Raiders of the Lost Archives

There’s been some recent well-earned publicity for a language project at the State Library of NSW. Prof. Michael Walsh from Sydney Uni’s linguistics department has been in charge of the effort. The team have been working for two years now on identifying previously unknown or lost materials from Aboriginal languages in the State Library’s archives and contacting researchers and communities about them. Some of the books were misfiled, others were known about but it wasn’t clear that there were language records in them. For some Aboriginal languages, there’s lots written about them, but for others, only a few words were recorded, and so every new find makes a big difference. The Nawo language of South Australia, for example, is known from only about 10 words.

This is great. You know the feeling. You’ve got a favourite pair of socks, but one of them’s gone missing in your room. You’re pretty sure it’s there, but it’s not in any of the usual places. It’s not under the bed, it’s not on the floor in the cupboard. And then, one day when you’d given up looking for it, you find it rolled inside an old T-shirt. I’ve done work like this as part of my job researching the histories of Australian languages. At one point, I felt that archival research was more dangerous than fieldwork. Ever had a papercut from a manilla folder? Almost took my finger off. And the compactus shelving sometimes feels like that scene in Indiana Jones and the Temple of Doom where the walls are closing in, about to send him to a squashy death.

A few years ago I spent a month at AIATSIS working through some tapes. They had been recorded in the 1960s from languages all over northwest Australia, by Anthony Peile, who was a missionary at Balgo in WA. He’d done a reasonably good job at summarising which languages were on the tapes, by writing on the sticky labels, keeping notes about the languages and talking about the languages on the tape. However, the glue on the tape labels had dried out, so many the labels had fallen off and were sitting at the bottom of the box. At some point, the notes had become from the tapes, so while we had a good idea what was in the tape collection, the gems were mixed in with a lot of other material that probably wasn’t going to be very useful. (One of the tapes had German drinking songs on it…) Peile also had a small speech impediment and pronounced “r” like “w”. This wouldn’t be problem except that some Aboriginal language names differ only in whether they have an “r” or “w”! In Peile’s pronunciation, Jawi and Jaru sounded practically identical, and we knew he had recorded both. Jaru is pretty well documented and still has speakers, but Jawi has only a few people who know a bit about the language, and the records are very slim. It was definitely worth listening to all the tapes to see if I could identify the languages.

The recordings had been made outside, and there was a lot of wind noise. I was feeling a bit seasick at this point; the tapes were stereo and the microphone hadn’t been held too steadily, so there was a lot of rocking back and forth. Stick a pair of headphones on and slide the balance meter back and forth to get a sense of what this feels like. I’d been listening to tapes for many hours, including some German drinking songs, and was just about ready to call it quits for the day. One more tape, I thought. I stuck the reel* on the machine and cued it up. I heard Peile ask “What’s the name of that language? Nindi nindi?” The speaker replied, “Nyindinyindi.” Hmmm, I thought. That’s a new name on me. So I did what all good academics do when they come across something new – googled it. Nothing.** Then the speaker started telling a story in the language, and I could understand most of it. It was close to Bardi, the language I did my PhD on (and can speak pretty well). I went back to the audition sheets for that tape, and I saw it had been listed as recorded at “Tinder Bay.” There’s no Tinder Bay in the right area, but there is “Pender Bay.” A few years later I was able to play the tape to Bardi speakers. No one knew the name “Nyindinyindi,” but they confirmed that the language on the tape was similar to Bardi.

That’s just one example of that week of work. In the end, I found recordings of 4 languages that were otherwise thought to be either unrecorded or sparsely recorded. The Jaru materials did turn out to be Jawi, which was another great find for the descendants of the speakers on the tapes.

It doesn’t surprise me at all that there’s a lot of materials in the State Library that weren’t known about, and it’s fantastic that they are working to remedy that. Some of the early records are now online, such as Larmer’s vocabularies from the mid-19th century.

*These tapes hadn’t been digitised at the time, so I was working from archival reel-to-reel copies. Most of Peile’s tapes have now been digitised.

**If you google Nyindinyindi now, the only relevant hits are my blog where I talked about finding this tape, and the Table of Contents of my Bardi Grammar.