find expression through the characters of the scene, are not allowed to
be circumscribed by them; they are his own deepest convictions.

[Footnote 8: No one would attempt to infer Stephen Phillips' eschatology
from the setting of his _Christ in Hades_.]

[Footnote 9: Vergil indeed was careful to warn the reader (VI, 893) that
the portal of unreal dreams refers the imagery of the sixth book to
fiction, and Servius reiterates the warning. On the employment of myths
by Epicureans see chapter VIII, above.]

[Footnote 10: See Heinze, _Epische Technik_, pp. 82 ff.]

[Footnote 11: This Vergil indicates repeatedly: _Aen_. V, 737; VI, 718,
806-7, 890-2.]

It has frequently been said that Vergil's philosophical system is
confused and that his judgments on providence are inconsistent, that in
fact he seems not to have thought his problems through. This is of course
true so far as it is true of all the students of philosophy of his day.
Indeed we must admit that with the very inadequate psychology of that
time no reasonable solution of the then central problem of determinism
could be found. But there is no reason for supposing that the poet did
not have a complete mastery of what the best teachers of his day had to
offer.

Vergil's Epicureanism, however, served him chiefly as a working
hypothesis for scientific purposes. With its ethical and religious
implications he had not concerned himself; and so it was not permitted
in his later days to interfere with a deep respect for the essentials of