ly the pictorial image becomes less vivid,
while the association of the nature and habits of the animal is more
distinctly perceived. The picture passes into a symbol, for there would
be too many of them and they would crowd the mind; the vocal imitation,
too, is always in process of being lost and being renewed, just as the
picture is brought back again in the description of the poet. Words now
can be used more freely because there are more of them. What was once an
involuntary expression becomes voluntary. Not only can men utter a cry
or call, but they can communicate and converse; they can not only use
words, but they can even play with them. The word is separated both from
the object and from the mind; and slowly nations and individuals attain
to a fuller consciousness of themselves.

Parallel with this mental process the articulation of sounds is
gradually becoming perfected. The finer sense detects the differences of
them, and begins, first to agglomerate, then to distinguish them. Times,
persons, places, relations of all kinds, are expressed by modifications
of them. The earliest parts of speech, as we may call them by
anticipation, like the first utterances of children, probably partook
of the nature of interjections and nouns; then came verbs; at length the
whole sentence appeared, and rhythm and metre followed. Each stage in
the progress of language was accompanied by some corresponding stage
in the mind and civilisation of man. In time, when the family became a
nation, the wild growth of dialects passed into a language. Then arose
poetry and literature. We can hardly realize to ourselves how much with
each improvement of language the powers of the human mind were enlarged;
how the inner world took the place of outer; how the pictorial or
symbolical or analogical word was refined into a notion; how language,
fair and large and free, was at last complete.

So we may imagine the speech of man to have begun as with the cries of
animals, or the stammering lips of children, and to have attained by
degrees the perfection of Homer and Plato. Yet we are far from saying
that this or any other theory of language is proved by facts. It is
not difficult to form an hypothesis which by a series of imaginary
transitions will bridge over the chasm which separates man from the
animals. Differences of kind may often be thus resolved into differences
of degree. But we must not assume that we have in this way discovered
the true