Collaborating in Gwich’in on the Web: Khahłok Gwitr’it T’agwarah’in

By Craig Mishler, Kenneth Frank, and Allan Hayton

Alaska Native Language Center, University of Alaska Fairbanks

moosejawblue@yahoo.com; kefrank1@gmail.com; notyah.nalla@gmail.com

This essay examines a reflexive methodology that has emerged in our ethnographic and ethnolinguistic work on Gwich’in caribou anatomy and cultural ecology in northeast Alaska and the Yukon.  It follows the work of several others who have closely examined and participated in linguistic collaboration (Yamada 2007,  Leonard and Haynes 2010; Schulist 2013) but extends that collaboration to the web.  Our project, which began in the spring of 2012, has been generously supported by a grant from the National Science Foundation program in Arctic Social Sciences, Award No. ARC-1162600. 

Logistics

We generally spend between one and two hours in the cloud each day editing and translating language texts generated by field recordings we made with Gwich’in elders in the communities of Arctic Village, Venetie, Fort Yukon, Fairbanks, and Old Crow between 2012 and 2013.  Some of these texts are in the form of traditional stories, while others are open-ended descriptive accounts, recipes, memorates, or extended dialogues.  Kenneth Frank, an elder and fluent speaker of Gwich’in, conducted most of the interviews while Craig Mishler operated audio and video recorders, made still photos, and took field notes. 

With our homes separated by anywhere from three hundred miles up to four thousand miles, depending on the time of year, and air travel costing as much as $800 round trip to visit one another face-to-face, we have gradually worked out a method to collaborate closely on Gwich’in language materials starting with Skype, the Internet voice calling service.   

Back in the 1990s, before the villages had Internet service, we spent a tremendous amount of time sending manuscripts back and forth in manila envelopes via the U.S. postal service.  Telephone calls certainly helped, but in those days they were quite expensive, and  the normal turnaround time for editing back then was seven to ten days. But this is how we produced the volume Neerihiinjìk: We Traveled from Place to Place: the Gwich’in Stories of Johnny and Sarah Frank (2005).

Protocols

We begin each cycle of the process by uploading digital audio field recordings to Dropbox, a popular cloud-based file storage system.  To conserve space and to speedup uploading and downloading, we generally convert the original large .wav files to smaller .mp3s.  There is some loss in quality associated with this codec conversion, but in most cases it has not resulted in a barrier to listenability and comprehension.  Next we alert our transcribers to the fact that these audio files are available for sharing and downloading.  Once downloaded, the audio files are imported by the transcribers into Express Scribe, a free software utility for controlling and playing back audio. 

Express Scribe does digitally, with an adjustable rewind controlled right down to the millisecond, what we once used to do with a dedicated cassette transcriber and foot pedal.  With Express Scribe and Word open at the same time in different windows, the original Gwich’in audio is typed on a laptop and transformed into a raw transcript.  Sometimes our transcribers turn this raw transcript into a raw translation, and sometimes when they have difficulties they forward it to the editors for further translating. This step is always negotiated.  Although it is possible to hear recordings being played back on Express Scribe through a Skype connection, the audio is generally muted or badly distorted for everyone except the person controlling the playback.

It should be said that although the base language for our online verbal discourse is English, sometimes Kenneth brings his wife Caroline into the conversation, and their discourse is almost entirely in Gwich’in. This latter technique is very effective for getting at nuances in the texts and for making spelling checks.  

At any rate, we believe our evolving methodology has great potential for working with other endangered languages, Dene or otherwise.  Although not without its problems, including drop outs, voice distortions, echoing, hissing, and sudden disconnects (which often seem to be weather-related), Skype works better than any telephone could because it frees our hands to type and edit texts and proofread translations of texts for hours on end.  In addition, there are no per minute costs.  Sometimes we have to hang up and retry several times before establishing a stable connection, but our patience is generally well rewarded.  

We would like to show here how we have adapted our schedules to accommodate this technology and demonstrate how it has made our teamwork go much faster and more accurately than it would have otherwise.  When the senior author is living in New Mexico during the winter months, we have to allow for a two-hour Alaska time zone difference. When the junior author is residing in Arctic Village in the summer months, we have to allow for slower Internet connections with more disconnects. When transcribers Crystal Frank or Allan Hayton join us online, we schedule joint work sessions for Saturday mornings or evenings because of their full-time employment obligations during the work week.  It’s all about flex time.

Screen sharing, a Skype feature, allows us to see each other’s documents as we edit.  In fact, Skype screen sharing allows us to see two Word documents at once, with the Gwich’in original and the English translation arranged either side by side, or above and below.  Before we discovered screen sharing, we would just e-mail Word text documents and open them separately and simultaneously on our own computers.  We could still talk about the texts that way, but we could not see each other’s edits.  And we also had trouble locating each other’s place within a document using this system until we started inserting line numbers in the Word documents.  Placed unobtrusively at five line intervals, line numbers helped us focus our energies and attention on the same words and the same sentences at the same time. 

More recently, however, our system has evolved further.  We still use Skype for voice communication, but Skype screen sharing allows only one of us at a time to make edits to a text.  For the other editor to make edits, the person sharing a document would have to end screen sharing and e-mail the revised file, so that the other person could also make edits, a tedious procedure.  It was manageable to some degree when Kenneth wanted to edit the Gwich’in and Craig wanted to edit the English translation because Kenneth was the active editor on the Gwich’in and Craig was the active editor on the English, but it still forced one of us into a passive role and therefore proved to be limiting.

However, we discovered that if we used Google Drive, the cloud-based server and storage service, for collaborative editing, multiple participants could make corrections simultaneously in real time. We are indebted to Allan Hayton for this suggestion.  After Word files are uploaded to Google Drive, they may be opened with Google Documents, a basic but serviceable word processor.  By sharing these documents, we can witness our own and others’ edits as they are being made.  Each collaborator viewing the document is recognized by a different color cursor and an icon containing the first letter of his or her name at the top of the document.

Although normally only the two authors of this essay collaborate, we have had a good measure of success connecting third parties.  The senior author signed up for Skype Premium, which allows multiple persons to converse live, as in a telephone conference call.  Thus we have at times invited Allan Hayton or Crystal Frank, our principal transcribers and two of the youngest speakers of Gwich’in, to join in and help us translate and edit texts.  Whenever we do this we effectively create an informal tutoring environment, a mobile school (see Figure 1 below).

Substance

Through the Caribou Anatomy Project collaboration, Allan has been able to deepen his understanding of Gwich'in language. Transcribing interviews with elders takes a great deal of close listening, rewinding, listening again, and careful typing to arrive at the correct representation of their stories. The process of then translating those stories into English is another step in the process that requires a nuanced understanding of both Gwich'in and English. Gwich'in concepts and ideas are not always directly translatable into English. The following passage, taken from a description of traditional caribou hunting methods recorded from Venetie elder Macarthur Tritt Sr., illustrates exactly what transpires in the editing process.  Allan had a challenge following Macarthur, who speaks very quickly and often elides his words.  Allan’s raw draft transcript and translation are here shown in italics, followed by Kenneth and Craig’s edited transcript and translation in standard font.  The words here in brackets are understood:

Aiintł'ee hee khan, dinjyaa! Yee’ee k'iinaa zhyaa ch'iji' daa choh zhyaa ah'al, khaiints'an choo kwaii. Viji' choo kwaii [dinjyaa tsii?] gwiinzii yiich'ya' t'ihnyaa! Aii shitsii Elijah t'ee zheh gwadhah kwantee dhidii ts'a' yaha' k'iinaa ah'al it'ee zhyaa 30-40 haa vadzaih choo nilii, yaha' zhat reh. Yaha' zhat k'ii'an ah'al. Izhik geh'an jidii t'ii'in łyaa vik'yaljik.

Aiintł'ee hee khan dinjyaahtsii! Yaa k'iinaa zhyaa ch'iji' daa choh zhyaa ah'al, khaiints'an choo kwaii. Viji' choo gwanlii, dinjyaahtsii!  Gwiinzii vi’yiich'ya' t'aihnyaa! Aii shitsii Elijah t'ee zheh gwadhah gokwantee dhidii ts’ą’ yagha' k'iinaa ah'al it'ee zhyaa .30-.40 haa vadzaih choo nilii, yagha’ zhat reh. Yagha’ zhat k'ii'an ah'al. Izhik geenjit t’ee t'ii'in łyaa vik'yaaljik.

And then quickly, (exclamation)! Coming this way at us were some big bloody horns, the velvet hanging off the red horns. Big horns, [dinjyaahtsii!], they looked really good! My grandfather Elijah was sitting inside the tent, and there were maybe 30-40 big caribou making their way towards our tent. They were coming straight at us. That is how I caught on to what my grandfather was doing with those stick people.

And then quickly, wow!  Coming this way right at us were some big red bloody antlers,  a lot of bull caribou just before the rut. They had big antlers, wow!  They looked really good! My grandfather Elijah was sitting outside the tent, and he was just knocking them down with his .30-.40 [calibre rifle] right there. They were coming right by us. That’s how I caught on to what my grandfather was doing [with those stick and moss scarecrows].

--------------------------------------------------------------------------------------------------------------------

We all chuckled when Allan misunderstood the English number “30-40” for the number of caribou rather than the calibre of the antique rifle used to kill them. This passage clearly requires skill and experience beyond mere fluency, since Macarthur also makes use of the poetic devices metonymy and synecdoche when he says, “Yaa k'iinaa zhyaa ch'iji' daa choh zhyaa ah'al. . .” (‘Coming right at us were some big bloody red antlers. . .’).  This sentence could easily be taken literally rather than figuratively.  However, if one gets bogged down in translating word for word, it can also make the English sound rather clunky.  The translator has to decide if it is important to gloss a sentence down to the morpheme level, or to grasp the overall meaning of what is being said.

The mentorship of working with Kenneth on such translations as this helped Allan discover where he held misunderstandings and also helped identify gaps in his language abilities.  For example, the Gwich'in word for the aurora borealis is yakaih or zheekaih (‘sky lights’), which Allan had always heard as zhee k'aii (‘sky willows’). The slight difference of the glottal stop on the “k” changes the meaning of the term. Allan also had a general understanding of caribou anatomical terms, ch'atth'an ‘leg’, ch'iki'  ‘head’, etc., but did not have a knowledge of the level of detail for body parts this research project has documented. In addition, Allan has also noted idiolectical differences in individual speakers, slight variations in speech unique to geographical areas or even family groups.

Mechanics

While we still use Skype for voice communication, Google Drive is now much preferred for visual screen sharing and editing since it is more dynamic and interactive.  We can even see each other’s cursors. These two programs, Skype and Google Drive, can be used in tandem to great advantage.  As part of our preparation, we now upload Word transcript files into Google Drive and open them with Google Documents for editing purposes.  When we finish a session, Google Drive allows us to download and backup newly edited files into one or more common formats: such as a Word (.docx) file, an Open Document file (.odt), a rich text file (.rtf), a plain text file (.txt), or a Web Page (.html zipped).  Google Drive nevertheless has one disadvantage compared to sharing documents on Skype, which is that line numbers of texts formatted into Word disappear as soon as they are uploaded and imported into Google Drive. 

To translate Gwich’in texts in Google Drive, the only practical option then is to double space and insert interlinear English translations underneath the Gwich’in originals.  When exporting and downloading jointly edited files back into Word and onto our hard drives, some additional housekeeping duties are required: namely, the interlinear English needs to be stripped out  and pasted onto a fresh page or pages or into an entirely new document, so that the Gwich’in text is continuous and not broken up.  Our preferred formatting is to keep all of the Gwich’in together and all of the English together in separate blocks within a single document.  This re-arrangement produces what are for us more aesthetically pleasing reader-friendly texts.  It also preserves the integrity and flow of both languages. 

Conclusion

Whenever there is a need for close collaborative decision-making, the process we have described here (summarized in Figure 2 below) offers great opportunities to meet in a common workplace, a virtual office.  Each partner in the collaboration brings a unique set of creative and critical skills to the table, so that we all learn from one another.  That learning has taken place even in the writing of this paper.  Again, it is not just Skype that makes this happen.  It is the conjunction and combination of Skype, Dropbox, Express Scribe, Word, and Google Drive that lead us forward.  Much like electricians, carpenters, and plumbers, we applied linguists require a complete set of tools to get our hands-on work done.

While the production of primary scholarly texts will always be our primary goal, our secondary goals, achieved through praxis, are to work efficiently on an equal footing and to teach the nuances of the Gwich’in language to a new literate generation.  The suite of software technologies we have adopted enable us to collaborate and support one another over great distances in ways inconceivable only a few years ago.  We recognize that these technologies may quickly go out of date, but at least for now, they work. Ultimately, we feel that we are collaborating not only with one another but with the elders and ancestors who passed on these stories.

References Cited

Leonard, Wesley Y. and Erin Haynes. 2010. Making “Collaboration” Collaborative: An examination of perspectives that frame linguistic field research.  Language Documentation & Conservation 4:268-293.  http://scholarspace.manoa.hawaii.edu/handle/10125/4482

Mishler, Craig, ed. 2005.  Neerihiinjìk: We Traveled from Place to Place: the Gwich’in Stories of Johnny and Sarah Frank.  2nd ed. Fairbanks: Alaska Native Language Center, University of Alaska.  Includes audio CD.

Schulist, Sarah.  2013. Collaborating on Language: Contrasting the Theory and Practice of Collaboration in Linguistics and Anthropology. Collaborative Anthropologies 6:1-29.

Yamada, Raquel-Maria. 2007.  Collaborative Linguistic Fieldwork: Practical Application of the Empowerment Model.  Language Documentation & Conservation 1(2):257-282. http://scholarspace.manoa.hawaii.edu/handle/10125/1717