Birch Bark Letters and the Second Slavic Palatalization, part 3

Sep 14, 2011 by

At the end of the previous posting, we’ve reached a conundrum: the Old Novgorod dialect must have diverged from the rest of the Slavic family early enough to avoid the application of the Second Slavic Palatalization, yet not too early so that close resemblance to other Old Russian dialects can be still accounted for. In other words, the Old Novgorod dialect must be placed, on the one hand, on a distinct branch of Slavic tree (say, we call it “North Slavic”), yet on the other hand, on the same brach as other East Slavic languages (cf. the chart below). But how could this be?

A solution suggests itself if we compare this problem to the one concerning Romance languages and the place of French in the Romance tree, more specifically. In a standard chart of the Romance family (see below), French is placed inside the Gallo-Romance grouping, which is inside the Western Romance grouping, which is inside the Italo-Romance grouping, which is inside the Continental Romance grouping.

Yet despite being so deeply embedded inside the Romance tree, French stands out from the rest of the Romance languages diagrammed here in one significant way: it is the only non-null-subject language (except Northern Italian dialects, but I will leave these aside for now). If some linguists a millenium from now discovered French documents and found out that it is a non-null-subject language, they would be just as perplexed about the place of French on the Romance tree as we are about the place of the Old Novgorod dialect on the Slavic tree. In this one way, French sticks out like a sore thumb, but in many other respects it is closer to Occitan and Catalan, then to Spanish and Portuguese, etc., which justifies its placement on the chart as shown above. How would our imaginary future linguists resolve this conundrum?

Hopefully, they would know the answer that we know now (perhaps by some coincidence they would also discover my blog posting on this topic!): French is a non-null-subject language because of the influence of a number of Germanic languages, which tend to be non-null-subject as well (think about English, for instance: we cannot say *Rains even if it in It rains doesn’t refer to anything).

Could we possibly find a similar solution to our Old Novgorod problem? I think so. If it’s “the neighbor who did it”, who were the immediate neighbors of the Old Novgorod? The answer is the now-lost “middle Finns”, that is groups who spoke some now-extinct Finnic languages, such as Merya, Meschera or Murom, or the surviving Finnic-speaking groups, like Votic and Veps. As you can see from the following map, the “lost middle Finns” lived immediately to the east of Veliky Novgorod, which is located just north of the Lake Ilmen (find the northern tip of the Slavic speaking territory and you will see a small light blue patch, the Lake Ilmen). And, conveniently for us, Finnic languages do not have the same phonotactic constraint that disallows velar consonants next to front vowel.

As I mentioned in an earlier posting, it is very difficult to determine with any degree of certainty what the influence of such Finnic-speaking groups on the Russian language (or Slavic varieties in general) might have been, even though we are quite certain that some such influence must have happened. It is a bit like the problem of finding a black cat in a dark room, except in this case we don’t know what animal we are looking for and we are not even sure which room we should search. In a typical situation, there exists a lag in time between the the coexistence of the two languages in the same area and the penetration of the substrate’s grammatical features into the superstrate language. Given that, we are not quite sure if the influence of Finnic languages should be obverved in East Slavic (vis-a-vis West and South Slavic), in Russian (vis-a-vis other East Slavic languages), or in northern dialects of Russian. To make matters worse, we cannot be sure that any of that influence survived to modern languages/dialects or can be attested in written documents that we have found. And since the languages of the “middle Finns” are themselves extinct, we do not know for sure what they were like either; we can only reconstruct them on the basis of other Finnic languages. Therefore, all our hypotheses about the Finnic influences, including mine (relying on the Finnic substrate to explain the non-application of the Second Slavic Palatalization in the Old Novgorod dialect) must remain just hypotheses, at least for now. But like other scientific hypotheses, it is worth entertaining.

Subscribe For Updates

We would love to have you back on Languages Of The World in the future. If you would like to receive updates of our newest posts, feel free to do so using any of your favorite methods below:

  • Ran

    Sorry, this is probably a stupid question, but — what exactly is the timeline here? It seems like, if the Second Slavic Palatalization was shared by all the (attested) Slavic languages, then chronologically it must have taken place long before Finnic influence could have prevented it, no? Are you hypothesizing that Finnic influence actually reversed the palatalization, causing /ts/ to change back into /k/? Or is it that the Second Slavic Palatalization was still "active" (if that's the right term?) even until a time when Finnic influence could have affected it? (O.K., I guess that was actually five stupid questions . . .)

  • John Cowan

    A very cool and convincing solution!

  • Steven Lubman

    Very interesting hypothesis! Hope that new birch bark writings will help solve this mystery!

  • Asya Pereltsvaig

    @Steven Lubman: Thank you! They keep unearthing new documents every summer — who knows what will be found. Maybe some day I will get into this dig too! 😉
    Actually, it would be quite an honor to work with Andrey Anatolyevich Zalizniak!

  • Asya Pereltsvaig

    @John Cowan: Thank you! I am pleased to know that I've convinced (some of) my readers.

  • Asya Pereltsvaig

    @Ran: not a stupid question at all but I am not a 100% sure of the answer. Here's my thoughts on this. In theory, for the Second Slavic Palatalization (SSP) to be "undone" once it's happened, people generations down the road should have been able to recognize which of the words with /ts/ were a result of that change (and undo the change back to /k/) and which were not (and the original /ts/ must remain a /ts/). In some cases, SSP applied across morpheme boundaries and could be easily detected through alternations like (Ukr.) ruká 'hand', plural rucí. But in other cases, SSP applied within roots (as with 'whole'), so it could not be as easily detected. As far as I know, all instances of /ts/ in Slavic arose through a palatalization process; there were no /ts/ or /dz/ or /z/ inherited from Proto-Indo-European (but there was a /s/). I might be wrong on that, though…
    This matter is further muddled by the fact that the Old Novgorod dialect had the process of tsokanie, that is /tch/ is pronounced as /ts/ (as in tsto for /tchto/ 'what'). This feature is preserved in many northern Russian dialects. We are not sure when that started either.
    As for the alternative hypothesis, it is quite possible too that SSP happened before the differentiation of Slavic languages, but it didn't affect some dialects or even subdialects, those that were in direct contact and coexistence with Finnic people. This non-application of the SSP might have started among Finnic-speaking people trying to speak the Old Novgorod dialect (say, the Finnic wives of Slavic men). It might have continued as a substandard feature for a long time, until it became common enough to be reflected in the birch bark documents (or until there were such documents to reflect it). This does not contradict what we know about substratum feature penetration from other languages (such as Celtic substratum features in English): they tend to remain "under the radar" for centuries before gradually growing in importance and becoming "standard"…

  • David

    A comparable case is with the Swedish dialects of Finland. Many of these have several phonological features that distinguish them from Standard Sweden Swedish: generalised accent 1 (instead of a distinction between accent 1 and 2), no participation in the Middle Swedish vowel shift ("den stora vokaldansen"), preserved short syllables in stressed position, etc. All of these features would have been part of the variety of Swedish brought by the first Swedish-speaking colonisers to Finland in maybe the 13th century or something like that, but they also correspond to the situation in Finnish. (The one unexpected feature is generalised accent 1, since scholars generally agree that the origins of the accent distinction in Swedish and Norwegian go back to the Viking Age, ie 9-11th c., but we note that generalised accent 1 can be found in the traditional dialects of Roslagen, the Swedish coastline opposite Åland, from where many of the colonists are likely to have come.)

    In other words, we may be dealing with a situation where contact with Finnish helps resist changes that affect other varieties of Swedish. For Ostrobothnian (and possibly for the Överkalix dialect in Norrbotten, abutting traditionally Finnish-speaking areas, and mutatis mutandis for the now-all-but-extinct Swedish dialects in northwestern Estonia), Finnish language contact just may also be one factor explaining the survival of Old Norse diphthongs, which were monophthongised in more southerly varieties as early as the 11th century (a process which we can track by spellings on runestones). (Once again we can compare with Roslagen, where the isolated Hållnäs dialect preserves Old Norse diphthongs. So although most varieties in areas from where the Swedish colonists to Finland are likely to have come have undergone monophthongisation, it is fully possible that they had not yet done so by the time of the colonisation, although that process was under way in other areas nearby.)

  • nycguy

    Could this not be merely an artifact of the spelling system? Similar to the "now you see it, now you don't" nature of the umlaut in old German writing?

    As I understand it, in German, the writing system noted the loss of the umlaut-motivating vowel earlier than it noted the change in the preceding umlauted vowel, so that there is an intermediate period during which the spelling system doesn't indicate umlaut in any way.

    In this Russian example, assuming that the change in the vowel is reflected in the spelling system before the forced change in the spelling of the consonant would provide an explanation of the phenomena without the need for a large theoretical superstructure.

  • Asya Pereltsvaig

    @David: Thank you for your detailed comment! This seems to be a similar situation of Finnic substrate affecting the Swedish superstrate. Very interesting indeed. Would Roslagen also be the area from which Vikings/Varyangians came to Russia?

  • Asya Pereltsvaig

    @nycguy: Thank you for your interesting comment. In theory, it could possible that the writing system used in Novgorod didn't represent the pronunciation faithfully in this particular respect. But two things speak against that hypothesis: (1) the Second Slavic Palatalization is supposed to have happened long (centuries, perhaps) before the writing starts to be used, so it's hard to imagine that the spelling simply didn't catch up with the pronunciation changes (as it didn't in English, for example); and (2) the non-application of the Second Slavic Palatalization is preserved in some northern Russian dialects, so unless it's a later undoing of the change (on which see my earlier comment), we can surmise that this is really how they said it, with a /k/, not a /ts/.

  • Anonymous

    How do you explain an archaism with a substrate effect? If the language contact happened already at time of Common Slavic, why is this the only great divergence in the Old Novgorod dialect?

  • Asya Pereltsvaig

    @Anonymous: Yes, I do think the contact is rather old and long-term. As for your second question, I don't think we can really predict which exact changes and how many to expect from a substrate influence or any other language-external factor…

  • gaming translation

    thanks for this information.

    gaming translation

  • Pingback: Influence of Old Novgorod Dialect on the Russian Literary Language - Languages Of The World | Languages Of The World()