16 LETTERS OF ST. AUGUSTINE. [LET. VIL
tr u tli; or when w e call up to our own minds a liv ely concep
tio n o f the things described w hile w e read history, or hear,
or compose, or refuse to believe fabulous narrations. Thus,
according to m y own fancy, and as it m ay occur to my own
m ind, I picture to m yself the appearance of .^ neas, or of
M edea w ith her team o f winged dragons, or o f Chremes, or
Parmeno.^ To th is class belong also those things which have
been brought forward as true, either by w ise m en wrapping
up some truth in the folds of such inventions, or b y foolish
m en building up various kinds o f superstition ; e.g. the P hle-
gethon of Tartarus, and the five caves of the nation of dark-
ness,“ and th e N orth Pole supporting the heavens, and a
thousand other prodigies o f poets and of heretics. Moreover,
w e often say, when carrying on a discussion, “ Suppose that
three worlds, such as the one w hich we inhabit, were placed
one above anotheror, " Suppose the earth to be enclosed
w ith in a four-sided figure,” and so o n : for a ll such things we
picture to ourselves, and im agine according to the mood and
direction o f pur thoughts. A s for the third class of images,
it has to do chiefly w ith numbers and m easure; w hich are
found partly in the nature o f things, as when the figure of the
entire world is discovered, and an image consequent upon this
discovery is formed in the mind o f one thinking upon i t ; and
partly in sciences, as in geom etrical figures and m usical har
m onies, and in the infinite variety of num erals: which, although
they are, as I think, true in. them selves as objects o f the imder-
standing, are nevertheless the causes of illu sive exercises of
the im agination, the m isleading tendency of w hich reason itself
can only w ith difficulty w ith stan d ; although it is not easy to
preserve even the science of reasoning free from this evil,
since in our logical divisions and conclusions we form to our
selves, so to speak, calculi or counters to facilitate the process
of reasoning.
■ 5. In th is whole forest of images, I believe that you do not
think that those o f the first class belong to the mind previous to
the time when th ey find access through the senses. On this w e
need not argue any further. As to the other two classes a ques
tion m ight reasonably be raised, were it not m anifest that the
1 Dramatis persona iu Terence. * Eefen-ing to Manichaean notions.