speaks of the cases which express one or many, though
with him _case_, or _ptosis_, had a very different meaning from what it
has in our grammars. The terms singular and plural were not invented till
they were wanted, and they were first wanted by the grammarians.
Zenodotus, the editor of Homer, was the first to observe the use of the
dual in the Homeric poems, and, with the usual zeal of discoverers, he has
altered many a plural into a dual when there was no necessity for it.

The scholars of Alexandria, therefore, and of the rival academy of
Pergamus, were the first who studied the Greek language critically, that
is to say, who analyzed the language, arranged it under general
categories, distinguished the various parts of speech, invented proper
technical terms for the various functions of words, observed the more or
less correct usage of certain poets, marked the difference between
obsolete and classical forms, and published long and learned treatises on
all these subjects. Their works mark a great era in the history of the
science of language. But there was still a step to be made before we can
expect to meet with a real practical or elementary grammar of the Greek
language. Now the first real Greek grammar was that of _Dionysius Thrax_.
It is still in existence, and though its genuineness has been doubted,
these doubts have been completely disposed of.

But who was Dionysius Thrax? His father, as we learn from his name, was a
Thracian; but Dionysius himself lived at Alexandria, and was a pupil of
the famous critic and editor of Homer, Aristarchus.(69) Dionysius
afterwards went to Rome, where he taught about the time of Pompey. Now
here we see a new feature in the history of mankind. A Greek, a pupil of
Aristarchus, settles at Rome, and writes a practical grammar of the Greek
language--of course, for the benefit of his young Roman pupils. He was not
the inventor of grammatical science. Nearly all the framework of grammar,
as we saw, was supplied to him through the labors of his predecessors from
Plato to Aristarchus. But he was the first who applied the results of
former philosophers and critics to the practical purpose of teaching
Greek; and, what is most important, of teaching Greek not to Greeks, who
knew Greek and only wanted the theory of their language, but to Romans who
had to be taught the declensions and conjugations, regular and irregular.
His work thus became one of the principal channels through which the