![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Entry tags:
in which fox has a hard time believing what she's told ...
... an epistolary tale.
June 11, 2003
12:07 pm EDT
[Dear] Prof. FormerProfessor:
... I was a student at [your university] from 1995-99, in Intensive Russian for my first three semesters. I'm working on an MA now, and find myself back in Russian classes, and something came up the other day that raised a question nobody here could really answer: word-final devoicing.
As it was presented to us, voiced consonants devoice word-finally. No problem; easy concept. But when our instructor, a native speaker, demonstrated, it really sounded to me -- especially in words ending in fricatives -- as if the sound began voiced, and then devoiced while in progress. In other words, she insisted that the "v" in "Chekov" devoiced completely, and then said "Chekovf." Also "jazs" for "jazz" and "garazhsh" for "garage" (apologies for not being able to reproduce any of this in Cyrillic). I asked her to repeat it several times, and this never seemed to change, but both she and the other instructor (a non-native speaker) were confident that the devoicing was total.
So it occurred to me that there could be a paper in this, if there hasn't been one already. I mean, in English I hear a difference between the ends of the word "svelt" and "veldt", or "bunt" and "bundt" -- granted, those are both borrowed Dutch words, but I'm sure I could come up with other examples -- and it may be the same in the case of Russian word-final voiced fricatives. What I'd like to do is find some more native speakers and hook everyone up to machines and get spectrographs, but if this is a non-problem (that is, if someone's already investigated it; if other native speakers report complete devoicing, I'm not inclined to believe them any more than I believe this one, without listening to them and measuring the voicing), obviously there'd be no point.
The professor here didn't know of any research in this area, but he admits he's not a phonologist, still less a phonetician, so the fact that he doesn't know of any research doesn't mean that none exists. Are you aware of investigation into this phenomenon? And if not, does it sound like something a person could get a paper out of (just a paper, not necessarily a master's thesis type thing)?
... Thanks for your help!
best,
fox
June 11, 2003
3:55 pm EDT
Hi, Fox,
... You ask a great question, but alas, there's no paper in it. In normal speech, at least, the de-voicing is total, or at least it is in standard Russian. I think what's happening is: 1) your instructor is just slowing down the pronunciation to demonstrate the de-voicing and, therefore (even if she is unaware of doing so herself!), starting out with some voicing. She has probably found that she has to do this, because (contrary to what you might think) de-voicing is NOT straightforward to most American students. Usually they have to be explicitly taught that, say, [d] and [t] are the same sound but for voicing. So I think she's showing (again, whether unconsciously or not) what the consonant "should be" according to orthography, and what it actually is. 2) Even in normal speech, de-voicing may not be carried out if the first sound in the next word is a voiced consonant. This lack of de-voicing is obligatory if the next word is a particle (e.g., "zhe" or "by"), which by definition is phonologically a part of the preceding word, but the de-voicing may also be prevented before any word starting with a voiced consonant. Russian has what linguists there refer to as "slitnoe proiznoshenie" (literally, "flowing-together pronunciation"), where there the word boundary is not distinct -- unlike English, and especially German, where there may be a slight glottal stop between one word and the next. Compare, for instance, "I have an aim in life" with "I have a name in life." It would be possible for an English speaker to confuse the two, but most people can tell them apart even in normally paced speech. But in Russian, this is possible only from context. Thus, "Ia govoriu ob Ane" (I'm talking about Anya) and "Ia govoriu o bane" (I'm taking about the banya) sound just alike.
It's not surprising for a native speaker to be completely unaware of these things. Your average Russian probably couldn't tell you that he/she de-voices word-final consonants (also, voiced consonants before a voiceless will devoice, even if not word final -- "vodka" is pronounced [-tk-]; and, although it's less frequent than voicing, also remember that voiceless consonants will voice before a voiced consonant, hence "sdelat'" is [zd'-] -- or actually [z'd'] because of softness assimilation, but that's a whole different story!). Your teacher, of course, is consciously aware of this, but perhaps only became so when she was trained to teach foreigners.
One final thought: In English, the difference between, say, "bus" and "buzz" is not only in the final consonant. In "buzz", the vowel is also longer, although it's a subsidiary and automatic phenomenon before the final voiced consonant and therefore we don't "hear" it. It may also be that your instructor is elongating the stressed vowel in "dzhaz" and "garazh" for emphasis, and that this elongation is reminding you unconsciously of the elongation before a final voiced consonant in English. Thus, you may also not be hearing exactly what you seem to be. (But I think my first explanation is the main reason here.)
Hope this helps.
Best,
FormerProfessor
September 5, 2003
9:45 am EDT
So my 9 am class MWF is Russian: Advanced Grammar and Syntax. (The "Advanced" may or may not actually be a part of the title -- I forget -- but it's required by the department for all incoming grad students in Russian, on the grounds that there's technical stuff they don't know but should if they're going to be teaching the language to freshmen. It's not a course in linguistics, but I got dispensation to take it as such if I write an extra linguistics-y paper for it.) We're talking about consonants today, and devoicing, and the subject of word-final (and consonant-cluster) voice assimilation comes up, and sure enough, the professor (same non-native speaker I mentioned in the summer) pronounces the word-final devoicing as total, and the native speaker in the class (which is not the same woman who was the native-speaker instructor over the summer) really really really sounds to me like she doesn't. Like the word lyubov, which according to the "rules" should be pronounced "lyubof", is actually coming out "lyubovf". And she's not there to teach foreigners how to pronounce things; these are foreigners who know already.
In fact, we're now told, voice assimilation applies only to obstruents: t, d, p, b, f, v, k, g, sh, zh, s, z. A cluster containing an obstruent and a resonant doesn't count as a cluster for assimilation purposes. And actually, [v] behaves like a resonant when followed by a vowel or another resonant (y, l, m, n, r) -- so in the word tvoi 'your', for example, the [t] does not voice -- the word is never, ever pronounced "dvoi", ever.
You see where this is going. What do I hear? "tfvoi."
Assuming you're still with me (hi,![[livejournal.com profile]](https://www.dreamwidth.org/img/external/lj-userinfo.gif)
ellen_fremedon; anyone else still here?), does it seem sort of like FormerProfessor didn't completely address my question? Just a bit, it seems like. I mean, he's not himself a native speaker, so I'm more inclined to believe him in cases like this (because native speakers are right about how to pronounce things in their own languages, but not always so objective about what's going on phonetically), but he didn't quote me any scripture. And also, if the issue is "resolved" in standard Russian, maybe there's some interesting dialect work to be done there. On the other hand, maybe it's not interesting at all. Maybe I'd write the paper and people would go "So there's not complete devoicing. So?"
But I really want to hook some people up to spectrograms and get some readouts (and learn how to read them, is the other thing). If only to determine whether my own ear is right, or whether I'm going bananas.
Thoughts?
June 11, 2003
12:07 pm EDT
[Dear] Prof. FormerProfessor:
... I was a student at [your university] from 1995-99, in Intensive Russian for my first three semesters. I'm working on an MA now, and find myself back in Russian classes, and something came up the other day that raised a question nobody here could really answer: word-final devoicing.
As it was presented to us, voiced consonants devoice word-finally. No problem; easy concept. But when our instructor, a native speaker, demonstrated, it really sounded to me -- especially in words ending in fricatives -- as if the sound began voiced, and then devoiced while in progress. In other words, she insisted that the "v" in "Chekov" devoiced completely, and then said "Chekovf." Also "jazs" for "jazz" and "garazhsh" for "garage" (apologies for not being able to reproduce any of this in Cyrillic). I asked her to repeat it several times, and this never seemed to change, but both she and the other instructor (a non-native speaker) were confident that the devoicing was total.
So it occurred to me that there could be a paper in this, if there hasn't been one already. I mean, in English I hear a difference between the ends of the word "svelt" and "veldt", or "bunt" and "bundt" -- granted, those are both borrowed Dutch words, but I'm sure I could come up with other examples -- and it may be the same in the case of Russian word-final voiced fricatives. What I'd like to do is find some more native speakers and hook everyone up to machines and get spectrographs, but if this is a non-problem (that is, if someone's already investigated it; if other native speakers report complete devoicing, I'm not inclined to believe them any more than I believe this one, without listening to them and measuring the voicing), obviously there'd be no point.
The professor here didn't know of any research in this area, but he admits he's not a phonologist, still less a phonetician, so the fact that he doesn't know of any research doesn't mean that none exists. Are you aware of investigation into this phenomenon? And if not, does it sound like something a person could get a paper out of (just a paper, not necessarily a master's thesis type thing)?
... Thanks for your help!
best,
fox
June 11, 2003
3:55 pm EDT
Hi, Fox,
... You ask a great question, but alas, there's no paper in it. In normal speech, at least, the de-voicing is total, or at least it is in standard Russian. I think what's happening is: 1) your instructor is just slowing down the pronunciation to demonstrate the de-voicing and, therefore (even if she is unaware of doing so herself!), starting out with some voicing. She has probably found that she has to do this, because (contrary to what you might think) de-voicing is NOT straightforward to most American students. Usually they have to be explicitly taught that, say, [d] and [t] are the same sound but for voicing. So I think she's showing (again, whether unconsciously or not) what the consonant "should be" according to orthography, and what it actually is. 2) Even in normal speech, de-voicing may not be carried out if the first sound in the next word is a voiced consonant. This lack of de-voicing is obligatory if the next word is a particle (e.g., "zhe" or "by"), which by definition is phonologically a part of the preceding word, but the de-voicing may also be prevented before any word starting with a voiced consonant. Russian has what linguists there refer to as "slitnoe proiznoshenie" (literally, "flowing-together pronunciation"), where there the word boundary is not distinct -- unlike English, and especially German, where there may be a slight glottal stop between one word and the next. Compare, for instance, "I have an aim in life" with "I have a name in life." It would be possible for an English speaker to confuse the two, but most people can tell them apart even in normally paced speech. But in Russian, this is possible only from context. Thus, "Ia govoriu ob Ane" (I'm talking about Anya) and "Ia govoriu o bane" (I'm taking about the banya) sound just alike.
It's not surprising for a native speaker to be completely unaware of these things. Your average Russian probably couldn't tell you that he/she de-voices word-final consonants (also, voiced consonants before a voiceless will devoice, even if not word final -- "vodka" is pronounced [-tk-]; and, although it's less frequent than voicing, also remember that voiceless consonants will voice before a voiced consonant, hence "sdelat'" is [zd'-] -- or actually [z'd'] because of softness assimilation, but that's a whole different story!). Your teacher, of course, is consciously aware of this, but perhaps only became so when she was trained to teach foreigners.
One final thought: In English, the difference between, say, "bus" and "buzz" is not only in the final consonant. In "buzz", the vowel is also longer, although it's a subsidiary and automatic phenomenon before the final voiced consonant and therefore we don't "hear" it. It may also be that your instructor is elongating the stressed vowel in "dzhaz" and "garazh" for emphasis, and that this elongation is reminding you unconsciously of the elongation before a final voiced consonant in English. Thus, you may also not be hearing exactly what you seem to be. (But I think my first explanation is the main reason here.)
Hope this helps.
Best,
FormerProfessor
September 5, 2003
9:45 am EDT
So my 9 am class MWF is Russian: Advanced Grammar and Syntax. (The "Advanced" may or may not actually be a part of the title -- I forget -- but it's required by the department for all incoming grad students in Russian, on the grounds that there's technical stuff they don't know but should if they're going to be teaching the language to freshmen. It's not a course in linguistics, but I got dispensation to take it as such if I write an extra linguistics-y paper for it.) We're talking about consonants today, and devoicing, and the subject of word-final (and consonant-cluster) voice assimilation comes up, and sure enough, the professor (same non-native speaker I mentioned in the summer) pronounces the word-final devoicing as total, and the native speaker in the class (which is not the same woman who was the native-speaker instructor over the summer) really really really sounds to me like she doesn't. Like the word lyubov, which according to the "rules" should be pronounced "lyubof", is actually coming out "lyubovf". And she's not there to teach foreigners how to pronounce things; these are foreigners who know already.
In fact, we're now told, voice assimilation applies only to obstruents: t, d, p, b, f, v, k, g, sh, zh, s, z. A cluster containing an obstruent and a resonant doesn't count as a cluster for assimilation purposes. And actually, [v] behaves like a resonant when followed by a vowel or another resonant (y, l, m, n, r) -- so in the word tvoi 'your', for example, the [t] does not voice -- the word is never, ever pronounced "dvoi", ever.
You see where this is going. What do I hear? "tfvoi."
Assuming you're still with me (hi,
ellen_fremedon; anyone else still here?), does it seem sort of like FormerProfessor didn't completely address my question? Just a bit, it seems like. I mean, he's not himself a native speaker, so I'm more inclined to believe him in cases like this (because native speakers are right about how to pronounce things in their own languages, but not always so objective about what's going on phonetically), but he didn't quote me any scripture. And also, if the issue is "resolved" in standard Russian, maybe there's some interesting dialect work to be done there. On the other hand, maybe it's not interesting at all. Maybe I'd write the paper and people would go "So there's not complete devoicing. So?"But I really want to hook some people up to spectrograms and get some readouts (and learn how to read them, is the other thing). If only to determine whether my own ear is right, or whether I'm going bananas.
Thoughts?
no subject
i'll think about it next week. :-)