May 10, 2012

Unknown language discovered in tablet from the Governor’s Palace in Tušhan

The Independent covers this:
Archaeologists have discovered evidence for a previously unknown ancient language – buried in the ruins of a 2800 year old Middle Eastern palace.

The discovery is important because it may help reveal the ethnic and cultural origins of some of history’s first ‘barbarians’ – mountain tribes which had, in previous millennia, preyed on the world’s first great civilizations, the cultures of early Mesopotamia in what is now Iraq.

Evidence of the long-lost language - probably spoken by a hitherto unknown people from the Zagros Mountains of western Iran – was found by a Cambridge University archaeologist as he deciphered an ancient clay writing tablet unearthed by an international archaeological team excavating an Assyrian imperial governors’ palace in the ancient city of Tushan, south-east Turkey.

The tablet revealed the names of 60 women – probably prisoners-of-war or victims of an Assyrian forced population transfer program. But when the Cambridge archaeologist – Dr. John MacGinnis - began to examine the names in detail, he realized that 45 of them bore no resemblance to any of the thousands of ancient Middle Eastern names already known to scholars.


Typical names, born by the women – the evidence for the lost language – include Ushimanay, Alagahnia, Irsakinna and Bisoonoomay.

Journal of Near Eastern Studies
Evidence for a Peripheral Language in a Neo-Assyrian Tablet from the Governor’s Palace in Tušhan (pp. 13-20) John MacGinnis DOI: 10.1086/664450 Stable URL:


cris said...

Very exciting and awesome finding, it gives me the chills!!
I think it may be the Kassite or Gutian or Lulubi languages

Onur Dincer said...
This comment has been removed by the author.
Onur Dincer said...
This comment has been removed by the author.
berry k. said...

The name Bisoonoomay has an Australian Aboriginal ring, somehow

Onur Dincer said...

But when the Cambridge archaeologist – Dr. John MacGinnis - began to examine the names in detail, he realized that 45 of them bore no resemblance to any of the thousands of ancient Middle Eastern names already known to scholars.

That means that in all probability they belong to a language not previously attested in ancient West Asia. This seems to rule out the Anatolian IE, Hurro-Urartian, Hattian, Semitic, Sumerian, Armenian, Indo-Iranian, Greek, Phrygian, Elamite, Kassite and Egyptian languages. They may indeed belong to a previously unknown language.

andrew said...

I'd love to read more analysis from someone whose seen more. It is indeed hard to match.

wagg said...

There were probably more extinct languages of no known language family anyway, in this general region, like the hypothetical "banana languages".

Unknown said...

Very exciting, I can't wait for further development! The Proto-Euphratean possibility would indeed be fascinating.

Onur Dincer said...

I think it may be the Kassite or Gutian or Lulubi languages

Perhaps. But I think there are enough known Kassite, Gutian and Lulubi personal names recorded in ancient texts to make them recognizable in such a case. BTW, Kassite is thought to be a Hurro-Urartian language; this makes it even worse as a candidate.

Casper said...

"their men young and old I took prisoners. Of some I cut off their feet and hands; of others I cut off the ears noses and lips; of the young men's ears I made a heap; of the old men's heads I made a marinet. I exposed their heads as a trophy in front of their city. The male children and the female children I burned in flames; the city I destroyed, and consumed with fire."
Ashurnasirpal II, 850 BC
Assyrian King

eurologist said...

The names have a Pakistani/Indian sound to me. What is known about pre-IE languages between Iran and Pakistan?

AdygheChabadi said...

Hi, Dienekes!

Omg, I nearly wet my panties when I read very exciting this discovery is!

I am inclined to believe it is more one the unwritten languages spoken of, but never Hamazian, Lullubi, Mannaean, Kaskian, and Gutian.

I am inclined to think of the Kassites as a Hurro-Urartian people as based on onomastics...their names make sense in Hurrian. Hamazian, Lullubi, and Mannaean may also be Hurrian-related.

As for the so-styled, 'Banana Languges'...Many ancient languages had a reduplicative quality, such as Elamite, 'Minoan', and, to a lesser extent, Hurrian. I defer to Mr. Rubio's paper on the 'Sumerian Problem' and Mr. Francfort's paper relating Marhasi to the Oxus Civilization (BMAC) and with some type of eastern Hurrian language.

I am extremely interested in what the language could be especially since the tablets were found in the sphere of Hurian Mittani. A caveat would be that the Assyrians are well-known for displacing peoples,

AdygheChabadi said...

To finish my thoughts on Mr. Francfort's paper...I believe the BMAC was Burusho-related and Marhasi to be found between the Indus Valley Civilization and Elamite Shimashki, placing it in the area of Gedrosia/ Makran/ Balochistan.

Furthermore, the paper has an interesting section on the so-styled, 'Banana Languages'.

Andrew Oh-Willeke said...

Eurologist's comment about a South Asian sound to the language, while vague, seems right. The names sound aurally much more like Indian/Pakistani names than they do like Aramaic, Elamite, Minoan, Hattic, Hittite, or Sumerian names.

The notion that some South Asian names, perhaps from a Harappan substrate, could have made their way to this region with the Mittani, and persisted in the local language is as plausible as any.

Harappan is one of the best attested languages known to have existed at some point which is completely unknown in lexicon, phonetics or grammar, but probably has some kind of substrate influence on Rig Vedic Sanskrit and Avestian Old Persian and would contribute to how modern Indo-Aryan languages sound relative to other Indo-European languages.

Obviously, the Indus River Valley civilization has collapsed a couple of millenia before by 800 BCE, but proper names could easily survive much longer than the language itself, and our Rig Vedic source for early Sanskrit (and the Avesta) could easily have omitted many proper names for commoner women causing them to disappear from direct evidence of a substrate Harappan language.

Ovibos Nunivak said...

Irsakinna sounds like a name from a greater Mesopotamian Sprachbund. I agree with other commentator, more information needed. Would like to see a paleolinguist's analysis.

pconroy said...

I know an Indian whose last name is Bishoo, and Usha is a Vedic Goddess:

So my guess is that they were Mitanni - an early Indo-Aryan language!

Unknown said...

Anybody has access for full list?

cris said...

When googling various variants of the women's names (before that we must assume that the -a and -ay in the end of those words are suffixes of the feminine or sthg like that), the best matching points to Hindi-Armenian
1/Ashima Bhalla, a Hindian actress

2/Kahnia is a hindi word

3/Arshakian is an Armenian surname

4/Finally shunum is also a Bengali word (bi is a prefix that could mean without as the name bi-nazeer wich means "without equal" (to her)

It could be Mitanni!

Belenos said...

If they were Mitanni, I would have thought more of the names would have been identified as such, and the paper would have tentatively suggested that identity. We have quite a stock of very Indo-European names associated with them.

Could be IE, but -a endings for feminine names are so common across language families that this can't really be used on its own as evidence of IE, much less Indo-Iranian.

eurologist said...

So my guess is that they were Mitanni - an early Indo-Aryan language!


Mitanny means nothing outside IE unless the substrate is isolated. My guess would be Elamite - which has been suggested to have a Indian/ Dravidian connection, before.

Kurti said...

Some people seem to think it might be Mitanni related. Well let me tell. Kurdish historians were sure about this even some years before. Some Kurdish historians claim that the Guti were some sort of Proto Indo-Iranians (Before they did split into a Iranian and Indo Aryan branch) which mixed with Hurrians and created the Mitanni Kingdom.

Kurti said...

Findings belong most probably to Gutians, which I see as the name givers of Kurds. According to some archeologists a mountain people living all the way between Zagroß and Taurus mountains were called Gutians. God knows what kind of language they spoke. Iranic or another proto Indo-European? or maybe isolated? Another fact a very famous mountain in Kurdish literature and history is called "ciyaye Judi" and translated means the mountain of Judi(which is a loudshift from Gudi). Its located in North Iraq on the border to Southeast Anatolia.

Anonymous said...

I have got a gift for the friends of this blog. Not long ago, I was having a look at the statues representing the wives of Pharaoh Akenaten (spelling varies). So anyway, it appears that Akenaten (an obviously lucky man) had many wives. It would seem that each alliance pact he did, he would cement by marrying a woman from the newly allied people, typically a noblewoman (ofc). Ladies and gentlemen please meet Kiya, wife of Great Egypt Akenaten, and a Mittani Princess. Kiya lived ca. 1350 BC, abt 100 years before the Trojan War. (I wish I was Akenaten).

Onur Dincer said...

Kurti, Mount Judi is located in the Şırnak Province of southeastern Turkey, not in northern Iraq. It is significant not just for Kurds, but for all Muslims, as it is the location where Muslims traditionally believe Noah's Ark landed after the deluge.

Nathan said...

pconroy : " So my guess is that they were Mitanni - an early Indo-Aryan language!"

Mitanni was an empire with an Indo-Aryan ruling elite or an elite with significant Indo-Aryan presence but the language of the Mitanni empire was Hurrian (a non Indo-European language).

So it was a case of elite dominance by Indo-Aryans but one that didn't succeed in replacing the language of the locals.

Andrew Oh-Willeke said...

The possibility that the source community, which could very well have been in an mountainous endorheic basin, could have been non-Indo-European (and in particular not Indo-Aryan) with similarities to Hindu names arising from substrate influences shared with Indo-Aryan languages, is quite real. I've seen enough Elamite and Hurrian names to doubt a connection between them (or with bannana languages for that matter), and I would think that Mittani king and priestess lists and geneologies would have provided enough insight into proper names from those languages to rule out that possibility, although it could be that Mittani language records are short on female as opposed to male names.