On technology training in the speaker community – Andrea L. Berez

[From Andrea L. Berez, University of California, Santa Barbara]
A few weeks ago in Uppsala, Nick Thieberger and I gave a talk on the need for digital standards and training in language documentation. During the Q-and-A, a distinguished member of the audience asked us, “How do you suggest we go about making communities do all the things you’ve been talking about?”
He was referring to the examples I had just been discussing regarding Alaska, where local (i.e., non-university) efforts at documentation and archiving are underway in several villages. He wanted to know how we, as linguists, can convince speaker community members to take up arms in the race for documentation and revitalization. After a moment’s consideration, I could only reply, “I’ve never had any luck making the community do anything.”
Although it may have seemed like a flippant response at the time, it was also a true one: any time I have been involved in proactively bringing technological standards and digital language-related activities to Alaskan communities, the result has always been different from what I expected.
Don’t get me wrong. I’m a huge proponent of local training in language technology, and I’ve actively participated some of that training myself. I believe knowledge is power, and I don’t subscribe to the notion that technology is somehow harmful or hegemonic. I consider it my responsibility to pass on my technical skills to anyone who wants to know. What I’m saying is that in my experience, attempts to introduce academic ideals about the “proper” way to do language documentation into the speaker community when nobody’s asked for it has led to frustration.

Take, for example, the Qenaga.org website, a US NSF-funded project to put Dena’ina language on the web. It’s an extensive website (check it out!) and includes a metadata catalog of all the Dena’ina language materials at the Alaska Native Language Center in Fairbanks. After spending two years constructing the site, we (the university-based project developers) assumed the Dena’ina people would be so excited by it that some eager young advocate would step up and take over maintenance and expansion of the site. Not so. While Qenaga.org gets frequent hits from the Dena’ina region, four years after the start of the project the website still resides on a university server and is maintained by a university staff. In an unexpected twist to the plans of us academics, nobody from the Dena’ina community volunteered. This is not a criticism–often life is simply too busy to administer a website.
As another example, I took part in a program to teach Dena’ina language workers to create time-aligned texts for dissemination on the web and on CD. Our lofty training goals were about long-term sustainability, interoperability and open-source software. Our students, however, were more interested in the display of the texts and wanted to know how to customize fonts and colors. Again, no criticism is intended: language products that are attractive and user-friendly are important too.
In both of these cases, we approached community technology training from the standpoint of digital best practices for long-term data preservation. My goal as a documentary linguist, like that of so many others, is to create an enduring record of language data that’s in danger of fading away. Revitalization, language pedagogy, language policy–those are, to my mind, community issues, and it’s not my place to step in and “make” a community undertake them. In the words of a friend, “Let linguists do what linguists do best.” Community members may have other goals, language-related or otherwise.
But linguists and communities are not necessarily forever destined to follow divergent paths when it comes to language technology. In Southcentral and Interior Alaska there has been a recent groundswell in several villages to to do something meaningful with the piles of old cassette tapes of Athabaskan language that people have in their attics and offices. There is a growing awareness that the contents of the tapes are precious. People are not satisfied to simply hand the tapes over to the Alaska Native Language Center for archiving anymore, preferring instead to make the contents available locally and for the long term.
Here’s what’s been going on. In one Alaskan village (let’s call it Village A), the teachers at the local school have undertaken to digitize some 350 hours of cassette recordings of language. They’ve done a remarkable job. I myself might have proceeded a little differently, digitizing everything as .wav and storing them on hard drive until such time that I could deposit them with a proper archive who would look after the collection in the long term. But that’s because I’m a linguist, and I have my own goals. Instead, the group digitized everything as mp3s, and are storing them (for now) only on a pile of CDs (which, to their credit, are held in a locked fire-proof safe). This does not reflect a mistake or ignorance on their part, but instead reflects a different set of goals: this group’s main goal is local dissemination. They want to be able to burn CDs quickly, for play in any standard CD player, and making CDs for on-demand copying seemed the best way to do it.
Now the teachers in Village A are ready for an archive. They want to build a database of their CDs, searchable by speaker, topic, date, or location. That sounds like standard metadata! They’re also curious what the tribal council in Village B three hours up the road has in THEIR audio collection, and they’re willing to work out a trade agreement. Now this sounds like discoverable metadata delivered on the web, plus widespread dissemination.
And wouldn’t you know it, that tribal council in Village B–well, they are ALSO thinking about how to digitize and catalog their tape collection, which arose independently from any thoughts in Village A. And everyone in A and B assumes that people in Village C, up in the mountains, would also be interested in some kind of trade agreement for their collection.
So it seems that in Alaska there is an upsurge in interest in high-quality, locally-administered, interoperable language archives. Sounds familiar, like the goals of OLAC and DELAMAN. Some questions will need to be addressed: do the Alaskan archives have any long-term potential for data migration? Who will administer the archive when the funding runs out? These are issues that are at the forefront of my mind, because I’m a linguist. How the villages choose to answer these questions remains to be seen, but I doubt pressure from linguists to “make” the community answer them in a particular way will do much good.
I fully support the development of digital standards for our profession and the widespread education of our colleagues in following these standards, but I think the issue is much more subtle when it comes to community training. We can be responsive to the requests for information and offer our experience and our knowledge about digital standards. And of course as linguists we should still carefully archive our own data. But to my mind nobody’s going to have much success “making” anyone do anything. It may happen, as in Alaska, that a responsibility toward data sustainability in the academic sense will evolve naturally, and with it the desire to learn what we as a profession can offer. In technology and anything else, the community-linguist relationship is a partnership. We don’t have to have the same goals in order to work together.

Follow