[Advaita-l] Authorship issues - 3

Vidyasankar Sundaresan svidyasankar at hotmail.com
Tue Sep 23 15:52:56 CDT 2003

Parts 1 and 2 of this series may be found at


To summarize, within academic scholarship, Sankara is taken to be, by 
definition, the author of the commentary on the brahmasUtra. The 
commentaries on gItA and the principal upanishads are also taken to be 
genuine texts. Among the independent texts, only upadeSasAhasrI has been 
accepted as genuine, while the other works traditionally attributed to 
Sankara have been generally set aside.

The acceptance of upadeSasAhasrI as a genuine text is largely a result of 
the researches of Sengaku Mayeda, the Japanese scholar, who applied Hacker's 
criteria of style and language usage to this text. These criteria are based 
on Hacker's study of the brahmasUtra commentary. Chief among these are -

1. Sankara is much more 'theistic' than many later authors. Thus, he often 
uses brahman and ISvara interchangeably, whereas later authors tend to make 
a distinction between nirguNa brahman and ISvara as saguNa brahman.

2. Another key observation is the usage of the terms avyAkRta nAmarUpa and 
tattvAnyatvAbhyAm anirvacanIya. Now, the adjective anirvacanIya or its 
variant, anirvAcya, is very notable in advaita literature. As such, it 
simply means "indescribable" and is often used to directly describe avidyA 
(ignorance) by many later advaitins. However, Sankara himself does not seem 
to do so. When he uses the term anirvacanIya, it is always as 
tattva-anyatva-AbhyAm anirvacanIya, i.e. indescribable as this (tat-tva) or 
the other (anya-tva), and as a qualifier of avyAkRta nAmarUpa.

3. In the brahmasUtra bhAshya, avidyA is used roughly thrice as often as 
mAyA. And when mAyA is referred to, it is often described as Sakti, a power 
of ISvara/brahman.

Applying these text-internal criteria to upadeSasAhasrI, Sengaku Mayeda 
finds this text to be genuine. This text satisfies Hacker's external 
criteria of manuscript attribution to Sankara bhagavatpAda, as opposed to 
SankarAcArya, and the presence of early quotations from it, as sureSvara 
quotes it in naishkarmyasiddhi and some early post-Sankaran Buddhist authors 
also refer to it as an upadeSa grantha of vedAnta. Mayeda has also applied 
these criteria to the commentary on mANDUkya upanishad and gauDapAda kArikAs 
and the pada and vAkya commentaries on kena upanishad. He concludes that all 
these texts are genuine. It should be noted that other authors had doubted 
the genuineness of the mANDUkya commentary, and it was mentioned in the 
earlier thread on vivekacUDAmaNi that doubts have been raised about the 
vAkya commentary on the kena upanishad even by people within the advaita 
tradition. Nevertheless, the vAkya commentary satisfies the criteria 
suggested by internal evidence from the brahmasUtra commentary.

Other than the three criteria listed above, another point is the notion that 
Sankara never refers to Ananda (bliss) with respect to brahman unless 
required by the source text he comments on. In particular, the stock 
expression sac-cid-Ananda as a characterization of brahman seems to 
originate with much later authors.

These criteria have never been applied in any great detail, so far as I 
know, to study the commentaries on the gItA and the other eight principal 
upanishads (ISa, kaTha, aitareya, taittirIya, chAndogya, brhadAraNyaka, 
praSna and muNDaka). The criteria have been used to reject a number of 
independent texts, however. The next and last post in this series will 
discuss pros and cons of the criteria described above.


Instant message with integrated webcam using MSN Messenger 6.0. Try it now 
FREE!  http://msnmessenger-download.com

More information about the Advaita-l mailing list