This page has not been fully proofread.

60
 
यत्वेन
 
परिपालयेत्
az afga ॥ "May this book, painfully written with broken back, loina,
and neck, forehead cramped down upon the fist, be guarded with an effort".
 
INTRODUCTION
 
III. Analysis And Grouping of The MSS
 
3.1. The chart. The most noticeable feature of Bhartrhari MSS is
their variability, particularly in the order of the slokas as well as the choice of
the stanzas included. It is generally possible to divide the MSS into two
classes at sight, the amorphous northern and the logically arranged southern
recension. Within the recensions, however, the variability is still enormous,
making it necessary to have finer grouping. The uses of MSS are two: first, as
testimonia for the inclusion and order of stanzas; and this can be served by any
MS purporting to be of the complete śatakatraya no matter how corrupt, or
even by a translation; for the scribes always give generous measure, well over
the 300 as a rule. The second purpose is the determination of the precise
text of any given stanza, for which even a MS fragment may help if it be
reasonably correct. Of course, with increasing evidence, it is possible to
make use even of corrupt MSS for determining the readings, if the principles
on which the scribe's pen slips can be established; moreover a fragment
gives evidence about the omission of a stanza if the possible displacements
of a śloka in its version are known. However, the totality of MSS is so
large that the problem is to choose a representative sample of convenient
size. To this end, therefore, it was decided to base the actual edition as
far as possible upon MSS which could fulfill both functions, being fairly
correct copies of all the three śatakas together. In this, we are helped by
the discovery that MSS which report stanzas in the same order generally
tend to have the same readings for individual stanzas.
 
The first step in the examination of such a complete satakatraya
MS is to make a synoptic chart of the stanzas, with the pratikas written
down in alphabetical order. Unfortunately, even this apparently simple
task is unusually complicated for Bhartṛhari. In the first place, it is
necessary to have a fairly good acquaintance with the stanzas themselves,
for there are different beginnings possible for the same stanza-as for
example să ramyā nagarī — bhrātaḥ kaṣṭam aho [169]. Grouping these under
different pratikas would obviously give misleading testimony, Secondly,
no chart can keep pace with the extra ślokas which appear almost with
every new MS, and certainly with MSS from every new centre. The
method finally adopted was to use a supplementary card index, and to
leave plenty of gaps in the chart proper where the pratika of a new śloka
could be noted for the MS in question. There is no possibility of exhausting
all such extras, and ignoring them is equally out of the question as they
have an annoying habit of turning up in the middle of some other Bhartṛhari
text. In charting, the śataka and the number of the stanza has to be
entered, but this is again complicated because almost every MS makes
some slip in the numbering, and many of them make such mistakes
regularly. The final procedure was to enter both the actual number,
and in brackets the number given in the MS where it differed from the
one in my count. Otherwise, one misses valuable parallels and fortutious
omissions. The neglect of this precaution makes it impossible to use