University of California, Santa Barbara
I am a PhD student in Linguistics. Early on in my program, I volunteered to assist at a workshop held at a university in Papua New Guinea that was supposed to teach the local participants how to document their own languages. There I witnessed what I considered to be at best misguided and at worst unethical behavior on the part of the foreign organizers of the workshop vis-à-vis the Papua New Guinean attendees, most of whom were poor villagers. There were many problems with the way in which the workshop was managed. It was conducted in English rather than the language all the participants spoke, Tok Pisin. Instead of making audio recordings, the standard practice in language documentation, participants were directed to produce written texts. This is because the organizers sought to amass a corpus of texts in the participants’ languages that would be useful for machine translation research, and it was claimed this would help document the participants’ languages by automatizing parts of the documentation process. Those who did not speak English were at a disadvantage, and those who were illiterate could hardly participate at all. The participants spent two weeks unwittingly producing data for the organizers’ research in computational linguistics, with no real benefit to themselves. In my conversations with participants as well as faculty at the university, there was a great deal of discontent with the workshop for these and other reasons.
When an article extolling the success of the workshop appeared in the Journal of Language Documentation and Conservation a year later, I was appalled. There was a great chasm between what I had witnessed and how the article represented the workshop. The article described it as an outstanding success that contributed to language preservation and was a real hit with the participants. To my mind, this was dishonest and unethical.
I struggled to decide whether and how to respond. The prospect of a conflict with a senior researcher worried me; nor did I want to earn a reputation for being a troublemaker. I also felt uncomfortable speaking for others. However, I did have a strong sense of how negatively the participants I had interacted with experienced the workshop, and some had even expressed their discontent to me directly. It was my correspondence with a faculty member at the Papua New Guinean university that finally convinced me that I should speak up. When I expressed my moral concerns about speaking on behalf of the villagers I was told that not only was I in the perfect position to do so, but I was in fact the only one who possibly could. The local Papua New Guinean faculty had given up and ceased attending after the first few days of the workshop, whereas I had been there the whole time and was informed enough to comment. The power difference between universities in PNG and those in the United States and Australia surely also played a role. Local faculty were not well positioned to criticize outside researchers who were bringing prestige and funding to their university.
It seemed there were two main avenues I could pursue. One would be to lodge a complaint of ethical misconduct with the sponsors of the research, the funders and/or the PI’s own university. The other would be to submit a response piece to the journal where the article had been published. I sought advice from my friends and family. I consulted with representatives of multiple IRBs. I even met with someone in my town who has a degree in ethics to get another perspective. I discussed the issue with my adviser and with another professor in my department. There was no unanimous support for any course of action. Some strongly supported my intention to respond in some way, while others urged extreme caution lest I irreparably damage my nascent career.
I finally decided to submit a response to the journal. The response genre is one that is established and respected in academia. Whereas an ethics misconduct claim would be behind the scenes and might in the end have no effect, a published response would bring the story to the awareness of the readership of the journal. Submitting a manuscript to review would mean that, if published, my perspective had been vetted by a panel of respected scholars in my field. And submitting a response would require me to forge my concerns into something that was at once scholarly and productive. My response could help positively shape future research, and might even raise the ethical bar for linguists working with marginalized communities. However, I would have to focus my argument and leave out some details I felt were critical. I would have to remove anything that came too close to conjecture, anything I did not directly perceive, or anything that came too close to ad hominem criticism. I would not be able to tell the full story.
In the end, I submitted my response, and it was published. So far, I have received only positive feedback from colleagues. In retrospect, it seems obvious that I should have done exactly what I did, but there were many points along the way where I considered different paths or where I nearly chose inaction over action. I hope that in writing my response as well as this blog post that I have provided the beginnings of a roadmap for other young or unestablished scholars who find themselves in similar difficult positions and are afraid to speak up.
I would like to conclude with a few thoughts about the broader context in which the workshop occurred, and what I think it means for research ethics in interdisciplinary fieldwork. Within linguistics, the last few decades have seen an increased emphasis on the need to document and describe endangered languages because they have a great deal to teach us about things like the range of possible linguistic structures and what happens when languages come into contact. But one effect of this has been to bring endangered languages to the attention of linguists and others who work outside the area of traditional field-based language description, and who see a way to connect themselves to this “hot topic”. The workshop I assisted with in Papua New Guinea was born out of this trend: computational linguists wanted to see if their methods might be used to help speed up the task of documenting endangered languages. This is a reasonable question to want to ask, but in my assessment, the workshop it resulted in was ethically questionable because the organizers were taking an experiment-based approach to research and had little awareness of research as a social process. Those who have been trained in another discipline’s ethics run the risk of engaging in extractive if not exploitative practices when they bring those ethics to local, often marginal, communities. Funders and collaborators have a responsibility to help educate such researchers, so more response articles like mine will not need to be written.