sources for person & number agreement

[Xephyr]

Also he says in his own conlang page that he first heard of Novgorodian on Wikipedia, hardly a scholarly setting ...

The key word there is "first". Do you think most astrophysicists first learned what "space" was in college? More than a few linguists first got interested in linguistics long before their first linguistics class.

It's also a funny accusation coming from the guy who was quoting "Professor Wikipedia, PhD" back on page one.

[R.Rusanov]

If you have a big interest in language contact I don't fault you for seeing it in places where it doesn't actually exist. As they say "if all you have is a hammer, everything looks like a nail"

You argue for two different examples of Finnish contact,

1. "mihi type" constructions
These are what Indo-European originally had for possession and in this regards Russian is conservative, not innovative. I don't think you've studied much Proto-Indo-European and Proto-Slavic if you can make such an egregious claim

2. copula dropping
Bearing in mind that copula dropping is present - as native speakers have explained many times ITT - in Slavic languages far divorced spatially from the Fennic realms there is no justification for your argument. Besides "finnish and russian both drop copulas and are more or less near each other QED copula dropping comes from Finnish" which is obviously illogical.

I don't recall Hungary ever being part of an organized slavic state, unless you count being a Soviet puppet state.

Slavs used to live in Pannonia (under Avar leadership mostly) before the magyars came. Many Slavic place names and words survive - Balaton and Szolnok, for example, come from Slavic Blatnograd and Solnograd respectively - and 1/5 or more of Hungarian roots are Slavic in origin even today

[Xephyr]

If you have a big interest in language contact I don't fault you for seeing it in places where it doesn't actually exist. As they say "if all you have is a hammer, everything looks like a nail".

Yeah man, like, "experts", what do they know, am I right? TEACH THE CONTROVERSY

[R.Rusanov]

He's hardly an expert, though, just another guy with a penchant for one specific area of linguistics which he sees everywhere. In fact I forewent sleeping to do some research into his actual writings and it's very ... well, ideological would be a nice way to put it. Given that he seems to have made up terminology - "Leskien's problem" garners a whopping 4 hits on Google one of which is his own paper - and written so much about supposedly non-Indo-European features without even reading up on Proto-Indo-European possessives - one read of which would turn his "mihi" vs "habeo" nonsense on its head ... again, I wouldn't call him an expert

This is what he says so y'all won't have to slog online looking for it:

... In Indo-European languages, the most common means of indicating predicate possession is what will be termed here as 'have'-type possessives, that is, the use of a specific verb to link a possessor subject and a possessed direct object, such as the English verb have, Latin habēre ...

Whereas this is what all IE experts say:

... Instead of a verb 'have'. Proto-Indo-European used a predicative construction, the mihi est construction mentioned above, which included a third-person form of the verb 'be' and an oblique case, referring to the possessor...

When someone makes an elementary mistake like that their credibility goes out the window

[Xephyr]

There you go again with the unlabeled quote boxes.

[Radius Solis]

Give it up, Rusanov. The discussion's been won long since. You're the only one left who can't see how full of shit you are.

[Mecislau]

If you have a big interest in language contact I don't fault you for seeing it in places where it doesn't actually exist. As they say "if all you have is a hammer, everything looks like a nail"

*sigh*

Really? So you're saying people who have formal training in language contact are incapable of actually identifying examples of their field of study?

You argue for two different examples of Finnish contact,

Those are just two that have been brought up so far. There are plenty of others. I'll give you a longer list in just a moment.

1. "mihi type" constructions
These are what Indo-European originally had for possession and in this regards Russian is conservative, not innovative. I don't think you've studied much Proto-Indo-European and Proto-Slavic if you can make such an egregious claim

It's related, but not the same. Of course I know a similar construction existed in IE—why else would it be called "mihi" or "mihi est" possession? You see some remnants of this some of the Baltic languages and in the benefactive dative of West and South Slavic. But note some key differences:

1) The IE mihi construction used the dative case. The Russian one makes use of a locational construction ("at me there is..."). There is clearly an innovative force here.
2) The Finnic languages also make use of a locational construction
3) The other Slavic languages are very strongly habeo-type. They have a verb (from Common Slavic *mati/jьmati "have") used for almost all types of possession. The Russian reflex of this verb is far, far more restricted, with the mihi-type being the default.
4) If you look at the history of the Russian possessive construction, you can actually see its emergence from what was originally a locational construction; that is, it emerged divorced from the IE dative. The oldest layer of birchbarks, for instance, make use of the possessive construction only in "ambiguous" cases, where both a possessive and a locative meaning are valid: "посъли къ томоу моужеви грамотоу ели оу него роба" ("Send that man a birchbark letter: does he have [another] slave woman?" or "... is there another slave woman in his home?" — 11th century). As time goes on, however, you can see a much clearer possessive meaning emerging: "а ѧ вьде ожь ѫ васъ есте тъваръ ольскы" (“But I already know that you have Oleska's [Alexy's?] merchandise.”)

2. copula dropping
Bearing in mind that copula dropping is present - as native speakers have explained many times ITT - in Slavic languages far divorced spatially from the Fennic realms there is no justification for your argument. Besides "finnish and russian both drop copulas and are more or less near each other QED copula dropping comes from Finnish" which is obviously illogical.

Did you completely ignore everything I've written?

Yes, other Slavic languages have limited degrees of copula dropping. It's an IE feature in various marked constructions. Only in East Slavic do you see copula dropping being the unmarked construction. This can clearly be traced back to the time of greatest Finnic contact, and the effect is most pronounced the further north you go in old East Slavic territory.

What makes more sense? That by a remarkable coincidence Old Russian decided to go rogue from the perspective of the other Slavic languages, beginning a process of copula loss, and that by pure chance the further into historical Finnic territory you go the more pronounced this effect was in Old Russian literature? Or that there was a Finnic substrate influence in an area that we know was once heavily Finnic, given how well we know how easily substrate effects can influence syntax. Your theory is ignoring evidence, ascribing it to coincidence; I'm actually putting together pieces of evidence.

And I like how you completely ignored the text I typed out for you giving numerous examples of copula-less constructions in Old Russian, which you repeatedly stated don't exist.




Now, some other examples of Sprachbund-like effects in Russian. Not surprisingly, you see the most pronounced effects in Old Russian and modern dialectal Russian; the modern literary language is based on Moscow speech, which in turn was heavily influenced by Southern speech in the 16th-19th centuries, so the modern literary language preserves much fewer of these, but nevertheless they are there.

These come to mind off the top of my head:

1) Locational possession

2) Copula usage

3) Direct/Indirect case usage with numerals — when the Slavic dual number was lost, each branch of Slavic dealt with restructuring around it in a different way. Russian ended up with a very curious direct (nominative/accusative) vs indirect (all other cases) contrast, where in direct cases you see numeral governance (the numeral assigns a case to the noun), and in indirect cases you see numeral agreement (the numeral takes the same case as the noun). You don't see this breakdown anywhere outside of East Slavic (you see a weak direct/indirect opposition in Polish, but it doesn't extend to all numbers). Guess where else you see a system identical to the East Slavic one? Yep, Finnic. In Finnish, for instance, numerals assign the partitive case in direct case environments and show agreement in all other cases (also bear in mind that the Slavic genitive is also a partitive). You don't see this in Uralic outside of Finnic. So you do you think these rather specific and odd breakdown completely by chance emerged in these two completely unrelated families that were in close contact?

4) Depalatalization in Old North Russian — although central Russian influence has largely undone most of this, in Old North Russian we see some very prominent tendencies toward depalatalization, with certain soft consonants hardening and a resistance to things like the second and third palatalization. The Finnic languages in the region were simultaneously going through a very similar process (compare, say, Proto-Finnic, with lots of palatalized consonants, and modern Finnish, with none. A number of minority Finnic languages in modern Russian (e.g., Karelian) do have palatalized consonants again, but that has been due to more recent Russian loans.

5) Cokanye in Old North Russian/modern North Russian dialects — this refers to a confusion of /ts/ and /tʃ/. I'm not going into all of the details, but you see in both ONR and NRus dialects a tendency to merge these two phonemes (and in ONR, the softness patterns of the two are quite interesting and indicative of issues of perception of palatalization). And guess what, this contrast also didn't exist in the local Finnic languages of the time. In fact, much of Old North Russian phonology essentially looks like Proto-Slavic phonology passed through a Finnic sieve, a very common theme in Sprachbünde all over the world.

6) Šokanye in Old North Russian/modern North Russian dialects — a related phenomenon, a confusion of /ʃ ʒ/ and /sʲ zʲ/.

7) Nominative objects — in both ONR, NRuss dialects, and Finnic, you see the nominative used to mark the direct object of non-finite verbs rather than the accusative. One ONR example: "И тобѣ емꙋ исправа оучинити." ("And it is for you to do justice to him", with исправа in the nominative; Timberlake 1974:10)

8) Possessive perfects — ONR and NRuss dialects can form perfects using a possessive construction, just like in Germanic and some Germanic-influenced Finnic languages. What's funny, of course, is that Germanic uses habeo possession and Russian mihi-est, so you end up with forms like NRuss "У него уехано" ("He has left" — Kuteva & Heine 2004:61). Again, you can trace the development of this in Old Russian, and I can give you a bunch of examples from Zaliznjak.

9) Leskien's Problem — I'm not really going to go deep in depth here because this issue is rather complicated, but in ONR you see a reorganization of the nominative and accusative case endings along Finnic lines in such a way as to obliterate the nominative/vocative contrast (which Finnic lacks) and to boost the nominative/accusative contrast (which was even stronger in Finnic than in Slavic), resulting in a set of endings completely foreign to the rest of Slavic. The most notable example of this is the use of the nominative singular ending -e rather than -ъ for Ŏ-stem nouns, giving you things like ONR брате "brother" (versus Old Southern Russian братъ).

[Mecislau]

Given that he seems to have made up terminology - "Leskien's problem" garners a whopping 4 hits on Google one of which is his own paper - and written so much about supposedly non-Indo-European features without even reading up on Proto-Indo-European possessives - one read of which would turn his "mihi" vs "habeo" nonsense on its head ... again, I wouldn't call him an expert

Have you even dealt with academic papers before? They're mostly locked-up behind paywalls, and thus often aren't very Googleable, especially when it's an issue of narrow focus. I can upload a more in-depth paper on the issue if you want.

This is what he says so y'all won't have to slog online looking for it:

... In Indo-European languages, the most common means of indicating predicate possession is what will be termed here as 'have'-type possessives, that is, the use of a specific verb to link a possessor subject and a possessed direct object, such as the English verb have, Latin habēre ...

Whereas this is what all IE experts say:

... Instead of a verb 'have'. Proto-Indo-European used a predicative construction, the mihi est construction mentioned above, which included a third-person form of the verb 'be' and an oblique case, referring to the possessor...

Uh, there's nothing wrong in either of those quotes. Yes, Proto-Indo-European used the mihi est construction. Yes, modern Indo-European languages mostly use "have"-type possession. Those are both true. In almost all of IE the mihi est construction was marginalized or eliminated entirely, and it's mostly a feature of Old IE. Even in Latin, where you had both coexisting, the habeo type had already become far more common.

When someone makes an elementary mistake like that their credibility goes out the window

... just like your mistakes on Proto-Slavic verbs? And the use of "be" in Old Russian?

[Drydic]

I don't recall Hungary ever being part of an organized slavic state, unless you count being a Soviet puppet state.

Slavs used to live in Pannonia (under Avar leadership mostly) before the magyars came. Many Slavic place names and words survive - Balaton and Szolnok, for example, come from Slavic Blatnograd and Solnograd respectively - and 1/5 or more of Hungarian roots are Slavic in origin even today

You're not telling me anything I didn't know, apart from Balaton/Szolnok (I'm skeptical about them being descended from the Slavic forms you give, but that isn't a point I really care about arguing or looking up, so whatever). Also remember you said Slavic state, and the Avars and their nomad allies were the ones in charge, not the Slavs.

[Mecislau]

Balaton/Szolnok (I'm skeptical about them being descended from the Slavic forms you give, but that isn't a point I really care about arguing or looking up, so whatever)

(Meh, I know it's not important, but I was curious)

It looks like Balaton is almost certainly Slavic in origin, related to Common Slavic *bolto "swamp", or more specifically, *boltьno město "swampy place" or maybe *boltьno jezero "swampy lake". The form R.Rusanov cites (Blatnograd) includes the same stem, but is certainly not the original form, since a) the -grad "city" element is nowhere to be seen and b) in old Slavic the form *boltьno gordъ (or Old South Slavic *blatьno gradъ) is infelicitous; if the stem is an adjective like blatno is, it would be masculine (in agreement with grad), not neuter.

The etymology of Szolnok is a bit more unclear. A quick web search brought up two theories: one that it was derived from the name of the city's first steward, or that is had something to do with salt trade (Hungarian só /ʃoː/ "salt", Slavic *solь "salt"; note that this isn't a recent loan, as this stem is seen through Uralic and thus strikes me as more likely a borrowing from an older IE language). I have no idea whether there was actually a salt trade there or not, however.

[R.Rusanov]

It wouldn't if it was a dithematic nominal construction like Dobromir or Blagoslav (not ˣДобъръмиръ, ˣБлагъславъ/Блажьнъславъ)

[gach]

Sorry for bringing this thread back up. I'm not particularly enjoying doing this for all the nationalism the thread reeks of.

Thanks for the nice list of contact features Mecislau. I have a long history of spending some time learning Russian and maybe also some Russian linguistics in the near future and a lot of this isn't in my active knowledge. I have some corrections to add though:

3) Direct/Indirect case usage with numerals — when the Slavic dual number was lost, each branch of Slavic dealt with restructuring around it in a different way. Russian ended up with a very curious direct (nominative/accusative) vs indirect (all other cases) contrast, where in direct cases you see numeral governance (the numeral assigns a case to the noun), and in indirect cases you see numeral agreement (the numeral takes the same case as the noun). You don't see this breakdown anywhere outside of East Slavic (you see a weak direct/indirect opposition in Polish, but it doesn't extend to all numbers). Guess where else you see a system identical to the East Slavic one? Yep, Finnic. In Finnish, for instance, numerals assign the partitive case in direct case environments and show agreement in all other cases (also bear in mind that the Slavic genitive is also a partitive). You don't see this in Uralic outside of Finnic. So you do you think these rather specific and odd breakdown completely by chance emerged in these two completely unrelated families that were in close contact?

This actually has a bit wider distribution and you find versions of it in Saamic as well. In North Saami, for example, the numerals from two up modifying nominative nouns impose the genitive-accusative case for the noun.

dállu => guokte dálu
house => two house.GEN/ACC
"house" => "two houses"

In other cases the noun retains the expected case and you could say that the numerals agree with the case, although the paradigm they follow is defective.

dálus => guovtti dálus
house.LOC => two.GEN/LOC/ILL house.LOC
"in a house" => "in two houses"

In East Saamic (in practice you'll find information for Inari and Skolt Saami) you find a more complicated system where in the core cases the numerals 2-6 impose singular genitive for the noun and 7 and up originally singular partitive, though this is being replaced by the genitive in Skolt. Oblique cases are retained as in the rest of Saamic and Finnic.

In conclusion, because of the wider distribution on the Uralic side and because of the variations the system has had time to develop, it seems better to think that this phenomenon is older on this side. But this only really means that you should be talking here about substrate influence rather than a common innovation.

And for Rusanov, as others have told, you shouldn't confuse "Finnish" and "Finnic" with each other. They are clearly distinct entities. Mixing them is pretty much the same as mixing "Russian" and "Slavic". During the period of time that has been discussed here the languages on both sides of the contact were already well differentiated. If you want to use specific names for the languages on the Finnic side, you are far better of talking about Old Karelian or Old Veps. The Novgorod birch bark letters actually include documents of what looks like some form of Karelian.

[Basilius]

Mecislau:

5) Russian linguistics programs will teach you the same thing.

I've been lucky not hearing much of such programs. Or maybe I misinterpret which "same thing" you meant.

At any rate, it seems unfair to me to pretend that there's an enthusiastic consensus in academic circles about the type of theories discussed in this thread. From my limited experience, the prevailing attitude is cautious scepticism (about profound Finno-Ugric contact influences in Russian, and about contact influences as explanations in general).

And as usual, things tend to be a bit more complicated than needed for the "language A changed because language B was around" type of reasoning.

To begin from where the flame started. Rusanov's examples are indeed irrelevant here (unlike Niedokonany's), but...

(1) I don't think there is anything controversial with the claim that dropping the copula was the unmarked option in Proto-Slavic for 3rd person present, i. e. that the copula was used only when emphasized. This seems to be a retention of a PIE feature, too.

(I can point to what I consider some pieces of evidence, although I don't think it will be new to you; in particular, Zaliznjak as quoted by you doesn't speak of just "a mix of forms with and without the copula", what he says basically implies that copula was *disallowed* in many, if not most, environments).

What is controversial is dropping the non-3rd-person present forms of copula; one can argue that these forms differed in their behavior from both the "aorist" forms (never omitted) and the 3rd person present forms; yet it is probable that omitting them was freer than dropping the copula in a typical Romance or Germanic language.

If this controversial bit is discarded, then what Russian actually innovated was dropping the non-3rd-person forms of copula. Here, as I understand, the influences from most Finno-Ugric and Turkic neighbors would support the conservation of the Proto-Slavic situation, i. e. would work against this innovation. (I am implying, for example, that conjugated forms of nouns with unmarked 3rd person can be seen, historically at least, as forms with encliticized copula in all persons except 3rd, and would support the analogous use of Slavic cliticized copulas.)

Do you agree?

(Dropping the copula in all persons is a feature of e. g. <modern> Lithuanian, but it looks like innovation there too.)

(2) The similarity between Finnish and Russian in constructions with numerals is indeed spectacular.

However, the Slavic simple numerals above 4 are historically singular nouns; construing them as heads with the counted noun in genitive plural looks natural and was probably the original state of affairs. Therefore, Russian is simply conservative in the way it treats the nom./acc. forms of numbers above 4; it altered the construction only in the oblique cases, and most other Slavic langs did so in all cases.

The difference in treatment of nom./acc. and obl. seems to be supported by other quantitative constructions, e. g. with original adverbs like много 'many/much' (resorting, essentially, to suppletive forms in oblique cases). Here, the current Russian usage is an innovation; at any rate, Old Russian doesn't seem to restrict constructions like мнози людие (nom.), многы люди (acc.) 'many people', with an adjective for 'many', which in modern Russian is used (in this sense) only in oblique cases. I don't know (and I am curious) if this has any parallels in Finno-Ugric.

And the construction used with numerals 2-4 is indeed an incomplete transformation of the original dual forms (of Proto-Slavic and Old Russian). The form of noun used in this construction is not (contrary to what school grammars say) gen. sg.; synchronically, it is best termed "counting form", and historically it's a blend of forms from different subparadigms. The use of gen. pl. of adjectives (even in today's Russian, obligatory only with m. and n. nouns) is, probably, due to a contamination with the construction used for numbers above 4.

To sum up: Russian constructions with numerals look like an intermediary stage of a change that was more complete in most other Slavic languages, i. e. they look conservative for the internal Slavic standards.

External influences could still be a factor here, but potential sources of such influences are confined to Baltic Finnic: AFAIK, the nom.+acc. vs. oblique split is absent in e. g. Mordvinic. What is much more common in the neighbor languages is singular nouns with numerals (Turkic, and partly Mordvinic), but it seems that Russian has managed to somehow escape this potentially much more massive influence.

Does this make sense for you?

(3)

In the early days when the region was still being colonized Russian and Finnic languages often had a closer-to-equal role socially in parts of the territory, with bilingualism (and thus contact + influence) being frequent and so contact-related influences were bidirectional. As Russian began to dominate and bilingualism become one-way, so did contact-related influence, so most Finnic languages acquired features from Russian while Russian gained very little from Finnic.

This is where your use of "Finnic" (instead of e. g. "Finno-Ugric") may indeed lead to confusion.

"Finnic" influence more pronounced in the North seems to imply that you're still speaking primarily of Baltic-Finnic.

Baltic-Finnic influence could be indeed important in Novgorod and in the other "Krivich" dialects, which had been exterminated by the time when modern Russian started to take its current shape.

In (what is now) Central Russia, most important contacts must have been with Baltic (IE Baltic) groups like Galindians (Golyad), and non-Baltic-Finnic Finno-Ugric groups like Merya, Meshchera and Mordva. Of languages spoken by them, only the Mordvinic languages have survived, and we know very little about the rest. But it is clear that none of them was *closely* related, or structurally very similar, to Baltic Finnic.

If there was a symmetry of bidirectional influences like you say, it left no trace in today's Russian.

Baltic Finnic loans in Common Russian are more-less absent.

(Non-Baltic-Finnic) Finno-Ugric loans in Common Russian do exist, but they are much less common than e. g. Turkic loans.

Turkic influence has been very important throughout the history of Russian, but it is odd to claim that it was more profound in Russian than e. g. in Bulgarian.

Therefore, even for someone who is (unlike me) enthusiastic about explaining major diachronic changes by language contact, it is odd to claim that the difference of East Slavic from the rest of Slavic is mainly due to Finno-Ugric, or even specifically Baltic Finnic, influence.

Do you disagree?

There are alternatives, like Terrence Kaufman's idea about Balto-Slavic (or Baltic + Slavic, but definitely not just East Slavic) having been profoundly influenced by Finno-Ugric. At any rate, it allows for Finno-Ugric influence to be used as explanation for features that are Slavic-internally conservative. (I am actually rather sceptical, partly because of the lines of reasoning like "Slavic languages have more cases than other IE languages in Europe because they contacted with Finno-Ugric", but at least some explanations would logically run in the right direction with Kaufman's theory.)