Language Leakage: An Interview with Sarah Thomason


On Language

The linguist discusses how technology shapes culture and culture shapes words.

A uniform for the Spokane Indians in Salish.

The first time Sarah “Sally” Thomason and I spoke, she’d just completed her annual two-day, eighteen-hundred-mile drive from her home in Ann Arbor, Michigan, where she teaches, to rural northwestern Montana, where she spends her summers studying Montana Salish. For thirty-four years, Thomason has been assembling a dictionary of this Native American language, which is spoken fluently by fewer than forty people. Thomason, a linguist, is fascinated by what happens when one language meets another, and how those languages change, or don’t. I had contacted her because I was interested in how certain words—say, e-mail, or google, or tweet—had been exported worldwide by American-born technology. I’d already called several linguists, and they all said I had to speak to Sally. No one, they said, had more insight into how linguistic traits travel, how pidgins and creoles are born, and how languages interact and change over time.  

The French government tried very hard to resist American loanwords like e-mail, promoting in its place messagerie électronique or courriel. They’d formed a whole agency for this purpose. Laws were passed and enforced. And yet e-mail prevailed—it was simply more efficient. But Sally was especially excited about languages that resist such borrowing, even in the face of extraordinary cultural influence and dominance. Montana Salish was one such language. Our conversations followed a pattern: I arrived expecting one thing and ended up somewhere entirely distinct, thinking differently about language and human culture.

Is it fair to say that you study what happens when languages meet? Is meet too friendly a word? I suppose there’s a whole range of things that happen, and sometimes it’s friendly and sometimes it’s not.

Right, but having a language disappear because all the speakers got massacred is actually really rare. There are a couple of examples where all the speakers of some language got wiped out by a volcanic eruption on an island. And there are a couple of examples, at least one in this country, where almost everybody was wiped out by smallpox and then the remainder was lynched by a mob.

What languages are those? 

Let’s see, Susquehannock is the language that died when all the speakers got lynched by a mob. It was an Iroquoian language. First they got devastated by smallpox. And there are cases where languages were almost killed, or at least there was an effort to kill them, out of genocide. You know, as a side effect of genocide in El Salvador, when they had all those wars some years back, the government decided that Indians were dangerous and they should kill them all. And so they actually killed an awful lot of Indians, but I think maybe only one or two languages completely disappeared. The most famous story is in the Bible—the people at the bridge saying shibboleth. And that was a case where they were both speaking dialects of Hebrew, I guess it was, and if you couldn’t say shibboleth because you didn’t have the sh sound, they’d kill you. But that wouldn’t have killed the whole language either, because the people who were trying to cross that bridge were all warriors, all men, and there would have been women and children who weren’t in the battle, I assume.

We’re living in an era when jargon, especially from the tech world, tends to bleed into the culture at large. Is that something you pay attention to?

Yeah, and that’s obviously important for social history, too, right? Things that get to be mainstream used to be very specific to a particular subculture. One aspect of that is what happens to slang. Every generation of teenagers will invent their own words because the whole point of teenage slang is to have in-group vocabulary that outsiders, like old people, can’t understand very well. And a lot of those words are ephemeral. The next generation comes along, gets their own words, the old words disappear. But some of them don’t, some of them hang on—and predicting which ones will hang on is a mug’s game.

But looking at the ones that did hang on is interesting because it tells you that they turned out to be useful. Mob—the word mob used to be a slang word. It’s a reduced form of a Latin word, mobile. But now it’s a really useful word. It’s interesting to see which words turn out to be useful. There must have been a time when a computer mouse was confined to a very small subculture.

Sure. Although, even now, we have fewer mice. Now we have pads, touch things, screens, styluses. So I guess it’s how long that tool remains in use.

Right, and if the tool goes, the words for it might also go, or they might get a new meaning. A hundred years ago you and I would probably have been able to name quite a few parts of a horse’s harness. Nowadays I know what a bridle is, but there isn’t much else. If you were now inventing a term for what we call horsepower you’d never settle on horsepower, because horses are not in the public consciousness anymore.

Are there languages that are better at adapting? When languages meet, does one “win”?

Sure. But that comparison has nothing to do with the structure or the vocabulary of the language, it has to do strictly with social factors. It’s not as if people come into contact and one crowd says, Boy, your language is a lot more efficient than ours! It depends on who’s got the power. The world I live in, the world you live in, Western Europe, the United States, highly industrialized countries, the paradigm we’re used to is colonialism—and then the indigenous languages are threatened. A lot of them have disappeared and the ones that haven’t are at great risk, so that seems like the norm.

But imagine a society—and again, these are mostly hunter-gatherer societies, but there are still a lot of those around—where the people practice exogamy, meaning you have to find a marriage partner outside your own group. Often the criterion is whether they speak the same language as you. If you have a society like that, you’re in contact with at least one other group and typically several relatively small groups—and it’s greatly to your advantage to maintain different languages, right? You don’t want to change your whole culture, you value your culture, exogamy seems like the way the world ought to be, and you certainly want to get married and you have this view that you shouldn’t marry your sister—then you preserve the languages.

That’s one reason languages get preserved. You find another phenomenon—it’s particularly common in and around Papua New Guinea, where there are about a thousand languages. That means that they’re close together, they’re small groups. Some of them are related to one another, so they’re pretty similar, and in that part of the world it’s probably not accidental that there are so many languages in such a relatively small area. It’s fairly common for groups to deliberately change their languages so they’re not so much like the guys next door. And the most spectacular examples are where you’ve got dialects of the same language and oh, we don’t want to be too much like those guys. It’s an identity-preserving thing, it’s a distancing phenomenon.

Road signs in Salish.

How did you get started on Montana Salish?

I was working on language contact, and the Pacific Northwest—Washington, Oregon, neighboring parts of British Columbia, particularly—is one of the best-known linguistic areas in the world. There are languages in that area, some of them totally unrelated as far as we know, that share all sorts of structural features. Not vocabulary, so much, but structural features that they didn’t inherit from their ancestors, that have traveled from one language to another. There’s also a phenomenon where you’ve got three or more languages in the same area trading features through multilingualism.

My family was already spending summers one mountain range to the east of the easternmost Salish language—most of the Salish languages are on the coast. I thought I could find out about this linguistic area if I started studying this language, and the tribes wanted somebody to come and help them get used to the writing system, a new linguistic device for them, so I was going to be useful. I thought I’d find out about this language and then I would find out about the whole family, and then I would be able to study the histories—how these features got from one family, where they started, how they got from one family to the next. So in 1981, I started trying to learn about this language, and it took about ten years before I realized I need maybe another 150 years for that project, and then I’d only need another century or so to understand the linguistic area.

But in the meantime, I really got hooked on the language. It’s a wonderful language. I like consonants, and they have thirty-eight consonants. I like big, long, complicated words, and they have huge, long, complicated words.

And it’s, this is a funny way to put it, it’s a stubborn language, right?

They don’t borrow from English or French. French was the other European language they were exposed to, but that stubbornness—it’s not unique, that’s what’s interesting. That is an areal feature, it’s part of the linguistic area. I don’t know if it spreads all the way to the coast but it certainly spreads to the Nez Perce, an unrelated language.

When you say it doesn’t “borrow,” do you have an example?

The word they use for automobile means “that it has wrinkled feet,” which is, incidentally, an example of how the words you have reflect your culture. If you’re a tracker, you’re going to be noticing the tire tracks—the focus of that particular word. And the word for telephone means “you whisper into it.”

Do you have an idea of why this language doesn’t borrow?

I know the reasons are cultural because they could easily borrow from other languages. But that’s as far as I can go. I did ask an elder some years ago, he was about ninety. And somebody had asked him what the word for television set would be and he said, Oh, I don’t know, I guess it would be … I forget what he said exactly, but something like, You look into it—you see things, something like that. And I said, So why do you do that? Why don’t you just use the English word for television set? He said, I don’t know, we just don’t.

Native American Flatheads (Salish) on the Flathead Reservation in western Montana, ca. 1900.

Are there any language interminglings that are particularly fascinating to you today?

There’s one in Canada—although actually I think it’s now spoken, to the extent it’s still spoken, on the Turtle Mountain Reservation in North Dakota. It was clearly created by bilinguals because there’s no distortion in either half of the language. All the verbs and verb phrases and the basic sentence syntax are from Cree, which is an Algonquian language. All the nouns and noun phrases, including adjectives and articles, are French. There’s a little bit of leakage from the Cree into the French and none in the other direction. The context where this arose was, as you might guess, mixed marriages. You had French trappers who were Scots, too, but French was the dominant language among the Europeans, trappers, and traders, and they married Algonquian women or had children with them, and the children were legally as well as socially separate. They formed their own community and they became buffalo hunters instead of, say, traders. They had a rebellion in the nineteenth century and their leader was hanged and then the people kind of dispersed, but the language …

What was their name?

It’s called Michif. It’s spelled different ways, but mostly Michif. It’s a phonetic distortion of Métis—it literally just means “mixed.”

The other spectacular example of this phenomenon is, or was, spoken in the Aleutian Islands. The Aleutians themselves belong to the United States but the westernmost islands in that necklace, just off the coast of Russia, belong to Russia, and they’re called the Commander Islands. One of them is Medny Island, which means “copper” in Russian. It was uninhabited until the early nineteenth century. The Russians were gearing up for their fur-seal-trade operation and they moved a bunch of Aleuts to the island and set them hunting fur seals. They produced children with Aleut women, the Russian traders did. So again, you had a mixed-blood population. And in this case, too, you had bilingualism—they knew Russian, they knew Aleut, but their social and economic positions were definitely separate. They were the middlemen between the Aleuts, who went out and got the fur seals, and the Russians, who brought the fur back to the mainland. So they were richer than the Aleuts, not as rich as the Russians, and socially stigmatized by both for being illegitimate.

So again, you have this language mixture, but it’s very different from Michif. Whereas Michif is Cree verbs and French nouns, this is mostly Aleut, except the entire finite verb morphology—mostly endings for tense and subject and the like—is just lifted wholesale from Russian. And Russian has a complicated verb morphology, but Aleut has a much more complicated verb morphology, like 450 different verb forms, whereas Russian has some dozen, three dozen maybe.

These are not pidgin languages—the kind you’d need to trade with the other guys—because these groups spoke the root languages, too. These languages occur for in-group purposes. When your group is separate, either willingly or unwillingly, a language turns out to be very useful. I think they must have come about at least in part deliberately. Although it’s hard to prove, because I wasn’t there. But there are examples of changes that are clearly deliberate. I told you about the—stop me if I’m babbling too much.

No, no, I love this.

I told you about the distancing changes in New Guinea. There’s an island called Bougainville—which is famous if you’ve read a lot about World War II—but it’s a big island and it has a language called Buin. Buin has several dialects, and one of them is Uisai. There are about fifteen thousand Buin speakers in all, and maybe fifteen hundred Uisai speakers. And Buin has, including all its dialects, a very elaborate gender system, sort of like what you find in French or Russian or German but more elaborate because each noun is either masculine or feminine, and then the verb will agree in gender with the noun, and the adjective will agree in gender with the noun, and so on. So in a sentence you’ve got a lot of markers indicating the gender—it’s part of the syntax as well as the lexicon. But in Uisai, all the genders are reversed. Every noun that’s feminine in Uisai is masculine in all the other dialects of Buin.

Now, this just isn’t conceivable as any kind of ordinary, natural, gradual linguistic change. I mean they have to have sat down and said, We’re too much like those guys, we’ve got to do something. How about this? A lot of linguists, maybe most linguists, would say this isn’t even a possible linguistic change. My belief, which has gotten more radical the older I get—which is nice, you don’t want to get intellectually fossilized—is that anything you can become aware of in your language, you can change if you’ve got a powerful enough motive. And of course, it’s not going to affect anybody’s language but yours, unless everybody else changes, too.

It’s hard to do that for a language like English. There’s too much inertia, too many speakers. But if you’ve only got fifteen hundred speakers and you really want to be different …

Sally Thomason in Montana, 2012.

Is there a concern that technology has altered language contact—that it’s happening all over the world, all the time, and that the primary language tends to be English?

Well, for those of us who value linguistic diversity, yeah, it’s a serious concern. On the other hand, it’s offset by the knowledge of language endangerment. There’s been a lot of publicity in the last decade, and if you look on the Internet there are dozens, if not hundreds of websites for revitalizing endangered languages. Then of course there are countries that don’t want English to take over anything—but whether it’s completely successful for any language is an open question. The only totally successful case of language revival is still Hebrew. Having your language serve as a vehicle for a major world religion is very useful if you’re trying to revitalize it.

Are there others that, to your mind, have a good shot at returning?

There are two cases that seem promising, although it’s not clear how much further they’ll go. One is Māori, the indigenous language of New Zealand. I mean, I say “indigenous”—they were only actually there for about three hundred years before the Brits got there, but still, they were there first. The other is Hawaiian. Both used a strategy called Language Nest where you get kids, preschool kids, in a setting where nobody speaks to them in any language but the language they want to revitalize, and because you’re getting them so early you’ve got a good chance of making them fluent. They did it in New Zealand first, and Māori is now one of the official languages there, but there are doubts about whether the revitalization effort is too late, because in New Zealand, too, everybody needs English. I published a textbook last year on endangered languages and the last chapter was on revitalization. I wanted it to be accessible to people who want to revitalize their language, so I tried to say optimistic things, but it got kind of hard to say optimistic things without lying.

And there are other programs. There’s a successful one run by a retired professor at Berkeley named Leanne Hinton—it’s called the Master-Apprentice Program. You use it when the language is on its last legs but there are a few fluent speakers left. You pair them with young people who are desperate to learn the language, they get trained to work together, and then you have this two-person team. It’s really impressive. But it’s small. You can get depressed thinking too much about this, if, like me, you find English kind of boring. But there are languages that are disappearing and being replaced by languages other than English. Cultures shift, languages change, and there are a lot of things other than force that make people give up their language and start speaking somebody else’s.

Ryan Bradley is a writer and editor. He lives in Los Angeles.