> Don’t believe everything you’ve heard, however: there are not hundreds of words for snow.
Skier’s learn a large number of words for different types of snow: everything from death cookies to boilerplate to corn snow. Vocabulary comes from interaction.
I am guessing different sports have wildly expanded vocabulary for certain elements of the relevant environment?
e.g. if I say I'm going to "bounce the box", I don't mean that I will throw a cardboard container at the floor, I mean that I will reboot a computer. And the sysadmins will know exactly what I mean.
Isn't the language almost extinct and the community almost wiped out? I remember reading a while back that there are only a few hundred speakers and as more canadians move into the area, the inuit community and language were being endangered.
Is there a systematic campaign to preserve the language and/or community? At the very least to set down the vocabulary/grammer/pronounciation/etc?
The decline of Inuktut wasn't principally caused by displacement or natural population decline, but rather the residential school system that attempted to completely assimilate young Inuit (and other indigenous) kids. Children were systematically removed from their communities and sent to English (sometimes French) religious boarding schools. They were physically punished for speaking their mother language. [0]
Inuktut is mostly stable, though "vulnerable," and strongest in Nunavik (the region in the article), I suspect because the somewhat-devolved region employs it as the primary language of instruction. [1]
Every tribe is working very hard on language preservation and some are getting much better results than others. Some have been lost, but the folks in Alaska are putting a lot of effort into it.
I asked a question on HN for a grant that was directed about using ML to aid in language preservation. I didn't get any responses. We were trying to expand the tools available.
It sounds to me more like you already had a tool in mind (ML) and were just looking for some nails ;)
If you approach the problem instead from the language preservation perspective and ask what is necessary to make it succeed, there are essentially two tasks: documentation (recording current speakers' knowledge of the language) and community maintenance (allowing people to keep using the language and teaching it to the next generation).
Documentation basically boils down to making a lot of recordings. Since text is usually nicer to work with, transcriptions are very useful, and in theory ML would be able to help with that, but to train a speech-to-text model, you first need transcribed training data, leading to a chicken-and-egg problem. Making the data easily available and easy to contribute to is more likely to be impactful.
For maintaining a community, ML can be useful by providing all the services that ML offers for speakers of more popular languages: text-to-speech, speech-to-text, predictive smartphone keyboards, machine translation, ... but all of those depend on the language being well-documented already. Otherwise, non-ML solutions are likely to be the only options (so formant-based speech synthesis e.g. with espeak-ng, dumb keyboards, rule-based translators, ...)
To give a specific example, I know of a community of Kabyle speakers who want to be able to use OpenStreetMap with voice navigation in Kabyle, so they've been making a lot of recordings (using https://tatoeba.org as a platform for the documentation part) and also have someone working on a speech-synthesis model (using Mozilla's DeepSpeech, I think). AFAIK, the project is coordinated by Muhend Beqasem, whom you might want to contact if you're interested in what they're doing: https://tatoeba.org/eng/user/profile/belkacem77
Supposedly there's an Inuttitut Rosetta Stone project although I don't see much else about it online. [0]
I tried to get access to the Iñupiat Rosetta Stone at one point but failed even though I have the "unlimited" "all languages" Rosetta Stone subscription. It seems like the tribe manages registration and acts as a gatekeeper. [1] I'm sure it has to do with who funded the project or something, but you'd think for endangered languages you'd want to make it as easy as possible to access language learning resources.
The practice of drawing conclusions about people or their cultures based on properties of their language has been controversial since at least the original Sapir and Whorf stuff.
If you admire or appreciate a culture, you might argue that the culture's values or history or distinctive outlook are directly reflected in grammatical features or vocabulary of an associated language. While this might be true in some ways and to some extent, it's something that's also very easy to deceive yourself about. You can easily make up a story about how distinctions that one language does (or doesn't) draw reflect the excellence or more appropriate priorities of its speakers' culture. Like "oh, this language makes distinctions that English doesn't -- because the people who speak it are so attuned to perceiving fine details!" or "oh, this language doesn't make distinctions that English does -- showing how its speakers are not distracted by the pointless minutia that occupy us in settled industrial civilization!".
A challenge to this kind of analysis is that studies of linguistic typology (deep structural features that languages do or don't have in common) show languages with, and without, particular characteristics all over the world, in languages spoken by people from all different kinds of cultures and societies. Most generalizations that might be given a higher significance just sort of fall flat when you look at typology data. Oh, that cool thing that we like in this language -- is also present in this other language from a different culture and part of the world! Oh, that thing we find annoying or irritating -- is present elsewhere too.
A kind of counterexample that I learned from John McWhorter's books is that complicated irregularities in languages (e.g. a very large number of noun classes with distinctive classifiers and agreement rules, or a very large number of irregular verb forms) are likely a sign of smaller and more isolated speaker communities, whereas simplification that removes these irregularities is associated with circumstances in which more people have to use a language non-natively (and therefore have a harder time mastering these details), which could include migration, social disruptions, or the widespread use of a language as a lingua franca, trade language, language of imperial administration, etc. That does mean that indigenous languages (especially if they have almost exclusively been used by a native speaker community in a relatively small area) can often be harder to learn for non-native speakers.
But you still can't necessarily count on cultural generalizations based on those observations. One exercise you can try to do (maybe this is also an idea from McWhorter?) is pick a random set of characteristics that some language might have, and then try to come up with an essay explaining why those features show how awesome or admirable the speakers' culture must be. If it sounds deeply plausible, then maybe that's a sign that people aren't actually good at being objective about this. :-)
Your comment is a very good counterpoint to Sapir-Whorf-hypothesis-like opinions, but unfortunately it's also so general that it could've been posted on any article about language and culture, since you don't reference any specific content.
The submitted article doesn't talk about grammar and morphology or other typological features, so criticizing it on the grounds that there's no relationship between typology and geography is rather beside the point.
What it does talk about is specific vocabulary like amiraijaut for September, the "time when antlers lose their velvet". That's reflective of Inuttitut speakers insofar as that they must've been living close enough to animals with antlers to observe the loss of velvet at the time when that translation was coined.
That kind of connection between language and culture is rather uncontroversial. For example, it was inferred that proto-Austronesian speakers already had pigs, dogs and chickens based on the words for these animals across different branches of the language family being cognates that allowed reconstruction of those words in the proto-language. That hypothesis also squares with the archaeological evidence http://press-files.anu.edu.au/downloads/press/n2320/html/ch1...
Of course vocabulary can be learned (as the author herself demonstrates) and transmitted outside the original context (e.g. this comment), so the cultural link is rather weak. People who've never seen a velvet-covered antler can still talk about amiraijaut in the same way that people who do not worship a god of war can still talk about Mars and his month March.
Readit News
Skier’s learn a large number of words for different types of snow: everything from death cookies to boilerplate to corn snow. Vocabulary comes from interaction.
I am guessing different sports have wildly expanded vocabulary for certain elements of the relevant environment?
You're looking for the phrase "term of art"
https://www.lexico.com/definition/term_of_art
https://dictionary.cambridge.org/dictionary/english/term-of-...
e.g. if I say I'm going to "bounce the box", I don't mean that I will throw a cardboard container at the floor, I mean that I will reboot a computer. And the sysadmins will know exactly what I mean.
And it's not just sports. Meteorologists have many different words to describe clouds.
Is there a systematic campaign to preserve the language and/or community? At the very least to set down the vocabulary/grammer/pronounciation/etc?
Inuktut is mostly stable, though "vulnerable," and strongest in Nunavik (the region in the article), I suspect because the somewhat-devolved region employs it as the primary language of instruction. [1]
[0] Excerpt from the Truth and Reconciliation commission's report: http://wordalive.wycliffe.ca/stories/what-residential-school..., also https://www.clo-ocol.gc.ca/en/timeline-event/indian-resident...
[1] https://www.cbc.ca/news/canada/north/inuit-inuktitut-stats-c..., https://www.kativik.qc.ca/school-board/
I asked a question on HN for a grant that was directed about using ML to aid in language preservation. I didn't get any responses. We were trying to expand the tools available.
If you approach the problem instead from the language preservation perspective and ask what is necessary to make it succeed, there are essentially two tasks: documentation (recording current speakers' knowledge of the language) and community maintenance (allowing people to keep using the language and teaching it to the next generation).
Documentation basically boils down to making a lot of recordings. Since text is usually nicer to work with, transcriptions are very useful, and in theory ML would be able to help with that, but to train a speech-to-text model, you first need transcribed training data, leading to a chicken-and-egg problem. Making the data easily available and easy to contribute to is more likely to be impactful.
For maintaining a community, ML can be useful by providing all the services that ML offers for speakers of more popular languages: text-to-speech, speech-to-text, predictive smartphone keyboards, machine translation, ... but all of those depend on the language being well-documented already. Otherwise, non-ML solutions are likely to be the only options (so formant-based speech synthesis e.g. with espeak-ng, dumb keyboards, rule-based translators, ...)
To give a specific example, I know of a community of Kabyle speakers who want to be able to use OpenStreetMap with voice navigation in Kabyle, so they've been making a lot of recordings (using https://tatoeba.org as a platform for the documentation part) and also have someone working on a speech-synthesis model (using Mozilla's DeepSpeech, I think). AFAIK, the project is coordinated by Muhend Beqasem, whom you might want to contact if you're interested in what they're doing: https://tatoeba.org/eng/user/profile/belkacem77
I tried to get access to the Iñupiat Rosetta Stone at one point but failed even though I have the "unlimited" "all languages" Rosetta Stone subscription. It seems like the tribe manages registration and acts as a gatekeeper. [1] I'm sure it has to do with who funded the project or something, but you'd think for endangered languages you'd want to make it as easy as possible to access language learning resources.
[0] https://www.rosettastone.com/endangered/projects/ [1] http://www.north-slope.org/assets/images/uploads/RS_Request_...
https://en.wikipedia.org/wiki/Hopi_time_controversy
If you admire or appreciate a culture, you might argue that the culture's values or history or distinctive outlook are directly reflected in grammatical features or vocabulary of an associated language. While this might be true in some ways and to some extent, it's something that's also very easy to deceive yourself about. You can easily make up a story about how distinctions that one language does (or doesn't) draw reflect the excellence or more appropriate priorities of its speakers' culture. Like "oh, this language makes distinctions that English doesn't -- because the people who speak it are so attuned to perceiving fine details!" or "oh, this language doesn't make distinctions that English does -- showing how its speakers are not distracted by the pointless minutia that occupy us in settled industrial civilization!".
A challenge to this kind of analysis is that studies of linguistic typology (deep structural features that languages do or don't have in common) show languages with, and without, particular characteristics all over the world, in languages spoken by people from all different kinds of cultures and societies. Most generalizations that might be given a higher significance just sort of fall flat when you look at typology data. Oh, that cool thing that we like in this language -- is also present in this other language from a different culture and part of the world! Oh, that thing we find annoying or irritating -- is present elsewhere too.
https://en.wikipedia.org/wiki/Linguistic_typology
A kind of counterexample that I learned from John McWhorter's books is that complicated irregularities in languages (e.g. a very large number of noun classes with distinctive classifiers and agreement rules, or a very large number of irregular verb forms) are likely a sign of smaller and more isolated speaker communities, whereas simplification that removes these irregularities is associated with circumstances in which more people have to use a language non-natively (and therefore have a harder time mastering these details), which could include migration, social disruptions, or the widespread use of a language as a lingua franca, trade language, language of imperial administration, etc. That does mean that indigenous languages (especially if they have almost exclusively been used by a native speaker community in a relatively small area) can often be harder to learn for non-native speakers.
But you still can't necessarily count on cultural generalizations based on those observations. One exercise you can try to do (maybe this is also an idea from McWhorter?) is pick a random set of characteristics that some language might have, and then try to come up with an essay explaining why those features show how awesome or admirable the speakers' culture must be. If it sounds deeply plausible, then maybe that's a sign that people aren't actually good at being objective about this. :-)
The submitted article doesn't talk about grammar and morphology or other typological features, so criticizing it on the grounds that there's no relationship between typology and geography is rather beside the point.
What it does talk about is specific vocabulary like amiraijaut for September, the "time when antlers lose their velvet". That's reflective of Inuttitut speakers insofar as that they must've been living close enough to animals with antlers to observe the loss of velvet at the time when that translation was coined.
That kind of connection between language and culture is rather uncontroversial. For example, it was inferred that proto-Austronesian speakers already had pigs, dogs and chickens based on the words for these animals across different branches of the language family being cognates that allowed reconstruction of those words in the proto-language. That hypothesis also squares with the archaeological evidence http://press-files.anu.edu.au/downloads/press/n2320/html/ch1...
Of course vocabulary can be learned (as the author herself demonstrates) and transmitted outside the original context (e.g. this comment), so the cultural link is rather weak. People who've never seen a velvet-covered antler can still talk about amiraijaut in the same way that people who do not worship a god of war can still talk about Mars and his month March.
Dead Comment