individuality in the universal cause or nature. In like manner we might
think of the words which we daily use, as derived from the first speech of
man, and of all the languages in the world, as the expressions or varieties
of a single force or life of language of which the thoughts of men are the
accident. Such a conception enables us to grasp the power and wonder of
languages, and is very natural to the scientific philologist. For he, like
the metaphysician, believes in the reality of that which absorbs his own
mind. Nor do we deny the enormous influence which language has exercised
over thought. Fixed words, like fixed ideas, have often governed the
world. But in such representations we attribute to language too much the
nature of a cause, and too little of an effect,--too much of an absolute,
too little of a relative character,--too much of an ideal, too little of a
matter-of-fact existence.

Or again, we may frame a single abstract notion of language of which all
existent languages may be supposed to be the perversion. But we must not
conceive that this logical figment had ever a real existence, or is
anything more than an effort of the mind to give unity to infinitely
various phenomena. There is no abstract language 'in rerum natura,' any
more than there is an abstract tree, but only languages in various stages
of growth, maturity, and decay. Nor do other logical distinctions or even
grammatical exactly correspond to the facts of language; for they too are
attempts to give unity and regularity to a subject which is partly
irregular.

We find, however, that there are distinctions of another kind by which this
vast field of language admits of being mapped out. There is the
distinction between biliteral and triliteral roots, and the various
inflexions which accompany them; between the mere mechanical cohesion of
sounds or words, and the 'chemical' combination of them into a new word;