This page has not been fully proofread.

INTRODUCTION
 
78
 
the verb form "could be" being understood. I have not gone out of my
way to impose hiatus or starred readings. The particle hi is the general N
method of avoiding hiatus across the caesura in 13°, and has been retained
also because its emphasis is not out of place in the context. For the rest,
some of our scribes have seen a hiatus where none existed, as in 42 where
ratyanta- has been taken as a mi-lection of sort for atyanta. There may
have been a real hiatus in 28 concealed by payusä svatma, but the evidence
is insufficient. Possibly, a case for hiatus may also be made out in place of
jarayā yāty ujjvalam in 1972, but I prefer to see confusion due most
probably to haplography or the occasional substitution of jūty- for yaty-;
moreover, it would be more natural to have some verb here in its explicit form.
 
It was gratifying to find several of these restorations supported by
version Q, such as 15%, 61, 70, 115³, 148%, 249%.
 
V. Who was Bhartrhari?**
 
5.1. The problem of identification. We are now left with a set
of verses grouped in order of probability, which show slow growth over
the centuries, but with a substantial nucleus which must have been original.
A characteristic of the oldest portion is its compression, with a philosophy
which makes a strong impression upon the reader of a marked literary
physiognomy. That Bhartrhari was mere anthologist is a view not
uncommon among scholars, but an anthologist or for that matter any
successful writer would have left us a better unified collection, as for
example happens in the southern versions of archetype delta. The solution
which I have proposed is that the collection is an anthology of verses
believed to have been Bhartrhari's by later compilers. Stanza 63 occurs
in Kālidāsa's Śākuntala, though as genuine Bhartṛhari as any other known.
Others are found in the oldest layer of the Pañcatantra as restored by
Edgerton. If, therefore, one man wrote these verses, he must belong to
the opening centuries of the Christian era; one of his stanzas current
as a subhāṣita could therefore appear without stigma of plagiarism as part
of the advice given to Kalidasa's heroine. One type of critic, of course, compares
the two poets, concluding that Kalidasa being the greater Bhartṛhari must
necessarily have borrowed from him, hence be an anthologist. An extension
of this argument may be seen in the suggestion thrust upon me that my
type of text-criticism was futile; to edit Bhartrhari, all one should have
to do would be to select the best stanzas, the second rate bein; discarded
as unworthy of the great poet. Such a procedure is difficult of execution
for the tremendous stauza 301, as great as any other in the whole volume,
is omitted in two well-determined versions, while the beautifully written
re re cätaka [721] is found ouly in W.
 
Three stanzas seem to furnish some evidence for a date not earlier
than that proposed above, i. e. the opening centuries of the Christian era.
No. 257, discussed before, mentions the pale complexion of Saka maidens;
less easy to date would be the hermaphrodite Siva of 224%, and the ten
 
This section condenses the views expressed in two articles of mine: 1] J. Oriental
Research [ Madras ] XV, 1946 pp. 64-77; 2] Bharatiya Vidya VII, 1946, pp. 49-62.