potters, brahmins, and RSis (contd.)

[Sudalaimuthu Palaniappan]

In my earlier posting, I should have mentioned that the CT texts have the
word "aGki" which is pronounced as "aGgi".

In a message dated 97-10-05 05:38:27 EDT, bhk at HD1.VSNL.NET.IN writes:

<< This is imagination running riot.Too far-fetched and unscientific. Bh.k.
  >>
I beg to differ from Dr. Krishna Murthy. I was not being unscientific. As I
said, I do not subscribe to the theory of general deliberate inversion in
Dravidian. What I was trying to do was scientifically trying to understand a
possible unscientific process of deliberate intervention by human agency in
altering names without any natural linguistic processes doing it. The
sociological phenomenon of Sanskritization is replete with such occurrences.
Let me give some examples.

The goldsmith community in Tamilnadu are known as  taTTAr or AcAri. In their
normal day-to-day conversation, they will use the form "AcAri" pronounced as
"AsAri".  The VaiSNavite brahmins of Tamilnadu used the caste title "AccAri",
pronounced as "AcchAri". For instance, the full name of the first
Governor-General of India, Rajaji, would be written "irAcakopAlAccAri" in
Tamil. The goldsmith community has a Sanskritized identity "viZvakarma
brAhmaNa". To emphasize that, in formal documents like marriage invitations,
they will always use the form "AccAri", even though their own dialect has
only "AcAri". It is hard to explain the extra "c" except by the need to move
up socially. I think a similar but reverse process is at work in Karnataka. A
Kannada brahmin told me that they drop the final "i" in "AchAri" and use
"Achar" to differentiate themselves from "AchAri" the name used by the
goldsmith community in Karanataka. Similarly, the sea-going fisherman are
called "paratar" or "paratavar". I have heard people of that community
claiming descent from "bharata" of Sanskrit epics. (The correct etymology has
to do with "para" of DED 3255) While Tamil orthography has prevented the
clear delineation of the distinctly different forms, if Tamil had the letter
"bha", today the caste will go under the name "bharatar". Of course,
wholesale name changes in caste names are common too.

Another source of evidence of deliberate manipulation of names is the
inscriptional material. The Chola inscriptions provide a scientific way of
studying this process. The Sanskrit portions of the Anbil Plates of Sundara
Chola and the Leiden Plates of Rajaraja show that in trying to create a
Sanskritic geneology for the Chola kings, the scribes converted "vaLavan2"
one of the names used by the Chola kings  such as in "kiLLi vaLavan2",
"karikAl vaLavan2", etc. into "vaLabha" as the name of a  mythical ancestor
in the solar race. Again, here, if one were to restrict oneself to tracing
the origin of this name purely based on natural sound change, we will not be
successful unless we consider the Sanskritization process.

That is what I was trying to attempt. Staal and others were looking at the
etymology of aGgiras without considering the Sanskritization phenomenon. To
quote Staal: "Who were these AGgirases, after whom Agni is frequently called
"AGiras" and first AFGgiras?" Hillebrandt concluded from a general survey of
their occurrence in the Rgveda that "the AGgirases were originally a family
which was rather outside the main Vedic tradition, as shown by their lack of
prominence in books ii-ix" (Keith 1925, I, 224). Following this suggestion
there has been a great deal of scholarly discussion on their identity and
provenance, and on the possible etymologies of their name. Earlier,
Macdonnell had defended the view that the AGgirases were intermediaries
between gods and men, and that their name is related to
..........,"messenger" English: angel. Evidence for their intermediary
status, however, is slight, and this view has been generally rejected. More
recently, the possibility of this etymology was revived by H. W. Bailey
(1957, 52-53), who postulated a root *ang, "sing, enunciate," which occurs in
the Vedic word AGgUSa, "song of praise." But Schmidt (1968, 51-52) has
pointed out that there are several problems in the derivation of aGgiras from
such a root (including the fact that a suffix -iras is not known anywhere
else.

   Outside of language, etymologies prove littele, "for usage is stronger
than etymology" (yogAd rUDher balIyastvaAt, as the MImAMsA philosophers have
it). Even if a word is Indo-European, like Asura, for example, it might refer
to things Indian and pre-Vedic. It is possible that the AGgirases were
singers, and reasonably certain that they were priests of a fire cult."

One cannot deny the occurrence of the form aGki pronounced as aGgi to denote
"agni" in Classical Tamil texts. One cannot deny the CT addition of -ar to
even borrowed words to form names for people. ParipATal 11.84 has the word
'vEtiyar" referring to the brahmins as "people of the Veda". So I see nothing
wrong in proposing a hypothetical "aGkiyar" pronounced as "aGgiyar", for a
people involved in a fire cult, all based on attested linguistic grounds. The
change of aGgiyar to aGgisar also is defensible on linguistic grounds. It is
the final step of "sar" > "ras" where I hypothesize a deliberate
intervention. It is not as outlandish as one might think. I have been told
that in California temples, during Sankalpa ceremonies, they refer to
California as kapilAraNya based on the derivation of California given by the
late ZankarAcharya of Kanchi. Considering this tendency of Indians to
manipulate the language, if we have to reconstruct the original forms, we
have to give due consideration to these "unscientific" tendencies which might
have been present from very early on.

In any case, it is just a possible process I theorized fitting all the known
facts. If somebody could come up with a far simpler and elegant etymology, by
all means I shall accept it. After all, that is the scientific method.

Regards

S. Palaniappan