What Do You Mean You Don't Speak My Language?
July 20, 2017 10:34 AM   Subscribe

 
oh no
posted by Phatty Lumpkin at 10:40 AM on July 20, 2017 [6 favorites]


"In our model, when a population grew larger than a maximum threshold... it divided into two populations, each speaking a distinct language."

So why should the two populations end up speaking different languages, when they started out with only one? That's the interesting question IMHO.

(The answer, again, IMHO, is that the question is wrong. The question isn't "what causes langauges to diversify" it's "what keeps them from diversifying." Human language is prone to a natural genetic drift which sends it constantly off into new directions, and the thing that reins it in is a community in constant contact needing to talk to each other. The wider the community that needs to stay in contact, the more language diversity will be inhibited. When you have a lot of tiny, largely self-sufficient communities, their language diversity is going to be off the charts. They don't *all* need to talk to each other, each just needs to get on with its neighbors, at most. If you have a huge empire which needs to be in constant communication, it's going to be a powerful force for uniformity. So you can have populations all over Europe speaking Vulgar Latin, more or less the same, for centuries, but once the Empire falls, you're on a path towards dozens and dozens of distinct Romance languages.)
posted by edheil at 11:19 AM on July 20, 2017 [9 favorites]


I was kind of mystified by the tropical islands have multiple languages (several hundred, say) and a huge country like Russia only has 105 or so. The article tried to explain it but didn't quite do it well enough for me to understand.
posted by MovableBookLady at 11:28 AM on July 20, 2017


I gather, from the way that they're talking, that they mean "how many languages a place has or had before one dominant group started murdering everyone else or using the state to force people to speak their language?"
posted by ROU_Xenophobe at 11:37 AM on July 20, 2017 [3 favorites]


This is super interesting, though I think there are many more factors in play. For one: when do you call something a language? When I was a child, I could discern between and speak dozens of dialects (I went to country fairs, some with very wide attendance, and a group of us met several times during the season and became friends). I've forgotten how I learnt, but also most of the dialects are now almost completely lost. Someone my age can understand the "dialect" but not speak it. Our children are clueless. What's important in this context is that some of those dialects were as far apart as languages. My parents couldn't understand some of the dialects I spoke relatively well.
The reason my friends' "dialects" were defined as dialects rather than languages was national education politics. They'd have entirely separate words, conjugations and grammar, but at school and through national media, they were told they were speaking dialect and needed to "adjust" to official language, when it was obviously not even close to the same. Two of my friends strongly objected to this and saw themselves as bi-lingual, which is why I ever noticed.
I imagine something similar could be observed in Germany, in the Alps and in Italy. So — while I am intrigued by the idea that some areas have more languages than others and that the tropics might be richer in languages, this cannot be seriously examined without focus on the politics of language.
posted by mumimor at 11:39 AM on July 20, 2017 [5 favorites]


It's a crude and old-fashioned model, and I'm sure a genuine linguist will happen along shortly to explain how limited it is, but administrative and geographical unity tend to discourage language divergence, whereas the opposite encourages divergence. You can have dozens of little islands that are, or were, basically self-governing and with only limited contact with their closest neighbors. There's little pressure to speak a common language, as long as a pidgin sufficient for trade survives. Whereas you can have a big ol' wide-open Russian steppe but, if it's ruled from end to end by one great Khan or whatever, there will be pressure to maintain a common administrative language, and the degree of contact with your neighbors will drop off less steeply over distance than if you have to negotiate the trip in a longboat, so you continue to be exposed to their speech.
posted by praemunire at 11:42 AM on July 20, 2017 [1 favorite]


Maybe it also has to do with one group's own resources. The more you have, the less likely you need to interact with others to obtain what's missing. In tropical climates, you may have all the "necessities" one needs, cutting down the requirement to trade or search elsewhere to obtain them. In a desert, those necessities may be spread out across wide spaces, requiring constant interaction between groups to create a self-leveling effect in the language pool.
posted by Atreides at 12:55 PM on July 20, 2017 [1 favorite]


Their test case, Australia, is atypical for modern humans: the people were hunter-gatherers not organized into states. So maybe raw resources is a pretty good predictor for them.

Agricultural states do have a pattern where one language family spreads out over a large area, like northern India or the Algonquian area of the eastern US. You get more language diversity in the mountains, places like the Caucasus or California.

Almost all humans speak imperial languages, but it's not quite as simple as the conquerors "imposing" their language. They're not providing schools; there's no government services to speak of. Empires work just fine if only the elite speaks the language of the capital. Besides, no empire has ever been able to arrest language change or division.

What works in modern times to extend standard language, is mixing of populations. And this was probably the case in China and Arabia and Rome too. Grab a bunch of Romans from all over Italy and settle them in Gaul, and they'll spread Vulgar Latin.
posted by zompist at 1:23 PM on July 20, 2017 [1 favorite]


I thought that was going to be an interesting article, but then it turned out to be nonsense.

So they spent 6 months fine tuning an algorithm to reasonably approximate a single region in which multiple languages are spoken (Australia)... and then claim their model "predicted" the right answer... ?? you have to be kidding? And not only that they have a caveat for why this "particular" model only works for Australia, as though that doesn't completely undermine the results?!?

Seriously?
posted by mary8nne at 1:58 PM on July 20, 2017 [10 favorites]


Atendu dua, tio signifas ni ne ĉiuj parolas Esperanton de nun?
posted by ejs at 2:13 PM on July 20, 2017 [2 favorites]


i had just been thinking (like literally two minutes before) about how racism plays into people referring to or thinking about chinese as a language instead of a language family when i saw this post and thought it was a really huge coincidence and that the title was a clickhole or onion headline for an article satirizing something like what i had just been thinking ?
posted by LeviQayin at 4:27 PM on July 20, 2017 [2 favorites]


i had just been thinking (like literally two minutes before) about how racism plays into people referring to or thinking about chinese as a language instead of a language family

No, people talk about Chinese as being one language because they are (consciously or unconsciously) following the Chinese government's line that it is one language, which is pushed for nationalist/political reasons.
posted by jb at 4:43 PM on July 20, 2017 [4 favorites]


I thought that was going to be an interesting article, but then it turned out to be nonsense.

So they spent 6 months fine tuning an algorithm to reasonably approximate a single region in which multiple languages are spoken (Australia)... and then claim their model "predicted" the right answer... ?? you have to be kidding? And not only that they have a caveat for why this "particular" model only works for Australia, as though that doesn't completely undermine the results?!?

Seriously?


Yeah, this is like p-hacking, but like ten times easier to bullshit.
posted by leotrotsky at 5:16 PM on July 20, 2017 [1 favorite]


No, people talk about Chinese as being one language because they are (consciously or unconsciously) following the Chinese government's line that it is one language, which is pushed for nationalist/political reasons.

sorry but i find your comment pretty pedantic and condescending and mansplainy. i am well aware that the chinese government pushes that line, but to claim that as the reason why seems highly dubious to me, especially in that, firstly, western discourse on chinese policy almost always seems to focus precisely on the discrepancy of what they are saying/pushing and what is actually the case (thereby serving as a foil to western democracies - look, we can see what they can't because they are submersed in ideology yadayada), and secondly, seeing that, if you crank the racism knob a tad you see people using chinese as short-hand for looks/sounds/appears "east asian." i think it's pretty obvious that there is racism involved in thinking or at least taking for granted that here we have Languages , and there merely Dialects , and i even get wanting sometimes to sort of flippantly be like "no, not everything is / is explainable by/as racism!"

but here i think it is racism!
posted by LeviQayin at 7:04 PM on July 20, 2017


Given that I have not identified my gender as male, I would appreciate it if you called it accurately as "jb-splaining".
posted by jb at 7:54 PM on July 20, 2017 [2 favorites]


i think it's pretty obvious that there is racism involved in thinking or at least taking for granted that here we have Languages , and there merely Dialects

It seems hard to tell that from simple ignorance of other countries. How many people in the USA know that there are language groups in Spain besides Spanish? Are they more likely to think of the various languages in China as "Dialects" than they are the various languages in Italy?
posted by straight at 9:46 PM on July 20, 2017 [3 favorites]


So they spent 6 months fine tuning an algorithm to reasonably approximate a single region in which multiple languages are spoken (Australia)... and then claim their model "predicted" the right answer... ?? you have to be kidding? And not only that they have a caveat for why this "particular" model only works for Australia, as though that doesn't completely undermine the results?!?


I'm guessing this is mostly covered under "the reporter not quite accurate reporting of what they did and how far the researchers want to take the results".

If you look at the actual paper, the story is more like this:

People have claimed that various factors explain the number of languages we have in places like Australia. We built a lot of models including those factors in different combinations, and tested them out on Australia. The model that worked best involved rainfall, population size, and a mobile population. This model is really freaking simple, but also really accurate, which is good: this means that this general technique might be able to be used to look at other places. The exact model will differ, but we will still be able to compare the effect of, e.g., rainfall, across areas of the world.
posted by damayanti at 4:59 AM on July 21, 2017 [1 favorite]


"The reason my friends' "dialects" were defined as dialects rather than languages was national education politics. They'd have entirely separate words, conjugations and grammar, but at school and through national media, they were told they were speaking dialect and needed to "adjust" to official language, when it was obviously not even close to the same. Two of my friends strongly objected to this and saw themselves as bi-lingual, which is why I ever noticed. "

The adage in my Linguistics 101 course in college was "A language is a dialect with an army."

"It seems hard to tell that from simple ignorance of other countries. How many people in the USA know that there are language groups in Spain besides Spanish? Are they more likely to think of the various languages in China as "Dialects" than they are the various languages in Italy?"

I just learned this like six months ago. A confluence of curiosities about Italian pasta names, Spanish states and South American accents led me down a weird hole where I learned that there are, what, like 40 Spanish languages in the Iberian peninsula, that whether Portuguese is a Spanish is an ongoing controversy, and that "Italian" is Tuscan like "Spanish" is Castilian. I had no idea how widely the languages differed, influenced in part by how much of the pre-Roman language any general locale retained, to the point of mutual unintelligibility. I took German in school, and my family speaks Northern German/High German, so I'd already had the experience of trying to talk to an Austrian and being told that German is his second language too.

But jb is right about the Chinese government's influence: Within China, Mandarin/Jing, Wu, Yue, Min, Xiang, Hakka and Gan, etc., are all considered part of a single Chinese language; foreign linguists are much more likely to classify them as separate languages. That's distinct from Italian or Spanish, where the Spanish government explicitly recognizes multiple Spanishes, and Italy has an official "Italian," then recognizes multiple others as languages e.g Sardinian.
posted by klangklangston at 2:36 PM on July 21, 2017 [3 favorites]


"What's the definition of a language?" I once asked my mother.

"A dialect with a navy."
posted by rum-soaked space hobo at 10:09 AM on July 23, 2017


« Older "I only think about the next mile"   |   Three things cannot be long hidden: the sun, the... Newer »


This thread has been archived and is closed to new comments