10 ms·

The fifty shades of Latin

131 points by drjohnson 9y ago

schoen 9y ago
If you'd like to see a few of the lesser-known Romance languages in action (especially from Italy, because that's apparently a strength of this company -- I think they're based there, judging by their motto Non solo parole 'not just words'), take a look around Logos Quotes, where they used to translate a different quotation every day into as many languages as they could manage. There are a LOT of Romance languages out there! http://www.logosquotes.org/ http://www.logosquotes.org/
- JoeDaDude 9y ago
  Italian itself has an interesting spectrum of shades seeing as how it was "standardized" a relatively short time ago (mid 19th century), at the same time there was massive emigration from the country, with the result that the Italian spoken by the expats is quite different from the standard Italian. [0] http://www.atlasobscura.com/articles/how-capicola-became-gabagool-the-italian-new-jersey-accent-explained http://www.atlasobscura.com/articles/how-capicola-became-gab... [1] https://news.ycombinator.com/item?id=10523661 https://news.ycombinator.com/item?id=10523661
  - toyg 9y ago
    And because of its geographical position and history, indigenous languages show influences coming from worlds as geographically distant as Arabia and China.
samch 9y ago
From my dad, linguist: "This is the classic example of an L-complex. A chain of mutual intelligibility can be established from Normandy to Sicily and over to Portugal, and in fact across all of Latin America, though Brazil is linked through Portugal and the rest of Latin America through Spain. An L-simplex is a grouping of dialects that are all mutually intelligible. In an L-complex there may be dialects that are not mutually intelligible. But on occasion you find a maximal L-complex that is also an L-simplex. Israel is one. The Basque country is another. But it's rare."
- unscaled 9y ago
  Interesting terms. I've never came across them in my linguistic study, but I see they were heavily employed by C.F. Hockett, so that might be less exposure to American Structuralism. I want to make a small correction here. Israel is a special case not because all regional dialects are mutually intelligible, but because it has no regional dialects at all (unless you're talking about Arabic, but then it's a far cry from an L-Simplex). This is due to the history of Modern Hebrew in Israel being relatively recent. It only reached a stable grammar with a sizable generation of native speakers in the 1920s and 30s, and until the 60s at least non-native speakers probably outnumbered the native speakers. By then Television was unleashed, and killed whatever budding dialects that existed. Practically speaking, there used to few minor dialectal variations in Jerusalem. It pretty much amounted to about 20-30 words, a default feminine form for the noun 'cat' and a different pronunciation for the number 200. But even these features are barely left in Jerusalem, which had seen large population exchanges.
panglott 9y ago
The term of art is "language shift", but it's nearly always a case of internal colonialism. Nearly any time you see a linguistic minority abandon its language for the majority language, you can bet that the government was taking kids away from their parents and beating and shaming them if they dared utter the language their parents spoke to them as babies. Occitan in 1860 was spoken by 39% of the population of France, and now has about 100k speakers. https://en.wikipedia.org/wiki/Vergonha https://en.wikipedia.org/wiki/Vergonha
- umanwizard 9y ago
  This is true sometimes but not "nearly any time". For example rate of Spanish fluency in Arizona among second-generation residents is much lower than among their parents, without any kidnapping and beating going on.
  - PhasmaFelis 9y ago
    Certainly plenty of shaming, though, if you've seen how many Arizonans feel about Hispanics, and heard the rhetoric about immigrants who "refuse to assimilate" (by failing to act, dress, and speak exactly like the locals).
  - panglott 9y ago
    The situation can be a bit different with 2nd generation immigrants, especially based on the parents' attitudes towards English. I do see occasional reports in the US of punishing and shaming students for speaking non-English languages, though, and many Americans are very hostile to any accommodation of languages other than English (Spanish signage, "press 1 for English"). And there is a clear, unstated language policy in the US of making immigrant children into English speakers (with no regard to preserving/increasing their mother-tongue proficiency), as well as a general policy of making everyone English-language monolinguals (if we wanted bilingual kids we would have bilingual elementary schools). Minority-language communities really refers to communities where there is a long, stable history of intergenerational language transmission, and where functional bilingualism is possible or likely. Such as any Native American language, or something like Louisiana French. A sane educational policy in this case would be something like "most instruction is conducted in the majority language, but kids aren't punished or shamed for speaking their mother tongue at recess, and there's maybe even an occasional minorty-language literature class".
data_hope 9y ago
I wonder how different / similar these romance languages are, compared to the reference frame I have: German dialects. German dialects can be mutually unintelligible, young germans typically know standard german and thus have a "common ground" for communication, also they usually speak a form of the dialect that is already considerably closer to the standard "high" language of newspapers and televisions, than what their grandparents or their great grandparents speak / spoke. Sometimes (typically in documentaries), they even subtitle dialect speakers. So yeah, I wonder if depending on the context, the classification of languages and dialects differs.
- XaspR8d 9y ago
  Well there's a notorious adage, "a language is a dialect with an army and a navy". I think academics shy away from attempting to make the distinction except when extremely obvious, and instead talk directly about quantitative measurements and feature overlaps (isogloss is a search term that may be useful here). Dialect/language lines will often have completely different shapes when you look at different distinctions in lexicon, phonetics, syntax, etc. If I had to generalize though, in a particular language "chain", linguists seem to identify an order of magnitude more separable languages than non-academics do. (Consider the cases of huge macrolanguages like "Chinese" & "Arabic", or even "Italian", whose singular labels by laypeople are pretty universally rejected.) It doesn't help that people are generally unaware of the incredible political pressure most nations put on presenting a singular linguistic front, when the truth is much much more muddled. As a result, the common parlance distinction between dialect & language often verges on meaningless.
- jackjeff 9y ago
  My experience is that it's possible for people to understand each other. But the farther you go and the more difficult it gets. It's mostly the accent and the word endings that change. My family is from Aveyron/Tarn (near Albi). We can understand texts from Frederic Mistral, written in Provençal (near Marseille) even though it sounds weird. My uncle says he had some success speaking Occitan in the Italian Piedmomd. However neither my parents nor my uncle understand any of the Catalan spoken in Barcelona. (I do, but I'm fluent in Spanish and not in Occitan...) I think my parents (born in the 1950s) are the last generation fluent in Occitan. In France, even though it's now being taught as a second language, it's essentially gone. My mom told me she used to be punished for using Occitan at school whether in the classroom or during recess. I remember when I was a child, the farmers used to speak it among themselves (or more likely to their elders). The same people today only really speak French, even among themselves.
gumby 9y ago
This variation was quite common in France until the commencement of an active program of linguistic assimilation (beginning with Luis XIII but really accelerating in the 1930s). Outside the Romance Domain, Britain held out well into the 50s (I still remember being shocked by the Yorkshire "thou" as a kid) but TV wiped it out -- the accents remain but the grammar and, mostly, the vocabulary, have all converged. Ironically I hear much more variation in Germany where there is an "offical" German that everybody learns at school but almost nobody speaks at home. America and Australia are, to my ear, in the same boat as the UK: accents abound but if you were to transcribe people's speech they would all appear to be Californian. That's definitely due to TV and the fact that US/UK TV doesn't need translation. After decades in the US I go back to Australia and people tell me, "wow, you've really kept your accent." Actually, it's just that they're very much used to hearing US speech on TV.
- dom0 9y ago
  > Ironically I hear much more variation in Germany where there is an "offical" German that everybody learns at school but almost nobody speaks at home. Nope.
  - gumby 9y ago
    I don't know what you mean by "nope." Certainly there are grammatical differences between Hochdeutsch, Plattdeutsch, Allemanisch, (much less the varieties spoken in Switzerland!) in addition to accent and (minor) vocabulary differences so I presume you aren't disagreeing with that. I have been to plays in Munich that were explicitly in Bayerisch and were deliberately incomprehensible (have had the same "problem" in Edinburgh). So I doubt you disagree with that. Do you disagree that people use Hochdeutsch at home? All I can say is that my kid grew up speaking Hochdeutsch at home because his mother (from south of Hannover, near Goslar) speaks it in the house, as do the rest of her family, neighbors, etc so it's what he learnt when learning to speak. In Berlin and in the overseas German School the kids all commented that he was the one who didn't have to learn the accent (in the overseas school it was very clear that all the kids had different accents).
    - 9y ago
vondur 9y ago
All of these various languages that sprang fort from Latin after the dissolution of the Roman Empire illustrate just how much the former Roman world shrank into small kingdoms.
fortes 9y ago
As a native Portuguese speaker, it's fun to look at the dialectic continuum you can construct as you go across the Iberian peninsula. It's imperfect, but you can travel with a very soft "language gradient": Extremaduran -> Portuguese -> Galician -> Asturian -> Castillian (Spanish) -> Valencian -> Catalan -> Aragones -> Occitan -> French ... Each individual step is pretty easy, but it adds up to a huge difference. Unfortunately, there isn't a great analogue for English. You can kinda construct a small jump by looking at Scots, but it ends quickly and is (at least to my ears) a jump on par with Portuguese to Galician.
- dangerbird2 9y ago
  English is member of a language continuum, albeit a disjointed one by virtue of the English Channel and the Norman conquest. The Ingvaeonic languages is a "gradient" including Scots, English, Frisian, and Low German/Plattdüütsch. Frisian is certainly legible to an Anglophone with a wide exposure to English dialects and historical periods. https://en.wikipedia.org/wiki/Ingvaeonic_languages https://en.wikipedia.org/wiki/Ingvaeonic_languages
  - jacobush 9y ago
    Don't forget https://en.wikipedia.org/wiki/Old_Norse https://en.wikipedia.org/wiki/Old_Norse and its influence on English. (I say) some of the contemporary Norwegian is very close to some contemporary Scots.
  - TheOtherHobbes 9y ago
    The English continuum also includes pidgins of various kinds, and Engrish.
- personlurking 9y ago
  After years in Portugal I still can't differentiate where anyone is from within the country, but put me a Brazilian event and I can tell you where each person is from in Brazil (where I also lived) by accent and word-choice alone. I'm impressed by my lack of ability when it comes to Portugal. Perhaps it's more nuanced. However, the language gradient you speak of is one of the reasons I love Iberia. If only modern English, as you note, would be so interesting. By the way, the region the author speaks of is beautiful, in case anyone plans to take a lazy drive through it. If yes, don't forget Andorra.
gabaix 9y ago
Growing up in the French side of the Pyrenees, I concur with the article. Traveling on the other side of the border was barely special. The road signs would have a dialect very similar to our own Gascon, and there was no need to switch to Spanish. One thing not mentioned in this article is how much these dialects have been losing ground over the past 50 years. My grand-mother spoke almost exclusively in her local dialect (Bearnais), my mom spoke it with my dad when they didn't want me to understand, and I barely know a few words. There are schools and local initiatives to revive the old languages. The reality is that less and less people speak the dialects. Most people stick to Spanish, French, and are now expected to know English well.
umanwizard 9y ago
The traditional languages of France are basically an infinite fractal to the extent that many of the forms are basically unheard of outside of the areas in which they're spoken. I lived in a (largely rural) department of France called Mayenne for a year. Old people there universally believe that they grew up speaking something called "patois mayennais" (let's translate this as "Mayennese dialect") before they had to learn French to get jobs in the city. However, I can find very little reliable information about this Mayennese dialect on the internet, in French or English. There are a few "local color" type newspaper articles that reference it, but that's about it. French dialect maps I've seen don't mention Mayennese dialect at all, and claim that Mayenne spoke either Gallo or Angevin.
- schoen 9y ago
  French Wikipedia seems to have an article about it https://fr.wikipedia.org/wiki/Mayennais https://fr.wikipedia.org/wiki/Mayennais including some purported vocabulary, plus a few books that discuss it! Some of them discuss it as "bas-mainiot" or "patois du Bas-Maine", of which Wikipedia says "Il [Département de la Mayenne] correspond essentiellement au Bas-Maine, qui formait la moitié occidentale de la province du Maine"... I guess you're probably very familiar with these geographical designations, but I'm definitely not! Google has a scan of a 1975 reprint of an 1899 book about "les parlers du Bas-Maine" which includes tons of references to speakers and lexical items observed in Mayenne. https://books.google.fr/books?id=mcykCmekgioC https://books.google.fr/books?id=mcykCmekgioC (It's sad that some pages are missing from the preview... maybe Google algorithmically concluded that the book might still be in copyright because it was "published" in 1975.) But it's true that a Wikipedia article isn't the same thing as finding a thriving language community online.
  - umanwizard 9y ago
    Not sure how I missed that Wikipedia article... thanks for digging up those references!
microcolonel 9y ago
It would be interesting to construct a many-dimensional language map of vectors. You could add two spherical (or three cartesian!) dimensions to your word vectors, and interpolate languages based on the known data. Computational interpolation between english dialects, or chinese languages, or indian languages, sounds like a heck of a lot of fun.
- schoen 9y ago
  Or extrapolation! https://en.wikipedia.org/wiki/Brithenig https://en.wikipedia.org/wiki/Brithenig (although that was manual, not computational)
  - microcolonel 9y ago
    And actually, with good enough data, you could probably shove in some text and figure out where it was written (or at least where the writer learned to write) within pretty good margins.

131 points by drjohnson 9y ago

If you'd like to see a few of the lesser-known Romance languages in action (especially from Italy, because that's apparently a strength of this company -- I think they're based there, judging by their motto Non solo parole 'not just words'), take a look around Logos Quotes, where they used to translate a different quotation every day into as many languages as they could manage. There are a LOT of Romance languages out there! http://www.logosquotes.org/ http://www.logosquotes.org/

JoeDaDude 9y ago
Italian itself has an interesting spectrum of shades seeing as how it was "standardized" a relatively short time ago (mid 19th century), at the same time there was massive emigration from the country, with the result that the Italian spoken by the expats is quite different from the standard Italian. [0] http://www.atlasobscura.com/articles/how-capicola-became-gabagool-the-italian-new-jersey-accent-explained http://www.atlasobscura.com/articles/how-capicola-became-gab... [1] https://news.ycombinator.com/item?id=10523661 https://news.ycombinator.com/item?id=10523661
- toyg 9y ago
  And because of its geographical position and history, indigenous languages show influences coming from worlds as geographically distant as Arabia and China.

samch 9y ago

From my dad, linguist: "This is the classic example of an L-complex. A chain of mutual intelligibility can be established from Normandy to Sicily and over to Portugal, and in fact across all of Latin America, though Brazil is linked through Portugal and the rest of Latin America through Spain. An L-simplex is a grouping of dialects that are all mutually intelligible. In an L-complex there may be dialects that are not mutually intelligible. But on occasion you find a maximal L-complex that is also an L-simplex. Israel is one. The Basque country is another. But it's rare."

unscaled 9y ago
Interesting terms. I've never came across them in my linguistic study, but I see they were heavily employed by C.F. Hockett, so that might be less exposure to American Structuralism. I want to make a small correction here. Israel is a special case not because all regional dialects are mutually intelligible, but because it has no regional dialects at all (unless you're talking about Arabic, but then it's a far cry from an L-Simplex). This is due to the history of Modern Hebrew in Israel being relatively recent. It only reached a stable grammar with a sizable generation of native speakers in the 1920s and 30s, and until the 60s at least non-native speakers probably outnumbered the native speakers. By then Television was unleashed, and killed whatever budding dialects that existed. Practically speaking, there used to few minor dialectal variations in Jerusalem. It pretty much amounted to about 20-30 words, a default feminine form for the noun 'cat' and a different pronunciation for the number 200. But even these features are barely left in Jerusalem, which had seen large population exchanges.

panglott 9y ago

The term of art is "language shift", but it's nearly always a case of internal colonialism. Nearly any time you see a linguistic minority abandon its language for the majority language, you can bet that the government was taking kids away from their parents and beating and shaming them if they dared utter the language their parents spoke to them as babies. Occitan in 1860 was spoken by 39% of the population of France, and now has about 100k speakers. https://en.wikipedia.org/wiki/Vergonha https://en.wikipedia.org/wiki/Vergonha

umanwizard 9y ago
This is true sometimes but not "nearly any time". For example rate of Spanish fluency in Arizona among second-generation residents is much lower than among their parents, without any kidnapping and beating going on.
- PhasmaFelis 9y ago
  Certainly plenty of shaming, though, if you've seen how many Arizonans feel about Hispanics, and heard the rhetoric about immigrants who "refuse to assimilate" (by failing to act, dress, and speak exactly like the locals).
- panglott 9y ago
  The situation can be a bit different with 2nd generation immigrants, especially based on the parents' attitudes towards English. I do see occasional reports in the US of punishing and shaming students for speaking non-English languages, though, and many Americans are very hostile to any accommodation of languages other than English (Spanish signage, "press 1 for English"). And there is a clear, unstated language policy in the US of making immigrant children into English speakers (with no regard to preserving/increasing their mother-tongue proficiency), as well as a general policy of making everyone English-language monolinguals (if we wanted bilingual kids we would have bilingual elementary schools). Minority-language communities really refers to communities where there is a long, stable history of intergenerational language transmission, and where functional bilingualism is possible or likely. Such as any Native American language, or something like Louisiana French. A sane educational policy in this case would be something like "most instruction is conducted in the majority language, but kids aren't punished or shamed for speaking their mother tongue at recess, and there's maybe even an occasional minorty-language literature class".

data_hope 9y ago

I wonder how different / similar these romance languages are, compared to the reference frame I have: German dialects. German dialects can be mutually unintelligible, young germans typically know standard german and thus have a "common ground" for communication, also they usually speak a form of the dialect that is already considerably closer to the standard "high" language of newspapers and televisions, than what their grandparents or their great grandparents speak / spoke. Sometimes (typically in documentaries), they even subtitle dialect speakers. So yeah, I wonder if depending on the context, the classification of languages and dialects differs.

XaspR8d 9y ago
Well there's a notorious adage, "a language is a dialect with an army and a navy". I think academics shy away from attempting to make the distinction except when extremely obvious, and instead talk directly about quantitative measurements and feature overlaps (isogloss is a search term that may be useful here). Dialect/language lines will often have completely different shapes when you look at different distinctions in lexicon, phonetics, syntax, etc. If I had to generalize though, in a particular language "chain", linguists seem to identify an order of magnitude more separable languages than non-academics do. (Consider the cases of huge macrolanguages like "Chinese" & "Arabic", or even "Italian", whose singular labels by laypeople are pretty universally rejected.) It doesn't help that people are generally unaware of the incredible political pressure most nations put on presenting a singular linguistic front, when the truth is much much more muddled. As a result, the common parlance distinction between dialect & language often verges on meaningless.
jackjeff 9y ago
My experience is that it's possible for people to understand each other. But the farther you go and the more difficult it gets. It's mostly the accent and the word endings that change. My family is from Aveyron/Tarn (near Albi). We can understand texts from Frederic Mistral, written in Provençal (near Marseille) even though it sounds weird. My uncle says he had some success speaking Occitan in the Italian Piedmomd. However neither my parents nor my uncle understand any of the Catalan spoken in Barcelona. (I do, but I'm fluent in Spanish and not in Occitan...) I think my parents (born in the 1950s) are the last generation fluent in Occitan. In France, even though it's now being taught as a second language, it's essentially gone. My mom told me she used to be punished for using Occitan at school whether in the classroom or during recess. I remember when I was a child, the farmers used to speak it among themselves (or more likely to their elders). The same people today only really speak French, even among themselves.

gumby 9y ago

This variation was quite common in France until the commencement of an active program of linguistic assimilation (beginning with Luis XIII but really accelerating in the 1930s). Outside the Romance Domain, Britain held out well into the 50s (I still remember being shocked by the Yorkshire "thou" as a kid) but TV wiped it out -- the accents remain but the grammar and, mostly, the vocabulary, have all converged. Ironically I hear much more variation in Germany where there is an "offical" German that everybody learns at school but almost nobody speaks at home. America and Australia are, to my ear, in the same boat as the UK: accents abound but if you were to transcribe people's speech they would all appear to be Californian. That's definitely due to TV and the fact that US/UK TV doesn't need translation. After decades in the US I go back to Australia and people tell me, "wow, you've really kept your accent." Actually, it's just that they're very much used to hearing US speech on TV.

dom0 9y ago
> Ironically I hear much more variation in Germany where there is an "offical" German that everybody learns at school but almost nobody speaks at home. Nope.
- gumby 9y ago
  I don't know what you mean by "nope." Certainly there are grammatical differences between Hochdeutsch, Plattdeutsch, Allemanisch, (much less the varieties spoken in Switzerland!) in addition to accent and (minor) vocabulary differences so I presume you aren't disagreeing with that. I have been to plays in Munich that were explicitly in Bayerisch and were deliberately incomprehensible (have had the same "problem" in Edinburgh). So I doubt you disagree with that. Do you disagree that people use Hochdeutsch at home? All I can say is that my kid grew up speaking Hochdeutsch at home because his mother (from south of Hannover, near Goslar) speaks it in the house, as do the rest of her family, neighbors, etc so it's what he learnt when learning to speak. In Berlin and in the overseas German School the kids all commented that he was the one who didn't have to learn the accent (in the overseas school it was very clear that all the kids had different accents).
  - 9y ago

vondur 9y ago

All of these various languages that sprang fort from Latin after the dissolution of the Roman Empire illustrate just how much the former Roman world shrank into small kingdoms.

fortes 9y ago

As a native Portuguese speaker, it's fun to look at the dialectic continuum you can construct as you go across the Iberian peninsula. It's imperfect, but you can travel with a very soft "language gradient": Extremaduran -> Portuguese -> Galician -> Asturian -> Castillian (Spanish) -> Valencian -> Catalan -> Aragones -> Occitan -> French ... Each individual step is pretty easy, but it adds up to a huge difference. Unfortunately, there isn't a great analogue for English. You can kinda construct a small jump by looking at Scots, but it ends quickly and is (at least to my ears) a jump on par with Portuguese to Galician.

dangerbird2 9y ago
English is member of a language continuum, albeit a disjointed one by virtue of the English Channel and the Norman conquest. The Ingvaeonic languages is a "gradient" including Scots, English, Frisian, and Low German/Plattdüütsch. Frisian is certainly legible to an Anglophone with a wide exposure to English dialects and historical periods. https://en.wikipedia.org/wiki/Ingvaeonic_languages https://en.wikipedia.org/wiki/Ingvaeonic_languages
- jacobush 9y ago
  Don't forget https://en.wikipedia.org/wiki/Old_Norse https://en.wikipedia.org/wiki/Old_Norse and its influence on English. (I say) some of the contemporary Norwegian is very close to some contemporary Scots.
- TheOtherHobbes 9y ago
  The English continuum also includes pidgins of various kinds, and Engrish.
personlurking 9y ago
After years in Portugal I still can't differentiate where anyone is from within the country, but put me a Brazilian event and I can tell you where each person is from in Brazil (where I also lived) by accent and word-choice alone. I'm impressed by my lack of ability when it comes to Portugal. Perhaps it's more nuanced. However, the language gradient you speak of is one of the reasons I love Iberia. If only modern English, as you note, would be so interesting. By the way, the region the author speaks of is beautiful, in case anyone plans to take a lazy drive through it. If yes, don't forget Andorra.

gabaix 9y ago

Growing up in the French side of the Pyrenees, I concur with the article. Traveling on the other side of the border was barely special. The road signs would have a dialect very similar to our own Gascon, and there was no need to switch to Spanish. One thing not mentioned in this article is how much these dialects have been losing ground over the past 50 years. My grand-mother spoke almost exclusively in her local dialect (Bearnais), my mom spoke it with my dad when they didn't want me to understand, and I barely know a few words. There are schools and local initiatives to revive the old languages. The reality is that less and less people speak the dialects. Most people stick to Spanish, French, and are now expected to know English well.

umanwizard 9y ago

The traditional languages of France are basically an infinite fractal to the extent that many of the forms are basically unheard of outside of the areas in which they're spoken. I lived in a (largely rural) department of France called Mayenne for a year. Old people there universally believe that they grew up speaking something called "patois mayennais" (let's translate this as "Mayennese dialect") before they had to learn French to get jobs in the city. However, I can find very little reliable information about this Mayennese dialect on the internet, in French or English. There are a few "local color" type newspaper articles that reference it, but that's about it. French dialect maps I've seen don't mention Mayennese dialect at all, and claim that Mayenne spoke either Gallo or Angevin.

schoen 9y ago
French Wikipedia seems to have an article about it https://fr.wikipedia.org/wiki/Mayennais https://fr.wikipedia.org/wiki/Mayennais including some purported vocabulary, plus a few books that discuss it! Some of them discuss it as "bas-mainiot" or "patois du Bas-Maine", of which Wikipedia says "Il [Département de la Mayenne] correspond essentiellement au Bas-Maine, qui formait la moitié occidentale de la province du Maine"... I guess you're probably very familiar with these geographical designations, but I'm definitely not! Google has a scan of a 1975 reprint of an 1899 book about "les parlers du Bas-Maine" which includes tons of references to speakers and lexical items observed in Mayenne. https://books.google.fr/books?id=mcykCmekgioC https://books.google.fr/books?id=mcykCmekgioC (It's sad that some pages are missing from the preview... maybe Google algorithmically concluded that the book might still be in copyright because it was "published" in 1975.) But it's true that a Wikipedia article isn't the same thing as finding a thriving language community online.
- umanwizard 9y ago
  Not sure how I missed that Wikipedia article... thanks for digging up those references!

microcolonel 9y ago

It would be interesting to construct a many-dimensional language map of vectors. You could add two spherical (or three cartesian!) dimensions to your word vectors, and interpolate languages based on the known data. Computational interpolation between english dialects, or chinese languages, or indian languages, sounds like a heck of a lot of fun.

schoen 9y ago
Or extrapolation! https://en.wikipedia.org/wiki/Brithenig https://en.wikipedia.org/wiki/Brithenig (although that was manual, not computational)
- microcolonel 9y ago
  And actually, with good enough data, you could probably shove in some text and figure out where it was written (or at least where the writer learned to write) within pretty good margins.