LINGUIST List 10.420
Fri Mar 19 1999
Sum: Articulatory posturing of Tuvan throat singers
Editor for this issue: Brett Churchill <brettlinguistlist.org>
Anthony M. Lewis, Articulatory posturing of Tuvan throat singers
Message 1: Articulatory posturing of Tuvan throat singersDate: Thu, 18 Mar 1999 23:36:32 -0600
From: Anthony M. Lewis <am-lewisstudents.uiuc.edu>
Subject: Articulatory posturing of Tuvan throat singers
Recently I posted the following message regarding the articulatory posturing of Tuvan throat singers. I am pleased with the large number of informative responses I received to my posting which I have summarized below, along with my original question(s). Thanks again to all of you who took the time to reply to my inquiry, and my sincere apologies to anyone who I may have missed in my summary.
My question has to do with the vocal configurations assumed by the traditional Tuvan throat singers of Siberia. For those of you who may not be familiar, the Tuvan Autonomous Republic is a central territory of the Russian Federation situated on the border of Russia and Mongolia. This particular style of singing (also known as “overtone” singing, and attested in other parts of the world) is most notably characterized by the singer’s production of (at least ) a single, fundamental musical note, accompanied by the corresponding overtones (presumably, harmonics, of the fundamental tone). The perceptual effect is that of a robust, whistling, almost chord-like nature.
I have two rather simple questions regarding this phenomenon: first, is this unique acoustic effect the result of a highly coordinated posturing of the vocal folds (e.g., a complex setting of register(s)… a la “soprano” in voice science terminology)?; purely supra-laryngeal in nature?; or a combination of both?. My suspicion is that the robust percept is purely the result of sustaining a configuration of the supra-laryngeal cavity which enhances certain (resonant) frequencies of the fundamental tone. I’d be most interested to hear alternative accounts (e.g., complex laryngeal posturing, contribution of the pharyngeal wall, etc.). My second question asks whether or not Tuvan singers (or any other “overtone” singers for that matter) are capable of producing more than a single “fundamental” tone (and, presumably, the corresponding harmonics) at the same time. This, as far as I can figure, would require the vocal folds to vibrate simultaneously at two different fundamental frequencies.
Summary of Responses:
James Kirchner writes:
I learned to produce overtones with a very different sounding voice timber. It was based on an imitation my sister and I used to do of our uncle’s speaking voice. The overtones began to be produced when we transferred this voice to singing. I once did it for fun in front of a group
of my high school kids when I taught in Europe, and within three hours the
kids in the other high schools in town knew about it.
Marc Picard writes:
that this topic was previously aired on Linguist List. Unfortunately, his extensive summary of responses to the previous inquiry was destroyed by my word processing program. His e-mail address is <<picardvax2.concordia.ca> for those who may wish to request his thorough and informative summary of the original posting.
Karine Megerdoomian writes:
The singers sing a fundamental tone, one harmonic and then a note one octave higher. And also, it sounded like it had to do mainly with supra-laryngeal configuration as you mention.
Dirk Elzinga writes:
I was interested in your query regarding Tuvan throat singing. I have been interested in this style of singing for several years now. A vocal group from New York, Toby Twining Music, also employs this technique (among others) in their performances, and about three years ago, I
attended a workshop conducted by ensemble members in which some of these techniques were demonstrated; we even got coaching on how to do it ourselves. Your speculation that the overtone effect is purely supra-laryngeal is consistent with how we were coached in the technique. I remember at the time coming up with an explanation similar to yours for
this effect. However, we may not have been doing it exactly the way the
Tuvans do! In answer to your second question, I believe that the vocal technique
known as “vocal fry” does get two simultaneous tones. It is a very rough sound, though, and a vocalist friend tells me that it is very hard on the vocal folds. If you add the necessary supra-laryngeal configuration to a vocal fry, you conceivably get three tones at once. You may want to
consult vocalists for details on vocal fry.
Paul Boersma writes:
Overtone singing is a combination of:
1. bringing two formants close together (by tongue posture etc), so they climb up together to form a single much higher peak.
2. positioning this peak (by the same means) at a harmonic of F0. Choosing a somewhat high F0 will help.
3. sharpening the peak by reducing energy losses into the trachea. This is done by lengthening the closure duration of the vocal folds. When these are closed no energy is lost into the trachea. This phenomenon corresponds to the pitch-synchronous changes in formant bandwidths during normal speech.
4. sharpening the peak by reducing energy losses at the yielding walls. This is accomplished by stiffening the pharyngeal wall and other structures, and (for lower tones) by using the nasal cavity.
5. sharpening the peaks by suppressing adjacent resonances. Nasality could help here as well. I have my doubts about this one, however.
The point is that every laryngeal and supralaryngeal trick that helps to improve the perceptual result, will be used to an advantage. Don’t believe everything I say, however. Ask Gerrit Bloothooft of Utrecht (gerrit.bloothooftlet.uu.nl), who did research on the articulation of overtones.
Gina Joue writes:
>From what I know, Tuvan throat-singing involves manipulating the false vocal folds and the aryepiglottic folds WITH the oral filter (tongue, etc) to intensify certain harmonics present when he vocalizes. In certain styles, the louder harmonics become the whistling melody, and the typical drone or bass line is usually the fundamental. Because the singer is manipulating the harmonics AVAILABLE in vocalizing, most singers tend to be male — by virtue that males tend to have lower fundamentals and hence have more harmonics to manipulate which are still within the audibility range. I don’t know of any throat-singers who can produce more than a single
“fundamental” at the same time. There are several styles in Tuvan throat-singing where the fundamental is used more like a lower melodic line, but that’s about it. A lot of this is taking advantage of the auditory system’s strategies (gestalt, figure-ground) and susceptibility to certain illusions. Anyway, you might want to check out Ted Levin’s entry in Garland Encyclopedia on Tuva monograph “Musical Representations of Nature among the Pastoral Herders of South Siberia.” or something like: “An Animist View of the World: Sound, Music, and Nature in the Lives of the Inner Asian Pastoralists.” (I’m not sure what eventual title he used) Dr. Anat Keidar at Vox Humana Voice Lab in NY who has done films of Tuvan-singers’ throats, Mark van Tongeren’s work, “A Tuvan Perspective on Throat Singing” in Oideion 2: The Performing Arts World-Wide, edited by Wim Van Zanten and Marjolijn Van Roon, Research School CNWS, Leiden, 1995
Ellen Gerrits writes:
You might want to contact dr. P.A. Lindestad. I heard his talk titled “Mongolian throat singing: Acoustical and vibratory characteristics”. He presented this at the 24th World Congress of the International Association of Logopedics & Phoniatrics in Amsterdam, august 23-27 1998. I think his conclusion was that in throat singing there are two voice sources. One generated by the vocal folds and the other by the ventricular folds. The ventricular folds vibrate in half the frequency of the vocal folds. I don’t have an e-mail address of Dr. Lindestad, but he works at Huddinge University Hospital in Sweden.
John Ohala writes
The definitive acoustic and physiological study of throat singing has not been done (unlike the case for the ‘double voice’ of the Tibetan monks’ chant). It is almost certainly a matter of supraglottal configuration and not a special state of the vocal cords or mode of voicing, except, of course, to produce an appropriate F0 and to avoid vibrato. I can do it to a limited extent and Richard Wright, a grad. of the UCLA Phonetics Lab (now at U. Wash.), can do it very well. Neither of us is doing anything special with the vocal cords. No, it does not involve two fundamental frequencies. The auditory effect conveys two notes, if you will, but one is the fundamental and the other is the harmonic that is “super”-reinforced. Or let me rephrase that: throat singing, per se, does not require two fundamentals. Whether the Tuvan singers can produce one is a separate question. The work on Tibetan chant appeared in J. Acoust. Soc. Am. more than 20 years ago. I believe Ken Stevens (MIT) was a second author on that.
Deborah D K Ruuskanen writes:
Not that I’ve studied this, but I *have* heard this done, there was an I guess exhibition is the proper word, and since I am musically trained I can tell you that none of the three men (aged around 20, and 35-40) produced more than one fundamental tone at a time, but that there were several overtones for each fundament – and it was sometimes hard to tell if they were simply overtones (harmonics) or if the top tone was a different halftone up or down from the octave harmonics. It was most impressive. We were also told that various fundamental tones represented different things, i.e. the water in brooks, the wind in the tops of the spruce trees, etc. Don’t know if that helps.
Brian Donahoe writes:
I saw your request for info on Tuvan throat singing, and forwarded it on to Prof. Ted Levin, ethnomusicologist at Dartmouth, who is in the process of writing a book about Tuvan singing. He passed along the following information for me to send you:
> Re. the query about phonation, you could tell this person to look out for the *Scientific American* article on throat-singing which Mike Edgerton (from the National Voice and Speech Lab at Univ. of Wisconsin) and I are now completing. It should be out in the spring. It should answer his questions in more detail. Finally, why don’t you check out the Friends of Tuva website. There’s a lot on throatsinging there. Just put “Friends of Tuva” into a search engine, and it should pop up.
Matt Walenski writes
Catford (1984) has a page or two (103-104) about it. He refers to it as ‘double voice’ involving simultaneous ventricular and glottal phonation. Pike (1947) refers to a ‘double whistle point’ (page147) which may or may not refer to the throat singing type of phenomenon. Kathleen Hubbard, here at UCSD, has mentioned someone whose name I can’t recall who’s worked on this acoustically (she may have measured it also), who claimed that it was an F2 manipulation. </fontfamily>
Anthony M. Lewis
Department of Spanish, Italian, and Portuguese
University of Illinois at Urbana-Champaign
e-mail to: am-lewisuiuc.edu