Tuesday, October 9, 2018

Can Languages Borrow Basic Grammar from Each Other? Feat. Germanic and Finnish, Not Forgetting Phrygian


Only one Germanic Language has Duplicated Preterites · Can Languages Borrow Basic Grammar from Each Other? Feat. Germanic and Finnish, Not Forgetting Phrygian

Some who know me since earlier also know, I consider commonalities between diverse branches of Indo-European (IE) language community to have risen the Balkanic way, as a Sprachbund phenomenon, not the Romance way as a Common-Protolanguage phenomenon.

Now, obviously, IE grammatical traits common to all and many groups would be contradicting this, if we knew that languages never borrow grammatical featers from each other.

Consider Spanish panadero, French boulang(i)er, where the modern form loses the (i) after the soft or palatal "g", German Bäcker, English baker, Swedish bagare .... what do these have in common, beyond meaning?

Well, they have in common a Latin ending -arius, which becomes -ero in Spanish, -ier in French, -er in English and German, and -are in Swedish. Oh, while we are at it ... in Finnish a suutari makes a kenkä when a cobbler makes a shoe, here the product has a Finnish stem, while the producer looks like a Latin loan word. So also a muurari makes a seinä when a mason makes a wall, where Swedish has a Latin loan for both producer and product (murare, mur). But the ending -ri is not restricted to these loans, look at a leipuri who makes leipä. Here the stem is a word also found in Germanic (including English loaf) and in Slavic (hleb in Polish), and so we do not get the -ri ending within the general context of a Latin loan. This to me looks like -ri is productive as a producer or agent ending all over Finnish, like -are in Swedish or like -er in English.

But producer names are perhaps not all that basic as grammar goes? Oh yeah, what about numerals?

Within the Indo-European context, it is accepted that numerals 1 to 10 are same because of common ancestry, which for 1 apparently have two different words. Latin "unus" is the same word as English "one" while Greek "heis, mia, hen" is apparently rather the same word as English "same". But for 2 to 10, the words are very clearly related all over the groups.

Just, within Finnish etymology, it is equally accepted that while "kymmenen" might be unrelated to "ten" (but I am not sure about that), the words for eight and nine respectively mean "two from ten" and "one from ten" except the part meaning "ten" is not "kymmenen" but related to "ten" in a much clearer way: kahdeksan yhdeksän ... beginnings explainably reverse yksi kaksi, but the ending deksan / deksän which should mean ten if kahdeksan means two from ten, and which therefore probably means ten, actually is the IE version of the concept, and presumed - not controversially, but commonplace - to be a loan from some form of Persian or other Iranian language. If we go to Hungarian, a known relative of Finnish, "eight, nine, ten" goes "nyolc kilenc tíz". Here it is rather "ten" which could be suspected of being related to Latin "decem" and therefore to "ten".

By the way, hundred in Hungarian is száz and in Finnish sata. Recall the satem languages, in IE philology? Yes, Hungarian and Finnish borrowed the word for hundred.

With thousand, Finnish tuhat, probably, but certainly not Hungarian ezer, is the same word as thousand, or in Lithuanian tūkstantis and in Polish tysiąc. Now Baltic involves Lithuanian, Slavic involves Polish, English is Germanic, and Slavic, Baltic and Germanian are all IE - but this word is shared between them and Finnish, and not with other "branches of IE" : Latin and Celtic have "mille", Greek has kilioi, Panjabi has Hazāra, which resembles Hungarian ezer.* And Finnish doesn't share tuhat with ezer.

Or sound changes. The Germanic sound change called Grimm's law is shared with Hungarian. Suppose both Finnish and Hungarian borrow "deksan" from Ossetic or Persian or whatever Iranian language may have had such a form. Finnish keeps it basically like that, but Hungarian would make d > t, k > h, so first syllable would go from deks to tehs, which easily turns into tíz ... intriguingly, Hungarian and Germanic are neighbours - these days along the Austro-Hungarian border, but they have been neighbours longer, especially if Etruscan is a version of Hungarian.

So, if we look again at the idea that basic grammatic material is not borrowed, and then forget what we just learned why this is not so, have a look at Finnish present and Old Greek imperfect verb endings ... enter the word "leave", and in Old Greek I'll take imperfect and in Finnish present for a reason I will later explain.

eleipon lähden
eleipes lähdet
eleipe lähdee
eleipomen lähdemme
eleipete lähdette
eleipon lähdevät


In first and second person singular and in first and second person plural, we find a fairly clear correspondence, perhaps a bit doubtful for second singular.

In third singular, there is a potential correpondence, but it could just be coincidence too. Only in third person plural there is no correspondence at all. Unless perhaps one would guess that "vat" had been "vnt" and "nt" is the standard IE ending for third person singular, but this is as moot as taking vocalic ending in third singular as a correspondence, the one case could be a coincidence, the other case the correspondence if such is unclear.

Now, what would have happened if I had taken present in Old Greek?

The first person singular would have been obscured, because -o in leipo is a clearly unrelated ending to -n in lähden. Likewise, if we go from verb endings to pronouns, the -n in both Finnish and Greek is reasonably presumed to be a previous -m, and related to the pronouns me in Old Greek, and minä in Finnish. Now, me is a good accusative, but in nominative you have ego. Which is clearly heteroclite in relation to me, and also unrelated to Finnish minä.

This is the reason why -s is suspected to be related to the pronoun tu, te, which in Old Greek was sy, se - and if so, it could be related to -t in lähdet.

If you are an Old Earther, don't care a whit about Adam and Eve, or don't care a whit about the generations between Adam and Abraham, or so, you might of course say that Finnish minä and Old Greek me are related because Uralic and IE both go back to a protolanguage even earlier than Proto-Indo-European and Proto-Uralic. To a Christian, this is not an option. Coincidence is not an option to a scientific mind. So, either IE is a Sprachbund, but some other Sprachbünder have involved some of them along with Finnish, or Finnish is a hybrid language, IE and Uralic, or IE is a hybrid language, Uralic and Semitic.

I favour the idea that IE is a Sprachbund, and that the original one was Anatolian-Aegean, and that Old Cretan was Aryan, i e an old form of Indo-Persian.

It was followed by later ones.

Now, what would be the place of Germanic in Anatolian-Aegean world? Phrygian. What would the stem "phryg" be in Germanic and what was it in Phrygian? "bruk" and "bryg". "Phryges in their own language are Bryges". And Greek ph = Germanic b, Greek (and presumably Phrygian) g = Germanic k.

Could there be a reason for a Germanic concept like "bruk" to be an ethnonym? Yes, since "bruk" means "use" (English has borrowed "use" from Latin via French). A bonnet which a Phrygian would have considered "usual" (in Danish "brugelig", in German "gebrauchlich") and also "useful" (Swedish "brukbar", German "gebrauchbar") would by a neighbouring people have been considered as "Phrygian" because they didn't usually make those bonnets themselves.

Phrygian does share parts of the Grimm's law sound changes with Germanic proper and - as mentioned - with Hungarian (and its old form or sister language to old form Etruscan). Nevertheless, Phrygian and Germanic are counted as different branches.

So, if Germans are Phrygians removed from Aegeo-Anatolian home, that would explain how they came to be part of an Aegeo-Anatolian Sprachbund, known as Indo-European, precisely as Aryans originating from Crete (which is disputed, so much that the linguist proposing it took his page down, he felt he could not support it) would explain why Sanskrit and Avestic came to share traits with an Aegeo-Anatolian Sprachbund.

Hans Georg Lundahl
Nanterre UL
St. Denis of Areopagus
and of Paris
9.X.2018

* For Persian, I don't get any pronunciation guide beyond Arabic alphabet, but Kurdish is written in Latin letters : hezar. Iranian languages are closer to Hungarian than Indic ones are. And in Armenian, it is hazar.

This word probably means heap, since one thousand pebbles or grains of whatever are often found in a heap of it, and it is probably related to "hazard" in the sense of chance, like drawing one item out of a heap./HGL

No comments: