Hello everybody,
>> But more seriously: I studied miscellaneous works on the subject of
>> Slavic languages similarities and classification. I think that
>> tree-based hierarchic models are wrong way. I propose to use
>> heptatochomic network model by V. V. Ivanov(*) filled by metrics by
>> Zhuravlev(**), which looks like this (please set monospaced font):
>>
>>        +---------------lechitic (polish)------northern(novgorodian)
>>    polabian                |                          |
>> (lusatian,...)          |                          |
>>        +-------------czechoslovak-------------------eastern(UA,BY,RUS)
>>                            |                          |
>>                            |                          |
>>                     slovenian-serbocroatian----macedonian-bulgarian
>>      
> Yeah, this is a model very similar to the one Slovianski uses: six branches, or
> subbranches, while the Polabian/Sorbian one is usually not taken into
> consideration. I'm not so sure about juxtaposing Czechoslovak and Eastern,
> though.
>    

Jan, maybe ordinary people do not feel this similarities because of 
different alphabets, accents and cultural and political background of 
eastern and western languages, but it is real, indeed:

This is based on the deep lexicostatistical test made by Zhuravlev (*) 
from the etymological dictionary of slavic languages in 15 books (ed. 
O.N. Trubachev, Moscow 1974).
There is a complex formula how to compute statistical distance of languages:

G(A,B) = SUM i is from 2 to n ((n + 2 - i) * V(A,B)i) / H(A) / H(B), where

A,B are tested languages,
n is total number of tested languages
G(A,B) is congruence of these two languages
H(A), H(B) is inherited dictionary from common root of these two languages
V(A,B)i is isoglossa

In this formula, the higher value has isoglossa limited to only two 
languages and the lower number has isoglossa, which is common for more 
languages.

The highest number of profimity is between upper and lower lusatian 
sorbian: 1.93,
Bulgarian-Slavomacedonian has 1.73,
Czech-Slovak has 1.36,
Bielorusian-Russian has 1.27,
Ukrainian-Russian has 1.20 (but ukrainians learn Russian as well) and
Polish-Bulgarian has only 0.93 etc.

If languages have proximity at least 1.1, they can be assigned into one 
group, but this approach goes intoa lot of such similar solutions.

For mutual understanding of ordinary people You need to have at least 1.2.
Old Slavonic has value 1.15 (to Polish) or better (1.27 to Czech, for 
example) to modern slavic languages.

> I tend to see the Slavic languages as a wheel, or circle:
> Russian - Belarussian - Ukrainian - Rusyn - Polish - Cashubian - Lower
> Sorbian - Upper Sorbian - Czech - Slovak - Slovene - Serbo-Croatian -
> Macedonian - Bulgarian - back to Russian
>
>    
I do not thing so. In circle You do not have inner relations.
Belarussian and Polish are not connected only through Rusyn and 
Ukraininan, for example.
> But it really depends how you look at it. Phonologically, the most obvious
> distinction is North - South (where Czech and Slovak are closer to the South);
> lexically, the East/West distinction is at least as important as the North/South
> distinction.
>
>    
Yes, there are many scientific models how to classify slavic languages. 
See references below.
> But tell me, does Ivanov actually call Old Novgorian a "North Slavic language"?
> I've been looking for sources for the hypothesis of a fourth (North Slavic)
> group, but couldn't find it.
>
>    
Yes, Novgorodian is assigned as the independent group. But total number 
of literary heritage is very small as far I know.
>> The ideal candidate for interslavic language from current modern
>> languages is probably Slovak or something shift bit closer to Rusyn. But
>> if we extend this model to extinct languages, the best candidate is Old
>> Slavonic.
>>      
> Well, there are in fact several candidates. Slovak and Rusyn are indeed pretty
> much at the epicentre of the Slavic languages, but they have a lot of their
> own specificalities as well. Another good candidate would be Old Ruthenian
> (a.k.a. Old Belorussian).
Belorussian? Why?

> And Church Slavonic, as you say. But the best
> candidate would of course be a modernised form of reconstructed Proto-
> Slavic! :)
>    
No, I say OLD SLAVONIC, not Church Slavonic. See another mail.

Protoslavic is yet more archaic than old slavonic. In protoslavic You 
have constructs, which are not present in any current slavic language 
except Polish.

>> There are also another not so scientific but also meaningful arguments
>> for the preference of modernized Old Slavonic:
>>
>> 1) Old Slavonic was already a conlang!
>>      
> I have heard that theory, but as far as I know, it is absolutely not a broadly
> recognised one.
>
>    
in another mail
>> 2) Old Slavonic is usualy assigned to the sub-group of southslavic
>> languages by its grammar and phonetics. Southslavic phonetics and
>> grammar (e.g. verbal system) is good candidate for an inter-Slavic
>> language. Southern accent is not so soft as eastern, and has very
>> similar phonetics with romance languages and Greek. This is not a bad
>> company, is it?
>>      
> Well, that's a matter of taste, really. To me, it is doubtful that South Slavic
> grammar (excluding Slovene) is such a good candidate, because it is so
> radically different from the remaining Slavic languages. IMO an Interslavic
> language should avoid forms that are not understandable at all to over 75% of
> the Slavic population.
>    
Who says? Simply say, this is not true. Maybe it is based on some polish 
perception of the surrounding languages, which I do not know. Southern 
people feel it differently.

Maybe this is the explanation: except Polish, any other slavic nation 
has its closely related and easy understandable language of another 
slavic nation. But who knows...

BTW. Our results with Neoslavonic in polish forums are also not so bad, eh?
> The greatest disadvantage of OCS, however, is not that it is archaic, but that
> it never covered the whole Slavic territory. There has never been any mutual
> influence with Polish at all, and its influence on Ukrainian has been very limited
> as well. It is not more understandable to a Pole than, say, Classical Latin is to
> a Frenchman.
Old Slavonic is not understandable to any modern man in the same way as 
Old Anglo-Saxon for anybody English speaking person. Jan, please, do not 
mix the apples with the horses.
You prefer proto-slavic, but this is yet much more "archaic" than OSl.

If Your simply transcript OSl into latin and pronounce it with modern 
accent, understandability dramatically improves. This is the approach of 
these canadian guys, whom I collaborate with: http://uss.utoronto.ca/

And moreover, if You fix the language (grammar, morphology, phonetics, 
orthography, add modern words, ) according to evolutionary changes which 
have also occurred in spoken slavic languages, You obtain well working 
Neoslavonic. This is it.
> That's something that always must be taken into account -
> especially if you consider that the number of Polish speakers is about as high
> as the number of Czech, Slovak and South Slavic speakers together!
>
> Therefore, I still think it is a great source of input, but it should be handled
> with care. Except for the soft consonants, the phonology of OCS is very
> South Slavic indeed. For example, the reflex of CSl. tj/dj as s^t/z^d is typical
> for Bulgarian, but doesn't exist anywhere else (not even in standard
> Macedonian), except for numerous OCS borrowings in Russian. I would strongly
> recommend you to consider this. Also lexically OCS is a bit of a mousetrap,
> because it contains lots of words that are understandable to South Slavs only
> (and not for the remaining 90% of the Slavic population). Therefore, to be
> useful as a real Interslavic language, it needs lots of adjustments.
>    
You are wrong again, sorry.
Any form of jazyk, jezik, jezyk, jazik or cviet, cvet, kviet, kvet or 
mezhdu, medzhu is well understandable at the almost same level within 
Slavic community. This is my practical experience.  I do not like 
neverending internet debates on it at our slavic conlang forums. We do 
not work really with our interslavic languages, but only solve again and 
again matter of secondary importance in English.

The biggiest problems we have are different:
1) false friends and
2) different grammars (mostly verbal tenses, conditionals, etc.)

This why we have our concept of language flavourization, receptive 
learning and passive understability. We strongly need to test all these 
ideas in the practice. This is why we organized the E.U. courses - open 
for anybody from slavic conlang community
(http://ec.europa.eu/education/trainingdatabase/index.cfm?fuseaction=DisplayCourse&cid=26829)
why we try to produce videos and speak in conlang
(https://sites.google.com/site/novoslovienskij/youtube)
and why I try to describe grammar and morphology of the entire 
Neoslavonic language in Neoslavonic.
https://sites.google.com/site/novoslovienskij/demonstracia-grammatiky-ns-jazyka
:o)

cheers
VM

references

Zhuravlev, Anatolij F. 1994. Leksiko-statisticzeskoe modelirovanie 
systemy slavjanskogo jazykovogo rodstva. Moskva: Indrik. Formula of the 
lexicostatistical model at page 63.

http://www.google.com/url?sa=t&source=web&cd=1&ved=0CBUQFjAA&url=http%3A%2F%2Fwww.phil.muni.cz%2Flinguistica%2Fart%2Fnovbla%2Fnob-001.pdf&rct=j&q=bla%C5%BEek%20klasifikace%20slovansk%C3%BDch%20jazyk%C5%AF&ei=EkD6TPmGM4Sv8gOHqbiLDA&usg=AFQjCNFhHFWgiDQk3a9bTbvjxqYfhWBUow&sig2=yvFXFir60GRbLBBuU2MAQw&cad=rja

http://uss.utoronto.ca/



-- 
------------------------------------------------------------------------
*assoc. prof. Vojtěch Merunka, Ph.D.*
[log in to unmask] <mailto:[log in to unmask]>
http://sites.google.com/site/vmerunka

*Department of Information Engineering*
/Faculty of Economics and Management/
Czech University of Life Sciences in Prague

*Department of Software Engineering in Economy*
/Faculty of Nuclear Sciences and Physical Engineering/
Czech Technical University in Prague
------------------------------------------------------------------------