Although not belonging to the Indo-European family, Finnish has personal pronouns that resemble (to a layperson, at least) the corresponding pronouns in Indo-European languages. For example, the consonant /m/ in the first person pronouns minä (singular) and me (plural) is also very common in first person pronouns in several Indo-European languages. Similarly for the second person sinä/se and for the third person hän/he. Interrogative pronouns kuka ("who") and kumpi ("which") also seem to share the consonant /k/ with Indo-European.

What is the relationship (if any) between these pronouns?

  • who and which are interrogatives, not demonstratives – Anixx Jun 10 '13 at 6:00
up vote 8 down vote accepted

Pronouns can be very tricky, so you should be very careful about using them as evidence of genetic relationship. At first, pronouns can be borrowed (and it seems to happen rather frequently, see Campbell and Poser 2008: 212-214). Secondly, similarity in form can be explained by what Nichols (2003) calls "resonance" (e.g. rhyme, alliteration etc.). That is why Antoine Meillet (1948) argued that

"Pronouns are similar in almost all languages, though it does not imply a common origin. [...] Therefore, pronouns must be used with caution in establishing relatedness of languages."

These pronouns are one of the arguments offered in support of the Indo-Uralic hypothesis. But that hypothesis is considered pretty shaky by most linguists: either because they disagree with it or because they think there's just too little evidence to tell.

Anyway, there's other possibilities. It could also be that PIE and Proto-Uralic weren't especially closely related, but that one borrowed pronouns from the other. Pronoun borrowing is rare but not impossible. It could also be that the whole thing is a coincidence. There are dozens of function words and morphemes in any given language. Between two neighboring languages, you'd expect a few tempting matches like this purely by chance.

Part of the trouble is, Proto-Uralic is harder to reconstruct than PIE, because the written record for Uralic languages is a lot shallower. The oldest Uralic writing, from medieval Hungary, is less than a thousand years old. (By comparison, the oldest IE writing goes back over 3,000 years.) So it's hard to be very confident about what Proto-Uralic looked like, and harder still to be very confident about its history going further back. We may just never have enough data to say anything certain about the Uralic/IE connection.

  • Indeed, not all scholars accept Uralic as a valid taxonomic node. See for example Marcantonio, Angela: The Uralic Language Family: Facts, Myths and Statistics. 2003. – Colin Fine Nov 5 '13 at 9:58

The similarity of Finnish hän and Scandinavian hann / English he / etc. is coincidental, or a case of later convergence. Germanic *h goes back to earlier *k (thus hann / he / etc. may be related to the -c seen in Latin hic "this"). By contrast, Finnish h usually reflects earlier , or sometimes another coronal consonant: a cognate of hän is North Sami son.

It is possible that the frequency of (or preference for) hän has been partly influenced over time by its similarity to Scandinavian hann, but there is no etymological relationship between the two. In Estonian, the 3rd person pronouns are from an entirely different stem: ta "he" and nad "they", related to Finnish tämä "this" / nämä "these".

The Finnish pronouns are minä, sinä, hän, me, te, he, and other Finnic languages have a similar pattern. Siberian Finnic languages have the pattern (examples are form the Erzya language) mon, ton, son, min, tin, sin. In Hungarian én, te, mi, ti (I, thou, we, you), Turkic men, sen (I, thou). Indeed, these are an "extraoridinary coincidence", and the verbal endings in Finnish are (-n, -t, -, -me, -te, -vat) and those in Teutish (aka IE) are -mi, -si, -ti, -me, -te, -nti.

  • 1
    Hi Conan, I edited your post to make it more readable. If you feel this changes the content of what you wrote feel free to correct or modify my edit. – robert Nov 3 '13 at 22:05

Similarities between the personal pronouns of languages of northern Eurasia (not just IE and Uralic) has been noted many times by different individuals. It's been claimed to be evidence for the proposed Nostratic family and other similar controversial macro-family proposals.

Since these proposals are not mainstream, there are several possible mainstream explanations:

  1. /m/ for first person and /t/ or /s/ for second person may represent a universal linguistic tendency like mama/nana and papa/tata words for "mother" and "father". This explanation doesn't seem plausible because the phenomenon seems to be largely confined to the Old World (except sub-Saharan Africa). For example in Americas the pattern /n/ - /m/ seems to be much more common.
  2. Pure chance. Widely used function words (like personal pronouns) tend to be short and chance resemblances are almost inevitable. But locality of the phenomenon seems to suggest a rather different explanation.
  3. Areal features due to language contact (borrowing in this case). It's often stated that pronouns are rarely borrowed and it seems to hold true for most historically known Eurasian languages. But it's also known that some other languages (like Pirahã) borrowed their whole set of personal pronouns. So, maybe, this constraint holds true in only certain cultural settings. It may well be the case that pronouns were borrowed in Eurasia in prehistorical times.

It is possible that certain pronouns were borrowed into Uralic languages from Indo-European tongues. Other borrowings that would seem rare have found their way into the family, such as the Hungarian "száz", Finnish "sata", and Khanty "sot" for 100, which night have been either from an Indo-Iranian root like "satem" or Slavic "sto/sotnya". Scythian, an Iranian language, existed in central asia thousands of years ago and surely interacted with the early Uralic tribes; thus it might not be inconceivable to say that there were some borrowings from a more advanced society, and that there might have been some bilingualism. Scythian could have been a supplier of valuable items from the rest of Asia, like gold and spices, and therefore, it would have benefitted tribal chiefs and noblemen to know Scythian or a related language. However, there is still the possibility of false cognates here, like Hungarian "ház" for "house" (Finnish: koti; Proto-Germanic: husan). Such might be the case in the relative pronouns. As for "sinä/se", there would not be enough data to prove that the voiceless dental in Indo-European becomes a voiceless sibilant in Uralic, that is to say that "*tuH > se" is inconclusive at best. Nevertheless, cognates false or not never definitively prove a genetic relationship. The grammar and morphology must be considered. Finnish has thirteen cases, like the essive, partitive, and illative, which do not correspond morphologically with the eight cases of Indo-European well. Furthermore, cases are not normally innovative, and are usually barrowed from other languages that have them, like the adessive and illative cases in Lithuanian. Moreover, Finnish observes vowel length, with is not common in Indo-European languages, which might employ pitch-accent.

The similarities between pronouns and other grammatical elements in IU and U are too many and too systematic to accept any other explanation than some kind of genetic relationship. But you also have the same similarities with the so-called Altaic languages, and it is not evident how the fundamental branchings go. Also the similarities can go back to some kind of prehistoric Eurasiatic dialectical continuum, rather than a very distinctly confined proto-language.

Your Answer

 

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

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