1 4 LETTERS OP ST. AUGUSTINE. [ l e t. VIL necessary. For m y purpose it suffices m eanwhile to have proved in th is way that memory can be spoken of as embracing also those things which have not yet passed aw ay: and now mark atten tively how th is supports m y opinion. Some men raise a groundless objection to that m ost famous theory in ­ vented by Socrates, according to which the things that we learn are not introduced to our minds as new, but brought back to memory by a process of recollection; supporting their objection by affirming that memory has to do only w ith things which have passed away, whereas, as Plato him self has taught, those things w hich w e learn by the exercise of the understand­ ing are permanent, and being imperishable, cannot • be num­ bered among things w hich have passed a w a y : the m istake into w hich th ey have fallen arising obviously from this, that they do not consider that it is only the m ental act of apprehension b y w hich w e have discerned these things which belongs to the p a st; and that it is because w e have, in the stream of m ental activity, left these behind, and begun in a variety o f ways to attend to other things, that w e require to return to them by an effort o f recollection, that is, by memory. If, therefore, passing over other examples, w e fix our thoughts upon eternity its e lf as something w hich is for ever permanent, and consider, on the one hand, that it does not require any im age fashioned by the im agination as the vehicle by which it m ay be intro­ duced into the m ind; and, on the other hand, that it could never enter the m ind otherwise than by our remembering it,— w e shall see that, in regard to some things at least, there can be an exercise of memory w ithout any image of the thing remembered being presented by th e imagination. Chap. II- — The mind is destitute of images jgresented hy the imagination, so long as it has not leen infwmed ly the senses of external things. " 3. In the second place, as to your opinion that it is possible for the mind to form to itself im ages of m aterial things inde­ pendently of the services of the bodily senses, th is is refuted b y the following argument:— I f the mind is able, before it uses the body as its 'instrum ent in perceiving material objects, to form to itse lf the imaged of th ese; and if, as no sane man can