Sign up ×
Japanese Language Stack Exchange is a question and answer site for students, teachers, and linguists wanting to discuss the finer points of the Japanese language. It's 100% free, no registration required.

In India, regional nationalism is strongly tied to language. This is particularly the case in the Dravidian-speaking south, especially among speakers of Tamil - Tamil nationalists trot out all manner of absurd claims about the ancientness and purity and divine inspiration of the Tamil language, and about the great power held by the ancient Tamil people (until they were oppressed by the wretched Aryans from the north, of course).

I point this out to provide context for the following question: is there any actual evidence that Tamil (or any other Dravidian language) has a genetic relationship with Japanese?

My cursory reading on the matter has pointed me to Ono Susumu, who appears to have put forth a hypothesis claiming a genetic link - but that was back in the 1980s, and it appears that his hypothesis has been refuted (that said, I'm basing this on secondary writings - I haven't read the actual papers in question). I am unaware of any recent work that bolsters Ono's conclusions.

Yet, Indian news outlets keep on putting out new articles claiming Tamil-Japanese links (e.g. here and here), to the point that many educated Tamil non-linguists take the claims of a Tamil-Japanese genetic relationship as almost-fact. These beliefs are usually of the form "A long, long time ago, the mighty Tamil people traveled through China, crossed into Japan, and brought with them their language, which mixed with or displaced the language already present in Japan at the time" - and you can certainly see how this fits in with the Tamil nationalist narrative I mentioned above.

All that said, I'm left wondering if there's some linguistic (or historical) evidence I'm unaware of that makes claims of a Tamil-Japanese link more plausible.

share|improve this question
Normally I would vote to move this to linguistics.SE but I see it has votes and no complaints so if the community finds this on topic then I won't object. – hippietrail May 22 '14 at 1:47

2 Answers 2

Just given the archaeological record, any such Tamil claims seem unlikely in the extreme, unless the proponents of this view also intend to make the Tamil the ancestors of the modern Koreans.

In terms of material culture, the Yayoi people that became the modern Japanese were pretty clearly from continental Asia, and they entered the Japanese archipelago from the Korean peninsula. They had metalworking and horses, unlike the Jōmon, and Yayoi pottery is more similar to mainland pottery than the distinct and lavish designs seen in late Jōmon pottery (see towards the bottom of the Wikipedia article). The beginning of the Yayoi immigration is often estimated at around 300 BCE, some 300 years after the creation of some parts of the Tamil-language Sangam literature.

Linguistically, Old Tamil and Japanese do apparently share the past / non-past distinction in verbs, with no future tense, and they apparently also share a vaguely similar negative construction. There is also some distant similarity in the syntax, in that both are SOV languages with verbs at the end. However, past there, the languages don't seem all that related. Someone who is more familiar with Tamil and the Dravidian family may be able to elucidate more, but in general, the big differences I've noticed are: 1) substantially different consonant inventories; 2) grammatical person in verb constructions in Tamil, not found in Japanese at any stage; 3) grammatical gender in nouns in Tamil, also completely alien in Japanese; 4) no distinction between adverbs and adjectives in Tamil, which is more like German than Japanese; 5) an apparent distinction between definite and indefinite in Tamil (compare English "the" vs. "a"), which doesn't really happen in Japanese.

Genetically, the Yayoi had more in common with the modern Koreans and Han Chinese than they did with the Jōmon people that had been in the archipelago for at least the preceding 10,000 years (some estimates push the Jōmon arrival back to around 30,000 years ago). Some genetic testing suggests a similarity between the modern Japanese people and some Southeast Asian and/or Melanesian populations. It bears noting that Denisovan hominin descendants have been found in similar geographical locations, and given the original Denisovan finds in central Siberia, a counter-claim could be made that it wasn't ever the Tamil moving into Japan, but rather a common ancestor population moving and branching, with some branches possibly winding up in the Japanese archipelago and some branches possibly winding up in southern India. That said, this Denisovan genetic link may well have been via the Jōmon rather than via the Yayoi.

In summation, a Tamil-Japanese link seems tenuous at best. Sure, humans get around, so I would be entirely unsurprised if some Tamil-specific genetic link surfaces in the Japanese genome. That said, it appears to be extremely unlikely that the Tamil language or culture is in any substantive way related to the Japanese language or culture, unless some solid link can be found between a migrating Tamil-related population and the Yayoi, who may have originated from somewhere around Jiangsu Province in China.

share|improve this answer
Your view of Yayoi seems the one before human DNA was tested. Currently it's getting reasonable that (Y-DNA haplotype) O3 was from Korean peninsula and O2b who brought wet rice cultivation was not via peninsula (due to a research of rice plant DNA). "Yayoi" usually stands for people who brought wet rice cultivation, so the definition is now waving. – user4092 Jan 31 at 14:07

Certainly at some point Japanese fishermen shared drinks/conversation with Tamil speakers somewhere along the coast of Malaysia & perhaps brought back some colorful new vocabulary (or spouses) ... but that's very different from a linguistic/etymological link on a grand scale.

As you say, Tamil (like many societies) has its proponents who claim it's the oldest root of language, has the oldest written works, etc. It's far more about ethnic/national pride than supportable claims. It is a very interesting language, but like other Dravidian languages it has distinctive grammatical, phonological, and graphical features not found in historical or modern Japanese.

  • Tamil has always had gendered (pro)nouns, plural/singular distinction, and declension/inflection of nouns & verbs for numerous grammatical cases in a sentence. These are all still very common features among Indo-European languages, but they don't exist in Japanese (or Chinese, or Korean), which uses particles instead & doesn't even have plural nouns, much less gender or declension.
  • Tamil has distinctive sounds not a part of Japanese, including retroflex consonants, at least 3 different "R" variations, and nasal sounds. These are all an important part of the language (vs. merely a sign of dialect as a nasal "が" or rolled "R" might be in Japanese). Retroflex sounds common in China haven't even had an impact on Japanese over the past 3000 years, so I'd argue Tamil fails on this count as well.
  • A key argument for a grammatical connection seems to be that both are agglutinative languages. So are Turkish and all its cousins across Central Asia, Korean (the closest actual suspected influence), Eskimo dialects and Quechua. Not a compelling argument for Tamil specifically, but more in line w/what's already suspected about Asian migration patterns.

  • Tamil doesn't distinguish between voiced/unvoiced or aspirated/unaspirated consonants, which is the exact opposite of Japanese (and Korean), where proper differentiation of "p/b", "k/g", "t/d", and "s/z" is very important.

  • Tamil's abugida/syllabary represents at least twice as many phonemes as Japanese, so of course it can claim some similarities - but so can Spanish. If Tamil was so influential why has Japanese historically been so phonetically limited? Even accounting for "lost" phonemes & vowel differences you don't approach anything like Tamil's 200+ combinations.

  • I once worked w/a very polite & well-educated Tamil man who claimed the oldest written documents in the world are in Tamil ... and yet it had zero influence on Japanese writing. The oldest known written Japanese was in Chinese. To my knowledge Japanese has never been anything like Manchurian, Tibetan, Mongolian, Thai, or any other written language with even a remote connection to Brahmic scripts.

The only way there's a link, beyond those imported through Buddhism across Asia like other Sanskrit-related words, is if you trace ALL languages back to a few shared ancestors - in that case, ok it's kind of like genetics, and they're linked ... But in that case so is everything else so it's a meaningless claim.

share|improve this answer
Correct me if I'm wrong, but I was under the impression that "genetic relationship" was a term of art in linguistics that didn't necessarily mean genetic-as-in-DNA. – senshin Mar 6 at 23:41
@senshin - You're right, and it was pretty clear that's what you meant - it just got me thinking along biological lines initially (how else would such a linguistic link have been established if not through personal interaction/migration/procreation?). Edited my answer to focus on the linguistic differences - doesn't change my opinion really that the conclusion of a link isn't well supported. – mc01 Mar 9 at 16:27

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.