Talk:Dravidian languages

Contents

1 Elamite 2 Past Dravidian Languages 3 Rewrite 4 Some comments 5 Australian Aboriginal languages 6 Indo-European link 7 Theories on the derivation of Dravidian languages 8 List of Dravidian languages

Elamite

On the contrary, they are perhaps related to the Elamite language spoken in Iran before the invasion of the Persians. I believe, though I can't claim any sources, that the current theory is that there was an Elamo-Dravidian continuum stretching from the borders of Sumer/Akkad all the way to the tip of the Indian subcontinent before the Persians invaded.

See: http://www.krysstal.com/langfams_dravidian.html
and HistoricalLinguisticsorPhilology

--Ben Brumfield

How many words of Elamite are known, aside from a couple of dozen personal names? Such sweeping linguistic theories really do need more solid backup than http://www.krysstal.com is offering. Wetman 20:32, 6 Jan 2004 (UTC)

Have a look at David W. McAlpin's Toward Proto-Elamo-Dravidian in Language, Vol. 50, No. 1. (Mar., 1974), pp. 89-101. Elamite has a written corpus comprising of thousands of words. --Imran 21:43, 6 Jan 2004 (UTC)

Isn't 'arasu' Tamil for 'king' or 'government'? (c.f. Singapore, which in Tamil is Singapura Kudiyarasu) --Xiaopo's Talk 20:30, Jan 2, 2004 (UTC)

It is. -- Sundar 10:35, Dec 28, 2004 (UTC)

Past Dravidian Languages

Indeed, it has been suggested that the language of the Indus Valley Civilization was Dravidian, but I don't know that anyone denies the possibility of other languages in the area of modern India. The widespread view that the language of the Indus Valley civilization was not Dravidian also needs coverage. --Xiaopo's Talk 08:43, Jan 3, 2004 (UTC)

When the script of the Indus Valley is eventually deciphered, opinions pro or con will have the beginnings of some substance. No one has the least idea of what phonemes go with what symbol, at present (Scientific American July 2003) Wetman 20:32, 6 Jan 2004 (UTC)

Rewrite

I rewrote this page because it was amazingly un-NPOV. The claim that Dravidian was the common ancestor of the Indo-European languages is believed by a very small minority, and I know many linguists who would go farther and call it a "crackpot theory". It obviously deserves mention, but NPOV allows less popular views to get less space, and this one deserves a few sentences detailing its claims and the response of the mainstream community -- certainly not the large portion of this article as well as another. These views were also mentioned on Aryan invasion theory as fact, in an extremely NPOV paragraph I rewrote. I moved some of the examples from Indo-European Dravidian words since those were greatly representative of the bunch. Also (almost done, promise!) let me present a page[1] (http://www.zompist.com/chance.htm) showing that chance resemblances amongst unrelated languages are indeed quite likely. Last, but not least, these repeated dismissals of mainstream linguists as having "only a passing familiarity with Dravidian languages" and citing an author's only qualification as being a native speaker of a Dravidian language don't mean much -- f'rinstance, I'm a native speaker of a Dravidian language and I believe these claims are rubbish. :-) Suggested reading for Codebytez: Comparative method and sound change. --Xiaopo's Talk 06:44, Jan 6, 2004 (UTC)

Some comments

We should really be discussing this on Indo-European dravidian words, but since we have started here, lets rock :)

• Applying the principle of regular sound change, the reconstructed proto-Dravidian word for "one" is *oru. English "one" on the other hand can be traced back to Old Germanic *ainaz and Proto-Indo-European *oynos.

• • Actually, dravidian has two words for "one". onnu which is used in the abstract numeric sense and *oru used for person. onnu corresponds to the IE root oi-no.

• Interesting that in all my years of speaking Kannada and Tamil I've never heard of these two words for "one" then. Nor can I find them in any grammars of the language. Since you marked *oru with an asterisk and not onnu, I presume onnu is a form in one of the modern Dravidian languages, and not a reconstructed form? Which language, then?

• • oru is used as an adjective indicating a single item or person in Tamil. onnu seems to be a slang for the number one, the more classical for being onRu (in Tamil), (and ondhu? in Kannada). -- Sundar 10:32, Dec 28, 2004 (UTC)

• English "attic" comes from Greek Αττική (Attiki), the name of a region of Greece (see: Attica, Greece). It has changed meaning over the years, and it is thus pure coincidence that its present meaning and pronunciation bears resemblances to the Tamil form.

• • I've heard about the above etymology and also at least another for attic: ut- [IE root for up] with the -ic suffix, literally upper. This sounds more probable than attic being named after a region of Greece simply because places named after things are common, things named after places are rare. It only happens when a new item is introduced by traders from a foriegn land. E.g. Indigo from India, Hamburger from Hamburg. However, in such cases, there is always a native word for the item. I could not find any non-tangential synonyms for attic. [If someone knows, please let me know]
    • Attic is a shortened form for 'attic storey,' attested in 1724. Another form attested in 1696, 'attic order,' refers to "a low decorative facade above the main story of a building." 'Attic' before this date refers to a type of column often featured in Classical architecture inspired by the Greek region Attica, a word attested in 1599. Before this, there is no written record of 'attic' in the English language. Please do not make generalized statements about Indo-European linguistics before you do basic research, especially if you're going to make things up to support your flimsy argument. Your fanciful derivation is laughable. 'Ut-' is not even the IE root for 'up.' It isn't a reconstructed root at all.

• The etymology I suggest is well-documented and appears to be accepted by virtually everyone, see [2] (http://encarta.msn.com/dictionary_1861587610/attic.html) [3] (http://www.bartleby.com/61/1/A0510100.html) and [4] (http://dictionary.reference.com/search?q=attic). "Late 17th century. Via French attique “Atticâ€? from Latin Atticus (see Attic). The word originally described a decorative structure (in the Attic style) above the main façade of a building." In fact, I can't find any source (reputable or otherwise) that suggests your etymology. Maybe you could help me out here? :-)

• • I'm dropping vocare and yellow (also other weak cases) and list only ones that have a solid IE root. Please understand that the page is still being worked on and I will attempt to rewrite in accord with Comparative Method.

• The thing is, the view of the vast majority of linguists is that there is no such thing as "Indo-European Dravidian words". Thus having a whole article for it wouldn't be NPOV. The NPOV policy states that we don't need to discuss minority views and beliefs in the field as much as majority opinions, and this theory is certainly believed by few enough people to not deserve an article of its own; in fact, it would mean the minority view gets more time and explanation than the majority. Not to mention that the very title, "Indo-European Dravidian words" is a viewpoint.

• • I understand you do not agree with the above theory, but Vfd'ing a new user's page that is in the process of heavy editing without comments on the page itself or at least some talk is just ... sad ;)

Codebytez | Talk 10:27, 6 Jan 2004 (UTC)

• Sorry if I made it seem personal or something, it's not. :-) This isn't meant to belittle your skills as a writer or Wikipedian, and you certainly can write well. I just don't think a page like "Indo-European Dravidian words" has any place in the NPOV policy. (Also, you're right, I should have left a note on Talk:Indo-European Dravidian words saying discussion was over here, sorry!) --Xiaopo's Talk 16:21, Jan 6, 2004 (UTC)

• Xiaopo: How would you feel if we do this:

I summarize and move contents of Indo-European Dravidian words into this page as a paragraph (fewer examples, will use only word for word equivalents). Then we kill the other page. If you still believe that Indo-European Roots and matching dravidian words/roots constitutes flawed methodology, we can paraphrase your lengthy analysis with a sentence or two. I can understand your argument about yellow/haladi, but would you still hold the same belief for IE root to Dravidian word/root matching? Given that no consonants are changed and the vowels are practically the same, they are practically the same word with the same meaning. -- Codebytez | Talk 22:30, 6 Jan 2004 (UTC)

I took a look at the page [5] (http://www.zompist.com/chance.htm) and trust me, its flawed. The math is OK, but the basic assumption of roots starting and ending in one consonant and having only one vowel in between is a major flaw. As you already know, roots can start and end with vowels and diphthongs. Consonants can flow to another consonant without a vowel in between. And roots that start in consonants can have diphthongs following them. The ending does not have to be a consonant. The model used by the paper would for example, leave out a vast number of the known IE roots and Dravidian roots/words. Developing independent models for different language groups and calculating the probability of the same word with the same meaning is a complex math problem.

The chances of two root words that sound exactly the same and mean exactly the same thing are RARE in languages that belong to different language groups. If you can find 25 root words that sound and mean the same among two unrelated language groups with no variations except the ending, I'd be mighty impressed. Indo-European and Dravidian pair is the only exception. Surely, they could be related? The requirement for 90% of roots to be the same is just a little too much. If they do, they are probably dialects :)

--Codebytez | Talk 05:31, 9 Jan 2004 (UTC)

---

It might be that theer is authority for draavida meaning south in Sanskrit, but the standard word for south is dakshina. draavida has other derivations related to the name Tamil, and the meaning 'south' is probably derivative. Imc 22:25, 28 Jan 2004 (UTC)

Agrees. Dravida is a word that means Southern, the context being anthropological (refers to the Southern culture and people) and not the direction.

It may be derivative, but I'm not sure. Do you know any place where we can look it up?

And Chancemill (who wrote the comment above mine), it's not always anthropological. c.f. Jana-Gana-Mana: "Punjaba, Sindhu, Gujarata, Maratha, / Dravida, Utkala, Banga" where it's obviously used in a geographical context. --Xiaopo's Talk 22:38, Jan 29, 2004 (UTC)

The online dictionary at http://www.uni-koeln.de/phil-fak/indologie/tamil/mwd_search.html provides no authority for draavida meaning south, either from English or from Sanskrit. So it would probably be safe to rewrite the article to say that the name Dravidian was derived from a word connected to 'Tamil', in this case meaning the south of the country. Imc 13:12, 30 Jan 2004 (UTC)

The original Prakrit term for Dravida is Damida, which has a dravidian etymology. Two possible derivations are Tamila -> Damida -> Dravida and Damida -> Tamila and also Damida->Dravida. Damida is a dravidian palindromic word, so its possibly an older form of both Tamila and Dravida. Codebytez | Talk

Is it related to theRku (meaning South in Tamil) ? -- Sundar 10:32, Dec 28, 2004 (UTC)

Xiapo: I know you went to a lot of trouble to investigate my word list, but it so happens that every word you have chosen was given as an example only by me. I am not aware of any other occurance of these four examples. You really should find other words incorrectly believed by lay people to be cognate. Also what shall we do about the many roots that are identical in PIE and Dravidian? Also, what reference do you use for the proto-dravidian language? Codebytez | Talk

Not much trouble, actually, some of the flaws in your examples were obvious at first glance. I don't see how we can give examples usually presented by lay people, since most lay people who comment on the subject (like you) create their own wordlists. Regardless of where you got yours, I think it's safe to say the examples I used are representative of the level of scholarship of such hypotheses.

If we really want to decide what goes in here, I would suggest that we have about a sentence or two pointing out that some people disagree with this classification. After all, we see far less discussion on Korean language as to whether Korean is an isolate, and yet that's far more controversial than whether Dravidian is related to IE! It is not our place to decide whether these theories are true or false, simply to represent the opinions of the people in the field -- and the consensus is overwhelming, this issue isn't even very controversial in linguistic circles.

If you're referring to *oru, I originally got it from here (http://zompist.com/euro.htm#dravidian). IIRC, however, it was mentioned in one of your Levitt articles as well.

--Xiaopo's Talk 03:19, Mar 3, 2004 (UTC)

I would prefer attic/atta, kill/kollu, chill/chali. Sound and mean almost the same but have different etymologies. Actually, it would help to know the etymologies of Dravidian words to make the point solid.

Unfortunately, I've been unable to find any reference on a reconstructed proto-dravidian reference/dictionary. I read somewhere on the internet that proto-dravidian has never been reconstructed. If anyone has an online or book reference to a complete proto-dravidian dictionary/reference, it would be of immense help.

Also I noticed you deleted the reference to Dravidian contribution of retroflex L to Sanskrit when you reverted. Was it accidental?

Codebytez | Talk

I actually think our purpose is not to go into length presenting arguments from either side, but just to include a sentence or two that this classification is disputed.

I don't know about any reference dictionaries; however, your Levitt article referenced several proto-Dravidian forms.

Whoops, yeah, that last bit was accidental. Erk, sorry. --Xiaopo's Talk 19:28, Mar 3, 2004 (UTC)

Australian Aboriginal languages

I'm quite surprised to find that there is no mention at all to the old, now discredited theory that Australian languages were related to Dravidian. When you read about Australian languages this is one of the first things you learn. Apparently it is the phonology and probably also the agglutinative qualities of the languages which once gave this impression. Maybe it's missing from this page because the perspective is reversed. Australian language literature always mentions Dravidian but I guess Dravidian language literature doesn't mention Australian. — Hippietrail 09:49, 9 Oct 2004 (UTC)

Indo-European link

I removed this pair of statements from the article:

The former claim that Dravidian and Indo-European share a common ancestor is generally based on more rigorous methods comparing Proto-Indo-European and Proto-Dravidian, and is generally held in higher regard by the mainstream linguistic community. This is statement is clearly false. I do not know of any mainstream linguist who has supported this. Refer to Bhadriraju Krishnamurti's recent book on Dravidian Languages.

I think dialogues like this belong on the Talk page, not in the article, until they are resolved. --Heron 14:23, 11 Nov 2004 (UTC)

Theories on the derivation of Dravidian languages

The "mainstream" theories are better dealt with in the "History" section, as I have now done ("History" should really discuss the history of the family, not just the history of its discovery). I don't know what to say about this section. Perhaps we could substitute this section with a new section (or separate article) dealing with the relationship between Dravidian and Indo-Aryan languages? That section could deal with issues such as the mutual lexical and structural influences, as well as the theories about the common descent of Indo-Aryan and Dravidian and the objections to these theories. ---- Arvind 17:15, 29 Jan 2005 (UTC)

List of Dravidian languages

I'd like to redo this list, so as to add information about the sub-groups and sub-sub-groups within each broad group. This could be done as a nice png tree, or through a multi-level list. Any suggestions as to which will be better? My impression is also that the South-Central and Central groups are considered to be two-subgroups of a broader Central group (the differentiation being broadly comparable to that between Tamil and Tulu). The Britannica agrees, but are there linguistic sources which indicate otherwise? ---- Arvind 17:27, 29 Jan 2005 (UTC)