I’ve mentioned the birch-bark letters of Novgorod a couple times (Birch-bark Mat, Birch Bark Books Online), but I’ve never given a full explanation of their origin and linguistic peculiarities. Now I don’t have to bother, because Asya Pereltsvaig of Languages of the World is doing it for me: Birch Bark Letters, part 1, part 2, part 3, and Birch Bark Letters and the Second Slavic Palatalization, part 1 (with a “part 2” on the Second Slavic Palatalization to come). It’s fascinating stuff; the letters “are scratched into the birch bark by using a sharp instrument, a stylos; typically, no ink is used and therefore there is no risk that the ink would fade during the long time since these documents have been written. And it is a long time indeed, since most Novgorod birch bark letters date from the period between late 11th and early 15th century.” And there’s a valuable lesson here:
In the early years since the first birch bark letter discovery, scholars thought they many of them were written by people who were not highly literate and therefore made many spelling errors. However, further careful study showed that there is a certain convention of writing birch bark documents that is simply different in minor ways from the system used to write other documents (books and the similar) around the same time. Essentially, this vernacular writing system is different in only three minor ways [which she explains]. No other “spelling errors” have ever been found in birch bark documents, indicating that these are not really errors at all!
If you look at language, or the world, with a view simply toward affirming your own experiences and prejudices and write off anything different as ignorant mistakes, you feel good and learn nothing. If you put aside your prejudices and examine the differences objectively, you learn things. Anyway, read the whole series; it’s enjoyably written and informative (and you’ll learn about medieval Russian cursing).
I don’t mean this in relation to the birch-bark letters, which sound fascinating, but as a writer I have to say: holding on to your prejudices can be stupid, but at some point, giving up all your “prejudices” also means giving up all your standards.
Just sayin’.
Very true. But for most of us, there’s little danger of giving up all our prejudices; I think it’s far more common for us to defend our prejudices with whatever intellectual sheathing we can muster, or if that fails to simply state them defiantly. Here’s a Louis Menand quote from a recent New Yorker:
“It doesn’t matter what Webster’s Third tells me: I will always feel superior to a person who says, ‘I am totally disinterested in that subject’ (though I will also strive to treat that individual with the dignity and respect owed to any human being). I can’t help it; it’s the way I was brought up.”
Of course he can help it; he simply chooses not to. He clings to his prejudices because they make him feel like himself. If he didn’t feel superior to a person who uses “disinterested” in a way he doesn’t, who would he be? But if you think about it, that’s a pretty sad form of self-definition, reminiscent of the adolescent who clings to his ranking of rock bands because they define who he is. Surely once we’ve grown up and acquired a good fund of experiences, we’ve found out enough about ourselves and developed enough confidence in who we are to dispense with such paraphernalia, especially in the case of something as widely shared as the “disinterested” shibboleth. I mean, how does that define him other than as a curmudgeon?
I am not sure if it’s widely known, but the birch bark ‘gramotki’ discovery in 1950s came after several centuries of convention that literacy in medieval Russia was non-existent outside the monasteries. It was quite a shock for scholars to see ordinary people writing numerous documents of secular nature on a daily basis. And, as you say, they contain a Russian different from the dialect of Muscovy which later began to dominate and suppress other strands.
Oh, and there are beautiful love poems among them.
The thing about the Second Palatalization is that all other documented Slavic languages have it, so one would think that Proto-Slavic already had it. One would be wrong.
Did Grimm’s Law operate in Proto-Germanic, or did that sound change spread through the dialect continuum later? Stay tuned. Same time, same channel… no, sorry, I’m bluffing (and of course the idea isn’t new, and of course there still isn’t any evidence).
Thanks for introducing me to the birch-bark letters. The evolution of y from oy and the change in its meaning was especially interesting.
I have to say, it seems to me that there’s room for scientific and artistic responses to language change. Scientifically, nothing but descriptivism can be defended: if people use words in a certain way their usage should be noted and thought about. But from a creative standpoint, some usage is more or less defensible. Best usage indicates broad knowledge of how the language was and is spoken or written. Wherever a distinction between words could or did exist, it should be taken into account when choosing which to use, and any linguistic change that results in broader, less precise definitions makes communication less appealing even if not harder to understand. I don’t look down on people who use “disinterested” as a synonym for uninterested or spell it “desinchristed” but I would never do either except to be silly. I imagine many people who use disinterested unconventionally do so because they wish to fancify their speech and would not if they knew how the word is typically used–which would make it the closest language can be to incorrect.
I imagine many people who use disinterested unconventionally do so because they wish to fancify their speech and would not if they knew how the word is typically used–which would make it the closest language can be to incorrect.
But that’s not true; they use it because that is what it means to them. And it is the closest such things can ever be to inevitable that it will, within (say) fifty years, be the meaning of the word; the traditional ‘unbiased’ sense will be marked “obsolete.” That’s precisely my point: all language change looks to the older generations, who grew up with the earlier form of the language, like degeneration, and rationales can always be found to support that essentially emotional view (“A valuable distinction is being lost!”). And yet once the change has happened, the language is found to be just as rich and expressive as before, and the folks who objected to the change just look silly in retrospect.
This is not to deny that there’s room for “artistic responses to language change,” but one should try one’s best to disentangle those responses from simple old-fartism.
David: Ah, but the third part is now posted proposing a solution, and a very clever one indeed.
Think French.
What I’m saying: the Second Slavic Palatalization happened when Slavic was already spoken in a large area, and it didn’t spread through all of that area. It didn’t reach the northern tip, where Novgorod is located. Maybe the Finnic substrate helped provide resistance, but that doesn’t change the conclusion that the SSP wasn’t present in the common ancestor of all documented Slavic dialects.
And yet once the change has happened, the language is found to be just as rich and expressive as before,
Perhaps overall the language is just as rich, but how can you deny that distinctions have been lost? Take ensure, insure, and assure. They used to have different meanings, now they are widely used interchangeably. I submit that valuable distinctions were lost there.
But if the distinctions matter, they’ll be recreated in other ways. Nobody actually winds up being unable to express what they need to.
Sure–but some distinctions aren’t essential. Words with complex meanings, more connotation than denotation, can be somewhat irreplaceable. Take “gay” for example. It was before my time, but apparently it could once be used to mean both lighthearted and colorful/showy. Now, obviously we can still express this idea, saying something has “cheerful colors,” or just by describing it, but no monosyllabic one word substitute has as of yet popped up. Languages do change, and the sum of these changes is positive, as people invent ways to speak about new phenomena and retain at least passively some of the language used to talk about the features of the past. But just as some changes enrich, some impoverish, and no one change is inevitable. It has always struck me as a shame that someone who knows so much and obviously cares about language should remove himerself from the fight. True, very little linguistic change is caused purposefully, but if Hebrew could be resurrected, imagine what we could do with English!
The thing about the Second Palatalization is that all other documented Slavic languages have it, so one would think that Proto-Slavic already had it. One would be wrong.
Interesting. Sounds like the discovery of the birch bark letters was to Slavic what the decipherment of Linear B was to Greek: various features shared by all previously known dialects, which had therefore been assumed to be inherited from the protolanguage, turned out to be absent from one dialect and therefore due to diffusion.
(For an outline of a similar argument about Germanic see this abstract: http://linguistics.berkeley.edu/bls/past_meetings/bls34/abstracts34/Garrett.pdf)
Uh … Didn’t Cranmer write a prayer in the Book of Common Prayer that magistrates would “truly and indifferently administer justice,” and it was modernized to “impartially” only in the 20th century?
Don’t exaggerate. What’s spoken in Israel today isn’t all that similar to any stage of Biblical Hebrew.
Is voice dissimilation really rare? Or is it just absent from Standard Average European?
Alternatively or additionally, what if this effect (if part of Verner’s Law as the abstract suggests) operated in Proto-Germanic, but has been obscured (like many other effects of Verner’s Law!) by analogical reformations based on other grammatical forms of the same stems in North and West Germanic? Verner’s Law created a /z/; Gothic preserved that /z/, except word-finally where it merged with /s/ because Gothic had word-final devoicing; German distributed it, seemingly at random, between /s/ and /r/; English acquired a new [z] as the word-internal allophone of /s/, merged some instances the old /z/ into this, and turned the others into /r/, again seemingly at random. I’ll stop here instead of flooding the thread with examples. :-]
Finally, the abstract clearly goes way beyond its own conclusions: it says there are wave effects and then goes on to suggest that that’s all there is, that there never was a Proto-Germanic language, that there is no tree pattern at all in Indo-European. This is just obviously wrong.
(Technically it didn’t. The subsequent loss of phonemic stress did… and it may well have been imported from Celtic. But I digress.)
Is voice dissimilation really rare? Or is it just absent from Standard Average European?
I don’t know, but I’d be interested to hear of other examples – I don’t think I’ve ever encountered any.
I used to have the handout from this talk, which may have addressed your objection about analogical reformation, but can’t find it at the moment. In any case I think the point is to suggest that diffusion/dialect convergence phenomena are more common than historical linguists tend to assume (just as you implied in your comment about Grimm’s Law), not that they’re the whole story.
Take ensure, insure, and assure. They used to have different meanings, now they are widely used interchangeably. I submit that valuable distinctions were lost there.
First, is there any evidence that most people ever applied the distinctions suggested by commentators, i.e., that those distinctions ever really existed? MWDEU suggests not. Second, even if distinctions were lost, weren’t these the least valuable sort of distinctions, applying as they did merely to usage itself? This isn’t a case where a word with interesting shades is being changed by common usage into something more boring; it’s a case of some commentators supposing the language would neater (and therefore I guess better) if we reserved this word for this instance and this word for that, when in fact nothing is lost or more poorly defined when more than one lexeme is available to fill a given space in a sentence.
Sashura, where can I find the birch-bark love poems?
Google took 0.1 seconds to find 74 results for “voice dissimilation”. Admittedly, the 9th is this page, but most of the others are papers. One of them is this abstract (pdf) which says voice dissimilation is productive right now in Kinyarwanda. This pdf, which I haven’t read, says “the voice dissimilation rule, very productive in Kirundi” in its preview. And this Google Books result shows voice dissimilation in Irish.
What can I say? I believe.
@David Marjanović:
>> if Hebrew could be resurrected
> Don’t exaggerate. What’s spoken in Israel today isn’t all that similar to any stage of Biblical Hebrew.
You are obviously entitled to your opinion, but plenty of experts in Modern Hebrew do regard it as the resurrection of Hebrew, despite its differences from earlier forms of Hebrew.
Surely that’s a matter of semantics; if everybody more or less agrees on the degrees of difference, what does it matter if you call it a resurrection or a rebooting?
OK, looks like there are a handful of other described cases of voice dissimilation; that’s interesting to know.
@languagehat: I agree; but in all fairness to David Marjanović, I don’t know if it’s true that “everybody more or less agrees on the degrees of difference”. Some have even gone so far as to claim that Modern Hebrew is more or less an Indo-European language remapped onto Semitic roots. So I can see how he might not consider it a “resurrection”; I just don’t think it’s fair of him to dismiss the popular view so cavalierly. (Especially since joe R’s comment did not depend on Modern Hebrew being particularly similar to Biblical Hebrew. Even if you consider Modern Hebrew to be an entirely new language, as some do, it’s obvious that it’s largely the result of conscious planning on the part of its speakers.)
Some have even gone so far as to claim that Modern Hebrew is more or less an Indo-European language remapped onto Semitic roots.
Oh yeah, those people. I guess I try to forget about them.
“Rebooting” sounds like a good way to put it!
As I now read it, it depended on planned change being much better than unplanned change by default.
While claims of “a Slavic language in search of a Semitic past” are clearly exaggerated, it’s quite evident – from past discussions on this blog – that many of the differences between Biblical and Modern Hebrew are due to more or less Standard Average European substrates, and that many of these were not planned.
@David Marjanović: I don’t think his comment depended on planned change being better, only on planned change being possible. His own proposals, of course, are ones that he considers better, but he raised Hebrew only as an example of how much could be accomplished if desired. (“You’re so good at [thing I consider useless], imagine how good you’d be at [thing I want you to do] if only you tried!”)
I certainly agree that Hebrew has had a lot of influence from European languages (phonologically, lexically, syntactically — even morphologically, sort of), but it’s also had a lot of continuity over the millennia. I don’t blame you for comparing it to Biblical Hebrew, since there were some efforts during the revival of Hebrew to recover some aspects of Biblical Hebrew, but really it’s much more similar to much more recent forms of Hebrew. (While we’re quibbling over terminology, by the way, I personally prefer “revival” to “resurrection”: the language never really died — it was always in use — but it wasn’t a full living language for a long time. It was on life support as a literary and liturgical language, and as a Jewish lingua franca when needed, but now it’s off the ventilator. “Reboot” is cooler, but to me it somehow makes it sound like it restored a much older form of Hebrew — like it “started fresh” by rolling back millennia of language change — and of course that’s not the case. But I don’t know why it sounds that way to me, since obviously both you and Hat are far too knowledgeable to have meant that.)
SAEness, like Balkanness, clearly spread by diffusion.