Loading ...
Sorry, an error occurred while loading the content.

linguisticising throat-singing

Expand Messages
  • Alex Fink
    A little while back, Daniel Bowman pointed me to http://www.youtube.com/watch?v=i0djHJBAP3U and wondered to what extent throat-singing-style overtones could
    Message 1 of 4 , Mar 10, 2013
    View Source
    • 0 Attachment
      A little while back, Daniel Bowman pointed me to
      http://www.youtube.com/watch?v=i0djHJBAP3U and wondered to what extent
      throat-singing-style overtones could serve a phonemic function in a
      language, with two different tonal tiers. Neither of us really know
      enough about the mechanics of throat-singing to design something
      plausible, though. What is it, in terms that a phonetician would use?
      It's clearly possible to sing lyrics through it, but the result
      sounds distorted; which sort of phonetic contrasts would it mesh well
      with, which poorly?

      Any cool ideas for how a language with overtone could use it?

      Alex
    • Robert Marshall Murphy
      As someone who throat-sings (moderately well), I am overjoyed to hear of other conlangers who are interested in this! I ve studied Tuvan throat-singing, so
      Message 2 of 4 , Mar 10, 2013
      View Source
      • 0 Attachment
        As someone who throat-sings (moderately well), I am overjoyed to hear of other conlangers who are interested in this! I've studied Tuvan throat-singing, so I'm not sure about other methods of analysis.

        In Tuva, they say there are three styles, conveniently called low, medium and high. Low is like Tibetan Buddhist monk chants, and gets the false-vocal-folds vibrating at half the frequency of the base note, i.e. one octave lower. Exceptionally talented throat-singers can get a fifth above the fundamental ringing as well, but this is very rare and I'm not even sure what anatomy is producing it. Cardinal vowel /o/ produces the overtone (in solfeggio) "do". /u/ produces a fifth below that, "sol". /ɔ/ make "re". /ɑ/ produces "mi". /a/ makes "sol" above "do". These are all four octaves about the fundamental. Speech-singing is possible, but unvoiced consonants either stop the singing or are realized as voiced.

        Middle style is like the low, but without the undertone. It is the easiest and free-est, most often combined with various techniques.

        High style (my best) is best accomplished with lips protruding. The vowels are /ɚ/ through /i/, but I'm not good at notating all those central-vowels.

        I've seen spectrograms of my throat-singing before, and it really changes the formants F1 and F2. I don't think that it could be considered an extension of ordinary speech, but would have to be its own thing.

        -Robert Marshall Murphy-


        On Mar 10, 2013, at 2:05 PM, Alex Fink <000024@...> wrote:

        > A little while back, Daniel Bowman pointed me to
        > http://www.youtube.com/watch?v=i0djHJBAP3U and wondered to what extent
        > throat-singing-style overtones could serve a phonemic function in a
        > language, with two different tonal tiers. Neither of us really know
        > enough about the mechanics of throat-singing to design something
        > plausible, though. What is it, in terms that a phonetician would use?
        > It's clearly possible to sing lyrics through it, but the result
        > sounds distorted; which sort of phonetic contrasts would it mesh well
        > with, which poorly?
        >
        > Any cool ideas for how a language with overtone could use it?
        >
        > Alex
      • Alex Fink
        ... Mhmm, the most straightforward way to make a throat-singing conlang would be as its own thing, I think, not trying to do the multi-modal trick of making an
        Message 3 of 4 , Mar 20, 2013
        View Source
        • 0 Attachment
          On Sun, 10 Mar 2013 15:34:36 -0500, Robert Marshall Murphy <mrandmrsmurphy@...> wrote:

          >As someone who throat-sings (moderately well), I am overjoyed to hear of other conlangers who are interested in this! I've studied Tuvan throat-singing, so I'm not sure about other methods of analysis.
          >
          >In Tuva, they say there are three styles, conveniently called low, medium and high. Low is like Tibetan Buddhist monk chants, and gets the false-vocal-folds vibrating at half the frequency of the base note, i.e. one octave lower. Exceptionally talented throat-singers can get a fifth above the fundamental ringing as well, but this is very rare and I'm not even sure what anatomy is producing it. Cardinal vowel /o/ produces the overtone (in solfeggio) "do". /u/ produces a fifth below that, "sol". /ɔ/ make "re". /ɑ/ produces "mi". /a/ makes "sol" above "do". These are all four octaves about the fundamental. Speech-singing is possible, but unvoiced consonants either stop the singing or are realized as voiced.
          >
          >Middle style is like the low, but without the undertone. It is the easiest and free-est, most often combined with various techniques.
          >
          >High style (my best) is best accomplished with lips protruding. The vowels are /ɚ/ through /i/, but I'm not good at notating all those central-vowels.
          >
          >I've seen spectrograms of my throat-singing before, and it really changes the formants F1 and F2. I don't think that it could be considered an extension of ordinary speech, but would have to be its own thing.

          Mhmm, the most straightforward way to make a throat-singing conlang would be as its own thing, I think, not trying to do the multi-modal trick of making an ordinary-speech mode of the same phonology.

          That said, I hadn't realized that it was just the formant differences of ordinary vowels that were being co-opted to produce the overtones here! Neat. Seems that, modulo not getting to have a voice contrast (unless "stopping the singing" is okay? I mean, stops already stop the airflow), a phonemic inventory of a throat-sung language might look positively unremarkable when described, if the overtone pitches are really just something like /u o O A a/ plus /@ @\ 1 i/.
          Though, hm, it'd be really neat if that meant the overtone tier was subject both to the sort of phonological processes that characterise tone (relative autosegmentality, obligatory contour stuff, drift and downstep and terracing oh my) as well as those that characterise vowels (fronting near palatal consonants, lowering near pharyngeal ones, rounding near round ones), and they got to interact in unusual ways...

          Alex
        • Paul Roser
          ... It s been a while since I d studied this part of acoustic, but I suspect that the reason some singers can achieve the fifth above the fundamental is
          Message 4 of 4 , Mar 20, 2013
          View Source
          • 0 Attachment
            On Wed, 20 Mar 2013 17:43:56 -0400, Alex Fink <000024@...> wrote:

            >On Sun, 10 Mar 2013 15:34:36 -0500, Robert Marshall Murphy <mrandmrsmurphy@...> wrote:
            >
            >>As someone who throat-sings (moderately well), I am overjoyed to hear of other conlangers who are interested in this! I've studied Tuvan throat-singing, so I'm not sure about other methods of analysis.
            >>
            >>In Tuva, they say there are three styles, conveniently called low, medium and high. Low is like Tibetan Buddhist monk chants, and gets the false-vocal-folds vibrating at half the frequency of the base note, i.e. one octave lower. Exceptionally talented throat-singers can get a fifth above the fundamental ringing as well, but this is very rare and I'm not even sure what anatomy is producing it.

            It's been a while since I'd studied this part of acoustic, but I suspect that the reason some singers can achieve the fifth above the fundamental is because their fundamental is so low.

            [snipped]

            >>I've seen spectrograms of my throat-singing before, and it really changes the formants F1 and F2. I don't think that it could be considered an extension of ordinary speech, but would have to be its own thing.
            >
            >Mhmm, the most straightforward way to make a throat-singing conlang would be as its own thing, I think, not trying to do the multi-modal trick of making an ordinary-speech mode of the same phonology.
            >
            >That said, I hadn't realized that it was just the formant differences of ordinary vowels that were being co-opted to produce the overtones here! Neat. Seems that, modulo not getting to have a voice contrast (unless "stopping the singing" is okay? I mean, stops already stop the airflow), a phonemic inventory of a throat-sung language might look positively unremarkable when described, if the overtone pitches are really just something like /u o O A a/ plus /@ @\ 1 i/.

            >Though, hm, it'd be really neat if that meant the overtone tier was subject both to the sort of phonological processes that characterise tone (relative autosegmentality, obligatory contour stuff, drift and downstep and terracing oh my) as well as those that characterise vowels (fronting near palatal consonants, lowering near pharyngeal ones, rounding near round ones), and they got to interact in unusual ways...

            I believe that the vowel changes are are directly linked to the overtone changes, so I don't think you can actually manipulate the overtones independently of the vowels... at least, not in a human vocal tract.

            If you're up for some fairly detailed phonetics treatments of overtone singing, there are two papers online you could check out:

            http://email.eva.mpg.de/~grawunde/files/GrawPhD3.pdf

            http://physastro.pomona.edu/wp-content/uploads/2012/09/R_Bryant_Foresman_Thesis.pdf



            >
            >Alex
          Your message has been successfully submitted and would be delivered to recipients shortly.