iffused in
sophistic circles by oral teaching, and that it became known to Critias
and Plato in this way. Its originator we do not know. We might think of
the sophist Thrasymachus, who in the first book of Plato's _Republic_
maintains a point of view corresponding to that of Callicles in _Gorgias_.
But what we otherwise learn of Thrasymachus is not suggestive of interest
in religion, and the only statement of his as to that kind of thing which
has come down to us tends to the denial of a providence, not denial of the
gods. Quite recently Diagoras of Melos has been guessed at; this is empty
talk, resulting at best in substituting _x_ (or _NN_) for _y_.

If I have dwelt in such detail on the _Sisyphus_ fragment, it is because
it is our first direct and unmistakable evidence of ancient atheism. Here
for the first time we meet with the direct statement which we have
searched for in vain among all the preceding authors: that the gods of
popular belief are fabrication pure and simple and without any
corresponding reality, however remote. The nature of our tradition
precludes our ascertaining whether such a statement might have been made
earlier; but the probability is _a priori_ that it was not. The whole
development of ancient reasoning on religious questions, as far as we are
able to survey it, leads in reality to the conclusion that atheism as an
expressed (though perhaps not publicly expressed) confession of faith did
not appear till the age of the sophists.

With the Critias fragment we have also brought to an end the inquiry into
the direct statements of atheistic tendency which have come down to us
from the age of the sophists. The result is, as we see, rather meagre. But
it may be supplemented with indirect testimonies which prove that there
was more of the thing than the direct tradition would lead us to
conjecture, and that the denial of the existence of the gods must have
penetrated very wide circles.

The fullest expression of Attic free-thought at the end of the fifth
century is to be found in the tragedies of Euripides. They are leavened
with reflections on all possible moral and religious problems, and
criticism of the traditional conceptions of the gods plays a leading part
in them. We shall, however, have some difficulty in using Euripides as a
source of what people really thought at this period, partly because he is
a very pronounced personality and by no means a mere mouthpiece for the
ideas of his contem