theories of the ancients respecting language put together.

The dialogue hardly derives any light from Plato's other writings, and
still less from Scholiasts and Neoplatonist writers. Socrates must be
interpreted from himself, and on first reading we certainly have a
difficulty in understanding his drift, or his relation to the two other
interlocutors in the dialogue. Does he agree with Cratylus or with
Hermogenes, and is he serious in those fanciful etymologies, extending over
more than half the dialogue, which he seems so greatly to relish? Or is he
serious in part only; and can we separate his jest from his earnest?--Sunt
bona, sunt quaedum mediocria, sunt mala plura. Most of them are
ridiculously bad, and yet among them are found, as if by accident,
principles of philology which are unsurpassed in any ancient writer, and
even in advance of any philologer of the last century. May we suppose that
Plato, like Lucian, has been amusing his fancy by writing a comedy in the
form of a prose dialogue? And what is the final result of the enquiry? Is
Plato an upholder of the conventional theory of language, which he
acknowledges to be imperfect? or does he mean to imply that a perfect
language can only be based on his own theory of ideas? Or if this latter
explanation is refuted by his silence, then in what relation does his
account of language stand to the rest of his philosophy? Or may we be so
bold as to deny the connexion between them? (For the allusion to the ideas
at the end of the dialogue is merely intended to show that we must not put
words in the place of things or realities, which is a thesis strongly
insisted on by Plato in many other passages)...These are some of the first
thoughts which arise in the mind of the reader of the Cratylus. And the
consideration of them may form a convenient introduction to the general
subject of the dialogue.

We must not expect all the parts of a dialogue of Plato to tend equally to