Brinton.] 224 (Oct. 5, 
most primitive tongues. On the other hand, we find in them a great 
many classificatory particles. These correspond only remotely to 
anything known in Aryan speech, and seem far more abstract than 
generic nouns. I will illustrate what they are by an example taken 
from the Hidatsa, a dialect of the Dakota. 
The word for sled in that dialect is m¢da-maidutsada. ‘The first 
part of this compound, mda, means anything of wood or into which 
wood enters. Fire is médé because it is kept up with wood. With 
the phonetic laxity which I have before noted, the first syllable mz 
may as correctly be pronounced @ or w?. It is a common nominal 
prefix, of vague significance, but seems to classify objects as distinc- 
tives. JZa designates objects whose immediate use is not expressed ; 
7 denotes instrument or material; dw, conveys that the cause of the 
action is not specified; ¢sa intimates the action is that of separa- 
ting; da, that this is done quickly (¢sa-da, to slide).* 
Thus by the juxtaposition of one classificatory particle after an- 
other, seven in number, all of them logical universals, the savage 
makes up the name of the specific object. 
This system was probably the first adopted by man when he be- 
gan to set in order his perceptions within the categories of his un- 
derstanding with the aim of giving them vocal expression. It is 
a plan which we find most highly developed in the rudest languages, 
and therefore we may reasonably believe that it characterized pre- 
historic speech. 
The question has been put by psychological grammarians, which 
one of the senses most helped man in the creation of language, or 
to express it in modern scientific parlance, was primitive man a 
visuaire or an auditaire 2 Did he model his sounds after what he 
heard, or what he saw? ‘The former opinion has been the more 
popular, and has given rise to the imitative or ‘‘ onomatopoetic”’ 
theory of language. No doubt there is a certain degree of truth in 
this, but the analysis of American tongues leans decidedly toward 
classing primitive man among the swaires. His earliest significant 
sounds seem to have been expressive of motion and rest, energy and 
its absence, space and direction, color and form, and the like. A 
different opinion has been maintained by Darwin and by many who 
have studied the problems presented by the origin of words from 
a merely physical or physiological standpoint, but a careful investi- 
* Washington Matthews, Grammar and Dictionary of the Language of the Hidatsa 
(New York, 1878). 
