Kevin James‘s “Vanishing Languages,” a musical performance that responds to and incorporates two endangered languages, was performed at Roulette last Thursday and Friday.

“83 languages with ‘global’ influence are spoken and written by 80 percent of the world population. Most of the others face extinction at a rate, the researchers said, that exceeds that of birds, mammals, fish and plants,” reports the New York Times.

What use does a recorded language serve after the speakers have passed away? Language can provide cultural or even scientific data (e.g. the Micmac language, the BBC tells us, can help track the effects of acid rain on trees in the area, as the word for each kind of tree changes based on the sound it makes in the wind). The Endangered Language Project is pretty cool to browse, though it’s hard for the average listener to glean much from individual recordings.

Composer and trumpet player Kevin James chose to respond to these endangered languages with music. In “Vanishing Languages,” he integrates the recorded speakers of Hokkaido Ainu (aboriginal language from northern Japan), Quileute (the language of a Native American tribe of the Pacific Northwest), and Dalabon and Jawoyn, two aboriginal languages from Australia. Two of these, Dalabon and Quileute, were included in the performance last Thursday.

James began research for the project six years ago, inspired by the Unesco Atlas of Endangered Languages. How do you even begin to track down the remaining speakers of these endangered languages, knowing only two to six are still alive? James answers in a recent interview:

I began by creating a list of potential languages by region then narrowed it down by selecting only language isolates (those for which no related languages exist) which were both on the verge of extinction, but likely to be “researchable” within the timeframe of the project. I then researched the regions of those languages I was considering (eliminating those whose not-likely-to-return factor was too high), and began contacting linguists for advice or introductions to potential peoples, tribes, and native speakers.

On the Kevin James general reliability scale, linguists fall somewhere between poets and philosophers, so access and response from the linguists themselves greatly influenced my opportunities and eventually led to my choices of languages and travel itinerary for the project. For example, key to the success of my time in the Pacific Northwest was was a quickly spoken conversation finalizing introductions to the Quileute Tribal elders, as my linguist contact’s mini satellite was slowly covered by the snows of a blizzard in the Inuit village he was visiting.

The first piece on Thursday was “Biyi-keninjh Kah-yenjdjung Dalabon,” performed by Speak Percussion. The title of the piece was given to James by a Dalabon speaker (no one in the audience dared to try to correct his pronunciation).

Dalabon has an incredible array of person, number, and kinship categories in its pronoun system. James shared in a Q&A after the piece that each person who stayed with the tribe for more than two days was assigned a “skin name,” which details your relationship to everyone else in the group and indicates how close you are allowed to be to others (“whether or not you could be in the same room”). It’s a complicated naming system that is designed to prevent inbreeding.

As the first piece began, a recording that featured six speakers played in surround sound, and the percussionists varied tempos and instruments to match, imitate, or harmonize with the recording. (Each percussionist was assigned two speakers to follow.) One performer characterized the speech pattern as “gentle, elegant, fast, and fluidly delivered.” The three percussionists played tin cans, bells, chimes, drums, and even zippers as they echoed the rhythm of the Dalaban speakers. A few moments of the performance were improvised.

I’m always a little skeptical of experimental music, but I’ll admit the percussion made me listen to the language in a different way, one that was searching for patterns and rhythms in the speech that I might not have heard without them. Listening to language without the chance to comprehend the content, or interpret the meaning from the speaker’s face, is a strange sort of listening.

Each percussion item was carefully chosen to represent a different piece of aboriginal culture—the tin cans, for instance, were used for the tribes’ lunches. In my favorite moments, rocks were thrown into large pitchers of water, tossed in a careful rhythm with the surface of the water amplified. Near the end of the piece, all three performers raced to the middle of the stage to open and close a series of wooden windows and zippers, in a rapid and theatrical manner.

Windows are very new to the tribe, James explained later, and no one is really sure what to do with them. The performers said it was a challenge to internalize the language, especially with so many changes in tempo, and that what James has created is “a unified percussion language that, even to us, sounds foreign,” according to one percussionist.

(James suggested that we check out Speak Percussion online after the performance, and you can hear them incorporate the sound of an onion being chopped in “Pasta Percussion,” from the Melbourne Food & Wine Festival.)

The audience was eager to know what the speakers were saying. James said that he was approaching the language as cultural artifact, and was interested only in its aesthetic value. While he recorded the chosen speakers, he told people they could talk about anything they wanted. In this piece, one woman told him how she had been taken from her family when she was a little girl, and how it felt to be whisked away to a new school. Another talked about buffalo hunting.

In the second piece, the meaning of the words was simpler and more significant. “Counting in Quileute” prominently features one speaker counting down from one hundred to two. Two because that is the number of Quileute speakers who remain alive today. (Both are included in James’s recording, though he shared that they don’t get along.)

Quileute is spoken with the tongue between the teeth, and the musicians imitated this speech pattern in a variety of ways: the brass section used thirteen mutes of various shapes and sizes to muffle their sound; woodwind players made a ferocious “tsk” sound over and over into their instruments; in the end, the percussionists threw gongs and bells onto the ground and let their sound die out.

Speak Percussion was joined by the [kaj] ensemble for this piece. A song or chant ran through the recording along with the counting, creating a more consistent rhythm than in the first piece. The variety arrived this time in the crescendos, and in the cacophonous energy brought by this ensemble. The cellist stomped the floor with his instrument; a viola player strummed like a rock guitarist. This time the audience could read the joy and despair in the musicians’ faces. The percussionists rang Tibetan singing bowls and gongs in quiet moments, performing a memorial for the words we are about to lose.

You can hear the recording in full here. The first speaker arrives at the two minute mark, and if you listen carefully you can tell it was first recorded on a Gramophone.

One piece varied between the Thursday and Friday performances, and you can read more about Friday in the Times:

In “Ainu Inuma” players also imitated the sound of a type of bamboo jaw harp played by the Ainu by striking the strings with the wooden side of their bows. The singing and moaning sounds of Dorothy Lawson’s cello were sometimes so close to the voice of of one female speaker that she became in effect her body double…. The imperfections of these old recordings, which Mr. James used alongside those he made recently in the field, show how heavily smudged the window is that we have on these vanishing cultures. And yet at times it seemed as if it were these voices who were willing the performance into existence.

I’m just pleased to see two new iterations of languages most of us will never get to know.