An Interview with Claire Bowern, Vice President of the Endangered Language Fund

Published Feb 24, 2016 by Len Epp

Claire Bowern is Vice President of the Endangered Language Fund and Associate Professor of Linguistics at Yale University. Her research includes working with the speakers of a number of languages from Northern Australia. In this interview, Leanpub co-founder Len Epp talks with Claire about her career, her research, and two books she has published on Leanpub, Learner’s Guide to Bardi and Learners Guide to Yan-nhaŋu.

This interview was recorded on November 9, 2015.

The full audio for the interview is here. You can subscribe to this podcast in iTunes or add the following podcast URL directly:

Len: Hi, I’m Len Epp from Leanpub. and in this Lean Publishing podcast, I’ll be interviewing Claire Bowern. Claire is associate professor of linguistics at Yale University, and Vice President of the Endangered Language Fund. In her research, Claire focuses on the languages and history of Indigenous Australia. Since 1998, she has conducted extensive field work with the last speakers of a number of languages from Northern Australia, and conducted extensive archival work with historical records from the 1770s to the mid 1960s. Claire is associate editor of the academic journals, Language and Diachronica, and co-author of An Introduction to Historical Linguistics, and has published and edited many other books and articles in the field of linguistics. Recently, she published two books on Leanpub. In this interview, we’re going to talk about Claire’s research, what it’s like to work on endangered languages, and the subjects of her Leanpub books.

So thank you Claire, for being on the Lean Publishing podcast.

Claire: Oh, no worries. Thank you for having me.

Len: Thanks. I usually like to start these interviews by asking people for their origin story, so I was wondering if you could tell me how you first became interested in linguistics, and in particular indigenous languages in Australia?

Claire: Yeah, sure. So, I started out as a classics major. I’m from Australia originally, and I started work at the Australian National University in 1995. There, I was pretty sure I wanted to do something to do with language, but I wasn’t sure exactly what. So I took a bunch of classes in Latin and Ancient Greek, and some classes in German and Old English, and Indo-European-type languages. And then I took a linguistics class or two because I thought, “I like language structure and I like those sort of problems.” It was pretty clear from very early on that it was the linguistics that was the sort of work I really wanted to continue with. So I ended up taking more and more linguistics classes, and fewer German and English classes and so on. I ended up double majoring in classics and linguistics.

At that point, I was still pretty certain I wanted to work on Indo-European ancient languages, and the linguistics of Latin or Latin language change, that sort of thing. But then I took a class on Aboriginal languages, actually taught by the same person who was one of the classics, the Latin people, at the University. He’d also done Indo-European and classics work as an undergraduate and graduate student, and had then shifted to work on Aboriginal languages. And I thought I should know something about Aboriginal languages, even though I wasn’t planning to continue any work at that stage. And then of course I found out that there wasn’t just one Aboriginal language, there were 300, 350 distinct languages. They constituted a number of different families. Now we think there are probably 25 or 26 different families in Australia.

If I wanted to work on Aboriginal languages I could do field work. I could work with speakers first hand, and I could make a difference for language documentation programs in those communities. And all of those things were things that I found very exciting about working on language. I very much liked working on Latin and Greek, but it’s a very different sort of work. You go to the library and you look at what people have said about these texts for the last 200 years, and then you try to find something else to say about it. Whereas, for a lot of the work that I was doing even as an undergraduate, it was impossible to look up the answer in the library, or it was impossible to ask someone because no one knew anything about the history of these languages. And the only way to find out about the languages was to ask the speakers themselves. So that seemed like the sort of thing I wanted to do. So I’ve been lucky enough to do a PhD … and then other work on some of these languages as well.

Len: Okay, I guess I should say that you actually cut out there for a moment. I’ll let you know if it happens again or anything like that.

Claire: Sure.

Len: Okay, so I guess my next question is: how did you end up at Yale eventually?

Claire: So I came to the US for my PhD in 1999. I went to Harvard and I did my PhD there. And then about the time that I was graduating, the Australian job market went through one of its periodic fluctuations when there were basically no jobs in Australia. My husband is also — he’s not a linguist, he’s also an academic and we were also both of us interested in staying in the US for a while. So I applied to jobs here, and ended up at Rice University for four years. And then in 2008, I joined the faculty at Yale and I’ve been here ever since.

Len: In your Leanpub profile, you say that your work “combines traditional methods of language documentation and reconstruction with computational methods more familiar from work in biology”. And I watched a video in which you talked about using cluster analysis, for example. I was wondering if you could explain a little bit about what your computational methods are, and how they’re useful for linguistics research.

Claire: Linguistics is quite similar to many other humanities/social fields, I guess you’d call them, where the use of computers has revolutionized the sort of work that we can do. When I first started going to the field, if I’d wanted to look at the properties of sound systems for the languages that I was working on, I would have needed to have taken a couple of tons of equipment to collect the information. And then analyzing the material would have been the work of months. And then I might have got a couple of plots out of a special computer that was designed to run that sort of material.

These days, I can just take my laptop. I can have speakers speak into the microphone on my laptop, and I can produce the results in under a minute, for some of the simple results. So, that’s just one area where computers have revolutionized the field.

In other areas too, particularly in historical linguistics, there are questions that we want to ask about language and language change that it’s really difficult to answer with pencil and paper. To take the example that you were talking about with the languages of Tasmania, there are documents from Tasmanian languages that span 150 years or so. But because Tasmanian Aboriginal people didn’t use language names, and because the recorders weren’t very good at recording where their materials came from — who they were talking to, where they were recording, and so on — we just have a collection of Tasmanian language data. We don’t know about how many languages were represented in that sample.

One potential easy way to answer that question would just be to count up how similar the words in each of the word lists are, and say, “Okay, these are the same language. No, this is something totally different,” and so on. But we know that many of the sources came from either the same speaker who was speaking different languages, or we have multiple languages represented in the same sample. So, just because we have a word in a particular word list, it doesn’t necessarily mean that that’s just a single language. In fact, we know in some cases that the words come from multiple different languages, and there were multiple people there, and that the recorder was switching languages from one word to the next.

That’s the sort of problem that a computer is really good at answering. So it’s a type of clustering problem, if we want to take the data we have and bin it, group it into groups. But we don’t know what constitutes those groups or how many groups we should have. There are many software programs that can do that sort of work. It’s not an easy problem, but it’s certainly a computationally tractable problem for a computer to do. It’s a very difficult problem for a human to do. So, there are other issues like this that make some parts of linguistics very amenable to doing work with computers.

Language change is another area where this is really interesting. This is somewhat a matter of debate, but it’s pretty clear that at least in some ways languages change pretty much the same way that species change. So, evolution of species with decent through modification as it’s called. You inherit the genes from your parents and grandparents and so on, but every now and again there’s a mutation in those genes and that’s a change. Those changes have mathematical properties that you can study through populations. Those sort of features also apply to languages, so you learn languages from your primary caregivers, so parents. You learn languages from the community you’re in, of course, as well. Most of the learning of that language is the same, so I speak pretty much the same way that my parents do. But there are some differences, and so these are the changes, as well. The tools that have been developed to look at species evolution in biology are also often very appropriate to look at language evolution, as well. So that’s the sort of thing I do.

Len: I found it really fascinating that given the assumption that things change in predictable ways, you can actually take data about language use, let’s say, in a region at a certain period of time and you can actually draw conclusions — historical conclusions — about what sort of social patterns existed prior to the point at which you have the data. So actually, you can do historical research by just looking — by only looking at the language. It’s just fascinating.

Claire: Right, there’s a lot of cases where, because for most of human history people were talking to each other face to face — writing is a very recent technology compared to the amount of time that people have been talking. So language has been around for probably about 100,000 years, writing’s been around for 6,000 or so. And even within the last 6,000 years, most of the world’s languages have not been written. And even within the languages that have been written, most of those populations have not been literate. So while we have a lot of information about the written record for the last couple of thousand years for many languages, we also have the — I guess the good point that — if you find, say, a loan from one language to another language, that loan is likely to have occurred because people were talking face to face. So the influence of written language on spoken language is minimal until the last couple of hundred years or so.

We can use that fact to trace population movements through the history of the words in individual languages. So for example, there are borrowings from Finno-Ugric languages like Finnish, and Estonian, and Hungarian, into Indo-European languages. And because of the sound changes and the stratigraphy, I guess you’d call it, like geological strata — because of the strata of the sound changes that we can identify, we can reconstruct when particular groups were in contact. And if there are a lot of learned words, like there are between Finno-Ugric and Indo-European languages, we know something about their culture as well. So for example, a lot of bee keeping terminology is borrowed from Indo-European into Finno-Ugric. So we know that there was a apiary and some bee cultivation and “honey” is a learned word into Estonian and so on, and so forth.

Len: I have a question about technology. One thing that’s emerged and become quite universal, at least in developed countries, is audio recording, and watching television, and listening to people talk. I was wondering within the linguistic community, if there are any conclusions people are drawing about the effect that’s having on language change? My instinct is that in some ways it will accelerate [change], but in some ways it will slow it. Because we can hear — we can actually listen to people, they can be sort of speaking to us, through a screen, but face to face from 10 years ago, or 20 years ago. Of course we also know that you can hear the way accents change from decade to decade. The 50s had an accent, the 60s had an accent. Are there any conclusions about how this kind of technology is internal to a language and a region? Is it slowing or speeding up change?

Claire: I think you’re right that’s it’s probably doing both. I also think that most of the effects are probably pretty ephemeral. So, for example, there’s been a long-standing debate about whether television is affecting language, and whether words are spread through the lexicon through television. And there are catch phrases, and phrases from sitcoms and so on that make their way into popular culture, but they tend not to survive very long. So they’ll be current for a couple of years — maybe five to ten years — and then the show will die, it’ll be replaced by something else. And those words will fall out of use as well.

I think it’s still pretty clear that face to face interaction is what has the main effect that drives language change and language preservation. On the other hand, there’s also the issue of prestige and the adoption of particular variants, both pronunciation variants and word variants that tends to go across social groups from more prestigious to less prestigious. Of course, defining prestige itself is a very difficult concept, so a model that I might look up to as being prestigious might be something that a person in the next office to me would not see as prestigious.

Len: Like “BBC English”, as they call it.

Claire: Right. A received pronunciation versus the Australian Broadcasting Corporation versus the American ABC. I think it is pretty clear that there is dialect leveling across the US, but there are also new regional identities and regional variants coming up as well. So, basically it’s all change and some things — some variation collapses, but new variation appears as well. So we needn’t worry that Twitter or TV or Skype is gonna cause the homogenization of American English. It’s going to change but we know that language changes anyway.

Len: I found a presentation online by you called “Ethics As Regulation, and Ethics as Morals in Linguistic Field Work” (download link). I was wondering if you could talk a little bit about why the Linguistic Society of America needs an ethics committee, and has ethical guidelines? Just because it might not be clear to everyone why linguistics professors would have ethical guidelines.

Claire: There are many different ways in which linguistics and ethics interact. So linguists tend to work with speakers of languages, and so as such our work is often subject to national legislation or international legislation overseas regarding the use of human subjects in research. So most of the experimentation that we do with language is pretty benign, but linguistics runs the gamut from putting people in MEG machines or doing — not exactly invasive tests — but quite boring and repetitive tests for several hours. We work with speakers of endangered language communities, or work with otherwise vulnerable populations where the potential for — at least historical, exploitation has been particularly high. So for example, with my work with Aboriginal Australian communities, there’s a long history of Aboriginal people being the subjects of research and not seeing very much benefit for that research.

Over the last 15 to 20 years or so, particularly in Australia and Canada but also in the US and some other countries, there’s been a push by both academics and community members to make sure that researchers are more responsible to the communities that they work with directly. So not just to the individuals who participate in the research, but to the communities who have a stake in the language as a whole.

So, for example, with the Bardi community that I work with, there are, these days, four or five people who speak the language fluently. But there are about 2,000 people who identify as Bardi, and who have a stake or an interest in outcome of that language. So they’re interested in whether I publish an article on the theoretical structures of Bardi, or whether I’m working on a dictionary that might be useful for a school program. If I’m recording oral history research, then not only the form of the language but the content of that recording is of interest to the communities. And of course that brings up privacy and confidentiality issues as well.

If I record a story about someone’s childhood, that’s personal information for them. And, of course, they may want that story published and widely available, but they also may not want that. So that’s an ethical issue that needs to be addressed. Historically, there’s been a relative lack of sensitivity by academics who work in this area. We’ve assumed that once someone gives us a recording, it’s up to us as to what we want to do with it. It’s increasingly recognized that that’s not the case. There’s been an increasing recognition by academics that the communities who have language records have an ongoing interest in the preservation and use of those materials beyond the strict academic context. And of course with the relatively widespread availability of internet materials, and more materials being put on the web, it’s possible for indigenous community members to have access to materials that were written in the 50s, 60s, and earlier through archival websites and so on. So there’s a lot less of a divide between the research community and indigenous communities these days. All of these topics, and many more, have ethical implications for linguistics.

Len: One of the really interesting problems that you bring up in the presentation is about having to choose — as I understood it — having to choose between time spent on revitalization of a language and time spent on preserving it. And I was wondering how those issues are resolved, generally? Is there a general way of doing that? Or does it depend on the strength of the government support for language programs around a community?

Claire: Yeah, I think there are a couple of different factors that we should take into consideration. One is the skills of the researcher and the community. Another is the level of interest in the community. I strongly believe that communities themselves should decide what should happen to their language. So if the community members don’t want the language to be written, or continue to be preserved and learnt by outsiders, then that’s the right of that community to make that decision. So that’s another factor in where one should channel energy, given the number of languages which are endangered, and the number of communities who would like to have linguists work with those communities. It doesn’t make any sense to work with a community who doesn’t want their language transcribed or written down. So those would be two factors.

The skill of the linguist, I mentioned, is another issue. If my primary skills are in language documentation rather than language pedagogy it makes sense for me to focus on what I’m best at, and hopefully get someone else to work on the pedagogy side. There’s also a translatability, I guess you’d call it, of materials. It’s always possible to make more simple materials from more complex materials, but the reverse is not true. So it makes sense to record a complex language and a variety of language, and make as good a record of the language as you can. And then derive the learner’s materials and beginning teaching materials from that source. Whereas if you start with alphabet books, or simple word lists, you’ll never be able to recover the structure because it just hasn’t been recorded. If the language is still relatively well spoken — then, if there are still a relatively large number of speakers and the time crunch is not quite as urgent, then it makes sense to make different priorities from what you would do if there were only one or two speakers of the language.

Len: Speaking of the number of languages that are disappearing, I saw on the Endangered Language Fund’s website that of the 7,000 languages currently spoken, half or even more than half are predicted to disappear this century. I was wondering if you could explain what the Endangered Language Fund is doing about that?

Claire: We’re a non-profit organization based in New Haven, Connecticut. We were founded by Doug Whalen, who’s a professor now at CUNY and Vice President of Haskins Labs, which is affiliated with both Yale University and a couple of other universities in the region. The ELF is partly a grant-administering body, so we have small grants for communities and researchers to work on language documentation and revitalization. We have twp programs, a language legacies program, and enduring voices program. One is for tribes who were on the Lewis and Clark expedition trail, and the other is a general worldwide grant program. And we also do language promotion and general endangered language advocacy work. But our main program is the grants program.

Len: And what would endangered language advocacy work entail?

Claire: It involves just raising consciousness about the issue of endangered languages. As you said, there are 7,000 or so languages in the world and roughly half of those languages are currently endangered. In some of the world it’s even worse, so in Australia between 90% and 95% of the languages are endangered. In Canada it’s somewhat similar. In the US, there are some relatively healthy languages, but there are also a very large number of Native languages, Indian languages, which are more or less endangered, losing ground to English. The same is true in many other parts of the world. There are, of course, other parts of the world which are gaining speakers, but we’re losing a language roughly every two weeks.

So part of our work is just raising consciousness about that. Another is to raise consciousness about the sort of information that is lost when we lose a language, and another is the social justice aspect of language loss. Language is a mark of a community in the way that other aspects of culture — with material culture and tangible culture — are. We, as English speakers, perhaps have a different view of this because our native language is so widely spoken. But for languages or communities where the language is much less widely spoken, and is very much a hallmark of being a member of that community, that’s a very important part of the social cohesion of that group. So there’s been work, for example in Canada, that recently looked at the levels of health outcomes for people whose languages had support versus those who did not. This was rates of diabetes and other chronic illnesses. And they found that communities that had relatively healthy languages, and relatively good language support also had lower instances of disease. [Here is a download link for a related article: “Language and Culture as Protective Factors for At-Risk Communities” — ed.]

This is on the one hand, a somewhat abstract argument given that language is intangible and it’s, in some ways, languages are all roughly equal. In that we can communicate the same ideas in English, or Mi’kmaq, or French, or Bardi, or Anindilyakwa. But on the other hand, the languages are also a monument to human cultural achievement of hundreds and thousands of years in many cases. And they are an important part of the communities that speak them, and they are a guideline for general community health. And so good language health is also related to good community health in other ways as well.

Len: I can imagine that the quality, I suppose, of inter-generational relationships would play a role in the transmission of language from one generation to another. And I guess one would see a healthier kind of relationship between — elders and younger people would correlate to the strength of a language. Does that make sense?

Claire: Right, absolutely. Although, ironically perhaps, some of the — some policies that have been designed to help native communities, in the US in particular, have also not been as friendly to languages. So for example, housing policies that have led to more housing on reservations have led to less overcrowding, but have also led to less inter-generational language transmission. Because rather than having grandparents, parents, children, or great-grandparents even in the same house, there’s more of the parents and children in the same house. And the primary caregiving for the children has been transferred to preschools, and kindergartens, and daycares, and somewhere where the staff may or may not speak the local language. So on the one hand, that’s led to great tangible economic gains and increases to overall quality of life that I wouldn’t want to gainsay for a moment. But it’s also had a by-product of making an already fragile situation more fragile for the languages.

Len: It’s really interesting, I hadn’t thought about that case before. I’d like to turn now, maybe to ask you on the topic of raising awareness, about some of the languages that you’ve worked with specifically. One of the books you’ve got on Leanpub is about a traditional language spoken by the peoples of the Crocodile Islands in Northern Australia. And I was wondering if you could tell me a little bit about the people, and what makes their language unique?

Claire: Sure, yeah. So that’s the Yan-nhaŋu language from the area north of Milingimbi, in the far north of Australia. If you picture a map of Australia, if you know where Darwin is, it’s basically in the center of the Northern Territory, on the coast — at the top of the top end, as it were. And you go east about 400 kilometres, the Milingimbi region and Arnhem Land is that part of Australia. And Yan-nhaŋu speakers live in that region, so the Crocodile Islands, the offshore islands. And many live on Milingimbi community, and Galiwin’ku (28:19 community name?) as well.

I started working with the Yan-nhaŋu community in 2004. I was approached to do a language — basically a language documentation project. I’d heard that some of the speakers of the language were interested in making a record of their language, and they wanted a linguist to come and work them. There was already some language work, Bentley James had been working there for a number of years. And there had been some other recordings with speakers going back to the 1970s or so. But there hadn’t been a fully sustained linguistic documentation of the languages before.

So I went for 6 weeks or so, and did a draft of the learners guide and worked on a draft dictionary that was already in progress. And then I came back a couple of other times to work with the community members again, to do more language documentation work. The community is — it’s quite small, even by Australian standards.

Perhaps I should give a little context about Aboriginal people in Australia. About 2% of the Australian population identifies as Aboriginal, and of that between 10% and 20% of Aboriginal people speak an Aboriginal language. So we’re talking about 250,000 people who speak a — sorry 250,000 people who identify as Aboriginal, and I think that’s right. Yeah, about 250 000 people and about — between 30,000 and 40,000 speakers of Aboriginal languages. Within that, within Arnhem Land, there about 5,000 people who speak Djambarrpuyŋu, which is the main lingua franca of that region — the common language of that region. But within that area, there are also Yan-nhaŋu speakers, and about 15 Yan-nhaŋu speakers. So a small language within a — the minority language within a minority community.

Yan-nhaŋu is one of 30, or so, different languages in that part of Arnhem Land and one of roughly 400 Aboriginal languages, as I mentioned before. So I worked with about five or six different speakers, about a third of the Yan-nhaŋu community. And we spent, I guess, about six months or so over four years working on different aspects of the language. So we’d do dictionary work, we did translation. We did material culture recordings, so I learnt how to weave baskets, and pandanus mats. I learned a bit about the bush foods, so what foods are good to eat, a bit about Yolŋu — so, Yolŋu is the Yan-nhaŋu name for Aboriginal people. So for Yolŋu peoples, I guess you’d call it worldview and some ceremonial information. And so part of the work was strictly linguistic, so, how do you say “dog”? How do you say “cat”? How do you say “snake”? That sort of thing. Part of it was how to document traditional practices, and part of it was how to document conversation and other aspects of language knowledge as well.

Len: It’s fascinating, you say that in the language words change shape depending on what part of the sentence they’re used in. I was wondering if you could explain what that means, and give an example if you have one ready to hand.

Claire: Yeah, sure. So that’s actually common in languages. Let me guess at a percentage — I’d say probably 85% of the world’s languages have that too, at least to some degree. It’s true in English, for example with “she” versus “her”. So we say, “She is running, but I have her book.” So that refers to the same person, but the form of the word changes depending on whether it’s a possessive pronoun or whether it’s the subject of the sentence. We have verbs in English that change depending on the tense — so if it’s past, present, or future, or if it’s a participle. So, you know “run” verus “ran” versus “running”, and so on.

And Yan-nhaŋu is just like that. Yan-nhaŋu has rather more inflection than English does, but it’s of a similar sort of type. So, for example, in Yan-nhaŋu, the word for “snake” is “mol’”. And if you want to talk about something, if I’m talking about seeing the snake I would say, “mol’-nha”, so with an NHA, “-nha” at the end. If I was talking about the snake seeing me I would say, “mol’-yu” with a “-yu” at the end. If I was talking about something on the snake I would say, “mol’-ŋa” with a “-ŋa” at the end. And if I was talking about something to do with snakes, if we were talking about the general subject of snakes, I would say, “mol’-pu”. So we have the root of the word, “mol’” in each case, but we have different endings on the word. And there are some other ones as well, but those are some of the main ones.

Len: Okay, I’d like to take a couple minutes, too, to ask you about the Bardi language, and the people who speak it. I have a specific question too about the language, where in the book it says that words change depending on who is doing what to who. I was curious to ask about what that means.

Claire: Bardi is not related to Yan-nhaŋu, Bardi is spoken in the far Northwest of Australia. So it’s, if you think about the west coast of Australia, the coast goes roughly north for a couple of thousand kilometers, and then it turns east. Bardi is spoken right at the tip of the Dampier Peninsula in that part of Western Australia. So it’s about as far away from my hometown in Australia as you can get — as my parents remind me from time to time. These days, there are about four to five Bardi speakers, but as I mentioned about 2,000 people who identify as Bardi, and a number of other people who have knowledge of the language, but are not full speakers of the language. So they understand it and they use words, but they don’t use it as an everyday language.

This learner’s guide was written for them. I have a long association with One Arm Point school at the tip of the Dampier Peninsula, where I’ve been going for — I guess more than 15 years now to do language work of a lot of different types. Bardi is, like many languages, a language where the words change depending on the position in the sentence. Bardi has very complicated verbs, so to say — so they have both prefixes and suffixes, so you change the beginning of the words and the ends of the words — so for example, “marra” is a word that means “cook”. If I am cooking something, it’s “nganmarran”. If you’re cooking something it’s “minmarran”. If I was cooking something, it’s “nganamarragal”, and so on and so forth. If I’m cooking something for you, it’s “nganmarranjiy”. So there are lots of different prefixes and suffixes that go on.

Bardi also has case marking, as it’s called. The forms of nouns change whether they are the subject or the object of the verb. So the way I would say, “I see the crocodile” is different from “The crocodile sees me”. So if it’s, “I see the crocodile,” it would be “ngayoonim” for I, and the “nim” part on “ngayoo-nim” tells the listener that I am doing whatever the verb is going to be. And then the verb would be “nganjalan” so the “ngan-“ on the verb would be, again, saying that it’s me that is doing that. And “jala” means “to see”. And then “linygoorr” is the word for crocodile. So that would be, “I see the crocodile.” Now, if it’s “The crocodile sees me”, we would put the “-nim” on “linygoorr” so it would be “linygoorr-nim” and then “ngayoo” for “I” and then “see” would be “injalan”. So it’s “in-jalan” for he, she, or it doing something, but “ngan-jalan” for I am doing something.

Now, the order of words doesn’t matter in Bardi. So for English, “I see the crocodile” and “The crocodile sees me”, the “see” part of the word changes depending on if it’s the crocodile or me doing the watching, doing the seeing. But mostly it’s the order of words that changes. So if it’s the past tense, “The crocodile saw me,” “I saw the crocodile”, it’s only the word order that tells us who is doing the seeing and who is being seen. But for Bardi, it’s all these word pieces that do that order, that give us that information. And the order of the words tells us what the most important part of the sentence is.

Len: And Bardi seemed to have quite a bit more complexity around the naming of family relationships than English speakers might be used to. I was wondering if you could explain a little bit about how that works.

Claire: Bardi has, let’s see, so for kinship terms if we take the — so there’s “ngayoo” which is “me”. Then if we think about my generation, there’s a term for “older brother” and “younger brother”, and then there’s a term for “older sister and “younger sister”. So English has “brother” and “sister” and then we can modify that to be older or younger. But for Bardi, there are different terms for each of these relations. There are terms for “mother” and “father”, of course. But then if we think about “son” and “daughter”, in English “son” is a male child and “daughter” is a female child. But for Bardi, the two terms, “aala” and “bo” are to do with the sex of the referent, so the sex of the person who’s talking. So I would talk about my kids as “bo”, a woman’s child. But my husband would talk about our kids as “aala”, i.e. a man’s child. So whether it’s a boy child or a girl child, it isn’t part of the meaning of that word, it’s who’s doing the talking. So that’s a bit of a difference between English and Bardi, but lots of languages have that sort of system.

Then just one other thing for kinship, probably the biggest difference is the grandparent terms. So English has “grandmother” and “grandfather”, and it doesn’t matter if it’s your mother’s father or your father’s father; it’s “grandfather” or “granddad” for both. But for Bardi, there are 4 different terms for mother’s father, father’s father, mother’s mother, and father’s mother.

Len: Okay, that’s really interesting. It’s a common observation, of course, but getting a window into entirely different ways of people relating to each other, or to the world around them, and even to animals and things like that must make your work endlessly fascinating.

Claire: Oh, absolutely. Yeah.

Len: I guess we’re running out of time a little bit here, but I just have one question about how is it that you discovered Leanpub, and why did you decide to publish these learners guides on Leanpub?

Claire: Initially I was looking for a traditional publisher for the learner’s guides. And I sent off a publication proposal to a couple of different local publishers in Australia. And I wouldn’t hear anything for a couple of months, a couple of years in some cases. And then I’d get back a very polite note that they were very sorry, but they didn’t think the print run would cover the costs of the book. They didn’t want to publish the book, given the likely low potential sales. So I started looking into online publishers, and I looked at a couple of different options — both academic publishers and self-publishing. And it looked like the Leanpub model was one that would be very workable for the sort of material I had. I was familiar with some of Leanpub’s publications from a couple of Coursera statistics and data analysis courses that I’d seen. So I’d seen the publication model, and the royalty model, and the “name your own price” type model. And that seemed like a really nice way to both encourage people who have the means to pay for the learner’s guides to pay for them, but not to restrict the materials to community members, or to other people who would really like a copy but for whatever reason can’t pay the full price, or can’t pay anything.

This is a long-standing issue in making materials accessible to indigenous communities and other communities where not everyone has the ability to spend a couple hundred dollars on books, just on a whim. And so it’s — I’m a great believer in making these materials as accessible as possible. On the other hand, it also seemed like if people are willing to pay for the materials, it would be nice to have those materials generate some revenue for the Endangered Language Fund. And so this seemed like a nice compromise between those two goals.

Len: Yeah, thanks very much for that. That’s what I was hoping to hear. Yeah, the “pay what you can” model — so in Leanpub we have a variable pricing model where the author sets a minimum price and a suggested price that the person, the customer pays for the book — and that minimum price can be as low as zero. And so some of the Johns Hopkins professors who have these Coursera statistics courses, data science courses, it was really important to them also, that they could have a minimum price of “free” for some of their books. Because their students are from all around the world, and they’re all with different currencies, and different levels of general wealth. And it was very important to them that people should be able to pay only what they want to.

Claire: Right, yeah purchasing power varies extensively — even within the US. If we think about faculty’s salaries, versus tenured faculty’s salaries, versus adjunct salaries, versus graduate student salaries.

Len: Yeah, and also we have a Leanpub for Causes program, which allows authors to share their royalties, and in your case it’s 100% of your royalties with a participating charity or a non-profit organization. And so you signed up the ELF and then proceeded to share your royalties, and we were very glad to see that happen too because that’s something that’s very important to us as well, to see books related to causes and the book can become a form of support to a cause.

Claire: Great, yeah. And I hope that in future we will have some other language documentation materials that might join the learner’s guides that are there at the moment.

Len: Great, great. Well, I guess our time is up. Thank you very much, Claire, for being on the Lean Publishing podcast, and for using Leanpub to publish your books.

Claire: No worries, thanks Len.

– Posted by Len Epp

Originally published at

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.