This longform post is about languages, roots, transmission directions in the context of Sanskrit & Tamil…

… and more importantly, it seeks to deal with suspect methodologies, premature optimizations, simplistic/superficial generalizations, giant logical leaps, self-referential statements, circular-reasoning, non-existing citations and the like, around some ideas – that have been plaguing the Tamil ‘discourse’ – or what passes for it.

I would also like to ascribe such seemingly unscholarly & infructuous approaches to a child-like eagerness/enthusiasm of some fundamentally good people – to ‘prove’ the ancientestest nature & origins of Tamil, in spite of all the evidence that stare at us in our face and associated kanging – and not to some deleterious motive. Okay – I agree that NOT all such folks are decent & ethical – leave alone being scholarly.

Normally, I do not indulge in major or sustained battles/discussions on X, because in my opinion, the context for the X-platform is not really about longform idea exchanges and also because I am not blue-ticked yet which would have otherwise allowed me to respond in longform. Also – ars longa vita brevis.

Again, I wanted to put in some significant time-effort in terms of approaching & analyzing this particular claim – as there are many many such types of uncalled-for claims that seem to be proliferating in the Tamil<->Sanskrit relationships – and I want to create some kind of a ‘template approach’ to deal with them. An exemplar, if you will.

Yes. I want to approach the claim as rigorously & as politely as possible, with a clear logic and marshalling of whatever evidence that I could muster; but of course I am not at all a Mr KnowAll and I do not have any scholarly pretensions. I continue to be an eager student of Tamil, Sanskrit & English – and assorted other things.

In any case, linguistics++ is only one of my minor interests, really. So, feel free to rebut/join etc in this discussion. Let us all learn together, if we can – with as much civility as needed and possible.

1

Background:

On July 29, 2026 this following discussion happened and I was not privy to that – am providing just the immediate context.

It was one rabid, frothing-at-the-mouth ‘Aravind’ who started off with an obtuse set of TamilSupremacist-Dravidian assertions. Pure unadulterated hatecraft. Usual story/fiction: Aryan invaders foisted themselves and Sanskrit on Tamilnadu post 13th century CE.

Here’s were Sri N Ganesan (henceforth NG) introduces the idea as in the subject/title of this post.

And ‘dagalti’ asks a follow-up query and NG answers, looping me in. As under.

I responded to him – and this post is the result of that undertaking from me to respond soon.

There was also another short exchange between me and NG:

Rest of the post is going to be about expanding on my observation:

~”Sanskrit bīja ‘seed’ derives from Dravidian vittu/viccu/vīca” is NOT established by DEDR 5401 & is phonologically weak as presented.

Ideally I would like to use ‘Dravidian’ instead of Dravidian, since it is a made up category-name chistened fairly recently, unlike the monikers Tamil or Sanskrit. It also has this heavy baggage of having been politically loaded. But, writing ‘Dravidian’ everytime could be jarring, so am avoiding that.

To the extent that my typing would allow for, I have used International Phonetic Alphabet (IPA), showing the correct pronunciation..

2

DEDR 5401

Dravidian Etymological Dictionary of 1984 (often abbreviated as DEDR), is a major academic resource painstakingly compiled and annotated by T. Burrow and M. B. Emeneau. Very reasonably comprehensive.

DEDR 5401 refers to a particular entry – that groups a large set of Dravidian forms meaning ‘sow’, ‘seed’, ‘semen’, ‘testicle’, and related notions. The distribution across South, Central and North Dravidian languages makes the underlying Dravidian family cohesive.

It is important to remember as a backgrounder that, DEDR is a comparative dictionary organised into proposed etymological groups, rather than a series of claims that every listed surface form must be inherited directly and regularly.

The important qualification occurs in the entry itself:

“Go., Kuwi, and Malt. words for ‘seed’ may be < IA.”

That warning matters. Burrow and Emeneau were explicitly allowing that forms such as Gondi vījā, Kuwi bica/bicca, and Malto bíci might be loans from Indo-Aryan, presumably from forms related to Sanskrit bīja.

Therefore, these particular forms cannot safely be used as independent evidence for a Dravidian development.

vittu -> viccu -> vīca -> bīja

That would risk circular reasoning. Some of the closest-sounding forms may already be Indo-Aryan loans.

3

What can reasonably be reconstructed for Dravidian?

The underlying family is usually represented approximately as something like:

• verbal base *vit- / *vitt-, ‘to sow, scatter seed’
• noun *vitt-, ‘seed’
• derivatives with suffixes, such as Tamil vitai

The exact Proto-Dravidian reconstruction depends on how alternations among vit-, vitt-, vitay-, vicc- are analysed.

Tamil forms such as:

• வித்து vittu, seed, semen
• விதை vitai, seed
• விதை- vita(i)-, to sow
• விச்சு viccu, seed or sowing-related form

…belong to the same general Dravidian lexical family according to DEDR. That much is reasonable.

But DEDR 5401 does not itself reconstruct Sanskrit bīja from this family.

So, the reason as to why this extract was used by NG is not at all clear!

4

Is vittu > viccu plausible?

Tamil and its sister/other languages exhibit historically conditioned palatalisation and alternations involving k/c, c/y, and consonants adjacent to front vowels or y. However, a specific development vittu > viccu requires its own comparative and morphological evidence and should not be justified by a general Tamil rule tt > cc.

One cannot simply posit a general rule: tt > cc, without specifying:

1. the earlier phonological environment
2. the chronological stage
3. parallel examples
4. whether the form is inherited, dialectal, analogical or expressive

The examples given in the claim of NG do NOT establish such a rule.

பச்சை : பாசி – paccai and pāci cannot simply be treated as an illustration of free cc : c alternation. Their vowel quantities, morphology and lexical histories also need explanation.

கொவ்வை : கோவை – This pair likewise involves more than geminate simplification. It has differences in vowel length and medial consonantal structure.

Similarity alone does not prove that it illustrates the development required for vittu > viccu.

விச்சு : வீசு – This comparison is particularly problematic.

Tamil வீசு vīcu, ‘to throw, fling, wave, blow, spread’, belongs in DEDR to a separate comparison associated with scattering or throwing, referenced by DEDR 5400 vitir- in the quoted entry. DEDR’s “Cf. 5400 Ta. vitir” indicates comparison, not an assertion that vīcu regularly derives from vittu.

A derivation such as vittu > viccu > vīcu would require, at minimum:

1. tt > cc
2. loss of gemination cc > c
3. i > ī
4. and a semantic development

It may possible that each step might conceivably occur somewhere, but the sequence needs evidence.

It cannot be inferred merely from resemblance as comfortingly assumed & indicated by NG.

In fact, this ‘resemblance’ issue is the one that is there in most of the takes of ‘folk etymologists’ like Mannarmannan & Paari Saalan, that plague the Tamil discussions.

5

The proposed vīca > bīja equation

This is the weakest part of the argument or the process for claim by NG.

Sanskrit bīja /biːd͡ʑa/ ‘seed’ and a proposed Dravidian form vīca or vīja look moderately similar, but historical etymology requires regular sound correspondences.

A borrowing from Dravidian *vitt- into Indo-Aryan as bīja (as claimed by NG) would definitely require explanations for:

Dravidian	Sanskrit	Problem
initial v-	b-	Possible in some borrowing situations, but must be demonstrated
medial -tt-	-j-	No obvious regular substitution
final consonant	final -a	Sanskrit nominal adaptation is possible
short i	long ī	Requires an account or an example
geminate stop	voiced palatal affricate	Major unexplained change, this is surprising

The change *vitt- > bīja- is therefore not a transparent or recognised sound substitution – UNLESS NG can satisfy every one of the above 5 criteria.

One might instead compare the later Dravidian forms vīja, bica, bíci, but DEDR itself warns that precisely these may come from Indo-Aryan.

They are consequently poor evidence for the opposite direction (Dravidian -> Sanskrit) of borrowing.

6

What is the accepted etymology of Sanskrit bīja?

Sanskrit bīja unquestionably means ‘seed’, and by extension ‘germ’, ‘source’, ‘cause’, generative material, semen and, in tantric usage, a ‘seed syllable’. The semantic explanation of bīja-mantra as a compact sound containing generative spiritual potency is traditional and unobjectionable.

I will NOT dispute that its deeper etymology, however, is uncertain at least in my limited knowledge. It seems to lack a straightforward and universally accepted Indo-European derivation. Such uncertainty makes a substrate or borrowing hypothesis possible in principle, but:

‘Uncertain within Indo-European’ does not automatically mean ‘borrowed from Dravidian!’

To establish Dravidian origin, scholars like NG would definitely need:

1. a reconstructible Dravidian source form predating Indo-Aryan attestation
2. regular or independently paralleled sound substitutions
3. plausible geographical and chronological contact
4. and exclusion of borrowing in the reverse direction

The proposed argument of NG supplies mainly semantic similarity and surface resemblance.

But one needs arguments that are really sound and substantive to present a scholarly take!

7

Is vit- a Sanskrit dhātu derived from Dravidian?

The wording of NG: “vit- is the dhātu-vēr in Dravidian” unfortunately mixes Sanskrit grammatical terminology with comparative reconstruction.

A better formulation would perhaps be: “Proto-Dravidian appears to have had a verbal root or base *vit- / *vitt-, ‘sow’.”

Calling it a dhātu is understandable informally, but Sanskrit grammatical roots and comparative Dravidian reconstructions belong to different descriptive frameworks.

Further, the existence of a (A) Dravidian root *vit- does not show that (B) Sanskrit bīja comes from it.

The two propositions A and B are separate which NG has to note of.

8

The Vināyakar Akaval claim

The contentious line is: “மூன்று கண்ணும் மும்மதச் சுவடும்..”

Project Madurai’s text (specifically cited by NG) and commentary interpret மும்மதச் சுவடு as the marks or traces produced by three kinds or streams of elephant rut-fluid. The particular commentary of the great Sri TP Meenakshi Sundaram Pillai, hosted there gives the names of those 3 as: கன்ன மதம், கபோல மதம், கோச மதம்…

…one would suppose that, after referring to the Project Madurai in the context of Akaval, NG would quote this exact reference, actually cited by him!

But NG ‘quotes’ something different! கர்ண மதம், கபோல மதம், பீஜ மதம்!! பீஜ and NOT கோச as in his reference! When asked about NG replied, as in the snippet from the Section1 above – “many scholars explain the பீஜ மதம்.”

I was slightly surprised by the defensive ‘I am my own citation’ kind of response, but that’s not a significant issue in the big picture of a host of other concerning lacunae in the claims of NG.

Of course one is aware that in a few glosses of the phrase mummdam, bīja-mada is given as a lexical explanation – but then NONE of them are of the incredible scholarly levels of Sri TPM Pillai!

Therefore:

1. மும்மதம் in the poem has no etymological bearing on Sanskrit bīja.
2. The reading பீஜ மதம் appears to be one modern anatomical or Sanskritising gloss among variants.
3. Even were bīja-mada an authentic technical term, it would merely use Sanskrit bīja in the sense ‘semen/generative fluid’. It would not prove that bīja originated in Tamil or Dravidian.

The poem ‘Vinayakar Akaval’ brought in by NG to prove his claim, is consequently irrelevant to the proposed etymology.

9

Conclusion

I hope, at least Dr N Ganesan has read this far… :-) In spite of this post is already some 1900+ words in length! Loooooong form indeed! But I think it would be an useful template or a framework to respond to such claims, hopefully… at least for me!

It also took me upwards of 7 hours to process claims, collect details, organize them and then respond comprehensively (one hopes so!) to the claims made by NG.

But.

I would be very eager to know NG’s response (and interested others) in terms of defending or revising his claims – point-by-point AND with tangible, verifiable & scholarly evidence.

Anyway. Here’s hoping this would encourage decent & evidence-logic based discussions on one of the vexatious issues with our Tamil. Thanks.

(END)