June 18, 2012

Is Burushaski Indo-European?

I try to shy away from linguistic controversies, and I doubt that the idea of the Indo-European origin of Burushaski will go down without a fight. The JIES is a little behind the times when it comes to online access, so I can't quite comment on the substantive part of Prof. Casule's theory.

We'll have to wait to see how much of this is press release hype and how much is new evidence. This idea has been in circulation for a while, but if there is indeed an issue of JIES dedicated to it, it probably deserves our attention.

Here is a paper critical of the thesis. Here is a seminar abstract by Prof. Casule with some more citations.

Cracking the code on the origins of a new European language
There is strong evidence to support the discovery of a new European language.

Macquarie University historical linguistics researcher, Associate Professor Ilija Casule, discovered that the language, known as Burushaski, which is spoken by about 90,000 people who reside in a remote area of North West Pakistan, is Indo-European in origin, not Indo-Iranian.

Professor Casule’s discovery, which has now been verified by a number of the world’s top linguists, has excited linguistics experts around the world. An entire issue of the eminent international linguistics journal The Journal of Indo-European Studies is devoted to a discussion of his findings later this month.

More than 50 eminent linguists have tried over many years to determine the genetic relationship of Burushaski. But it was Casule’s painstaking research, based on a comprehensive grammatical, phonological, lexical and semantic analysis, which established that the Burushaski language is in fact an Indo-European language most likely descended from one of the ancient Balkan languages. Professor Casule believes that language is most probably ancient Phrygian.

The Phrygians migrated from Macedonia to Anatolia (today part of Turkey) and were famous for their legendary kings who figure prominently in Greek mythology such as King Midas who turned whatever he touched into gold. They later migrated further east, reaching India. Indeed, according to ancient legends of the Burushashki people, they are descendants of Alexander the Great.

Tracing the historical path of a language is no easy task. Professor Casule said he became interested in the origins of Burushaski more than 20 years ago.

“People knew of its existence but its Indo-European affiliation was overlooked and it was not analysed correctly. It is considered a language isolate – not related to any other language in the world in much the same way that the Basque language is classified as a language isolate,” he said.

The remoteness of the area that was independent until the early 1970s when it became part of Pakistan, ensured Burushaski retained certain grammatical and lexical features that led Professor Casule to conclude it is a North-Western Indo-European language, specifically of the Paleobalkanic language group and that it corresponds most closely with Phrygian.

Dr Casule’s work is groundbreaking, not only because it has implications for all the Indo-European language groups, but also provides a new model for figuring out the origins of isolate languages – where they reside in the linguistic family tree and how they developed and blended with other languages to form a new language.

Image: Map of Burushashki speaking areas


Andrew Oh-Willeke said...

Just to be clear, Indo-Iranian languages are by definition, Indo-European. The Indo-Iranian language subfamily split from Proto-Indo-European is one of the more basal in the Indo-European linguistic tree.

The argument here is not over whether Burushaski is Indo-European, or even whether it is among one the Eastern side of one of several isoglosses (such as Satem v. Centem) with an East-West divided (but different precise boundaries) within the Indo-European languages.

It is also not over whether Burushaski is an example of a previously unknown branch of the Indo-European languages, and it is not a discovery of a previously unknown language as Burushaski has been known to exist and has been known to be in the Indo-European language family for many, many decades, maybe even a century. It has also long been known that it is not Indo-Aryan (i.e. Sanskrit derived). And, it has finally, long been known that it isn't a particularly close linguistic relative of the Indo-Iranian languages of its immediate geographic neighbors.

The controversy is simply over which of the mostly Eastern isogloss branches of the Indo-European language tree it belongs to, a Balkan-Western Anatolian one, as proposed, or an otherwise ill attested isolate branch within the Indo-Iranian group.

Yes, the story is interesting, as are all long distance migrations of whole peoples that are clearly known from history that lead to distant outposts of language families. A Phrygian Macedonian origin would also be a millennium or two more recent than the very basal Indo-Iranian language theory, which nixes the value of Burushaski as a window into very early PIE linguistics but adds detail to the Macedonian expansion story - in the later the Burushaski speakers would have conquered or displaced people who spoke an Indo-Iranian or Indo-Aryan language, rather than a non-Indo-European language.

I also concur that at least a good sampling of the evidence in favor of this theory is necessary to really evaluate it. Some linguistic theories are rubbish with flimsy and strained support, while others really are quite solid and quickly become consensus views.

Dienekes said...

has been known to be in the Indo-European language family for many, many decades, maybe even a century.

Not sure where it was "known", but it is not listed as IE in any standard reference work.

German Dziebel said...

Based on my reading of Casule's 1998 and 2003 works, I doubt the proposed connection. But since then he published a book that I haven't read. A short review thereof by Vajda (the linguist behind the Ket-Na-Dene proposal) is worth mentioning: http://elanguage.net/blogs/booknotices/?p=1863

Anonymous said...

Burushaski is overwhelmingly considered to be an isolate.

Also, the history of Balkan languages is too poorly known for any definitive comparison, and so any bold claims over their genetic relationship at this stage are necessarily speculative.

That news article obviously gets it wrong with "is Indo European in origin, not Indo-Iranian", which is already indicative of the quality we should expect from this kind of hype piece.

And, in fact, Casule's own theory is not new at all. He's been touting it for many years, since the 1990s, with little success. I don't think it would be too harsh to call his ideas "fringe", because, so far, that's what they appear to be.

Could he be right? Yes - but the evidence is so scant and the suspicion of specialists so pervasive, that it is really unlikely.
And I'm not even talking about the motives (the accusations that Macedonian nationalism is driving his theories), as that's another story.

As for Bengston's paper, his "Dene-Caucasian" is equally fringe, so to some extent we have there a case of pot calling kettle black.

Jaakko Häkkinen said...

Even other distant comparers do not see the hypothesis (promoted by Čašule for a long time) credible:

AdygheChabadi said...

This is bunkum...Burushaski is well-established as a linguistic isolate possibly distantly related to Northeast Caucasian languages.

Burushaski may have burrowed from IE via contact, but it, per se, is not at all an IE language. The core vocabulary is alien to any IE language and, again, this has been established beyond any reasonable doubt.

I typed something similar to that like 9 minutes after Dienekes posted this article. Strangely, Dienekes, either, chose to not post it or he, somehow, didn't receive it.

terryt said...

"Burushaski is overwhelmingly considered to be an isolate".

That is the way I understand the situation.

Nirjhar007 said...

Few things apart from Burushaki,
first, Satem v. Kentum.
The latest and most accepted view today is that 'PIE' wasnt Kentum at all! but had a particular palatal sound which evolved to Kentum and then Satem that is why Sanskrit have the palatal sibilant! and there is examples also that 'Kentum' languages evolved satemization like in Greek 'Syn' for Latin 'Cum' (Skt. 'sam'), there is contradicting examples also for Kentum location like Bangani have the word 'Koto' for 100 the language which is from 'Satem' South Central Asia.
Secondly, Sanskrit is the modified and grammarized form of Vedic! So by Skt practically you don't mean whole 'Indo-arya' but the part which was conserved by worlds first grammarians like Sakatayana,Yaska and Panini.
Have a good time.

eurologist said...

If it's an existing language spoken by 90,000 people - how can it be so difficult to classify? If it were some ancient IE, surrounded by lots of IE speakers, how can it become so isolated as to become unrecognizable? That doesn't even happen in a 4,000 - 5,000 year time frame.

Phrygian is ludicrous given how recent it is. Some other, ancient Anatolian IE >4,000 - 5,000 years old, I could believe. But then it still would need to have a huge amount of Caucasian and/or Turkic mixed in to make it so different. And a good linguist should be able to peel away such layers.

There were other, no-IE, non-Semitic, and non-Caucasian languages spoken in Anatolia. My guess is, it is more related to one of those.

Nirjhar007 said...

Sorry, I don't see any IE language crossing the barrier of ~2000b.c. Whether its Anatolia or anywhere else in parctical evidences pov:-( but you are quite right in the rest:-).

eurologist said...

I don't see any IE language crossing the barrier of ~2000b.c


PIE is widely considered to be 6,000 to 9,000+ years old. Just because nothing was written down until ~4,000 years ago doesn't make it 4,000 years old. It takes time for all the subgroups to form - a process that was largely completed by 3,500 to 4,500 years ago.

John Cowan said...

The error in the press release is trivial: a single-word repair will change it from "Indo-European in origin, not Indo-Iranian" to "Indo-European in origin, but not Indo-Iranian", which aligns with Čašule's claims.

Jim said...

eurologist, I understand your frustration:
"If it's an existing language spoken by 90,000 people - how can it be so difficult to classify? If it were some ancient IE, surrounded by lots of IE speakers, how can it become so isolated as to become unrecognizable? That doesn't even happen in a 4,000 - 5,000 year time frame."

But classification ids hard to settle. that happens to be the time range for both proto-Salishan and as it happens proto-Algonkian
and maybe also for the split form Kootenai (Ktunaxa) if a there really was a connection in the forst place. Those language groups all share some pretty tanatalizingly unusual syntactic features and as it happens their proto-homelands were basically adjacent. But oopsie - not much shared vocabulary, at least not that has been identified.
So for now they are not classified as related in some larger grouping.

If htere were no historical texts and if Welsh had gone extinct and was not available for comparison, there would be no reaosn to think Irish was not an isolate within western IE, just like Basque. You could easily expalin away, inaccurately, a whole raft of similarities as borrowings. That would not be because they were borrowings, but because the evidnec of affilaition had perished. Languages don't have DNA to compare; the only evidence is what you would call morphological.

Rob said...

I thought we had realtively substantive amount of data on Phrygian, as far as ancient non-Greek palaeoBalkan languages go

Charles Nydorf said...

I don't know enough to have an opinion about the affinities of Burushaski but I was delighted yesterday to hear Professor Casule being interviewed on National Public Radio in the US as part of the 'All things considered' evening news.

Nirjhar007 said...

Yes indeed, but there is a gigantic gap between what is considered and what really is;-).

Anonymous said...


NON-CAUCASIAN languages in Anatolia? That sounds interesting.