I have noticed that several "Altaic" languages have similar words for "language," but I do not know whether this is a coincidence, or due to historical language contact.

The word "kieli" in Finnish is similar to "khel" in Mongolian and "dil" in Turkish. All three of these words mean "language."

Has the original source of this word been identified in any historical linguistic studies?

  • 2
    If you're expecting a common origin, or a borrowing going back a millennium or three, you should be using the reconstructed proto-Uralic, proto-Turkic, and (when available) (pre-)proto-Mongolic, not modern languages (and especially not modern languages with as much recent borrowing as Finnish and Turkish).
    – abarnert
    Commented Feb 10, 2019 at 18:32
  • Also, your question title doesn't ask the same thing as the body. If you want to know whether and/or how the Uralic root is related to other language families, ask that; if you really do just want to know "From which language was the Finnish word for 'language' derived' the answer is just "Late Proto-Finnic" (aka "Gulf of Finland Finnic"), which is not a very interesting or useful answer, and could be looked up in any dictionary (even Wiktionary).
    – abarnert
    Commented Feb 10, 2019 at 19:13

2 Answers 2


The exact nature of the relationship between Uralic and "Altaic" languages has so far remained quite obscure. Most specialists of these languages tend to be negative, though this might be a bit too severe. Finnish kieli has good comparanda in the rest of Uralic and indeed somehow looks related to Mongolian. Turkic dil is obviously another word.

  • The "Altaic" language classification is often considered obsolete as a language family, though there is evidence of extensive language contact between the Turkic and Uralic languages, including some similar personal pronouns. Commented Feb 9, 2019 at 19:00
  • @AndersonGreen Wondering about the "obsoleteness" of the concept of Altaic languages: Can the Ghengis Khan empire have been a strong historical link which has kept or even enriched the closeness between Altaic, Uralic and Turkish langauges? In which case that would not be so irrelevant. Commented Feb 10, 2019 at 21:15
  • 3
    @StephaneRolland Anderson isn't saying that the family is obsolete (as in broken up into no-longer-related languages); he's saying that theory that such a family exists (the idea that the languages are genetically related) is obsolete. And if you're right that the Mongol Empire enriched the closeness of a bunch of languages, that would count against the Altaic theory—it would imply that much of the apparent similarity of Turkic and Mongol languages is do to borrowing/contact/areal features, so there's even less evidence of a common ancestor than at first appears.
    – abarnert
    Commented Feb 11, 2019 at 20:07

The "Altaic" languages are generally not considered a valid group, not to speak of Ural-Altaic, so the words you mention do not go back to one original source, but rather to two separate sources, one for each family.

The Uralic words you mention, go back to Proto-Uralic *käle: https://en.wiktionary.org/wiki/Reconstruction:Proto-Uralic/k%C3%A4le

The Mongolian and Turkish words, on the other hand, go back to Proto-Turkic *dɨl. You can find more cognates for that word here: https://en.wiktionary.org/wiki/dil#Azerbaijani


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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