Arbre Lingua
June 21, 2018 9:49 AM   Subscribe

 
Nitpick: the article makes it out to be a diagram of "the world's languages", but as the diagram itself says, it's only an illustration of the families of the Nordic languages specifically. World-tree would be much bigger and more confusing :P

I enjoy that the artist includes cats even in the tree diagram. A Redtail's Dream sounds familiar; I didn't realize it was by a Finnish artist.
posted by inconstant at 9:56 AM on June 21, 2018 [3 favorites]


Make sure to scroll all the way to the Nordic kitties. The cutest thing ever! (Though a bit wrong about Finnish, as 1) the Finnish cat should be saying miau, 2) it should not be alone! In reality it has a little Estonian friend. I SWEAR I have seen a version on Reddit where someone added an Estonian cat, but of course I can't find it now…)
posted by Vesihiisi at 9:58 AM on June 21, 2018 [2 favorites]


Also, if you like Minna Sundberg’s style, check out her other other comic, Stand Still. Stay Silent. It's rather dark at times, which is not surprising given its postapocalyptic setting*, but very beautiful.
* What's surprising is how cheerful it can be despite the setting.
posted by hat_eater at 10:15 AM on June 21, 2018 [4 favorites]


…and the Nordic Kitties make it unclear, but Finnish only is a Nordic language if you interpret "Nordic" as "spoken in a Nordic country", not in the accepted linguistic meaning, as used in the Nordic countries. I get linked to the Nordic Kitties regularly because they're right there in the intersection of my linguistic/Finnish/feline interests (I fully expect a friend to share this FPP with me!), and it bothers me every time :/
posted by Vesihiisi at 10:43 AM on June 21, 2018 [4 favorites]


Previously.
posted by languagehat at 10:56 AM on June 21, 2018 [3 favorites]


Gujarati, Rajasthani are shown as Central zone Indic languages, but I have always seen them shown belonging to the Western branch.
posted by Gyan at 10:57 AM on June 21, 2018


Curmudgeon here. Languages, such as "English", "Hindi", etc are complex entities created by nations, educations systems, socioeconomic pressures, and so on. That much is obvious.

Less obvious is that the characterisation of language as this kind of entity owes a whole lot to the industry and field work of the Summer Institute of Linguistics and the Ethnologue, from which this data is drawn.

Trouble is, these are faith-based entities, whose concern it so translate the bible as widely as possible. The reason they are so keen to document dying languages is to translate the bible into them. This has greatly coloured our very notion of what a "language" is. Scientists should approach such data repositories with great caution. Try to see, for example, what remains of the manner in which the Selknam or the Yaghan of Tierra del Fuego coordinated their affairs. You will find translations of the Gospels, of the Acts of the Apostles, and such. You will find no evidence of the manner in which their lives were conducted. The bible became a lens by which the very essence of a language was picked out. You will find little or nothing that is not presented through this distorting lens. Chant is my gig, and there is one Selknam chant left, but little of the manner in which it functioned.

This Christian agenda underpinning the science of language is not sufficiently widely known.
posted by stonepharisee at 11:40 AM on June 21, 2018 [6 favorites]


> Languages, such as "English", "Hindi", etc are complex entities created by nations, educations systems, socioeconomic pressures, and so on. That much is obvious.

Huh? It's obvious that language, like any human creation, is not independent of the rest of human culture. The rest of what you're saying is absolute balderdash, unless I'm misunderstanding you. The classification of languages into families has nothing to do with how lives are conducted, nor should it. It's a matter of consistent relations of phonemes and morphemes. That has nothing to do with "faith-based entities," it's science, and I'm not sure what you're trying to prove.
posted by languagehat at 11:51 AM on June 21, 2018 [6 favorites]


Ah languagehat, nice to see you. I'm not spouting bollocks. The SIL was founded in 1934 by missionaries. It publishes Ethnologue. That which counts as a language, that is to say the unit employed there, is that into which the bible can be translated. So if you hear that Papua has 1,345 languages, for example, that means that it would require that many distinct translations to get the bible across to everyone.

This complex intertwining of religious missionary and our understanding of language is important. When you lean on such distinctions as phonemes and morphemes, you are leaning on an intellectual construction that has been informed, inter alia, by this missionary activity that now puts itself forward as a scientific resource.

The old adage about a language being a dialect with an army and a navy is good, but does not go nearly far enough. The sciences of language have been built from a literate European Christian perspective. It shows everywhere. The IPA, for example, aspired to providing a structure within which the sound contrasts of all languages could be put, just as elements can be ordered in the periodic table. But unlike chemical elements, it turns out that the further you move from a European base, the less the system fits.

There is a whole lot of language missed if you use translatability and the bible as your yardstick. If you're interested, here's a talk I gave 2 weeks ago on the matter. It misses chant, joint speech, gesture, prosody, gaze, all kinds of things. I criticise the sciences of language from the inside, not from some random point.
posted by stonepharisee at 12:09 PM on June 21, 2018 [2 favorites]


If we want to get into the nitty-gritty, consistent relations of phonemes and morphemes are a signal of relatedness, but this type of classification specifically focuses on historical descent. Like evolutionary "family trees" and species definitions in biology, it's not always a clear-cut thing -- languages from different "branches" might intermingle and influence one another to a huge extent, "speciation" is a continuum rather than a binary toggle, and current linguistic science is still unable to discern the relations of many languages (the Australian languages being a really big example).

That being said, Ethnologue as a specific institution is hardly the progenitor of linguistics. Bit like attributing the idea of the nation-state to NATO or something.
posted by inconstant at 12:23 PM on June 21, 2018 [4 favorites]


But the family tree image employed evokes associations of biological growth, and natural development. The leaf labelled English, on the other hand, carries with it all the practices and politics of the British Empire. By virtue of what do we call the speech of Prince Charles proper English, but the speech of almost every PoC in the US a dialect, or a variant. It is important not to wash the politics out of the phenomenon.
posted by stonepharisee at 12:30 PM on June 21, 2018 [1 favorite]


I'd love to see the first version of this with more root bases, this seems mostly focused on the left half of the mega landmass we gave too many different continental designations too.
posted by GoblinHoney at 12:52 PM on June 21, 2018 [2 favorites]


This diagram says absolutely nothing about dialects or variants or properness.

GoblinHoney, it's focused that way because it's only looking at the families that contain the "Nordic languages" (Finnish included, though Vesihiisi has mentioned caveats with that). That's why it's so inaccurate to call it a diagram of "world languages".
posted by inconstant at 1:06 PM on June 21, 2018 [1 favorite]


I just really want this and the Nordic Kitties as posters. I love beautiful art and language, combining the two? Fantastic.
posted by angeline at 1:22 PM on June 21, 2018 [1 favorite]




> The SIL was founded in 1934 by missionaries. It publishes Ethnologue.

Yes, I'm very aware of that. Like inconstant says, Ethnologue is pretty peripheral to linguistics, and it's odd and fighty to act as if all of historical linguistics is somehow discredited because there are some missionaries counting languages.
posted by languagehat at 2:20 PM on June 21, 2018 [5 favorites]


> f you're interested, here's a talk I gave 2 weeks ago on the matter. It misses chant, joint speech, gesture, prosody, gaze, all kinds of things. I criticise the sciences of language from the inside, not from some random point.

I watched the start of it, up to the point where you claim that structuralism, starting with Saussure, was "heavily biased in favor of Indo-European languages" and that Saussure and Chomsky display a "frightening disregard for the primary data." The former is complete nonsense (the structuralist approach has been successfully applied to languages from everywhere; Bloomfield famously used it for Algonquian languages), and the crack about disregard for the data is true of Chomsky but not of Saussure or any of the great linguists who followed in his footsteps. To claim otherwise is either dishonest or ignorant.
posted by languagehat at 2:30 PM on June 21, 2018 [5 favorites]


It's from a comic book set in a post-apocalyptic world where only Nordic countries have survivors. Hence the references on the tree illustration to the "old world" and "year zero". And in that context, it is refreshingly accurate with respect to linguistics.

Also it is awesome because Minna Sundberg is like the best cat illustrator currently alive. I have both the language tree and Swadesh list cat illustrations as framed posters : )

And angeline, they are available to buy as posters here: https://hivemill.com/collections/stand-still-stay-silent
posted by lefty lucky cat at 2:51 PM on June 21, 2018 [1 favorite]


There is a whole lot of language missed if you use translatability and the bible as your yardstick.

The importance that you ascribe to SIL and Ethnologue in the development of linguistic theory (and historical linguistics) is odd. They're not nearly as fundamental as you seem to think. But there's also a missing part of the argument: How the theoretical concepts that linguists use to describe historical relationships don't work for non-European languages.

I work on a language spoken in West Africa, using primary data that I collect myself. The language is unwritten and some of the people that I work with are not literate in any language. The challenges for structuralism that this language poses are largely the same ones that European languages do. The phoneme is just as relevant - and it would be very difficult to describe the language without it. So, when you say that the phoneme works less well the farther from Europe you go, my eyebrows raise.

The obvious retort is that I'm just blinkered by the theoretical tradition that I come from. But I come face-to-face with the limits of structuralism every day in my research; I'm well aware of its limitations. My specialty is prosody, where structuralist and non-structuralist approaches are both common, and where neither really works.

I mean, it's absolutely true that structuralism can't describe all aspects of a language. It's also absolutely true that the contents of descriptive grammars are biased toward those aspects of the language that can easily be described in structural terms. However, limited in scope isn't the same as wrong. The fact that descriptive grammars generally neglect prosody doesn't mean that the phoneme doesn't exist, or that the comparative method isn't applicable to this family of languages.
posted by Kutsuwamushi at 3:35 PM on June 21, 2018 [8 favorites]


Hence the references on the tree illustration to the "old world" and "year zero".

It seems there are a couple of year reckoning systems in which Year Zero would be what is more traditionally referred to as 1 B.C. (Gregorian or Julian), if anyone else was wondering. Also, "Hindu and Buddhist calendars" apparently count years starting from zero, as would many computer programming languages, but I guess the Keplerian astronomical year is the more likely referent.
posted by sfenders at 4:43 PM on June 21, 2018


The referent is the web comic Stand Still. Stay Silent which these illustrations are taken from. The Old World is our present world, and Year Zero is the year when a disease kills off most of humanity. The main narrative begins in Year 90, after the survivors emerge from a period of quarantine into a radically changed world.
posted by lefty lucky cat at 4:59 PM on June 21, 2018


Ah, right. That would explain how "English" and etc. were around in year zero.
posted by sfenders at 5:18 PM on June 21, 2018


Semi relatedly, SSSS is nearing completion-- there's only one more chapter yet to go (in this story anyhow), so for those of you who haven't tried reading it for that reason, you might start now. Maybe you'll catch up by the time it finishes.
posted by nat at 6:00 PM on June 21, 2018


In these dark times, a MeFi post like this is worth its weight in gold many times over: Minna Sundberg, cats, trees of language families, and all the linguistics experts on the site arguing like cats about Very Inside Linguistic Baseball stuff much more heatedly than a casual reader might necessarily notice. This is basically my idea of a good time. (I'm a crappy date, sorry.)

Also there *are* no casual readers here :)
posted by motty at 7:15 PM on June 21, 2018 [2 favorites]


To add a tangent to the structuralism debate, structuralism is not limited to language but is a far broader thing, and structural anthropology was extremely impactful. Meaning that on this broader level too it has never been limited to Europe / the West / Indo-European anything. Levi-Strauss for example conducted his research among South American peoples.
posted by Pyrogenesis at 12:05 AM on June 22, 2018 [1 favorite]


If we want to get into the nitty-gritty, consistent relations of phonemes and morphemes are a signal of relatedness, but this type of classification specifically focuses on historical descent. Like evolutionary "family trees" and species definitions in biology, it's not always a clear-cut thing -- languages from different "branches" might intermingle and influence one another to a huge extent, "speciation" is a continuum rather than a binary toggle, and current linguistic science is still unable to discern the relations of many languages (the Australian languages being a really big example).

I think a lot of the cutting edge work focuses more on Syntactic structures (assuming that something like parameters have a cognitive reality), (see the work of Longobardi et al.). The main argument is that phoneme inventories are superficial indicators of a language's 'phylogenetic' heritage, easily borrowed and passed around. (It has been pointed out that using phonemic inventories to make this point is probably not very helpful, and that phonotactic structures would probably be more resilient. )

The point is that syntactic structures are more robust across generations and a better indicator of how language's relate diachronically.

As to the other debate in this thread, while I do think SIL deserves a great deal of side-eye I am not sure its fair to question their data out-of-hand. Ultimately, I do not think that the connectionest viewpoint which denies the existence of a mind can ever be reconciled with a theory of generative grammar, which I suspect is at the root of this 'cat fight.' Personally I find the arguments presented against the usefulness of the IPA for non Indo-European languages unconvincing. I am on board with the idea that presenting 'English' or 'Spanish' as a monolithic bloc is not useful for a great many linguistic endeavors, but not here. The notion of an E-language like 'English' is a useful fiction in many cases, especially here. It has nothing to do with I-language and shouln't be conflated with such -- and the same goes for any extra-linguistic behavior such as gesture, chant, etc.
posted by os tuberoes at 1:05 AM on June 22, 2018


Phoneme inventories aren't used for the determination of historical relatedness; systematic correspondences between phonemes in the lexicon are. That's why some Australian languages with nearly identical inventories are considered unrelated, while languages like English and Russian are considered related. As far as I'm aware, no one has made a convincing argument that systematic correspondences can be borrowed, short of a massive replacement of the lexicon - which can happen, but is rare.

A lot of the cutting edge work in historical linguistics doesn't use the comparative method because the comparative method has been mined already. It's been around for over a hundred years at this point. It's shown itself to be quite robust in that time, but it has limitations (i.e. systematic correspondences being obscured by various types of change). And of course there are interesting questions about change outside of phonology.

It's still being used as new data comes out (e.g. with the Dogon family in Mali), or to revisit preliminary classifications of understudied languages (e.g. many of the families in Africa), though.
posted by Kutsuwamushi at 9:31 AM on June 22, 2018 [2 favorites]


« Older Radicalism In The Ranks   |   Effects of Traumatic Separation on Children Newer »


This thread has been archived and is closed to new comments