We at FEL are deeply honoured to have our inaugural blog post written by Nancy C. Dorian, Emeritus Professor of Linguistics at Bryn Mawr College, USA, whose work has been formative in shaping endangered language research for decades,
and which serves as an inspiration for so many of us as we begin and/or continue our own work in endangered language communities.

In today’s research environment, there are some important requirements built into research proposals that are meant to safeguard the people whom the researcher is studying. For linguistic fieldwork, this means following certain ethical guidelines and offering evidence that the people whose language is under study agree to the presence of the researcher(s). Linguists who choose to go into the field to work on endangered languages sometimes make that choice with the welfare of the language and its speakers in mind. Certainly most would wish to do no damage, and some become activist advocates for the language they study. More and more often now, field researchers express a wish to do some service for the speaker population, for example by helping devise or improve an orthography; creating an archive; creating user-friendly written materials; or encouraging local speakers to preserve traditional material digitally, on film or in books. Since all of these positive developments seem to be growing more prevalent, why raise the issue of doing damage?

There are actually certain potential hazards built into the fieldwork undertaking, some more obvious, some less so. Deciding on which speakers become “sources” (if there are enough speakers to allow a choice, that is) sets up many opportunities for provoking rivalries of one sort or another. If we settle on one or two excellent speakers to work with, are we enhancing their status and leaving other fine speakers feeling diminished? Favoring only the most elderly, or the most traditional, or only speakers with 100% local parentage, or even – as I found in my own fieldwork– one spouse over the other (she was reliably local, he was subverted by Bible literacy), can cause ill-feeling. If there are already opposing camps within the speaker group, more serious factionalism may appear, with one group belittling information given by another or undermining any maintenance or revitalization efforts that the other group supports.

Dialect differences pose a related problem. Speakers in one location are often acutely aware – even hyperaware – of differences between elements of their own speech and corresponding elements in the speech of people in another location very nearby, whether in phonology, morphology, or syntax. Ideally, researchers would like to recognize clear-cut forms that represent local consensus. But they may instead have to decide to what extent they will acknowledge the linguistic untidiness that often surfaces.

Deciding to work in just one among a string of villages or hamlets with very closely related but locally distinct speech forms (from the local populations’ point of view, that is) may be a matter of practical necessity because of limited time or money. It may be advisable, too, because one location is more accessible than others or because the fieldworker has a certain entrée in one location but not the others (thanks for example to a dissertation adviser or colleague who knows the area and can provide some introductions). But dialect differences as a marker of local identity are so potent that any strategies that can reduce the sense of exclusion among other villages’ populations are worth adopting, to avoid the possibility that people from the “excluded” villages will resist orthographies, dictionaries, storybooks, or any other maintenance and revitalization initiatives that feel alien to them.

There is also the issue of the observer vs. the observed. The linguist or linguistic team comes to the fieldwork setting with a certain expertise that may or may not be recognized by the local speaker population, and once there asks people from a culture not their own to allow their speech (and usually some of its context as well) to be observed and recorded. Even if the request is formally granted, the uses to which specialists like linguists and anthropologists put the recorded materials are usually obscure to people who have no experience with scholarly literature. This very fundamental problem strikes at the heart of the “informed consent” issue, since it’s hardly possible to consider the consent “informed” if the outcomes are obscure to the people “consenting”. In East Sutherland, I found that simple unfamiliarity with the publication process led to an unforeseen problem. One of the sources who had agreed to be recorded for an oral history I was preparing asked for, and got, removal of one or two sensitive matters from a draft of the book’s manuscript. After publication, when it was no longer possible to remove anything from the book, she urgently wished to have two more matters removed.

Most fundamentally of all, linguistic fieldwork as a scholarly undertaking calls for a degree of detachment that is mostly absent in other protracted and intense social interactions. When local people have done months of close work with linguists or other fieldworkers, sharing their special knowledge and giving hours of their time, how fathomable can it be to them if the researcher(s) one day tell them that the time has come – from the scholars’ point of view, that is – to leave for good, possibly to take up the same sort of work in another place with speakers of a different language? Successful fieldwork relationships often have elements of friendship, and this is not the way friends treat friends. Especially if the fieldwork has been done with great intensity and apparently avid interest, perhaps by a team of researchers and nowadays almost invariably with a good deal of visibly expensive high-tech equipment, there’s likely to be a sense of shock and even betrayal on the part of speakers who have been reliable and cooperative sources and who consciously or unconsciously imagined a more ongoing connection.

This does not exhaust the possibilities for causing some sort of damage while doing well-intentioned linguistic fieldwork, but being aware of potential problems like these may help avoid them, or at least reduce the researcher’s surprise if they do crop up. Much of the advice and instruction available to prospective fieldworkers deals with how to get good information. But how to get good information without causing discomfort or inconvenience to our sources, and without causing social rifts within the speaker population, is worth special consideration.

What can we do to do better, then?

If the researcher is going to need a good deal of multisource information, it can help to make it known from the start that the kind of study being planned involves working with a number of sources. Forewarning the first source or two that the researcher will be interviewing other sources as well can avoid raising expectations that that those first sources are to be the only, or the favored, source of information.

We may be able to describe the speech form in question linguistically with only limited input from any larger group of speakers (dissertations have been written relying on one or two speakers who act as sources for a university class or project). But a grasp of how that speech form is used and the shapes it takes in a variety of different circumstances goes well beyond analyzing how it expresses the plural or the past tense (if it does). Recognizing that consensus is not always available in speech forms without a history of codification, and that individual renditions of what is offered in response to the same question or stimulus may vary systematically or unsystematically according to gender, or degree of acquaintance, or age, or style, may lengthen the time needed for the work but do the language (and other scholars) more service.

In the case of an endangered language, it can be especially important to learn how older people respond to any aberrations or simplifications that are in use among younger people. Finding out what differences the oldest and most fluent speakers take exception to, and how strongly, can suggest what the chances are that younger speakers will try to carry the language forward or will instead give up using it because older and more conservative speakers criticize the way they speak.

Researchers who aim to represent the language under study with the greatest degree of accuracy and depth have their best chance of achieving this if they make an effort to learn the local language, or so I would suggest based on one instance in which I learned to speak the local language and another in which I didn’t. Assuming that the researcher is received with something short of a deliberately distancing cautiousness or hostility, showing interest in acquiring the local language offers a degree of validation that nothing else quite matches. The researcher’s results may not be impressive in terms of reproducing native speaker models accurately, but as even tourists know, the effort is often appreciated. (A sense of humor helps, of course, for surviving local speakers’ enjoyment of our inevitable mistakes.) Whether or not the researcher is trying to learn the local language, fieldwork results can benefit a great deal from a researcher’s expressions of interest and appreciation as more of the local language’s structure and expressive potential comes to light.

The discipline and restraint needed to take the age and physical condition of potential sources into consideration can be hard for zealous fieldworkers to muster. Pressing for just a few more examples or just a little more explanation, a common temptation for all intense and eager fieldworkers, can have an element of bullying. If we’re lucky and are allowed to carry on working in spite of our failings, we usually find that older and less robust sources respond better to an interviewer who shows patience along with interest. Very elderly sources who are less exhausted are also more likely to be willing to agree to more sessions.

If they become well-wishers, linguists who describe often also advise, something that’s harder to do than linguistic analysis. In this connection we need to remember first and foremost that the language belongs to the community in which it’s spoken. As linguistic researchers we come, go, and to some extent participate, in so far as local speakers are willing to let us. But those local speakers (and often non-speaking but aspiring members of the ethnic group) are the ones who are in charge of the future their language will have, and the essential decisions that will impact their linguistic future are always theirs.