In May 2019, Eleanor Ridge and I sat down and had a conversation about the role technology plays when conducting linguistic fieldwork. Eleanor talks about using apps like SayMore for collecting metadata, language anotation tools like ELAN, and FLEx, as well as the importance of transcription and innovations in the field, and how connectivity (internet and phone) affects one’s ability to do fieldwork in remote island locations. This conversation is contextualised within her experience of conducting linguistic research on the Vatlongos language of Southeast Ambrym Island in Vanuatu. Eleanor is currently based at Massey University, University of New Zealand. Some of her recordings are available through her Language Landscape project page ‘Vatlongos SEA‘ You can follow her on twitter: @eleanor_ridge
ED: I am talking to Eleanor Ridge.
ER: That’s me.
ED: Hi Eleanor.
ER: Hi Ebany.
ED: Could you tell me a little bit about the type of field work you have conducted and where you have done your fieldwork?
ER: Yes, I certainly can. So, my main bit of fieldwork has been for my PhD and I’ve been working on Vatlongos language which is also known as South East Ambrym in the literature, and it is spoken in Southeast Ambrym island in Vanuatu. It’s an Oceanic language and it’s closely related to Paamese, another language on the island opposite. My project has been documentation and description, but particularly paying attention to variation between different speaker communities because it is spoken both in South East Ambrym and in a community called Mele Maat, which is just outside the capital of Port Villa. They relocated there in the 1950s following a volcanic explosion. So I’m looking at the differences between how language is spoken in those two communities.
ED: When was the last time you went and did fieldwork?
ER: The very last time I was there was in August last year (2018) because I was in Vanuatu for a conference and I managed to squeeze in a couple of weeks on the island and a week in Mele Maat which was really nice. I was finishing up my thesis at the point so it was a good opportunity to ask some final questions and get a couple more stories as well. Before that I did two chunks of six months in 2014-15 and 2016-17.
ED: So you started visiting them in 2014?
ER: Yeah, that was the first time I went to Ambrym but I had worked in Vanuatu before which is what originally got me into linguistics, because I was teaching on Pentecost Island and learned Apma language, the language of Central Pentecost at that point. That was when I decided that I really like languages and that they’re really fun, and that I wanted to come back and study that for my Masters.
ED: So what does fieldwork look like for you when you’re on site?
ER: Because I’m working on two different sites, it looks very different when I’m on the island compared to when I’m working with the urban community. So when I’m on the island it’s maybe more what people expect when you say ‘fieldwork’. I’m staying with a host family who are wonderful. Well two host families. I live with Madleen and Simeon in Moru and their kids, and they’re brilliant. They also do transcription work and organize sessions for me. They provide lots of their expertise as well as their hospitality. There what normally happens is that Simeon organizes for me to visit different villages, so I got recordings in every village that Vatlongos is spoken in in South East Ambrym. Simeon was organizing that for me, and arranging things in advance. He made sure that announcements were made in advance. A typical day would be, get up, have breakfast, set off to do some recordings, arrive in the village, make sure everyone knows we’re there, they come and share their stories. Certainly, that’s what it was like in the first fieldtrip, and increasingly I was also asking more guided elicitation questions about things that I wanted to see if they varied in different villages. Then we would go back home and I would upload my recordings in the evenings. The evening would also be used for some transcription work. I’m quite pleased about how the transcription worked.
ER: Because we were moving between two communities I wanted to make sure we had independent transcription so that I didn’t need to be there in order for transcription to be taking place. Because it just wouldn’t be very efficient. So what I did was use the auto-segment feature in SayMore to automatically segment, which saves a lot of time. If you’ve ever used ELAN to segment manually you’ll know that. Then I use that to generate mp3s that Madleen and Simeon had on their mobile phones. They could then listen to each track, each track had a code and a number, and that meant that they were writing it up down in their books and I could type it up so that it was all time-aligned. So that’s often how we spent the evenings which is fun, and we would have lots of discussions about things that come up.
ED: Yeah, sounds fun, and that reminds me that you were really good about uploading things to Language Landscape for one, but also keeping really on top of technologies and social media while you were in the field.
ER: Yes, I was good at that wasn’t I?
ED: You were excellent.
ER: Especially during my second fieldtrip and after that. I wanted to do some awareness raising. I’m lucky and it’s a thing about a lot of communities in Vanuatu are very open, certainly in the Vatlongos community, people are just really keen to share their stories. Often when you’re doing fieldwork you have to think very carefully about anonymity. Sometimes there are restrictions on who is allowed to hear certain stories and that sort of thing. In the community I’m in, people really want their stories to be shared and they want their authorship to be acknowledged while doing that. I put a lot of videos up on YouTube with people’s permission and people were really keen for that. It worked really well for getting some enthusiasm for the project in the urban Mele Maat community where people have more internet access. Especially with young people in Mele Maat who are less likely to speak the language. This was a way of getting them really engaged with the project so that worked really well. Increasingly I’m aiming to archive and I’m hoping that will be the means to spread it from now on, but that’s been a fairly slow process. It’s increasingly happening. I have a project up with Pangloss and I’m working with ELAR to get stuff up on ELAR as well as the moment. Sharing stuff on YouTube is very effective though and I was able to produce subtitles very quickly because of that transcription process.
ED: So you would say that definitely using the more up-to-date software like SayMore streamlined the whole process more?
ER: Definitely, it was just really time saving. It meant I could produce subtitles really easily. I could have subtitles that were both the language and English, or the language and Bislama. We were also using Bislama, the national language, for transcription as well. Especially for semi-speakers in Mele Maat, it’s really useful to show the worth of the language but also make it accessible for people who maybe don’t understand straight away.
ED: Yeah, so just a recap on the languages…you have Bislama…
ER: Yeah, Bislama is the national language. It’s a pidgin English. Nearly everyone in Vanuatu speaks Bislama. That’s a really good way to get recordings accessible to the entire community and also semi-speakers of other languages of Vanuatu who would be interested in the stories. Maybe family and friends as well. Then you have English and French which are both languages of education in Vanuatu but in the Vatlongos speaker community, nearly everyone speaks English as their language of education. Practically everyone.
ED: And when you’re there what languages are you using?
ER: Because I already spoke Bislama, I spoke Bislama as the language of communication when I first arrived. Increasingly I speak Vatlongos. On the second trip and last August I was mostly speaking Vatlongos with the people who wanted to. Sometimes in Mele Maat though, if people are not confident in speaking Vatlongos you can sometimes put them on the spot. So it’s good to speak Bislama anyway.
ED: So for the people who are quite comfortable speaking Vatlongos, does you speaking it change your interaction at all?
ER: Generally people are really really enthusiastic if I’m speaking Vatlongos, but there’s also that feeling about Bislama as well. I think if people are used to dealing with outsiders, especially white outsiders, they’ll often start in English and that can be kind of nerve racking for people and also they really appreciate that you’ve made the effort to learn the language of Vanuatu. Even more so when it’s the local language.
ED: And what about meeting people? You said people are quite keen to share their stories, is it easy to make friends and approach people?
ER: Yes, definitely. Vanuatu is a really great place for that. Everyone wants to know what you’re up to. Everyone is really keen to talk about what they’re doing. It’s really easy to meet people as long as you’ve got a language which you can have those conversations in. People are really keen to discuss those things, and there’s a lot of pride in the culture. A lot of pride in local stories, local crafts and those types of things. It’s great, it’s very easy for me. I don’t have to do much persuading or too much work to show why the language is worth recording. People absolutely believe it already. There’s never any kind of queries of “But why are you interested in that?” The worth of the language is really self-evident to people which is great.
ED: That is really great.
ER: I know!
ED: Yeah, that’s not always the case.
ER: Exactly, and I so feel for people who have to do more work than I do on that.
ED: But you just have to run with it. That’s why you can do such great documentation.
ER: It does mean that the whole process has been much quicker and I have pretty sizeable corpus for the amount of time that I’ve been spending on it for that reason. Also because of how amazing Madleen, Simeon and Bell have been on transcription.
ED: So going back to the technology theme, obviously tech is constantly evolving. What are some things that could come up in the future that you would want to work with to improve the fieldwork process?
ER: I’m really interested in some of the projects that are coming up where people are working with automatic transcription based on a test area. I want to speak to some of the projects that are working on that like ELPIS and PERSEPHONE. This is a process that is getting more and more efficient all the time and there are a few groups that have different ways of implementing it. Basically, you have to do transcription and transcribe these languages up to a certain point, and then with a certain number of hours of recording you would then be able to train software for that to happen automatically. Then you would be able to record. It’s never going to be perfect, but you can use that as a base and then tidy it up. So then if it’s getting, I think some of the figures I’ve heard is 80% right, and then you can just go through and tidy them up. That would be hugely time saving. Interesting in itself, I think it would help you spot things that you might not spot otherwise, which would be cool.
ED: There’s a guy in Germany who has been working on the software to do that with Uralic languages.
ER: Yeah and there are some people in Australia who have been working on it. There are lots of people working on that and I think it would be amazing. It would mean that any documentation project would immediately have transcripts. It would make those first 8 hours of transcription even more important and you could then expand and be able to access all of these other things that people produce themselves. You would be able to access data more easily. It would no longer be the case that you would have to make do with quite small corpora either. You could end up with big corpora and you could ask the types of questions we ask of larger languages which I think would be great.
ED: Yeah, and obviously time is a critical issue for these smaller languages. Is the technology in any way accessible to speakers? Or how accessible is it?
ER: You mean the technology I’m using now? SayMore?
ER: At the moment this will vary depending on where people are working and the access they have to computing and power. Many people have laptops. Some of these programs are quite power and memory hungry though so that doesn’t always work that well. ELAN is not very easy to access. There is a very steep learning curve with that. Whereas SayMore is much easier I think. I did take a laptop out with me that people are using for other things. It wasn’t really powerful enough to use SayMore and it wasn’t immediately obvious to people what it was useful for, so they weren’t that keen to learn. That’s why we kept to pen and paper most of the time for transcription. I can see that there would be situations where people would want to do that for themselves. Especially for subtitles and things. That would be clear what the use is in that instance. Mostly if you have some really enthusiastic people who are interested in the linguistics of the project and how that works, I think these are really amazing tools for teaching that. Things like FLeX like I’ve been using, which is a lexical database software, is brilliant for teaching those types of things. Unfortunately, I haven’t had anyone on the project who has wanted to take it to that level, but I would be really keen to take that on if anyone was keen. What it does instead is make it really easy to generate outputs that are accessible to the community. One thing that I’ve had success with which I was sharing when I was out in August was a dictionary app that you can generate automatically from FLeX, and that was something that when I was on the previous fieldtrip, a lot of teachers had been asking for because we had been trying to implement vernacular education for years 1-3. We now have the materials which is fantastic, but a lot of the teachers are not very confident with the spelling system, and they’re not very confident with their own reading and writing in Vatlongos. That was something they were asking for, a dictionary and in particular anything on a computer or a mobile phone. Many more people have mobile phones than have laptops, so anything on a mobile app is much more useful. That’s been really well received and is something I want to work on further to develop proper definitions and add photos and more of that sort of thing.
ED: So there are things like electricity and basic commodities?
ER: It’s all via solar power so you do have electricity. It’s improved drastically in the few years that I’ve been going out there. The first time I was out there were two stores that I knew of where I could take my laptop and get it charged, maybe in a few days. At the time I wasn’t able to upload everything straight away from the recording. Increasingly individual families have solar panels and car batteries for charging things. That’s also something that because my host family refused to accept any rent, I’ve been trying to give gifts of suitable value to show my gratitude. That is what I did on the second trip. I managed to get solar panels that were cheap enough that it was within the right bracket. Now Madleen and Simeon have solar panels and that’s worked really well and has made it much easier for me. It’s really useful for them as well.
ED: What about things like internet and communication?
ER: That is also something that has improved vastly since I first went there. The first time I went there was no internet access at all. Mobile phones worked which was good, but you would have patchy reception. They built a new communication tower recently between my two fieldsites which has been really good. During the second fieldtrip I could get mild 3G network which was really awesome. Especially because my dad got married and I was able to receive some photos of his wedding which was great.
ED: That’s lovely!
ER: Yeah! It makes a big difference to how connected you feel.
ED: Which is important when you’re doing fieldwork.
ER: Yeah, definitely, definitely. Because I’m moving between two communities, I haven’t found that too challenging because I’m not really on the island for more than 2-3 weeks. I’m on the island and then I’m back in town and the cycle repeats so I’m never that isolated. When I’m in Mele Maat, I stay in town in a guest house. Actually, last time I was out I was able to stay with a friend who was renting a house just by Mele Maat which was incredible good fortune. It was a bit harder to do some of the spontaneous recordings that I do when I’m staying on the island. Also, in Mele Maat lots of people have 9-5 jobs which makes it harder to pin people down. I found that I had to spend twice as much time in Mele Maat or Port Vila compared to the island and I still had far fewer recordings there than I had on the island. It’s just that it’s harder to make progress. It took a bit more effort and planning to get stuff done there.
ED: Yeah, it’s not about the amount of time you spend in a certain place, and more about the connections you make with people.
ER: Yeah exactly. There’s also limitation on things like, I can’t travel at night. I would need to have a car to be able to do that confidently because the buses are a bit dangerous at night. So that’s a big restriction because I know that I need to leave at about 4 o’clock in the afternoon. I can’t spontaneously follow up on those connections so building those connections takes longer on Mele Maat than on the island as well.
ED: I think we’re going to wrap things up now. Is there anything you else you would like to share along the same theme of technology and communication?
ER: I think it’s been really interesting to see how those technologies have been embraced, or not, and be able to see it within two communities at the same time. One that has had internet access for quite a while now and where Facebook is hugely popular now, and another where that’s just starting. It’s been good to see how that works with people and how it shapes the way people use language in terms of the language they choose to do Facebook status updates in Bislama or English, but then maybe the comments will be in smaller languages. It’s really fascinating. Also seeing how those platforms can serve as opportunities to use and support more language groups has been very interesting.
ED: Have you seen any age-related patterns within social media platforms?
ER: Yeah, I know lots of older people who are using facebook and who really love it, but it’s particularly popular amongst younger people in the urban areas. You also see a lot of people talking about how it’s really popular among young people in urban areas. There are plenty of joke posts and memes about people staying up late.
ED: Ah, they’re creating memes as well?
ER: Yeah yeah, there are Vanuatu memes. I’ve even seen things about specific people. The other thing that is great which wasn’t a thing when I first started, is that I now get updates when I’m not in the field about how people are doing and what is happening. The only time that I’ve seen Vatlongos used is after people have died, so I get to hear about those things, which is upsetting but also helps me feel more part of the community. People will do pictures of people with kin terms on them in Vatlongos, and certainly Bislama is used everywhere.
ED: Yea, it’s good that you’re able to feel connected even at a distance, just like we are able to do this interview at a distance! Thank you so much for taking the time to chat about your fieldwork experience. It’s been a real treat.
ER: Anytime! It’s been great to talk about my experiences!
Data Elicitation – A data collection technique used in the social sciences. In linguistics, the practice of ‘eliciting’ information or data, involves asking questions or using stimuli to collect linguistic data.
Interlinearization – The process of providing a morpheme by morpheme translation, or glossing, of a word or utterance.