Best and worst practice in language documentation: LIP discussion

Ruth Singer recaps some of the interesting points of last night’s Linguistics in the Pub, an informal gathering of linguists and language activists that is held monthly in Melbourne

The announcement for this month’s Linguistics in the Pub outlined the topic as follows:

“There is much discussion of best practice in language documentation but as we all know, no language documentation project is perfect: each is the result of collaboration between researchers and a community with restrictions on time, money and many unforeseen circumstances. There is always a gap between what we achieve and the most wonderful project of our dreams.

Come and tell us about your experiences. What aspects of your language documentation work are you most proud of? What will you do differently next time? And what are some of the great things you have planned that you just couldn’t get off the ground?”

The idea behind this discussion topic is that language documentation projects tend to aim high and this can result in those leading language documentation projects feeling disappointed. Spurred on by hearing about innovative projects, egged on by others in the language documentation field to follow best practice in an increasingly multiplying number of areas we sit at our computer concocting new language documentation projects that will create years of recordings, miles of transcripts and beautiful metadata as well as lovely outputs that suit the needs of the language speaker community. In the process we will develop wondrous collaborations with language speakers supporting them to develop the capacity to carry out language documentation work themselves and also meaningful collaborations with other academics such as musicologists, anthropologists and ethnobiologists.


While acknowledging the importance of continuing to be inspired by the work of others, this session, as Nick Thieberger predicted, focused on discussing problems we had had in our projects (the initial idea was to discuss achievements and great ideas for the future too). But as Nick pointed out, language documentation projects are rarely complete failures as anything that improves on the resources we have on a language is still an improvement. The rare project that flops completely is usually the result of a breakdown in relations between the linguist and the language speaker community. It was those relations, in particular realising the aim of capacity building that we discussed as well as the importance of sessions that go over ‘what went wrong’ in our field.

We discussed the various ways we had tried to build the capacity for language documentation in language speaker communities we work with and the problems we face. Certain parts of language documentation were much easier for speakers to take on. However there were a couple of overarching problems. One was the lack of any framework that we as linguists have of capacity building and the broader practice of facilitating community development. The other problem is that as linguists we view capacity building as a way for language speakers to carry out aims that we and our supporting institutions value, under our direction. However, speakers are likely to prefer to be in charge of the project and place greater value on making recordings and publishing them, than compiling metadata and creating transcriptions. This can create problems for linguists who are accountable to funding bodies and universities. Another problem is the difficulty of attracting language speakers to the project who have the necessary skills such as literacy and technological skills. Those who have the skills are likely to look for more long-term employment, which language documentation projects cannot usually provide.

The lack of comparability of our fieldsites can make it difficult for linguists to take part in constructive discussions of ‘what went wrong’. However there may be much value in the discussion of ‘failures’. Amos Teo mentioned that hospitals regularly have meetings to go over cases that did not receive the best treatment and what can be done to avoid the same situation recurring. He also pointed to the annual Failure Report ( http://legacy.ewb.ca/en/whoweare/accountable/failure.html) released by Engineers without Borders. They explain their reasons for releasing their reports below:

“EWB believes that success in development is not possible without taking risks and innovating – which inevitably means failing sometimes. We also believe that it’s important to publicly celebrate these failures, which allows us to share the lessons more broadly and create a culture that encourages creativity and calculated risk taking. This is a culture we value within EWB, and also try to work with our partners in Africa to create in their organizations.”

They then introduce the case studies by saying “We are excited to be able to share these Failure Reports from our program staff and volunteers.”. The annual Failure Reports of have been introduced by none other than Bill Gates and seem to be the inspiration for an incipient global failure movement (http://www.admittingfailure.com/). Perhaps Language Documentation as a discipline can learn from this movement? We may have unwittingly retained some of the macho spirit of early language documenters by spending so much more time reporting successes and so much less discussing failures. Regular discussions of ‘what went wrong’ can do much more than help others fall into the same trap. They can also bring people together who face similar problems and also provide a place for linguists to reassure one another that their project is still making significant achievements, even not every single thing they had planned has worked out.

One Comment

  1. James C. says:

    Retrospective field stories are a good way to communicate failures. The anthropology fieldwork literature is replete with stories of individual and group failures in communicating, hypothesizing, researching, and just getting along in other cultures. I learned far more about the basic issues in fieldwork from reading anthropological literature than I have ever learned from the literature on linguistic fieldwork. This is at least partly because descriptions of linguistic fieldwork are largely focused on either the technical details in gathering data (elicitation, recording, transcription, analysis) or the process of linguistic discovery. They have little to say about actually interacting with people as part of the linguistic fieldwork endeavour. I earnestly believe that linguists who have had extensive fieldwork experience have an obligation to the field to write about their experiences in an informal, confidential manner. This will provide literature for aspiring fieldworkers that can inspire them and teach them to consider repercussions of planning and assumptions. It will also help reduce unrealistic expectations and will help banish myths about fieldwork.

    The tradition in linguistic fieldwork so far has been to leave the interpersonal aspects of fieldwork to the field’s oral tradition. The advisor is supposed to chat with the student about all the various unpleasant things that could happen, sending the student out prepared to face fieldwork with a pragmatic, practical outlook. But relying on oral tradition to communicate this is disadvantaging people coming from situations where there is no extensive experience in fieldwork. It effectively denies ‘outsiders’ who haven’t received training in fieldwork but who might nonetheless make significant contributions, and serves as a filter between those who ‘know how’ and those who don’t. This filter is arbitrary, there’s no need for it other than to perpetuate the privilege of membership. This problem is compounded by the exclusionary attitude among fieldworkers, exemplified by the “it’s not real fieldwork unless you’ve had malaria” trope, among others. If we instead move from an oral tradition to a literary tradition about fieldwork then we invite people who have no academic lineage in it to participate. This will only enrich our situation, not diminish it.

Leave a Reply