Tuesday, 22 January 2019

Language classification

Some people claim there is no Proto-Dravidian and it is Tamil that gave rise to South Indian languages. Can you convince or dissuade someone about this theory?
This is a label problem. I hate label problems. No one likes label problems.

Greek is not an old language. Lots of people think it is, but it isn’t. It’s as old as just about any other language - like, say, Italian, for instance. Italian and Greek are the same age: one comes from Latin and one comes from Ancient Greek, and since Latin and Ancient Greek were the same age, so are their descendants.

Well, now, you might say, hang on a minute here. Ancient Greek is still Greek, just an ancient form of it! Yes, that’s true, but the same is true of Italian: Latin is still Italian, just an ancient form of it. You could call Italian “Modern Latin” if you really wanted to. We don’t do this, since there are plenty of other contestants for the category of “Modern Latin” - Spanish, French, Portuguese, etc. - but if you really wanted to do this, sure, you could.

The only reason anyone thinks Greek is an old language is that both Ancient and Modern forms of the language are called “Greek”. This gives the illusion that they’re the same language, or maybe just not that different, even though the difference between Modern and Ancient Greek is roughly the same as the difference between Italian and Latin.

Not that this makes Italian an ancient language, either. When Latin and Ancient Greek were being spoken, Ancient English - Proto-Germanic, it’s called - was around, too, but no one wrote it down and we don’t have many records of it. With the exceptions of some  and , every language spoken today has an ancient form like this, and so .

Notice how confusing that all was. Language names were not designed by linguists. It’d be great if they were, but they weren’t. It’s the people who speak the languages who decide when to change their name or when their language is different from another. Sometimes they never change the language name because they don’t see any need to; when that happens, you get problems like Greek.

And there are many, many, many problems that arise from these arbitrary name differences. Like, for instance, this one.

Tamil is the same age as any other language. . It was first written down sometime in the first millennium BC, and that form of the language was called . It is subject to the exact same label problems as Greek: Old Tamil and Modern Tamil are two distinct languages that happen to be related and share a name, but a speaker of Old Tamil couldn’t have a proper conversation with a speaker of Modern Tamil.

Along with Telugu, Malayalam, Kannada, and a number of others, Tamil belongs to the Dravidian family of languages, spoken mostly in southern India. In the same way that blood relatives are related because they descend from one ancestor, languages are related because they descend from an ancestor language.

At one point in the past, all the Dravidian languages were one single language. We can say that very confidently. Why? Because the Dravidian languages are clearly very similar, and the only way that similarity ever happens is when all those languages come from one language, like the Romance languages from Latin or the Indo-Aryan languages from Sanskrit.

We call this proto-Dravidian language, well, “Proto-Dravidian”. That’s all that “Proto-Dravidian” means: “the language that gave rise to South Indian (i.e., Dravidian) languages”. No one’s claiming that Proto-Dravidian didn’t exist. There are some people, though, who think that Proto-Dravidian is Tamil, which is what you’re asking about. You could reword your question as this:

Some people claim there is no Proto-Dravidian and it is Tamil that is Proto-Dravidian. Can you convince or dissuade someone about this theory?”

In that case, then, what if Proto-Dravidian is the same as Tamil? No. A modern Tamil-speaker couldn’t have a conversation with an average ancient Tamil-speaker from 2000 years ago, and that ancient Tamil-speaker certainly couldn’t talk to a Proto-Dravidian speaker from the millennia further back in time that it would have been spoken.

The Dravidian languages change more slowly than, say, English, but the differences between modern or even ancient Dravidian tongues and the Proto-Dravidian of 3000 BC are so great that to still consider Proto-Dravidian anything like Tamil would be silly. Languages change a lot; Tamil isn’t exactly immune. Normal language business, nothing to worry about, happens to the best of us.

To answer your question, Proto-Dravidian definitely existed for the reasons listed above, and since even Old Tamil and Proto-Dravidian are so far apart, there is no reason to think that Proto-Dravidian was Tamil in any recognizeable form.

Alright, new plan: if Modern Tamil is so different from Ancient Tamil From 2000 Years Ago, but they’re still both called “Tamil”, then why not call Proto-Dravidian “Tamil” too and continue claiming that Tamil is Proto-Dravidian?

And this is where we run into those pesky label problems. Exact same thing as calling  “Greek”, I tell you!

You could call Proto-Dravidian “Tamil” if you really, really wanted to. Go ahead. Knock yourself out. We don’t usually do this, though, because “Tamil” already refers to a language that still exists, so you’d have plenty of confusing conversations. “This Tamil word is ultimately from that Tamil word - but, you know, not the same Tamil.”

Besides that, the established practice for naming proto-languages is “Proto-(name of language family)”. In that case, we need a name for the entire family - the whole group of Tamil, Telugu, Malayalam, Kannada, etc., not just Tamil by itself. Oh, well, why not simply call them the Tamil languages, then? No! This makes it worse! Then you’re saying things like “This Tamil word is of Tamil origin - but, you know, still not the same Tamil”.

Let’s slightly modify the word “Tamil”, eg. “Tamilic”, and name the whole family that, then! It works for German and the Germanic family, so why not for Tamil? Yes, okay, sure, you could do that. There’s not even any problems with this. In fact, we do this already:

Dravidian: From  (drāviḍa), hypercorrection of dāviḍa, dāmiḷo, damiḷa, from tamil (mod. ).

So, uh, there you have it, folks. Congrats.

But again, that’s only a name. When Proto-Dravidian existed, all the Dravidian languages were one single language. Proto-Dravidian wasn’t something like 45% Tamil, 28% Malayalam, negligible amounts of Brahui, etc.; it was all of them, together, at once. It was just as much Tamil as it was Kannada, or , or . We call the family “Dravidian” for mostly historical reasons, nothing more.

To answer your question a second time, there was indeed a Proto-Dravidian language, which no one doubts. Contrary to what some of the other answers have said, though, it could not have been Tamil, or anything you’d recognize as it: just as today’s Tamil is very different from the Tamil of two millennia ago, the ancestor of Tamil three millennia further into the past would have been so far from Tamil that there’d be no reason beyond pure nationalism to claim they’re the same.

It’s a label problem. I hate label problems. No one likes label problems.

Thanks for asking!
By
K. Jagadeesh

No comments:

Post a Comment