Flutopedia - an Encyclopedia for the Native American Flute

Sign up for our Flute Newsletter

 


 
Previous PageNext Page
Flutopedia Symposium

The Warble

More than any other ornament or effect, the warble (or “warbling”) is the classic and distinctive sound of traditional Native American flutes. Only a small portion of present-day Native American flutes are designed to warble. But, historically, some makers went as far as to discard a flute if it did not warble.

Descriptions of the warble vary from “bubbling” ([Hensley 2002] and [Riggs 1869] Tah-koo Wah-kan or, The Gospel Among the Dakotas, page 480), to “a full, vibrating tone” ([Fletcher 1911] The Omaha Tribe), to a “natural vibrato” ([Crawford 2006]), to a “burble” ([Payne 1988]) and a “yodel” (Doc Green Silverhawk, personal communication, October 22, 2010). Reactions by listeners vary widely. Some find the sound of a warbling flute pleasing, some find it annoying, some consider it a mistake by the player. This makes it an excellent example of a culturally specific ornament.

This page was updated substantially in early March 2013 from research done over the last two years by Barry Higgins of White Crow Flutes. This research included the use of spectrograms to determine the acoustic attributes of a warble, as well as a classification of different types of warbles that is described on this page.

Turkey Legs, 1920s. Photo: Thomas Bailey Marquis

Turkey Legs, 1920s
Photo: Thomas Bailey Marquis Larger image

Recording Sample

One of the best examples of a steady-state warble from a traditional recording is this two-second sample from a recording by Turkey Legs (pictured at the right, from [Marquis 2007]). The original recording was provided by O. W. Jones of Wild Horse Mountain Flutes (where you can hear the full-length recording), with audio restoration by Barry Higgins of White Crow Flutes:

Audio Player disabled - visit Troubleshooting.

Note that it sounds like the flute is vacillating back and forth between different notes. However, this classic warbling sound is actually the sound of different harmonic components of the same note coming into dominance at different times.

Types of Warbles

In studying the acoustics of recordings of warbling flutes, Barry Higgins identified two types of warble:

Steady-state Warble

A steady-state warble is generated by some flutes when they are played with a forcefully, but steady breath pressure. This is almost always done on the fundamental note of the flute, with all finger holes covered.

The warbling sound is produced by the flute itself, without changes in breath pressure by the player. Although a steady-state warble does require increased breath pressure to initiate the effect, once in that “zone” the physics of the flute take over and the steady-state warble continues.

Flutes that can produce a steady-state warble are called warbling flutes.

This type of warble is sometimes called a “true warble”.

Vibrato-induced Warble

A vibrato-induced warble is a generated by the player using vibrato technqies.

By varying breath pressure, the player can edge the flute close to an overblow, without letting the flute fully transition into the second register. The vibrato techniques can approximately mimic the sound of a steady-state warble, but not precisely.

A warble produced with vibrato is sometimes calles a “pseudo-warble”.

Spectrogram Analysis

A spectrogram is a visual representation of the sound. This is a spectrogram of the two-second Turkey Legs excerpt above, showing a full cycle before, during, and after a steady-state warble. (Click on the picture to see a larger version ... you might wish to right-click or command-click and select "open in a new tab" or "open in a new window"):

Spectrogram of a True Warble by Barry Higgins: Turkey Legs Warble excerpt

Spectrogram of a steady-state warble by Barry Higgins:
Turkey Legs Warble excerpt Larger image

The horizontal axis is time – two seconds of sound in the case above.

The top half shows the sound wave (in the case above, a monophonic sound). The larger the wave, the louder the sound. Notice that the volume of the sound actually varies quite a bit within this two-second window.

The bottom half (the spectrogram “proper”) shows the spectrum of frequencies in the same sound wave. The vertical axis in this lower portion represents frequency, from 200 Hertz to 2,500 Hertz in this case. The loudness is represented by the intensity of the color – in this case, the depth of the black.

Notice that the spectrogram is divided into harmonic bands – identified here from the first to the fifth harmonic band. At the purple line, the flute has not yet begun to warble: the volume of each of the harmonics are in step (“in phase”) with each other. Then, further to the right, with an increase in volume (cause by an increase in breath pressure), the second harmonic begins to become out of phase with the first harmonic. The third harmonic also follows this pattern shortly afterwards. After that point the flute is in full steady-state warble, with the second harmonic alternating with the first and third. It appears on the spectrogram that the harmonic bands are “vying for dominance”.

The yellow and green lines are visual guides. As the second harmonic gains strength (higher density black), the first harmonic loses strength at which point they are phase inverted. Then we have a reversal of return of dominance to the first harmonic and weakening of the second harmonic.

Spectrogram of a True Warble by Barry Higgins: Detail of Turkey Legs Warble excerpt

Spectrogram detail by Barry Higgins:
Detail of Turkey Legs Warble excerpt Larger image

Steady-state Warble on a Contemporary Native American Flute

Here is a similar spectrogram showing a steady-state warble produced by a contemporary instrument:

Spectrogram of a True Warble by Barry Higgins: Tsunami Flute mid-G played by Barry Higgins

Spectrogram of a steady-state warble by Barry Higgins:
Tsunami Flute mid-G played by Barry Higgins Larger image

The flute is a mid-range G4 crafted by Dusty Moore of Tsunami Flutes. According to Barrry Higgins, Dusty did not know the flute could warble. It did not warble normally, but as the block is moved slightly toward the foot of the flute (i.e. closer to the splitting edge) and provides sufficient steady breath pressure.

Notice that the first three harmonics show the well-defined spectrogram pattern of a steady-state warble, but the fourth and fifth harmonics show a different, “honey comb” pattern.

Vibrato-induced Warble

This section looks at spectrograms of two recordings by the same player at the same concert, Timothy Archambault at the National Museum of the American Indian in 2007.

The first is from Tim's performance of Belo Cozad's Memory Song:

A spectrogram of a section of this recording shows the classic attributes of a steady-state warble. Since this is a stereo recording, the top section shows sound waves for both the left and right channel:

Spectrogram of a True Warble by Barry Higgins: Timothy Archambault playing Belo Cozad's Memory Song

Spectrogram of a steady-state warble by Barry Higgins:
Timothy Archambault playing Belo Cozad's Memory Song Larger image

In this second recording, Tim is performing the Kichesipirini Honor Song:

The spectrogram of a warbling section in this recording shows a very different pattern:

Spectrogram of a vibrato-induced Pseudo-warble by Barry Higgins: Timothy Archambault playing Kichesipirini Memory Song

Spectrogram of a vibrato-induced warble by Barry Higgins:
Timothy Archambault playing Kichesipirini Honor Song Larger image

Notice that the second and third harmonics (and the fourth, if you squint your eyes a bit) are lined up with eacth other. This is an indication that this warble is induced by Tim's vibrato techniques rather than by the acoustics of the flute itself.

Other Recordings

Jaspar Blowsnake – 1937

Frank Gouldsmith Speck recorded Jasper Blowsnake, Thunder Clan, Winnebago Indians in Wisconsin in 1937. Several short excerpts that demonstrate the warble are available at the web site of the American Philosophical Society, on their Native American Sound Recording page — in particular, Recordings 49,2 and 49,3. Details of their recordings are also available on this Recordings Detail page.

Richard Fool Bull, 1960s

Richard Fool Bull, 1960s Larger image

Richard Fool Bull – 1960s

Here is an excerpt from a set of recordings of Richard Fool Bull (also known as “Richard Fool's Bull”). They were made in the early 1960s by Bob DeMersseman at the Rapid City Museum and provided to me courtesy of Jaime Bethel.

Richard was quite a character, serving in the armed forces in World War I ([LynnSherow 2001]), acting in the movie A Man Called Horse, and well known as a flute player and maker. One of his flutes was recently used by Peter Phippen in his Grammy-nominated album “Woodnotes Wyld”. Bob estimates that “he was about 85 in 1965” and that “he participated in rodeo until way into his 60’s” (personal communication, December 19, 2011).

I believe that the photo at the right, courtesy of Jaime Bethel, was taken during one of the recording sessions set up by Bob DeMersseman.

The ownership of these recordings is not known to me, but I have received permission from Bob DeMersseman, the recording engineer, to post this excerpt:

Audio Player disabled - visit Troubleshooting.

Toubat Flutes

I learned about the warble one late night in November 2002 at the Flower Mound flute gathering. Doc Payne, Peter Phippen, Robert Gatliff, Mark Slater, and a few others of us were sitting around after a long day of workshops and a lot of flute playing. Somebody got the idea to see if the “new guy” (me) could “play like a real flute player”, so Doc went and fetched one of his Toubat flutes. I apparantly was able to push the flute close enough to a real warble to satisfy Doc, and he gifted me the flute. That flute is shown here:

Toubat Flute, collection of Clint Goss

Toubat Flute, collection of Clint Goss Larger image

Here is a recording on that Toubat flute, my version of the song Zuni Sunset recorded February 16, 2012, with my best shot at the warble on the low note:

Audio Player disabled - visit Troubleshooting.

Here is a spectrogram of the above recording.

Quiz: am I letting the flute produce steady-state warble, or am I using vibrato-induced warble?

Spectrogram of a recording by Barry Higgins: Clint Goss playing Zuni Sunrise on a Doc Payne Toubat flute

Spectrogram of the above recording by Barry Higgins:
Clint Goss playing Zuni Sunrise on a Doc Payne Toubat flute Larger image

The warbling sound was so important to traditional players, that they would not play a flute that did not produce the sound. From [Fletcher 1911] The Omaha Tribe, page 371, referring to the Omaha culture:

To be acceptable, a flute must give forth a full, vibrating tone when blown with all six holes closed. It was interesting to watch men, old and young, take up a flute to test it; they would readjust the stop piece, bound to the top over the opening and usually carved, and if after several trials the instrument could not be made to give this vibratory tone the flute would be laid aside and no words would avail to make the man take it up and play a tune on it.

Acoustics

From an acoustic perspective, precise analysis of the warble is difficult. Kevin Locke provides a general description as “a rapid alternation between two different pitches (acoustical beats)”, a description also found in [Wapp 1984].

Here are two descriptions by Dr. Richard W. Payne from [Payne 1988] and [Payne 1999], respectively:

Using a properly constructed Plains flute with all tone holes covered, diaphragmatic air will produce the tonic pitch F sharp in warm vibrato. This accentuated vibrato, known as the ‘warble,’ is an important feature of traditional Plains Indian flute playing. It is accentuated by directing the air blade slightly high over the fipple edge and further enhanced by adding support of the air column slightly compressed in the pressure chamber of the flute. Tone frequency of the vibrato can be increased by as much as a half-step [semitone] by pushing cold air to produce a multiphonic warble, the tempo and tonality of which can be controlled in an effective manner. This warble, scrupulously avoided by organ pipe builders who term it “burble”, is a prized attribute of the Plains flute which can be driven with considerable variation in air pressure, in contrast to the organ pipe.

The ‘toubat flute’ requires particular skills in playing technique. In the old style of Plains flute playing, the vigorous multiphonic ‘warble’ on the fundamental note was highly regarded. This requires careful alignment of the block, nest, and roost so that the air stream is directly slightly high on the fipple edge [splitting edge]. Breath control is critical to execution of the oscillating warble; this is accomplished with warm diaphragmatic breathing dynamics. Air compressibility is also aided by providing slight impediment to the air stream, resulting from a slightly narrowed embouchure hole and air vent.

And here are thoughts from R. Carlos Nakai, from a personal communication in 2002 with Tim Crawford that he published in [Crawford 2006], page 27:

In actuality, the warble sound which is sounded at the all-closed position only on all five- and six-hole flutes is, to my experience, merely an indicator that the sound-producing mechanism is well made and is correctly positioned for optimum air flow from the air chamber [slow air chamber] over the block and directed by the saddle/bird mechanism agains the distal edge of the body tube hole [sound hole]. The oscillations of air movement coincidence with the Coreolis effect and the standing sine wave in the body tube [sound chamber] helps make this effect possible. … As an effect, if one only played the flute's tonic pitch [fundamental note] then the warble effect would be a useful embellishment for modulating that singlular pitch. The variations of air intensity in effecting a more or less pronounced and sometimes faster or slower warble is also an indicator of effective use of embouchural air control by the flutist and adds to the quality of the resolving pitch.

In more cases than not, the subsequent quavers/vibratos that are performed at various states of one's performance are matched to the warble. So, upon returning to the resolving pitch, if it is the lowest one, it will be the use of the warble rather than the vibrator. It's simple, but difficult in practice!

J. W. Coltman, in a detailed analysis of flute acoustics, describes the warble as a situation “in which amplitude modulation occurs in all partials but with different phases”; but in the end concluded that analysis of the warble “is yet to be explained” ([Coltman 2006]).

Here are messages from Mike Jones and Russ Wolf of the Native Flute Woodworking Yahoo newsgroup on October 17, 2010 regarding designing a warbling Flute:

Mike Jones: Warble is an instability in the flute, and when present is almost always at the fundamental and once in a while on the first finger hole. Some flutes that don't normally warble can be coaxed into it by adjusting the totem block forward and aft. A high and deep chimney seems to help create warble. On the Flutemakers Yahoo Group (not a NAF-related newsgroup) they were discussing it a little and someone indicated that finger hole position and certain undercutting might make other notes unstable enough to warble. It seems that most NAF players no longer appreciate warbling flutes, even though for some tribes, historically, the warble was required trait for a flute. I have a couple of flutes that I can make warble by breath control. The warble happens as I increase blowing pressure until it is almost ready to overblow.

The TuneIt tuner program for the PC has a spectrum view that lets you see the relative peaks of the harmonics in a given note. I've noticed that if the 2nd octave harmonic is almost as high as the fundamental then the flute is easy to overblow and is also likely to be able to warble.

Russ Wolf: I do make the Toubat flute. Doc Payne designed it to be warbling flute. It seems that factors that contribute to a warbling flute are: a narrow flue; a lot of restriction from the block — that is, a high, deep or overhanging block; the air stream hitting the splitting edge high. The first couple of years I made flutes I was using a more narrow flue and often had a warble on my flutes. For the last number of years I have been using a wider flue, and work for (adjust the splitting edge) a more stable tone on the fundamental — I rarely get any warble at all these days.

 

Andy Cox, October, 2010

Andy Cox, October, 2010 Larger image

While some makers attempt to achieve a warbling flute by careful design, others seem to stumble on the right combination. Here's a email from Andy Cox (personal communication, October 25, 2010):

Oh, the warble … I stumbled upon doing so by hammering a nail flat and making a small cane flute one afternoon. We were riding bikes and had stopped under a tree to get a breath of air, next to us was a cane patch. Of course the little thing was crude, but it played with grass wrapped around the fipple. Came home and made several tests. It seems that if one does not have “the right tools”, things happen.


Origins

The origin of the warble is open to conjecture, but Kevin Locke proposes that it was “probably incorporated into the instrument's design to imitate vocal pulsation that is a characteristic of Indian singing” (see An Informative Essay on the Indigenous Flute Tradition).

James Stevenson lists two distinct flutes, a “Sacred flute” and a “Sacred warbling flute”, in his 1884 catalog of about 4,900 “archæologic and ethnologic specimens collected in Arizona and New Mexico during the season of 1881. These collections were all obtained from the pueblo of Zuñi in Northwestern New Mexico, and the pueblos comprising the province of Tusayan, in Northeastern Arizona.

The Zuñi names for the “Sacred flute” and a “Sacred warbling flute” are distinct: Shoh-k'on-ne and Tchá-he-he-lon-ne, respectively ([Stevenson-J 1884], page 583).

Much has been written about the warble, and here are some links to available information:

 
Previous PageNext Page