From Science (20 Dec 2019, 3666.472: 1517-1522) comes “Emotion semantics show both cultural variation and universal structure, by Joshua Conrad Jackson, Joseph Watts, Teague R. Henry, et al.:
Abstract
Many human languages have words for emotions such as “anger” and “fear,” yet it is not clear whether these emotions have similar meanings across languages, or why their meanings might vary. We estimate emotion semantics across a sample of 2474 spoken languages using “colexification”—a phenomenon in which languages name semantically related concepts with the same word. Analyses show significant variation in networks of emotion concept colexification, which is predicted by the geographic proximity of language families. We also find evidence of universal structure in emotion colexification networks, with all families differentiating emotions primarily on the basis of hedonic valence and physiological activation. Our findings contribute to debates about universality and diversity in how humans understand and experience emotion.
The conclusion:
Questions about the meaning of human emotions are age-old, and debate about the nature of emotion persists in scientific literature. The colexification approach that we take here provides a new method and a set of metrics to answer these questions by creating vast networks of how people use words to name experiences. Analyzing these networks sheds light on the cultural and biological evolutionary mechanisms underlying how emotions are ascribed meaning in languages around the world. Although debates about the relationship between language and conscious experience are notoriously difficult to resolve (28), our findings also raise the intriguing possibility that emotion experiences vary systematically across cultural groups. More broadly, our study shows the value of combining large comparative linguistic databases with quantitative network methods. Analyzing the diverse ways that people use language promises to yield insights into human cognition on an unprecedented scale.
It seems awfully short to be yielding insights into human cognition on an unprecedented scale, but what do I know? I await the decision of the jury. Thanks, Trevor!
Maybe I missed this, but do they compare collocations within the same culture to see if the ultracultural variation is larger than the infracultural variation? They talk about colours, where the effect is larger and more amenable to an empirical investigation, e.g., show people images of various (combinations of) wavelengths and ask what colour the image is.
(@PP: Colexifications, not collocations.)
our findings also raise the intriguing possibility that emotion experiences vary systematically across cultural groups.
Far from “unprecedented”, this seems pretty unsurprising. I would have thought that pretty much anybody with a more than superficial knowedge of a non-SAE language would find this very old news indeed.
Kusaal zɔ means “run.” It also appears as a transitive verb meaning “fear.” But it also appears as a ditransitive verb in e.g. M zɔt o nimbaanlig “I pity him”, where nimbaanlig by itself is a noun meaning “pity.” So “I fear him pity”? Well, no, not really …
Other Oti-Volta languages do the same thing with their “run” verbs, even when they are not cognate with zɔ.
Kis is the usual word for “hate”: M kis o “I hate him.” But the cognate adjective kisʋg means “ritually forbidden.”
“I’m angry” is M sunf pɛligya “My heart’s gone white.”
There’s nothing groundbreaking about this, and nothing remarkable about Kusaal using a quite different set of idioms for emotional states from SAE. This is a perfectly commonplace observation. I have no doubt but that other Hatters can produce long lists of their own.
“happy” is hard to translate even within SAE.
differentiating emotions primarily on the basis of hedonic valence and physiological activation
so, um, on the basis of how they feel.
like DE said, startlingly precedented.
there are interesting things to be said about the subject matter – i find the differences between the ways yiddish and english lexically slice up the emotional and sensory fields endlessly fascinating (and that’s just within west germanic!) – but moshke’s berdele i wish people would pay close attention to the specifics of languages on their own terms instead of doing this kind of (doubtless eminently fundable) english-mediated* statistical GIGOpalooza.
.
* i mean, “‘anxiety’ was closely related to ‘fear’ among Tai-Kadai languages”?!? so – setting aside the rather strong implication that languages in that family use the words “fear” and “anxiety” – how does presuming seamless translatablility into those english concepts in the way that sentence necessitates get past reviewer 1, let alone reviewer 2? and what colexifies, in any tai-kadai/kra-dai/daic language, with the emotion conveyed by “self-undermining methodology much?”?
I see many of the authors are affiliated to
So not actual Linguisticians, but some sort of Linguistics-adjacent? Can DM fill in more?
I’m ashamed to see a few affiliated with New Zealand institutions — again no actual Linguisticians. (What does a ‘Religion Programme’ bring to the table?) Worse, they used NZ taxpayer monies. Although no more:
Based on this startlingly ‘precedented’ bollix, they deserved it.
I have no doubt but that other Hatters can produce long lists of their own.
Not a Linguistic collocation so much as a cultural one: I’ve been sat in a cafe overhearing a conversation between an about-my-age bloke and his reticent dad. Clearly somebody close in the family has just died, and the son felt he had to show support. There was not one actual emotional word, but lots of “That’s all right, mmm”; “They’ll be ok about it”; “I’d better talk to …”. (A very familiar style to me. I don’t think I heard my father use any of the vocabulary analysed in the paper.)
Would these Cultural Evolution jokers like to stuff that into their algorithms?
@AntC: I see the supposed Kiwis among the numerous co-authors also concurrently hold positions at one or more academic institutions in Europe. A long commute. Unless they’re telecommuting? Or taking advantage of the time-zone differences to work around the clock?
Also a bunch of co-authors associated with various departments at UNC-Chapel Hill, but nary a one of them associated with that university’s Department of Linguistics. (I just checked the website and UNC does have such a department.)
To push back a little on a highly skeptical jury here. They didn’t use collocations, but colexification (like English palm meaning both a kind of a tree and a part of a hand). They did forthrightly said that they researched two alternative explanations for emotions which preexisted their research and found that their method supports one of them more than the other. Etc. Both reviewers 1 and 2 are supposed to read a paper before rejecting it. At the very minimum they should be able to find at least three typos.
The method is not new and the results are not surprising. Most of scientific publications (unlike blog comments) are like that. Whether the methodology is established enough to make inferences that they want to make is not my place to tell. The jump from a purely linguistic realm (how similar are similar languages) to psychology is highly suspect, but again, who am I to judge.
And as for the authors not being True Linguists, I am sure next time someone reports on discoveries of well-credentialled Chomskyites, the high commissioners in the comment thread will stand up and applaud.
I generally agree with DE and rozele here (as is my wont). But, don’t get mad at me for saying so, I’m afraid I’ll have to be a bit of a devil’s advocate here. Science is an endless series of approximations, and this is one of the first ones on this particular subject matter. A similar study done two hundred years ago comparing, say, “accusatives” in all the languages known to your Standard Average European Scholar, would be obviously flawed, but would still yield interesting results to build on.
This study is mining the six-year-old database of colexifications. The entire database rests on dictionaries old and new. Lexicography, even as practiced today, is unsatisfactory in that it fails to methodically outline semantic fields. The semantic map technique, something that could have been invented 200 or more years ago, was only invented 20 years ago, and is used only in limited specialist studies, not as a foundation for dictionary making. Even guides to near-synonyms for language learners are far and few between, and that even for the big European languages (another shout-out here to Keenan’s Breaking Out of Beginner’s Spanish). So, baby steps.
@AntC: I wouldn’t get near cheering what the NZ government is doing, even if you like some of its side effects. The right-wing war against universities is heating up worldwide, and humanities departments — all of them — are the main and obvious targets (plus, I suppose, environmental studies). They will go after the obvious bugaboos and liberal hotbeds, like Gender Studies and Sociology, and then someone will come and upstage them by going after linguistics and anthropology, all for the sake of the “taxpayer” (that is, the right-wing taxpayer). This past year the University of Alaska and the those of Leiden and Utrecht and who knows where else were saved at the last moment, as budgetary death-blows were converted into mere crippling wounds. Don’t chuckle at a spreading fire, even it will annoy some people you don’t like.
@Y thanks, but to be clear: I was not cheering nor chuckling, and I’m well aware of what agenda
thislast year’s change of government is smuggling in for ‘the sake of the taxpayer’. Given that atmosphere, the humanities departments could try harder not to be such “obvious targets”. I guess the project ‘won’ (if that’s the word) the funding under the previous government, but it wouldn’t take a genius in “cultural evolution” to see the winds have changed since NZ (and governments everywhere) got past COVID. Then publishing this seems like opening a gunpowder keg too close to a spreading fire.From the second paragraph of the abstract:
#
Although not all emotion words are common—the German word Sehnsucht refers to a strong desire for an alternative life and has no direct translation in English—there are many words that appear to name similar emotional states across the world’s spoken languages.
#
Try “longing” or “yearning”. What is this “alternative life” ? A personal parallel universe?
No need to read further. Thanks to the authors for this timely emotion experience ! I leave it to the reader to guess which one.
They didn’t use collocations, but colexification (like English palm meaning both a kind of a tree and a part of a hand).
Yeah. And I see there’s a whole Colexification database ” from more than 2400 language varieties of the world”, which is what these dudes plundered. So they haven’t themselves looked at any languages, they’ve just run some statistical analyses.
(I drafted an anger-soaked outburst about the dumbness of identifying “concept” with lexeme, but round here I’d only be teaching my grandmother … .)
——–
They didn’t use collocations, …
Which intel won’t stop me riding a hobby-horse.
And wikip goes on to the “revisions and critiques … ill-defined in a larger population … ill-fitting …”. It’s a mess (linguistically as well as pragmatically). And you’da-thought a professional with a wide experience of personality differences would demand a more coherent structure — for teaching purposes at least.
I guess collocations is what LLMs/GPT are relying on, too. With similar benightedness of understanding.
(I drafted an anger-soaked outburst about the dumbness of identifying “concept” with lexeme, but round here I’d only be teaching my grandmother … .)
Grandmothers are not necessarily know-it-alls. It’s just that nobody likes being teached at in anger. I recently read about something called “nudge words”. Maybe you could deploy some of those.
@AntC, I don’t think this particular kind of study is the sort of thing that draws fire. If anything, math (however pointless) makes studies appear more STEM and therefore less hippie.
Russell Gray and Simon Greenhill are the two Kiwis. They are the two who more than anyone started the fashion for Bayesian phylogenetics in linguistics (upon which I curl my lip.)
@Stu: For what it’s worth, two of the authors (List, Forkel) are German.
BTW of specific interest here, department of Celtic and Classical Languages and Culture at Utrecht is being discontinued. That is where Peter Schrijver is (as head of the department). They had just celebrated their hundredth anniversary, in 2023.
the dumbness of identifying “concept” with lexeme
Yup. Or with collocation, come to that.
M ya’am kpɛn’ɛf in Kusaal means “I’ve fallen in love with you.” Literal translation word by word yields “My gall bladder has entered you.” I’m going to go out on a limb here and suggest that this tells us nothing whatsoever about how the Kusaasi conceptualise romantic love. The idea that it might is just a variant on the ever-popular etymological fallacy.
Both reviewers 1 and 2 are supposed to read a paper …
There’s a typo? Somebody can’t count? One reviewer wished to remain anonymously anonymous?
The last paragraph just before Acknowledgments seems to have been written by the Marketing department. A load of blah-blah grand claims not supported by the really quite cautious/trivial “findings” in the previous few paras.
appears to be from the Department of the Blimmin’ Obvious. (And applies to most everyday semantic fields.)
@Stu, “nudge words” [according to a WaPo article] seems to be a term of art from the Personal Growth/Wellness industry. “Choose a nudge word to set a vision for 2025”. So AFAICT not about nudging others. Anyhoo, where’s the fun in letting off steam subtlely?
The grandmother who taught me how to eat soft-boiled eggs died a week ago now.
how the Kusaasi conceptualise romantic love.
So which vital organ for brotherly love vs love-of-country vs the many-splendoured thing vs all-you-need-is love? Does it extend to for-love-not-profit, “the word “love” thus implying “nothing”.” as in a tennis score?
@Brett, my sympathies. I have many fond memories of my grandmother’s wise teachings. (Secrets of her egg custards were never divulged, though.)
I haven’t read the paper, but a couple of comments.
The Mac Planck Institute in Jena (per Wikipedia):
Of the three departments, i know that both the Department of Archaeogenetics and the Department of Archaeology do great things. I know less about the Department of Linguistic and Cultural Evolution, but since it’s led by Russel Gray of Gray & Atkinson fame — the New Zealand connection — (and also because of the Institute’s surprising history in Economics) I’ll assume that their focus is on developing mathematical models for the development of human language and culture. That may lead somewhere eventually, but it’s still very much at the development stage, and unfortunately — as we’ve seen with the Bayesian phylogenetics work — obviously wrong results are being published and hyped as groundbreaking discoveries rather than test studies in developing methodology and requests for feedback from the scientific community.
I will say that this is not about the humanities being humanities. This is about humanities trying to be hard science without understanding its limitations, or about hard scientists imposing their methods on the humanities without bothering to learn them. It’s what we will have more and more of when the real humanities are starved to death.
@Antc:
Nɔŋ is “love” as in “family, spouse”; “love of country” is not a thing in the traditional Weltanschauung.
“I love you” as whispered to your SO is M bɔɔdif “I want you.” Naden’s dictionary says that the verb in question, bɔ “seek” means “fuck”, but I suspect that some confusion has arisen here. (All Western Oti-Volta languages for which I have data use cognates of Mooré yēbe for “fuck.”)
M ya’am kpɛn’ɛf in Kusaal means “I’ve fallen in love with you.” Literal translation word by word yields “My gall bladder has entered you.” I’m going to go out on a limb here and suggest that this tells us nothing whatsoever about how the Kusaasi conceptualise romantic love.
Neither does “I’ve fallen in love with you”, for that matter.
Is “love” some kind of ditch with slippery walls? Is this a request to help the speaker out, or to leave them there and bring them coffee and donuts forevermore ?
M bɔɔdif “I want you.”, bɔ “seek”
“Long for”. Is this the same as (or related to) the Scandi-Congo “dwell” word? If so, note “belong”.
It all fits. I tell you!
I was actually serious (except, obviously, when I wasn’t) about the Kusaal verbs, but I was misremembering the vowel of your bɛ “be in a location/exist”.
I’m no stranger to silly terminology, but “hedonic valence”…
How long is the “supplementary material”, i.e. the actual paper? The “paper” in Science (like in Nature and PNAS) is really just an extended abstract.
…so that is where Karl May got the idea that Muslims swear by the Prophet’s beard. I sit in awe.
That place does employ some actual linguists. I recognize the name of the fourth author, Johann-Mattis List.
Yeah, well, there are reviewers who take five minutes to skim the manuscript and write a review that strongly urges the authors to do exactly what they did…
That’s not their job. That’s the proofreaders’, i.e. authors’, job.
(I do it anyway, but that’s because I’m that kind of nerd.)
Yes, but Science (again like Nature and PNAS) tries very hard to pick only the most breathtakingly groundbreaking manuscripts and reject all others (often without review) as not interesting enough. There’s an Impact Factor to defend.
Yeah. To have Sehnsucht nach a person is a common way to express, in writing, that you’re missing them plain and simple.
The puzzling thing here is that, as Y pointed out, there are native speakers of German among the authors. I suppose this is the genetics/particle-physics model of authorship, where everyone who has contributed to any part of the research is automatically an author, as opposed to the model I’m used to, where being an authors means you’ve at least read and approved all of the manuscript.
And there was desking of heads and palming of faces.
Nah, just bad writing by scientists:
I’m sure this means they acknowledge
– the feedback of their editor and six anonymous reviewers;
– the assistance, whatever it was, by K. Gray, K. Payne, E. McCormick, J. Leshin, and N. Caluori;
– and the research assistance of R. Drabble, I. Khismatova, and A. Veeragandham.
Not that conservative blowhards trying to kill science is a new thing, but the strain is far more virulent now.
I recommend Oliver Butterworth’s The Enormous Egg (online here), a very fun young-adult book from the 1950s, complete with science denialists and the aforementioned blowhards (David Marjanović, you in particular might enjoy it, suspension of disbelief notwithstanding.)
I was actually serious (except, obviously, when I wasn’t) about the Kusaal verbs
Oops, sorry.
No, they can’t be cognate. The “seek” verb must have been *bàw- in proto-Western Oti-Volta (cf Mooré bào “seek”, baoda “want” (confirmed by the vowel correspondences in the two Kusaal dialects.) I can’t trace the etymon back beyond WOV at present.
The “exist, be somewhere” verb had the root *bè-, reflected in Kusaal bɛ, Mooré bè; the vowel of Farefare/Gurenne boe has been secondarily rounded by the initial b. A similar thing has happened in some other Oti-Volta languages: beside Yom be, Nawdm béé, Moba bé, Nateni péé there are Buli boro, Byali bo and Mbelime bō. But that kind of rounding after labial consonants is very common across the whole family.
The “exist” verb thus goes back to proto-Oti-Volta *bè-; this may be cognate with proto-Bantu *bá- “be” too, but I don’t have an explanation for the vowel difference, and I’m leery of such short forms in comparative work, especially with anything less than complete matches of all segments. The tones actually do correspond properly for cognates, though, in this case. POV Low goes with proto-Bantu High.
(If anyone is following this stuff about tone at home, I have changed my mind about the realisation of the tones of POV, so they now look more like WOV than e.g. Gurma; this is because I now think that this makes more sense from an OV-internal standpoint, and that I was previously too influenced by cart-before-the-horse comparisons with Bantu. But the actual correspondence sets are all just as they were.)
No, my fault. It wasn’t meant as (only) a joke, but when I based it on misremembering the vowel, the suggestion became ridiculous anyway.
But thanks for the explanation. The truth is that I’ll say anything to provoke another one.
I was going to reply about The Enormous Egg, but I see I already have. Uncle Beazley must be covered in snow this morning.
Does it extend to for-love-not-profit, “the word “love” thus implying “nothing”.” as in a tennis score?
No. However “gratis, for nothing” is Kusaal nɛɛm, which is an abstract-noun-used-adverbially from the adjective nɛɛr “empty”, which in turn appears in Mooré as néere “beautiful”, as in pʋg-neere “beautiful woman.” What this implies about Mossi taste in women I cannot say; one hesitates to accuse them of Trumpian values. More research is needed.
[Actually, the root is also seen in verbs, from which it becomes clear that the underlying meaning is actually “clear, light”; the Kusaal “empty” sense is not too distant from “clear”, and the Mooré adjective presumably has developed rather like English “fair” in “fair maiden”: opposite direction, but same sort of association of ideas.]
A new take on Vanity Fair.
@de
Are you perhaps missing the equation empty = not tattooed (or pierced?) = beautiful, an equation which will also become more apparent in Britain, when the “heavy coat” season eventually ceases?
Tattooing is not much of a thing in those parts, which go in rather for decorative facial scars. A woman without proper facial scars would be highly unattractive, as presumably not even being Mossi at all. (There are men who find barbarian women attractive, I suppose, but that’s hardly the sort of thing to go into on a family blog like LH.)
moshke’s berdele
i hate to disappoint DM, let alone herr may, but this is certainly not a traditional snarl, and may be original to me (and i am influenced by many literary and paraliterary prophets-beard oaths).
emotion words vary in meaning across languages, even if they are often equated in translation dictionaries
bold words, for a paper that appears to be entirely based on using shared translations into english to declare emotion words equivalent.
Is “love” some kind of ditch with slippery walls?
a ha-ha, one presumes.
“Love’s a ditch with slippery walls” sounds quite promising as the start of an emo pop lyric.
“I lost my gall bladder in Ouagadougou” perhaps less so. Maybe a power ballad.