Rivka Galchen’s New Yorker essay on the famous mathematical genius and dropout Alexander Grothendieck (May 9, 2022; archived) is absolutely fascinating and makes me wish I could retrieve my math-major mind of a half-century back so I could understand more about his contributions. But what makes me bring it here is this sentence:
Grothendieck’s discoveries opened up mathematics in a way that was analogous to how Wittgenstein (and Saussure) changed our views of language.
I just… I mean, it’s like writing “how Bergson (and Einstein) changed our views of physics.” For certain values of “our” it may make sense, but those values do not include actual mathematicians and physicists. It continually astonishes me how vanishingly small is the amount people who have not studied linguistics know about language and the science that deals with it.
One of AG’s accomplishments was to formulate a good notion of what a `family of things’ (for example colors or vowels; see perhaps
https://nilesjohnson.net/hopf.html
for a more formal example (a family, or `fiber bundle’, of circles parametrized by a sphere). This may be more a matter of logic or cog sci than linguistics, adjacent to Bill Clinton’s insightful “it depends on what the meaning of `is’ is” … but such questions were of interest to Sapir, Whorf, Saussure…
I suppose it depends who the ‘we’ in ‘our’ is. In the contexts of mathematics and philosophy, Wittgenstein probably was more influential on ‘our views of language’ than Saussure: they simply understand the concept of a ‘language’ differently to us.
Wittgenstein’s name seems to appear as a sort of reflex elicited by the word “language” in this kind of context, as a way of proclaiming that the writer is intellectually quite sophisticated, really. (Especially if they’ve been fleetingly exposed to philosophy at university, though I see that RG is actually an MD.) It’s really just a sort of curlicue or decoration.
The surprising thing is that someone who has also heard of Saussure would know no better. Probably just on autopilot at that point. Those column inches don’t write themselves …
they simply understand the concept of a ‘language’ differently to us.
Quite right, and I suppose I should be used to it by now, but I just can’t seem to get on board.
The surprising thing is that someone who has also heard of Saussure would know no better.
That’s what got to me. I probably would have shaken my head with irritation and moved on if it had just been a rote invocation of the Big W, but to then insert Saussure parenthetically…
I’m not saussure about all this.
thread won
Yes I’m astonished how many (seemingly) random byroads of math/logic I bump into Grothendieck.
Would that Galchen had actually understood some W (or S) rather than just getting “views of language changed”. These howlers just before and after the bit you quote:
“Relationships” are relationships (or we could say ‘predicates’ holding amongst the noun-phrases).
If it doesn’t make sense to take x from some Category before you translate it, then it doesn’t make sense to take the analog of x after translating — is the whole point of Category Theory. (The not making sense is an ‘invariant’.)
Whereof Galchen cannot speak …
Yes, I figured there must be plenty of journalistic fudging/incomprehension going on, which would bother me if I remembered more of my math…
OK, I’m going to step up and be stupid for all to see.
My understanding of Saussure’s achievements are that they are, a, figuring out PIE laryngeals, and b, the Course. The importance of the first is unquestioned, but what Galchen and y’all are talking about is presumably the Course. From what I understand (cue stupidity), it’s an important landmark in the philosophy of linguistics, not in linguistic practice, synchronic or diachronic. Can one look at a grammar written after Saussure and point out any new clarity achieved only through his elucidation of langue vs. langage, or the formal statement of the arbitrariness of sign?
The Grothendieck article falls together with A Beautiful Mind, the biography of John Nash, another revolutionary mathematician. As the joke goes, psychology is just applied biology; biology is just applied chemistry; chemistry is just applied physics; physics is just applied mathematics; mathematics is just applied insanity.
An article I read once about Terence Tao, another brilliant mathematical luminary, pointed out with astonishment how down-to-earth and un-eccentric he was, bucking the stereotype. He was modest when interviewed, making it sound like hard work and collaboration are all it takes. Euler was another unusually sane yet brilliant mathematician; but he also enjoyed being of a generation that came upon what was then a vast number of low-hanging fruit.
a vast number of low-hanging fruit
Well, it might seem like that now, but no one had noticed all those low-hanging fruit until that time.
I remember reading a silly article years ago comparing Newton and Einstein, and it downgraded Newton because math and physics were in their infancy so there was so much easy stuff to be discovered. Whereas I always thought the opposite, that Newton’s brilliance was in seeing what had never been countenanced until then.
there was so much easy stuff to be discovered.
Yeah, because obviously the motion of billiard balls, and the trajectory of cannonballs, is exactly the same as planets orbiting a star. Strange how neither Pythagoras nor Eratosthenes nor the Arab mathematicians spotted it.
Somebody should tell those flat-earthers. (BTW none of the ancient mathematicians thought the earth was flat. We had to wait until the C21st to ‘discover’ that.)
obviously the motion of billiard balls, and the trajectory of cannonballs, is exactly the same as planets orbiting a star. Strange how neither Pythagoras nor Eratosthenes nor the Arab mathematicians spotted it.
Billiard and cannon balls were invented much later. There was no similarity to be spotted. The Greeks were interested only in tiny balls on teenagers, and the Arabs had only falafel, which do not orbit the plate.
Saussure really changed our view of Mont Blanc!
https://en.wikipedia.org/wiki/Horace_B%C3%A9n%C3%A9dict_de_Saussure
@Popup. Horace Bénédict de Saussure (“changed our view of Mont Blanc”) was related to Ferdinand de Saussure (the linguist) but were not the same person.
Yes – the Saussures were an important family in Geneva for generations! (And there are still some of them around.)
Personally, I know Horace Benedict best, as I’m more interested in mountains than linguism, but there were lots of them!
https://fr.wikipedia.org/wiki/Famille_de_Saussure
Can’t we just be grateful the clueless journalist didn’t write “Wittgenstein (and Chomsky)”?
Oh, I am, believe me!
But Chomsky (unlike …?) changed linguistic practice…
Saussure changed everything. Not alone, and not directly, but his students learned from him how to look at language, and they spread out and created modern (“structural”) linguistics.
Wholehearted agreement with our host. I hesitated to bring it up, but Saussure interpreted things in terms of their oppositions and more general relationships, which is one of the things category theory in general (and Grothendieck and, say,
https://en.wikipedia.org/wiki/William_Lawvere
in particular) have taken the ball and run with.
PS apologies for missing closed parenthesis upthread, here it is ) finally.
(Whew!)
Speaking of PIE…
langue vs. parole
Both Newton and Einstein standed on the shoulders of the giants, but it takes some doing to climb up there and stand (you have to also choose a suitable giant) whereas for most of us it is an arduous task to groupe a giant by the shoulders and hang down there flailing hopelessly.
From that WiPe link:
# Lawvere continues to work on his 50-year quest for a rigorous flexible base for physical ideas, free of unnecessary analytic complications. #
Sounds like my kind of guy, I’m gonna take a little look.
But isn’t Reality already a flexible base for physical ideas (and everything else) ? Unfortunately, I think it is precisely “rigor” that usually leads to “analytic complications”. I would prefer a pleasingly-plump-but-firm base for physical ideas. You know, avec des fesses comme il faut.
Perhaps the intention of “Wittgenstein (and Saussure)” was “Wittgenstein (I would have used Saussure, a better example, but I’m worried some of you haven’t heard of him)”. Is that reading more charitable or less?
langue vs. parole
Indeed. I caught it too late and hoped no one would notice.
“Low hanging fruiit only in comparison to later problems that required years of study just to be aware of. Kepler’s work in particular required the solution of hundreds of equations with paper and pencil..
If you want to slag on Newton, his later alchemical career offers an opportunity. Chemistry then was not low hanging fruit.
Perhaps the intention of “Wittgenstein (and Saussure)” was “Wittgenstein (I would have used Saussure, a better example, but I’m worried some of you haven’t heard of him)”. Is that reading more charitable or less?
It’s plausible, at any rate. But before splurging on charity, I would run a background check on the recipients. You don’t want it to fall into the hands of layabouts who have not even heard of Wittgenstein.
Mollymooly, that’s how I read it. I make no claim about the degree of charity.
Charity never faileth: but your linguistics teacher is not afraid to give F’s. Study accordingly.
Fear is not in charity. [1 John 4:18, DRA]
Speaking of Euler, a paper published last month argues that the cause for Euler’s right-eye blindness (before a cataract and a botched operation got the left one), and for the cerebral hemorrhage which killed him, was brucellosis, contracted from drinking raw milk, and widespread in Russia in his time.
avec des fesses comme il faut
Or as Dr. Johnson said, “with a fundament of good sense.”
There is, by the way, a fine book titled Not Saussure, by Raymond Tallis, about the clueless appropriation of structuralist ideas by poststructuralists.
I expected that someone must have hit on that little funny before I did. Cornpone humor is well-trodden ground.
The Tallis title is witty, however.
Sorry, this comment is a long quote from reddit.
From Newbie Question: Why is it that many Linguistic departments tend to study Linguistics ahistorically?
Q (rather dumb):
I’m still not clear about this topic. I studied linguistic anthropology. For instance, I read a book about a linguistic who went to a community and he/she learned how that specific community’s “Language use.” However, I did not learn the “etymology” or “origin” of words.
I’ve heard that before there was a career in Philology which studied language in a very different way? For instance, you could study multiple languages and their history.
It seems this way of study language stopped due to the ideas of Swiss Philosopher Ferdinand de Saussure where he advocated for a different way of studying Linguistics.
Answer (rather interesting):
Oh, man. I’m a graduate student in linguistic anthropology in the US. This narrative about Saussure is very common in anthropology, but really undeserved: His career was principally as a historical linguist. His Course in General Linguistics is split between synchronic & diachronic sections. Within his historical context, he argues that attention should be given to language as a synchronic system, but that’s not meant to be to the exclusion of historical linguistics. What’s more, Saussure is of relatively little importance in the historical development of the tendencies in US linguistics that have led to the most widespread programs of structural linguistic research worldwide. If you do typological/functional stuff, Saussure is one of your intellectual ancestors; if you’re working in the Generative tradition, that’s really not the case.
In general, anthropologists’ ideas about what linguistics is, what linguists do, & how linguists think language works are pretty inaccurate & self-congratulatory. Sadly, this is almost as true among linguistic anthropologists as it is among socio-cultural anthropologists. If you listen to how anthropologists tend to talk about linguistics, you’d think that there’d been no development in linguistic theory since Saussure, or you’d hear a version of Chomsky that seemed to be borrowed from a bad summary from 1957. Plenty of linguists are engaged in historical linguistics. That’s been true for a couple centuries, without interruption. Many linguists do both synchronic structural research & historical research.
Now, with regard to that book you read: It is true that accounts of fieldwork are going to focus on the roughly synchronic documentation of the system of a living language rather than on its history, but that’s because different methods correspond to different kinds of research, and are appropriate for addressing different kinds of questions. (Additionally, for languages that don’t have long written traditions, documentation of the contemporary system of language is a necessary first step before any significant historical work can be done.)
If you’re still in university, I recommend that you take courses in a linguistics department in addition to any courses you’re taking in linguistic anthropology. If you’ve left school, you might take a look at an overview of historical linguistics such as Lyle Campbell’s textbook; alternately, you could look into work on the history of a particular language to get a sense of what sorts of things are going on in the field.
In general, I recommend against believing anything anyone says about Saussure: He mostly serves as a straw man today. (Plus very few linguists read him: He’s of far greater interest to anthropologists—who also don’t read him, but we anthropologists apparently don’t think we need to read scholars before citing them.) If you’re interested in Saussure, read him directly! The Course is enjoyable to read & mostly fairly clear. (The book is also more than a century old, & is not representative of modern thinking in linguistics.)
In general, anthropologists’ ideas about what linguistics is, what linguists do, & how linguists think language works are pretty inaccurate & self-congratulatory
It seems to me that there is a certain element of what the psychoanalysts call projection here …
a version of Chomsky that seemed to be borrowed from a bad summary from 1957
He’s got much worse since then …
Plenty of linguists are engaged in historical linguistics
Where?
All Africans on this forum (Lameen and David E.) are engaged in historical linguistics…
P.S. a half of them is also into ophthalmology but it is not as rare as one may think. Dahl was an ophthalmologist, a viking cossack ophthalmologist.
the most widespread programs of structural linguistic research worldwide. If you do typological/functional stuff, Saussure is one of your intellectual ancestors; if you’re working in the Generative tradition, that’s really not the case.
Is “Generative tradition” actually “the most widespread … worldwide” (as opposed to the English-speaking world)?
viking cossack ophthalmologist
Between us all, my fellow Hatters, we can carry forward the magnificent linguistic work of the great Viking Cossack Opththalmologists of yore! We have the vikings, we have the cossacks, we have the ophthalmologists …
Well, in a scholarly sense the “English-speaking world” more or less means “the world of those scholars who publish their journal articles in English (sometimes not very idiomatic English!) regardless of their personal L1 or the dominant local language in the country where they currently live and/or where the university with which they are affiliated is located.” How much linguistics scholarship occurs outside the English-speaking world, so defined, these days? (I do not say this in a triumphalist way. Surely it would be better for everyone to publish in Latin instead. Or perhaps Bahasa Indonesia. Welsh would be silly, but maybe Breton?)
Kusaal! What a question!
I suspect that that English-language publications do indeed account for the great majority of academic linguistic stuff currently being produced; indeed, this is precisely why the solipsistic generative dead-end has done some much damage to linguistics in general. It’s a that pity these people are so self-satisified that they are no longer able to learn from the anthropologists.
Chomskyan vapouring is, of course, of no earthly use to any anthropologist.
Welsh would be silly
Dunia dim la wʋsa wʋm Kʋsaal; dinzugo, li nar ka tinam banɛ vɛɛnsid pian’adnam la an si’em la sɔb Kʋsaal ma’aa. It’s obvious really.
[EDIT: A Hat pa’a niŋim ninja*.]
* Zapan tanpsɔbnam. Ba su’a kʋʋdi ba zugdaan dataas.
How do we measure the volume of stuff produced in a language?
In petasyllables.
He’s got much worse since then …
The difference between Structuralism and Generativism seems to be:
Structuralists look for distributional regularities to explain language.
Generativists look for processes. Mainly movement and constraints on movement, and “empty elements” that become necessary to track and tag movements. In accordance with universal principles.
I am a dinosaur in that I tend to feel more affinity with the former, although there is plenty of the latter in traditional / pedagogical grammar.
it’s how people think and speak about patterns.
Put a clock on the wall in such a way that 11 is on top, you will hear “turned”, write 1233345 you will hear “repeated”, ask someone to compare abracadarba and abracadabra and people will speak about permutations (or metathesis).
You don’t want it to fall into the hands of layabouts who have not even heard of Wittgenstein.
I’ve heard of Wittgenstein but my brain filed him as “one of those philosopher guys” so I get surprised when he shows up in a linguistics context. For Saussure at least I know he did the laryngeal thing which is clearly (and famously) linguistics.
I was actually surprised that he didn’t say Chomsky but maybe he had reasons not to. (The more I read of the actual article the more I feel like Chomsky might not have been that bad of an example.)
How much linguistics scholarship occurs outside the English-speaking world, so defined, these days?
A lot of Russian papers are in Russian. AFAIK a lot of Chinese papers are in Chinese, though maybe not because Chinese is really not a very good language to do linguistics in (the writing system is really bad at anything that isn’t an extant Chinese word).
AFAIK the linguistics papers in German are overwhelmingly historical linguistics (not necessarily of languages that have anything to do with German).
Oh, they just add nearly as many Latin letters as they might need.
It is unsporting.
AFAIK the linguistics papers in German are overwhelmingly historical linguistics (not necessarily of languages that have anything to do with German)
Good. Maybe I spoke too soon. A lot of the Chomskyite hegemony in Anglophonia seems to be the result of a triumphant march through the institutions, so the effect may be more limited than I feared. Like the Empire of Nicaea, surviving the the loss of Constantinople to the Crusaders until the Emperor Michael* recaptured the capital, so too genuine linguistics will one day reestablish itself even unto the English-speaking lands …
* Halliday is but sleeping, until he awakes to free his people from their oppressors.
Preach, brother!
@ David E.: Is there some Hallidayesque equivalent of Drake’s Drum or the Dord Fiann that can be played to summon him in time of need?
I do not know what the author means by “structural research”.
somewhat similar to what happened in MSU in 50s when our “linguists” emerged as a distinct community.
I did hear that “structural” was seen as synonymous to “modern” linguistics (also just “linguistics” as opposed to [that boring thing that others are doing]) so the next question is what was called “structural” in linguistic literature of 50s.
This way similar.
No, they are not hostile to history cf. the adjacent Moscow school of comparativistics. But there apparently was a difference between this circle of people and the background which they emerged from in 50s-60s. There is also a difference between what is taught to “linguists” and “philologists” today, but they are not isolated from each other.
I do not have a clear understanding how exactly the old ways were different, even in Moscow. I know much less than nothing about the global situation (I mean, I know there are some tensions between old and new in Arab universites but even this depends on the university…).
The question is not dumb (and is not trivial). A young (but rather poorly informed) person heard about something that I “heard about” as well and asked people to tell more.
Is there some Hallidayesque equivalent of Drake’s Drum or the Dord Fiann that can be played to summon him in time of need?
I am not permitted to say, beyond that it involves two splendid old electric trains with pantographs.
Michael Halliday on Wikipedia.
Weel, I tried the site of the Institute of Linguistics and their “publications” page.
Surprisingly in “periodicals” part they have two series – both from 90s – and one journal (in English) about Africa.
The series are “Основы африканского языкознания” (Basics of African Linguistics), “Исследования по языкам Африки” (Studies in the lnaguages of Africa), the Studies are open access so I clicked a random volume.
It does mention Kusaal (once), it does not mention Chomsky (once).
That shows a fine sense of priorities.
What does it say about Kusaal?
The paper is К оценке ареального фактора в грамматике соседей языков ква (to estimate of-areal of-factor in grammar of-neighbours of-languages Kwa).
Not groundbreaking:/ He compares Kwa languages to their neighbours. The 7 features are:
“VO” order ; “Poss N order ; “serial verbal constructions” ; a construction with “take” illustrated with “take book read” and “take knife cut rope” (read the book, cut the rope with a knife) ; benefactive with “give” ; objects of ditransive verbs lack specialized marking ; complex locative constructions with more than one adposition.
He dismisses Mande langauges (in contact but too different) and Defoid languages (those for which he has data are similar but are not in contact with Kwa) and Kru languages (not enough data). For Gur he has data and Gur is diverse enough. He considers Gur, its most similar-to-Kwa representatives, and finds that they are not always in contact with Kwa, and its least similar-to-Kwa representatives and finds that they are not always distant. More research is needed:-E
Kusaal is ++++-+- (while Dagbani, Moore, Ngangam, Bouna Kulango and many but not all Kwa are +++++++).
The “Gur” table is p 290, other tables and examples from Kwa are above. Examples from Kulango and maps are below. https://iling-ran.ru/library/africa/IssledovanijaPoJazykamAfriki_6.pdf#page=290
Their English journal is here
Thanks, drasvi!
There are a lot of syntactic similarities between “Gur” and “Kwa” langauages, unsurprisingly. A Sprachbund … I often find familiar syntactic features all over West Africa, even in unrelated languages; notably in the various creoles.
Eastern Oti-Volta throughout has SOV when the object is a personal pronoun; Baatonum/Bariba (if you call that a “Gur” language at all, which you shouldn’t) is straightforwardly SOV.
Properly speaking, Kusaal does not have serial verb constructions at all, though published papers erroneously claim that it does; however, not only do the constructions in question feature an overt linking particle* (so that they are not serial verb constructions in Alexandra Aikhenvald’s strict sense) but the allegedly “serial” parts can even follow non-verbal clauses. There are certainly parallels, though, and “take” and “give” are indeed used in a serial-verb-like way.
Mooré is much the same as Kusaal with respect to alleged “serial verbs.” I think there has been a tendency among those studying these languages to assume that they have serial verb constructions without investigating the issue properly, because that what they were expecting to find in this linguistic zone.
Kusaal doesn’t really have any ditransitive verbs, strictly speaking, either; when two “objects” occur, the first is always indirect, and such indirect objects can actually occur with virtually any verb, including stative verbs like “be sweet.” So “causatives” derived from transitive verbs are really applicative: diis “feed X Y” is properly “give X Y to eat” rather than “make X eat Y.” Other cases apparently featuring two objects are really object + predicative complement. It’s true that there are no differences of actual form among the various kinds of “object” or predicative, but that is hardly surprising given that none of these languages has a case system. Still, they don’t use prepositions or postpostions to mark indirect objects, so I suppose that is valid as far as it goes.
The possessor-possessum order thing is entirely valid (you remember that Meinhof made a big deal of it …)
* Often realised as segmental zero, which confuses the issue, as so often in Kusaal. You can nevertheless demonstrate that it’s still there by its effect on preceding words. Toende Kusaal actually does have zero instead in all cases, as does Dagaare, but that’s the exception, not the rule, in WOV. The usual realisation is syllabic n.
Surprisingly, I was able to benefit from reading it in my language. I do read fast in English, but I browse faster in Russian. There are not many examples in this paper, a quick glance was enough.
He does not say anything about pronouns. His “give” example is Akan kofi tɔ-ɔ bukuu ma-a ama Kofi buy-PST book give-PST Ama “Kofi bought book for Ama” and Kusaal is said not to have this. He is not very clear about bitransitive, but it seems as soon as the two objects follow the verb and are not marked, it is fine (but a preposition would count as marking).
I do read fast in English, but I browse faster in Russian.
Same for me, with the languages reversed. I can glance down a page and grab what I need in English, but with Russian I actually have to read every sentence (pretty much, unless it’s a very formulaic text).
Kofi bought book for Ama” and Kusaal is said not to have this.
Actually you can say
Awin daa da’ gbauŋʋ tisi Awimpuak.
Awini TENSE buy book LINKER give Awimpoaka
“Awini bought a book for Awimpoaka.”
where tis is “give”, and the verb is denatured just as it would be in a canonical serial verb construction: it doesn’t mean “Awini bought a book and gave it to Awimpoaka.”
However, there actually is a linker particle, for all that it is invisible: it’s the reason that gbauŋ “book” has acquired a mysterious “ending” ʋ. Gbauŋʋ is really a reduced form of the underlying gbauŋɔ, with the final vowel preserved in “liaison” before the linker particle n, now itself generally realised as zero, though still often written n in older texts. The full form of “book” turns up clause-finally in all its glory in
Li ka’ gbauŋɔ.
“It’s not a book.”
where the suppression of the usual apocope of final short vowels is due to the negative enclitic, which is also realised as segmental zero.
It’s a neat and consistent system once you see the underlying logic of it, but it’s not surprising it’s resisted analysis in the past. People are so prejudiced against invisible words …
People are so prejudiced against invisible words …
Chomsky isn’t.
Ooh, burn!