• This topic has 2 voices and 3 replies.
Viewing 5 posts - 1 through 5 (of 5 total)
  • Author
  • #346762


    A Russian linguist A. F. Zhuravlyov published a monograph called “Lexostatistical model of Slavic linguistic affinity”. The author developed a statistical model utilising large samples of Slavic words to evaluate lexical distances between Slavic languages. I obtained data from Zhurvlyov’s study (pp 129-133) and constructed MDS (multi-dimensional scaling) plot and phylogenetic network. See the diagrams below.

    Samples were taken from “Etymological dictionary of Slavic languages. Proto-Slavic lexical stock”. Currently, there are 39 volumes in the dictionary. Sample sizes (p.113).

    Old Slavic (Old Church Slavonic) — 1124 words
    Bulgarian — 3262
    Macedonian — 2035
    Serbo-Croatian — 4568
    Slovenian — 3519

    Czech — 4264
    Slovak — 2933
    Polish — 3350
    Upper Lusatian — 1895
    Lower Lusatian — 1574
    Polabian — 452
    Kashub-Slovin — 1683

    Old Russian — 2681
    Middle Russian — 3834
    Southern Russian — 3833
    Ukrainian — 3905
    Belarusian — 3288

    – Old Slavonic (Old Church Slavonic) was a southern Slavic language brought with religion to eastern Slavia. Hence it is in the same cluster with other southern Slavic languages.
    – Old Russian was the earliest written Slavic language in eastern Slavia used for writing manuscripts. Old Church Slavonic (southern Slavic language) greatly influenced written old Russian which explains the similarities.
    – Kashub-Slovinski language at proto-Slavic level gravitates towards Lustian and Polabian languages.

    It is important to note that the lexicostatistical comparison between the languages was done at “proto-Slavic level” using Slavic words only.

    Monograph : http://www.inslav.ru/images/stories/pdf/1994_Zhuravlev.pdf
    Etymological dictionary of Slavic languages: https://en.wikipedia.org/wiki/Etymological_dictionary_of_Slavic_languages

    In larger resolution : http://oi59.tinypic.com/anzbqs.jpg

    In larger resolution: http://oi59.tinypic.com/24ceizc.jpg



    Updated graphs with northern Russian (n=4168) is added.

    Tree : http://postimg.org/image/l79gqi40p/full/
    MDS plot : http://postimg.org/image/gzjx1cmy9/full/



    Never heard about “Kashub-Slovinski”. It is like to say “Polish-Belarusian”. These are two different languages, while “Slovinski” is exctint. Kashubian is still in use.

    When in some research you find such crucial mistake you can be sure that such research is not worthy of taking it seriously.

    PS. There was no “Slovinski” language, but “Slovincian”. Detail, I know.




    The comparison was done at ‘proto-Slavic” level using samples from the “Etymological dictionary of Slavic languages. Proto-Slavic lexical stock” (39 volume) as highlighted above. Many linguists stated Slovincian ( ‘Slovinski’) was a dialect of Kashubian. There was little lexical difference between Slovincian (‘Slovinski’) and Kashubian languages at ‘proto-Slavic’ level. So the samples for Kashubian and Slovincian were combined and labelled as Kashub-Slovinski . Similarly, linguists combined Serbian and Croatian labelling it as Serbo-Croatian in the past. There was also little difference between GreatRussian (no longer spoken) and one of the three Russian dialects used in research which are also no longer spoken. There are several dead languages used in this research.

    In my opinion, this is one of the best research on the subject published by a well known linguist whose area of expertise is Slavic languages and comparative linguistic published by the Academy of Sciences. Much better than simple comparison done using 200 words from the Swadesh list.

    When you think you may found a ‘crucial’ mistake then investigate if it is in fact a mistake.

    PS ‘Slovinski’ is the spelling used in Russian. I was not sure how it was spelled in English, so I used that spelling. https://ru.wikipedia.org/wiki/%D0%A1%D0%BB%D0%BE%D0%B2%D0%B8%D0%BD%D1%81%D0%BA%D0%B8%D0%B9_%D1%8F%D0%B7%D1%8B%D0%BA



    A new paper came out on Slavic and Baltic genetics that included a supplementary material on Balto-Slavic languages. http://journals.plos.org/plosone/article/asset?unique&id=info:doi/10.1371/journal.pone.0135820.s008

    Authors presented a lexicostatistical analysis using Swadesh list of 110 words. They are also addressed some issues with selections and interpretations of the words in past papers , many loan-words were included and inaccurate translations provided.

    There’s an interesting commentary on Slovenian language.


    3.2. The case of the Slovenian language

    Modern Slovenian belongs to the South Slavic clade according to the traditional classification of Slavic languages (Sussex, Cubberley, 2006). However, significant linguistic similarities between Slovenian and West Slavic lects have been observed earlier in number of studies. See, for example, on specific ties between Slovenian and West Slavic (e.g., Slovak) or even on support of the mixed South/West origin of Slovenian, e.g., Bezlaj, 2003, Sobolev, 2000, Bernstein, 1961, Stieber, 1972, Lekov, 1958. 19 Likewise, the Slovenian (Ljubljana koine and literary Slovenian) wordlist, available in our study (see sources below), possesses a substantial number of both South Slavic and West Slavic lexical matches (cf. similar observations in Novotná and Blažek, 2007: 195). Such a mix introduces enough incompatible characters into the input matrix to make the calculation of robust trees impossible. Due to this reason we have deliberately excluded Modern Slovenian from the current analysis. We suggest that one of the possible scenarios is that Slovenian is historically a West Slavic language being influenced by neighboring Serbo-Croatian during the last millennium. To demonstrate specific ties between Slovenian and South Slavic, on the one hand, and West Slavic, on the other, we calculated a set of NeighborNet phylogenetic networks (Bryant, Moulton, 2004; Makarenkov et al. 2006: 89–90) of Balto-Slavic languages with the use of SplitsTree4 software from the binary matrix described above; the non-parametric bootstrap test was performed with 10 000 pseudoreplicates in each case. Two additional taxa were introduced into the original dataset: Slovenian and Modern Demotic Greek, the latter as an outgroup
    for the Germanic-Balto-Slavic clade.

    Three networks without Slovenian (Fig. H–J in S2 File) reveal the same major clades of Balto-Slavic languages as phylogenetic trees do (Fig. B–G in S2 File) irrespective the outgroup used. Incorporation of Slovenian into network analysis reveals following: Slovenian appears to be an independent branch of Slavic languages which is nearly equally close to West and South Slavic, but distant from East Slavic (Fig. K–M in S2 File), thus supporting the putative mixed nature of Modern Slovenian. Further lexicostatistical investigation of Slovenian dialects, such as in progress in the GLD project, are needed to elucidate the place of Slovenian among Slavic languages.


Viewing 5 posts - 1 through 5 (of 5 total)

You must be logged in to reply to this topic.


5 User(s) Online Join Server
  • Glockamole
  • Tujev
  • TypowaPolskaKobieta
  • haver25
  • kony97