and in Mesopotamia, ¹ and logically implied in Hellen-
istic art, of which he will regard the Roman imperial
style as the final stage. A historian, moreover, will
note that when an artist's name is given in the Roman
period, that name is generally Greek; ² and an ethno-
logist will perceive a connexion between many of the
commonest features of Roman decorative art and the
symbols of Mazdaism and other ancient religions of
the East. ³ On the other hand, those who believe that
the true function of art is to give concrete expression
to abstract ideas will be able to point to real advances
in naturalistic portraiture, in the representation of
historic events, and in the aesthetic exploitation of
plant and animal life.

To decide between the merits of these claims and
counter-claims is not easy. We have to make up our
minds whether to treat Roman art as an original
activity, or as the last phase of Greek art, or as the
realisation in Mediterranean terms of ancient modes
of visualising and rendering form inherited from the
East. And if there is found to be some justification
for all these points of view, we have to judge which
standard of values is decisive, and, more specifically,
what was the quality in Roman art that could account
for its widespread and enduring popularity in the
Middle Ages and the Renascence.

In dealing with so fundamental a problem we must
take the widest possible view of Roman art. We must
compare it with other Roman institutions. We must

-2-