PROBLEMS IN HISTORY OF GREEK SCULPTURE 607 



make large use of hypothesis, and, however cautious his tempera- 

 ment, he can hardly fail at times to confound plausible hypothesis 

 with well-established fact. 



If this meant that we are doomed to endless, unprogressive guess- 

 work, it would be discouraging indeed. Fortunately nothing of the 

 sort is true. The advance which during the last hundred years has 

 been made in the understanding of the history of Greek sculpture has 

 been enormous, and is going on at the present day with accelerated 

 speed. This advance comes about in part through the constant acces- 

 sion of new materials. Even literary documents come to light, like 

 the fragment of a list of Olympian victors * found in Egypt and first 

 published in 1899, which has supplied us with valuable dates in 

 the careers of Pythagoras, Myron, and Polyclitus. New sculptors' 

 inscriptions continue to be discovered. And above all, the stock of 

 known sculptures is augmented each year by pieces which had been 

 hidden underground or sometimes even at the bottom of the sea. 

 Herein is one of the great, exciting compensations to the student of 

 Greek art. Every fresh discovery makes a problem. The new thing 

 must be studied and assigned to its proper place. It may become the 

 starting-point for a new set of hypotheses, and so lead to an extensive 

 readjustment of views previously entertained as to the history of 

 Greek art. 



To this accession of new material there must come an end, and that 

 end cannot be very far off. But the study of old material is only 

 less fruitful than the acquisition of new, and it is hard to foresee a 

 time when discoveries can no longer be made with the materials in 

 hand. 



Something has already been said of the part which the study of 

 copies plays in our reconstruction of the history of Greek sculpture. 

 Your attention is now invited to some of the more general questions 

 which that study involves. I realize as fully as any one that art-criti- 

 cism, to be profitable, must be exercised on the actual object. Abstract 

 discussions are likely not only to be dull, but also to miss the essential 

 point. Yet I venture to hope that a few considerations may be worth 

 putting forward, even without the help of visible illustrations. 



To begin with, we need a working theory as to how these 1 copies 

 were made. We know that in the Roman imperial period, to which 

 they chiefly belong, the practice of taking casts from statues, or at 

 least from bronze statues, was in use. Casts are easily multiplied 

 and easily transported, and from a cast or casts a workman or work- 

 men, in the same or different parts of the empire, could make any 

 number of copies in bronze or marble, agreeing with the original in 

 dimensions and in all principal features. But the opinion has recently 

 1 Grenfell and Hunt, Oxyrhynchus Papyri, part u, no. ccxxii. 



