses are found inchide in their form of 

 expression certain modifications of the tense concept which might be 

 called "semi-temporal," like in'choatives, which express the beginning 

 of an action ; duratives, which express the extent of time during which 

 the action lasts; transitionals, which express the change of one state 

 of being into another; etc. There is very little agreement in regard 

 to the occurrence of such tenses, and the characteristics of many 

 languages show that tenses are not by any means required for clear 

 expression. 



Wliat is true of tenses is also true of modes. The number of 

 languages which get along with a single mode, or at most with the 

 indicative and imperative, is considerable; although, in this case also, 

 the idea of subordination may be expressed if it seems desirable to 

 do so. 



The few examples that I have given here illustrate that many of 

 the categories which we are inclined to consider as essential may be 

 absent in foreign languages, and that other categories may occur as 

 substitutes. 



