trace words to the Saxon, nor to the Celtic, the Greek, the Dutch, the
Mexican, nor the Persian; nor is it his province to explain their
meaning in Latin, French, or Hebrew, Italian, Mohegan, or Sanscrit; but
it is his duty to explain their properties, their powers, their
connexions, relations, dependancies, and, bearings, not at the period in
which the Danes made an irruption into the island of Great Britain, nor
in the year in which Lamech paid his addresses to Adah and Zillah, but
_at the particular period in which he writes_. His words are already
derived, formed, established, and furnished to his hand, and he is bound
to take them and explain them as he finds them _in his day_, without any
regard to their ancient construction and application.

CLASSIFICATION. In arranging the parts of speech, I conceive it to be
the legitimate object of the practical grammarian, to consult _practical
convenience_. The true principle of classification seems to be, not a
reference to essential differences in the _primitive_ meaning of words,
nor to their original combinations, but to the _manner in which they are
at present employed_. In the early and rude state of society, mankind
are quite limited in their knowledge, and having but few ideas to
communicate, a small number of words answers their purpose in the
transmission of thought. This leads them to express their ideas in
short, detached sentences, requiring few or none of those _connectives_,
or words of transition, which are afterwards introduced into language by
refinement, and which contribute so largely to its perspicuity and
elegance. The argument appears to be conclusive, then, that every
language must necessarily have more parts of speech in its refined, than
in its barbarous state.

The part of speech to which any word belongs, is ascertained, not by the
_original_ signification of that word, but by its present _manner_ of