Our nightly reading these days is Paul Scott’s Raj Quartet, which I first read three decades ago and have very much been wanting to revisit; at close to 2,000 pages, it should occupy our bedtimes well into next year. The first novel, The Jewel in the Crown (also the title of the superb television serial made from it), is online at archive.org for anyone who wants to sample it; I thought I’d post this passage for its linguistic interest:
The teacher at the Chillianwallah Bazaar school, whose pupils were all Indian, was a middle-aged, tall, thin, dark-skinned Madrassi Christian, Mr F. Narayan: the F for Francis, after St Francis of Assisi. In his spare time, of which he had a great deal, and to augment his income, of which he had little, Mr Narayan wrote what he called Topics for the local English language weekly newspaper, The Mayapore Gazette. In addition, his services were available as a letter-writer, and these were services used by both his Hindu and Muslim neighbours. He could converse fluendy in Urdu and Hindi and the local vernacular, and wrote an excellent Urdu and Hindi script, as well as his native Tamil and acquired Roman-English.
“Roman-English” doesn’t convey anything to me; I’m guessing it might mean English written in the Roman alphabet, but how else would it be written? All suggestions welcome.
Also, just because it was preying on me and I’m pleased to have solved the puzzle: I’m enjoying my new Blu-ray of Béla Tarr’s Sátántangó, notorious for its 439-minute running time (I’m following my brother’s advice and taking it in chunks, easy to do since it’s divided into twelve parts), and today I watched the sixth part, “A pók dolga II (Ördögcsecs, sátántangó) [The Job of the Spider II (The Devil’s Tit, Satan’s Tango)],” which takes place in a bar where everyone is getting increasingly drunk. One character, Kelemen, keeps repeating the same phrases over and over until you want to slug him, and the most frequently repeated was subtitled “I was plodding and plodding” (it’s the first thing you hear in this YouTube clip). Of course I wanted to know what the Hungarian was, and I think I’ve finally figured out it’s vágtattam (see the conjugation here), which means ‘I galloped.’ I don’t know why the translator went with “plodding,” but it seems misleading.
Update. It would appear rather to be baktattam, from baktat ‘plod, trudge, walk slowly’; see Xerîb’s comment below.
There are plenty of google hits for ‘Roman English’, often along the lines of ‘Qur’an Transliteration in Roman English Script’.
It does seem to be specifically the script rather than the language (and it looks like it’s part of a list of scripts in your quote) – I found one page which talks about ‘writing the Urdu in Roman English Script’, which just seems to mean ‘in Roman characters’.
This seems like a faint possibility, but I’ll mention that with his namesake St. Francis, it seems possible that his is a Roman (Catholic) English.
But then I know little about Indian linguistic/religious history beyond what I learned from that classic of Bollywood Realism, Amar Akbar Anthony.
(Of course, his Roman-English ability follows on his ability to write in “Urdu and Hindi script”, so a script meaning seems more likely.)
One online Indian source offering “Roman English” versions of the Quran also offers a different volume in “Roman Urdu,” which means in context Urdu typeset in roman script rather than the usual script. See also https://en.wikipedia.org/wiki/Roman_Urdu, which is more focused on online use. There’s still the question of what non-Roman script Indian publishers might typeset an English text in, though … That would just make the phrase peculiar-because-redundant, though, rather than nonsensical.
László feLugossy (one of the actors) – I wonder what the story behind the surprising (non-)capitalisation is.
https://m.youtube.com/watch?v=eAEOfqF5p78
I am believing a knowledge of Hindi is helping understand this video, but it is not necessaried!
László feLugossy (one of the actors) – I wonder what the story behind the surprising (non-)capitalisation is.
Me too, but it seems to be a quirky personal addition; apparently he was born Lugossy László.
Hmm, and there are variants: Lugossy, feLugossy, Fe Lugossy László, fe Lugossy Laca.
Oh wait, I think I’ve solved the Roman-English thing; way down in the OED Roman entry we find:
I think “Roman-English” as describing a script, in a sentence that has already referred to Hindi and Urdu script(s), is an adequate and plausible interpretation of the passage and while a character like Mr Narayan may have written in a “stately stile” I’d like to see some post-17th-century usage (the 1948 Baugh doesn’t do it*) before entertaining that less obvious-in-context interpretation.
There are enough other references out there to “Roman-English script” to convince me that it was sort of an idiom, relevant in places where the dominant local script was not Roman but the primary Roman-scripted “outsider” language was English.
And an old New York lawsuit from around 90 years ago quotes (probably irrelevantly to the exact issue at hand) the monthly-pass ticket-selling policies of the Long Island Rail Road at the time, which had instructions for what the ticket seller was to do when “a purchaser is unable to sign his or her name in Roman (English) script.” Presumably they were worried about immigrants who were not illiterate-as-such but could only sign their name in some other script that the LIRR personnel might not be able to read?
Sorry, re the Baugh quote, it’s perfectly clear I think to a reader of modern English what a “noble Roman style” means in the context of someone writing English prose. But that’s no evidence of continuing use of “Roman English” to mean English-written-in-that-style. Or “stile,” if you like.
There was a Roman man
Who walked a Roman mile.
He found a Roman solidus,
Engraved in Roman stile.
…
Láci is the established nickname for László, FWIW.
I second Jen above. Mentions of both the “Roman English script” and “Roman English alphabet” don’t seem uncommon. I cannot get search engines to differentiate “Roman English” (which seems rarest) from “Roman-English”, “Roman/English”, or “Roman (English).” It works also with “Latin” instead of “Roman.”
My unsupported guess is that it’s merely a shortcut that our host would have opposed as an editor.
I don’t think it can be emended to: “He … wrote an excellent Urdu and Hindi script, as well as his native Tamil and acquired English.” I infer the wrong meaning of having a good command of written English, rather than good penmanship.
It also cannot be emended to: “He … wrote an excellent Urdu and Hindi script, as well as his native Tamil and acquired Roman.” Roman what? And “Latin” would be worse, suggesting he occasionally wrote in the classical language.
“He … wrote an excellent Urdu and Hindi script, as well as his native Tamil and acquired Roman scripts” is clear to me: he had excellent penmanship in the four different scripts of the four languages he was fluent in. However, this emendation clashes with an Italian obsession not to repeat the word script (or any other) so quickly.
More important, maybe for every native English speaker who’s stopped and puzzled by “Roman-English” there are two who’re stopped and puzzled by “Roman script” and need a double take to interpret it as the script used to write English?
Not sure if “English script” would work for native speakers, but it wouldn’t for me. I’d read it as an opaque allusion to some antiquated calligraphy, most likely related to copperplate script.
Consider the entry for te on p. 563 of this 1893 dictionary, which I think would be called a ‘Roman English’ dictionary of Hindustani:
And then there is this amusing example from the Census of India (1911), vol. 15 (Agra and Oudh), p. 262, which gives one an idea of the status of the collocation somewhat before the period in which The Raj Quartet is set.
For a straightforward example of present-day use of Roman English, scroll down to the blue box ‘Ayatul Kursi in Roman English Transliteration’ on this page, out of Hyderabad. (The language in Roman English script is Arabic.)
He could converse fluendy in Urdu and Hindi and the local vernacular,
The local vernacular of Mayapur is Bengali, a language spoken by over 240 million people with a rich literacy tradition. It seems odd to call it a “local vernacular.” Is that Scott being condescending or is the condescension part of the narrator’s character? (Or am I wrong in thinking the school is in Mayapur?)
The action is, to cut and paste from wikipedia, “in Mayapore, a fictional city in an unnamed province of British India. The province, which is located in northern India, shares characteristics with Punjab and the United Provinces. The names of places and people suggest a connection to Bengal, for example Mayapore is similar to Mayapur in West Bengal; however, the physical characteristics place the setting in north-central India, rather than in northeast India.” Presumably somewhere where the “local vernacular” is some Hindi “dialect” that may deviate substantially from the confected literary standard. A long ways from Madras, though.
I think it is Baktattam és baktattam és baktattam, from baktat ‘plod, trudge’.
(Some scenes of that…)
In Roman English ﻁ ﺖ of Urdu and त of Hindi are expressed by (t).
Ah, that’s thoroughly convincing, and I am convinced.
The action is, to cut and paste from wikipedia, “in Mayapore, a fictional city in an unnamed province of British India. The province, which is located in northern India, shares characteristics with Punjab and the United Provinces. The names of places and people suggest a connection to Bengal
So far it sounds rather as if it were in northwest India, but I’ll keep my eyes open for clues. I think the Mayapore/Mayapur connection is irrelevant.
I think it is Baktattam és baktattam és baktattam, from baktat ‘plod, trudge, walk slowly’.
That makes sense, and I thank you for the correction!
When Hari moves back to Mayapore and lives with his aunt, he starts taking Hindi lessons from Pandit Baba, not Bengali.
It is hard not to think that Scott chose Mayapore as a speaking name: माया māyā ‘illusion, deception, delusion, fraud’ + पुर ‘fortress; town’.
Xerib, you’ve won the thread three times over.
I think the Mayapore/Mayapur connection is irrelevant.
Apparently. Thanks to you and the others for the context.
I happen to be visiting Madras/Chennai at the moment; not worth a comment by itself, but perhaps a story from this morning is worth sharing. My tour guide asked me if I knew what the “oldest surviving language” was. For half a second I thought that he wanted me to say Sanskrit, but fortunately before opening my mouth and embarrassing myself, I remembered where I was and I realized that of course Tamil is the oldest language. He said yes, Tamil is the oldest surviving language in the world, you can look it up on the Internet, it is 5000 years old. I didn’t think of it at the time, but it makes sense when you think that the Dravidian languages are related to Basque, of course known to us all as the oldest language in Europe.
No, no, no! Welsh is the oldest language in Europe. It said so on the BBC, and everything!
But I thought Welsh was Dravidian! That’s what Zizka/John Emerson always told me…
Is Welsh even spoken in Europe, other than by a few stray tourists? That’s like saying Maltese is spoken in Europe – a sop to Continental expansionism and imperialism. I suppose the spoken-in-Europe Breton may be claimed by Welsh nationalists as a deviant dialect with odd orthography, or something like that.
Maltese is spoken in the EU, Welsh is not. Politics is geography.
Everyone in continental Europe knows Lithuanian is the oldest spoken language in the world, another „fact“ tour guides like to tell.
Everyone in Europe knows that tour guides are full of shit. (I’m not playing a tour guide on the Internet).
I don’t blame the guides too much though. Tourists want to be fed fake facts like “Lithuanian (Welsh, Tamil) is the oldest living language”. It’s easy to remember and something you can tell your friends when you get home. It’s vaguely impressive without requiring any additional context. What else are you going to tell people about Lithuania that is actually interesting but doesn’t require some historical or cultural knowledge to appreciate?
Litvak jokes?
For the EU-is-Europe crowd I guess Turkish is a European language, but only because it’s spoken on the non-European island of Cyprus rather than because it is spoken on the remaining bit of Continental Europe governed by Turkey. And Norwegian is a non-European language?
It would surely take a very hardline ideologue to declare that Switzerland is not in Europe.
(I gather, from an impeccable source, that the good people of Davos would all be speaking German now if it were not for the Americans.)
Turkish is also spoken in Bulgaria. Also Austria, Germany…
Turkey is in Europe for football purposes: I have not been able to escape the knowledge that the three soccer teams of İstanbul are, in alphabetical order, Beşiktaş, Fenerbahçe and Galatasaray.
Admittedly I don’t know about Ankara.
Being “European” for soccer purposes (a quality possessed at present by the national team of Kazakhstan – is Kazakh a European language?) is hardly a much more reliable guide than being “European” for Eurovision Song Contest purposes – an expansive vision of European-ness that has variously encompassed Australia, Azerbaijan, and Morocco.
@DE but can you answer one simple question: why not Abkhazian? Or Ubykh etc.
I think I began to think of Maghreb as a part of Europe.
It can of course be an entity on its own. But if it is not, it is in some ways Africa (but in it is not in many ways that will make locals disagree), it is in some (many) ways the Arab world but it is also in some (many) ways Europe.
Cultrurally Roman empire – then a long period of hostile contacts with Europe – then a period of cultural and economical Europeanisation.
Geographically there is a sea of water in the north (in some ways an obstacle which hinders migration, but in some it means “waterways”) and of sand in the south. And a land leading to the east, but very long land.
“Tourists want to be fed fake facts”
We do not. I mean, why say so? Have any tourist told you so? Is it introspection? Or maybe tourists (who you know have a good reason to understand they’re being lied to) respond with excitement to such facts? Say, Russians do want TV to lie them but I do have a good reason to say so.
As far as I know, all tourists want is to learn something interesting.
Also about Lithuanian it is a silly formulation of some actual fact about Lithuanian.
I can even say that semantic extension of “old” to this is not very strange.
Before the Arab conquests, all nations along the shores of the Mediterranean were often thought of as part of a single high-level cultural zone (as they still are in a historical context), but no one would have called that zone “Europe” since it obviously encompassed the Mediterranean shores of Europe AND Asia AND Africa.
Vanya, sorry for annoyed tone. Sometimes when poeple ‘want to be lied to’ it’s highly unpleasant to me. And sometimes it is not (fiction). Your words that tourists want it somehow reminded me about the first.
@JWB, yes, but it is also Europe in a certain geographical sense. Tunisia recembles south Italy in some ways it does not recemble Burundi.
(On the other hand, there are “African” species in Maghreb as well).
Also in the Greek and Roman times, “the Greek” or “Roman” world was not Europe.
Whose cultural orbit Maghreb is? I think both Arab and European.
Being “European” for soccer purposes (a quality possessed at present by the national team of Kazakhstan – is Kazakh a European language?) is hardly a much more reliable guide than being “European” for Eurovision Song Contest purposes
Well, at least Turkey and Kazakhstan have a territorial toe in Europe as it is usually defined geographically – Turkey has Eastern Thrace and Kazakhstan some land West of the Ural river. For the ESC, it’s not a problem if you are about as far away you can be from Europe without being covered in Pacific…
As for African cultural influence it is complicated. Those are different influences and many locals will see them as “low” culture. And I’m not speaking about conventions – I’m speaking about my own needs. I need to partition the world somehow in my head and sometimes it is convenient to think of Maghreb so, and other times it is convenient to think of it as Arab world, and other times as north of Africa.
Kazakhstan has been seriously Sovieticised and linguistically Russified. To the extent Russia is cultural Europe, Kazakhsan is largely so.
I usually tell Russian learners looking for immersive courses that they can also try Ukraine, Belarus and Kazakhstan. Of course with Muslim learners I also remembered that Kazakhs are Muslims as I was telling it. Kazakhs and Tatars in Russia. But Kazakhstan hardly recembles the Muslim cultural world.
Soccer’s Europe includes (to quote wiki): “the transcontinental countries of Russia (suspended), Turkey, Azerbaijan, Georgia and Kazakhstan, as well as the West Asian countries of Cyprus, Armenia and Israel.” I guess “transcontinental” is a clever word but there’s clearly an ad hoc mixing-and-matching of nominally geographic criteria and cultural/political criteria rather than a consistent methodology. By implication, I guess Malta is considered unproblematically European?
From a tectonic-plates perspective, Malta is apparently unequivocally African whereas Sicily is “transcontinental” in that it’s sort of top of the collision zone.
@JWB, “geographic” – note that geography is not only landmasses. It is everything (culture too) and a plenty of geogrtaphical lines and isolines (i replaced “isogloss” here with a Russian word, WP says in ENglish it is “contour line (also isoline, isopleth, isoquant or isarithm)“) are about rainfall or ranges of plants or, say, watercheds etc.
In terms of watersheds maybe northern slopes of certain North African mountain ranges will be the Mediterranean basin (or not? I looked it up in WP and a very large chunk of Algeria is “Mediterranean basin”)
I mean, we can say “Europe” is what the encyclopedia says in geographical terminology, but geographic criteria can be even economical.
Eurovision and soccer overlap pretty well. Probably not perfectly but I can’t be bothered to check.
David Marjanović : Ankaraspor. I mean I could not immediately recall an Ankara soccer team, but I recognized it when I duckduckgoed “Ankara soccer team”. I think they played some Bulgarian team at some point in the past.
Hm, their Wikipedia article does not seem to support that — as in playing against a Bulgarian team. Maybe my memory is faulty. https://en.wikipedia.org/wiki/Ankaraspor
Going back to languages rather than soccer, the wiki article on https://en.wikipedia.org/wiki/Languages_of_Europe finesses some of the edge cases by having a separate list of “Languages spoken in Armenia, Azerbaijan, Cyprus, Georgia, and Turkey” because “There are various definitions of Europe, which may or may not include all or parts of Turkey, Cyprus, Armenia, Azerbaijan, and Georgia.” OTOH it treats “Maltese” as unproblematically in, and points out that Kazakh is also spoken in Astrakhan Oblast outside Kazakhstan’s own borders. Dagestan is another bit of “European” Russia that conveniently picks up a lot of languages that thus get to be European. I frankly find it somewhat odd to think of the Caspian Sea as having a “European shore” as well as an Asian one, although I understand that in that part of the world everything is so fuzzy that any approach that permits the drawing of an actual line is going to yield a line that seems odd or counterintuitive somewhere along its length.
I’m reminded of a previous discussion in which I discovered that remarkably few Hatters seemed to share my view that Egyptians obviously are, and always have been, Africans.
(IIRC I had sparked the discussion off by remarking on the obvious falsity of the statement in Gerald Browne’s Old Nubian Grammar that “Nubian is the only indigenous African language whose development we can trace for over a millennium.” Browne may in fact have been implying, either that the ancient Egyptians came from Asia, or even that the entire Afroasiatic family is not “indigenous” to Africa, but those are even less sensible propositions.)
Not just African — Hamitic, even.
If I have an opinion on that, it’s that the africanness of Egyptian (or any other language) — and the relevance thereof — depends on the definition of ‘African’ and the purpose of the classification. It’s the same for ‘European’ and ‘Asian’.
@Y:
It occurred to me that Browne, who was really an old-style “philologist” rather than an Africanist linguist, may have been sufficiently out of the loop on African language study in general that he would have imbibed in his youth the notion of “Hamites” from Asia bringing grammatical gender, pastoralism, and all things nice in general to Darkest Africa, and never learnt any better.
https://en.wikipedia.org/wiki/Gerald_Michael_Browne
If one accepts that Egyptian descends from a hypothesized Proto-Afroasiatic, then presumably it’s only “indigenous” if the PrAA urheimat is in Africa. Which may or may not have been the case, with me personally have nothing to offer in judging between “it was in Africa” proposals and “it was in Asia” proposals. (Obv. the urheimat of something ancestral to the Egyptian language need not have been the residence of most of the people ancestral to the subsequent Egyptian population that spoke that language.)
Note that the PIE urheimat may have been in either Europe or Asia, and indeed if you say oh it was definitely somewhere on the “Pontic Steppe,” many candidates for the exact Europe/Asia boundary cut through that steppe.
Indeed, it appears that the location of the conventionalized land boundary between Europe and Asia has shifted somewhat over time, generally to assign formerly “Asian” territory (including some of that steppe) to “Europe” at roughly the same time that Muscovite political rule was expanding or had recently expanded into the reassigned territory. E.g. it was not too many centuries ago that the Sea of Azov was viewed as having both a European shore and an Asian shore such that the current bridge over the Kerch Strait that the Ukrainians keep trying to blow up would have been a transcontinental one.
If one accepts that Egyptian descends from a hypothesized Proto-Afroasiatic, then presumably it’s only “indigenous” if the PrAA urheimat is in Africa.
Just as Welsh is not indigenous to Britain …
Indeed, the logic leads inexorably to the conclusion that there are few – if any – indigenous languages anywhere at all (and none whatsoever outside Africa.)
Just as Welsh is not indigenous to Britain …
And English and French are not indigenous to Africa, a more acceptable conclusion. Now, Afrikaans…
However, I agree with you that all Egyptians are Africans. On the other hand, Egyptian-Americans are typically not African Americans. It’s that simple.
Does anyone think that Welsh is indigenous to Britain, for any useful sense of the word? That might tend to suggest that it’s a word of limited usefulness, of course. Separately, the only reason we can trace Nubian back so far is that we have texts written in it, due to an earlyish local adaptation of the perhaps-not-indigenous Coptic alphabet. That alphabet of course was an “African” modification of the Greek alphabet (European) which was in turn a modification of something Phoenician (Asian) …
There is a long-established theory that the potentially-traceable Urheimat of any currently-extant language ought to have been in or near the plain of Shinar (conventionally taken to be in Asia), with a sudden cusp/break in continuity that occurred in that location making tracing any further back impossible.
To one of Jerry F’s points, Egypt is yet another one of those “transcontinental” polities, with the Sinai typically being classified as part of Asia. This may make “all Egyptians” generalizations hazardous.
J.W. Brewer : “I frankly find it somewhat odd to think of the Caspian Sea as having a “European shore” as well as an Asian one, although I understand that in that part of the world everything is so fuzzy that any approach that permits the drawing of an actual line is going to yield a line that seems odd or counterintuitive somewhere along its length.”
I find it odd to not consider the western shore of the Caspian Sea as part of Europe, but I don’t think it matters ultimately. It’s all very arbitrary.
@DE, I want to note that by calling Maghreb “Europe” I don’t mean it is not “the Arab world”. Both.
And I also do see Maghreb as [cultural] Africa too. My problem with its cultural Africanness is that I poorly understand African cultural influences there. Much worse I think than Lameen, but maybe (or not) even for Lameen it will be difficult to tell how exactly Africa has influenced cultures on its [Maghreb’s] Mediterranean shore.
JWB, or Egyptian belongs to a sub-branch that moved out and then in Africa (as could – or not? Is this known? – happen to Ethiopic)
(And as is known Semitic alphabets may or not be derived from Egyptian)
@V: yes, it’s all arbitrary, and I had not previously spent much time musing about the history of various attempts to draw the line different places. But it does seem like prior to the arrival of Russian soldiers and administrative personnel along part of the Caspian shoreline noone had a definition of “Europe” that reached any part of that shoreline. Whether geographers gaining a better understanding of topography and drainage basins etc would have shifted the line even if the military history had been different is not clear to me. That the Volga would seem like a “European” river along most of its upper course could have been an obvious motivation for carrying “Europe” all the way to its mouth regardless of who lived there, I guess.
The fundamental problem may be that if boundary-drawing team A starts working their way southward from the Arctic Sea along the ridgeline of the Urals while boundary-drawing team B starts working their way across the Black Sea from the Bosporus on a vaguely northeastern heading, they’re just not destined to run into each other in any particularly obvious place.
Half a century, that is. A bit earlier than that, my mom was taught a boundary between Europe and Asia that cut through the lowest part of the area between the Sea of Azov and the Caspian Sea, so the entire extensive Caucasus foreland – Adygea, Chechnya, Dagestan, everything – belonged to Asia.
I was taught the ridgeline of the Caucasus instead, making the recently mentioned Elbrus the highest half mountain in Europe (just as the Zugspitze is the highest half mountain in Germany).
If you follow geology instead, Europe is a patchwork of numerous continents to begin with. Bohemia is one – some 400 million years ago it was surrounded by vast deep ocean on all sides!
Europe is a continent like Pluto was a planet: a historical error from an era of incomplete data and fuzzy definitions. The geographers need to take a leaf out of the astronomers’ book and strike Europe from their list.
@David M.: The even-earlier view (allegedly unquestioned until circa 1725) was that the Sea-of-Azov boundary between Europe and Asia then proceeded upstream along the river Don and never got to the Caspian at all, although the ancient geographers who developed this approach apparently lacked knowledge of where the Don went once you got a considerable ways inland. This approach turns up in the para-Biblical (quasi-Biblical? pseudo-Biblical?) Book of Jubilees, where it marks the original boundary between the descendants of Shem and those of Japheth.
If you follow geology instead, Europe is a patchwork of numerous continents …
That’s a slippery slope (probably literally).
These radicalleftwingrevisionist geographers want to claim Zealandia is a continent “almost entirely submerged mass of continental crust”, with only its mountain peaks (New Zealand) above the waves.
If being below (current) ocean levels is merely an inconvenience not a disqualifier, somebody will want to make the mid-Atlantic ridge a continent — and since that ultimately connects to all the ocean ridge “system which extends for about 65,000 km”, we’ll be adding continents willy-nilly.
No, the ridges are oceanic crust, not continental crust.
That includes all of Iceland.
Iceland is, uniquely, an oceanic island atop a mid-ocean ridge. Still not continental, though.
The Kerguelen plateau has a continental core.
I refer to the Book of Jubilees as a linguistic joke when talking about particle cosmology sometimes, because it is also known as the “Leptogenesis” (or “Lesser Genesis”). In physics, the same name, leptogenesis, was separately coined as a portmanteau of lepton and baryogenesis. One of the most important puzzles in fundamental physics is why there is more matter than antimatter in the universe.* The great Soviet physicist and dissident Andrei Sakharov pointed out that to have this asymmetry requires a number of conditions be met. There need to be: i) differences between the physics of matter and antimatter (C and CP violation), nonconservation of B (the total number of baryons—protons, neutrons, and heavier three-quark particles—minus the number of antibaryons), and nonequilibrium dynamics in the early universe.**
The standard model has plenty of C violation, and it quite plausible for the early universe to be far out of thermal equilibrium. However, the other conditions are harder to meet. The standard model of particle physics does contain both CP violation and B nonconservation, but both effects are too weak to account for the baryon asymmetry that we observe today. However, the discovery of neutrino oscillations a couple decades ago opened up the possibility for a whole bunch of new kinds of CP violation involving leptons—electrons, neutrinos, and their heavier analogs. The idea of leptogenesis is that a strong lepton asymmetry could be generated in the early universe by these new forms of CP violation, which is then transferred to a baryon asymmetry by (normally very weak) standard model processes that do not conserve B or L (the lepton number), but do conserve B − L.
So when giving a talk about symmetry to a knowledgeable audience, I will sometime say something about how the baryon asymmetry could be produced using the Book of Jubilees. I don’t know if anyone ever gets the joke without explanation, but it’s a good way to see if people are paying attention.
* We know there is more because there is more matter than antimatter in cosmic rays, which we can tell are coming from very far away (mostly from outside the Local Cluster of galaxies) since they arrive uniformly from all directions equally.
** The nonequilibrium behavior can be replaced by CPT violation, but that’s a much more exotic possibility—albeit one that I work on myself.
Y : “The Kerguelen plateau has a continental core.” Huh, I didn’t know that, but it makes sense. It’s out of the way of the mid-oceanic ridge system that I think AntC alluded to, that IIRC Marie Tharp discovered in the ’50s. Active plate tectonics is a really important part of why Earth is unlike other planets. Venus has _some_ but not much. It might have had more in the past.
Brett: That stuff is absolutely fascinating, and makes me wish I had the background to get more than a fuzzy distant glimpse of what it’s about.
@Brett: what hat said!
@mollymooly: luckily, we don’t even need to make a new category for it, since we’ve already got both “subcontinent” and “peninsula”! personally, i vote for the latter, since the type specimens (south asia and central america) have tectonic autonomy but europe does not (just as dwarf planets have gravitational rounding, while (as i understand it) asteroids and other small solar system bodies don’t).
Of course peninsulas can be somewhat ill-defined, not unlike continents, I suppose. Exactly where does the Great European Peninsula jut out from the Eurasian Mainland? Just eyeballing it on the map, you’d think the mainland gets notably pinched by the Black Sea on the south and White Sea on the north and some line in between the two of them would mark the beginning of Peninsular Europe. The Urals seem way too far east to be plausible.
I myself grew up either in or near the northern end of the Delmarva Peninsula, whose exact northern boundary is ill-defined not least because political, cultural, and geological criteria for marking it diverge a bit from each other yet are easily muddled up. And often the gap between the different potential lines seems pretty minimal in the grand scheme of things (mostly less than 10 miles unless you take an outlier-aggressive position trying to push the line unusually far south). But even a modest gap can create https://en.wikipedia.org/wiki/Debatable_Lands, and sometimes there are people who live there.
Then the Mediterranean sea will be the Mediterranean bay or gulf.
It is not that there is a principle that a gulf between something and a peninsula (as White Sea above) can’t be called sea, but that would be neat.
But… all gulfs are mediterranean.
So then it won’t have a name.
i’d argue (if forced) that western asia consists of three peninsulas: anatolia, scandinavia, and europe (or possibly europia, for euphony). straying from that more tectonically-based approach, which excludes the arabian subcontinent, to a more purely shoreline-based one, we could perhaps speak of two west asian peninsulas: one defined by a line roughly from rostov-on-don to arkhangelsk, the other by an arc running roughly batumi-baku-tehran-isfahan-basra. i suppose there’s a compromise position to be had here, with scandinavian, european, and arabo-anatolian peninsulas.
[edited to add] @drasvi: if the name weren’t taken already, it could just be the red sea, and we could call the caspian the green sea to complete the cardinal-direction set of colors*! but The
GulfFjord With No Name is much better..
* slavic version, per wikipedia, which credits the 1987 Ukrainian Soviet Encyclopedic Dictionary (and is perhaps based mostly on the ruthenias?)
Bokelji (people of Bay of Kotor) will disagree, they say that their bay is the “southernmost fjord in the world”…
@Brett cosmic rays, which we can tell are coming from very far away (mostly from outside the Local Cluster of galaxies) since they arrive uniformly from all directions equally.
I readily confess most of that post went past my ears. But I did pay attention when I got to “all directions equally”. I understand this claim to be disproven — as it is for the CMB. Here’s a couple of quick results:
When CMB was first discovered, the equipment wasn’t sensitive enough to detect fluctuations — neither angular coverage, polarisation nor distance (and therefore time depth) of source. There’s for example a temperature map at wikip CMB that show significant lacunae, as measured by the Planck spacecraft.
There’s quite a bit of cosmological presupposition deriving from the supposed uniformity of the Big Bang. I think it needs re-examining. Indeed perhaps the cosmological constants are not so uniformly constant, and perhaps models don’t need to introduce ‘dark matter’
That description has echoes of the Luminiferous Aether.
@AntC: I thought about mentioning* that at the very highest energies, the cosmic rays we see do not come from all directions equally. However, I decided not to, because the anisotropy of those cosmic rays turns out not to be relevant to the issue I was talking about. The determination that the highest-energy cosmic rays impinging on the atmosphere come preferentially from the direction of nearby** active galactic nuclei (AGNs) was the solution of another important puzzle in physics. Following the observation of the cosmic microwave background (CMB), it was realized that interactions between ultra-high-energy cosmic rays and the CMB photons would be substantial. At energies above a few times 10¹⁰ GeV, the cross-section for interaction between an energetic proton and a CMB photon grows very rapidly, blocking off the propagation of such protons on scales of tens of megaparsecs (Mpc). This leads to the Greisen-Zatsepin-Kuzmin (GZK) cutoff, an energy above which it was expected that no cosmic rays would be seen. Although the GZK cutoff does exist observationally, it is not quite as sharp as expected, and there are a small number of cosmic rays well above it. This was a puzzle for decades, until the Pierre Auger experiment showed a correlation between the super-GZK events and the locations of nearby AGNs.*** This indicated that the small number of events that were observed probably originated closer than about 100 Mpc—close enough that they had a decent chance of surviving passage through the CMB.
As to dark matter, see my comments here. It is a very ordinary hypothesis, which explains observations on galaxy, cluster, and cosmological scales quite well. Dark energy, in contrast, is entirely different. It is not readily explainable based on our microscopic understanding of fundamental physics, and the evidence for it comes only from cosmological measurements.
* Probably, it would have been a footnote to a footnote.
** The sources are nearby, in that they are still basically in the Virgo Supercluster (close enough that the observed matter distribution at the relevant is still quite anisotropic); however, that’s still pretty far away—well beyond the confines of the Local Group.
*** A linguistic note: The abbreviation for “active galactic nucleus” is “AGN”; for “active galactic nuclei,” it is most commonly “AGNs” with a regularized plural. I imagine this is the case because AGNs are so important in astrophysics, talked about all the time, that it feels natural to pluralize the term in the way one would pluralize a regular word.
On Afro-Asiatic: Isn’t the current thinking that it was spoken in the Green Sahara and split due to the Sahara’s un-greening? Or am I remembering that wrong?
Thanks @Brett, yes I had a feeling we’d touched on the subject before. I see that thread had already thrashed the subject of Dark Matter; it did go on to scientific theories that don’t pass what I’ll call the ‘sniff test’:
* In what way is Dark Matter “explaining” anything rather than a label for ‘our sums don’t otherwise work’?
* “observations” from a speck of flyshit orbiting an unremarkable star in an unremarkable galaxy;
* using puny instruments, and only over a period of a blink of an eye in cosmological terms; and
* certainly not based on even beginning to understand all the weird shit that was going on in the first microseconds after the Big Bang.
* Observer bias much?
* wrt “uniformity”, Dark Matter is 85% of the mass of the universe, allegedly; yet
* less than 0.1% of the mass of the Solar System.
* We lack an explanation of the nature of Dark Matter, and why/how it can be spectacularly absent locally.
* (Not that we’ve actually detected it locally, that <0.1% claim is equally dubious.)
* Then I don't see "a very ordinary hypothesis"; I see an unfalsifiable hypothesis.
A comparison to what I regard as sensible science:
* Logic/Mathematics has Gödel's incompleteness theorems.
* These aren't just some obscure purely-theoretical corner of algebra, we can point to familiar examples:
* It's unknown if there are infinitely many prime pairs;
* It's unknown if there are infinitely many Perfect Numbers, or any that are odd;
* The Kollatz Conjecture seems intractable.
* And that's without going into all the ramifications of Riemann's Zeta function, which might be built on sand.
So a Mathematician can honestly/humbly say 'we just don't know, we might never know'.
It's when cosmologists use seemingly-confident language like "which explains … quite well" (and yes Science journos might be failing to pass on the caveats), that I can practically smell the ectoplasm.
@Hans: the phrase “the current thinking” implies a dominant consensus view but my non-expert impression is that I’m not sure there is one. There are clever hypotheses about an African location and other clever hypotheses about an Asian location and not necessarily enough hard evidence to really choose among them other than by subjective gut intuition about plausibility. That PAA probably started breaking up into daughter languages several millennia earlier than PIE did is a further complicating factor.
@Hans:
No, I think that’s right: it certainly fits the linguistic evidence.
I think an awful lot of sheer speculation is involved, but I find it hard to see how any competent linguist looking at the linguistic facts available nowadays would conclude that the original AA homeland was in Asia, unless they had some kind of ideological bias (or had just never broken free of the ideological biases of their teachers.)
That doesn’t preclude the possibility that proto-Ancient Egyptian might have originated in Asia, I suppose. The earlier linguists who studied it seem to have taken it as being basically a very aberrant Semitic language deeply influenced by “Hamitic”, rather in the way Amharic really is. Partly this was no doubt because of the racist ideologies which were then the norm, but also because scholars familiar with actual ancient Semitic languages were naturally going to be struck by the many perfectly real resemblances, and not a lot was known about the rest of AA in those days anyway, so there wasn’t much else for them to go on. (And Chadic was only finally accepted as pukka AA really quite recently.)
Moreover, predynastic Egypt already seems to have been very mixed in population (though the evidence seems to be unnervingly dependent on things like craniometry of ancient skulls …) So the idea of an influx of proto-Egyptian-bearers from Asia is not obviously stupid, anyhow.
Incidentally, I was struck by the claim here
https://en.wikipedia.org/wiki/Prehistoric_Egypt#Pre-Dynastic_Egypt_(c._6000-3000_BC)
that the goat was only introduced to Africa as late as 6000 BCE, given that the word for “goat” is clearly reconstructable to proto-Volta-Congo. It’s all pretty much guesswork, but I’d say proto-VC has to be quite a bit older than, say PIE. Seems a tight timescale to me …
it did go on to scientific theories that don’t pass what I’ll call the ‘sniff test’
With respect, unless you have a pretty deep immersion in the science involved I don’t know why you think your sniffing has any substance to it. It reminds me of the “my five-year-old could paint better” reaction to “modern art.”
@AntC: Indirect evidence precedes direct evidence all throughout scientific history: gravity, atoms, continental drift, neutrinos. For some things there never will be direct evidence, like the interiors of stars and of the earth or proto-languages. And the explanations are never perfect to begin with.
Do you pick arguments because you’re bored?
In the same way Luminiferous Aether doesn’t pass the ‘sniff test’, nor Chomsky’s Language Gene, nor that there are etymons shared between Proto-Celtic/Old Chinese. Am I forbidden from questioning those because I don’t have a qualification in Linguistics? The story of the King’s New Clothes …
Go read wikip on Dark Matter at “invisible”, “hypothetical”, “does not interact”. “unless more matter is present than can be observed” reeks of the ‘diagnosis by exclusion’ of alleged medical complaints like Chronic Fatigue Syndrome*, as we’ve discussed here before.
* I hasten to add the symptoms/experience for CFS sufferers is very real and very painful, and severely impacts their lives. That doesn’t mean there’s a disease. Neither does it mean it’s a disease of the same causation as Long Covid. (I note there’s now ‘diagnoses’ for Long Covid in people who’ve not been diagnosed as ever having Covid, or at least not any severe form. This is worryingly like ‘diagnosing’ CFS in people who’ve never had glandular fever.)
Similarly, the observations of galaxy/cluster rotational movements, lensing, etc is very real. That doesn’t mean the explanation for them can only be matter of some heretofore unknown constituency. I suggest the puny instruments aren’t very good for purpose**/space is very big, this stuff is many millions of light years distant/it’d be prudent to get better observations before jumping to theories.
** As they weren’t at the time for detecting CMB. That is, they detected CMB is a thing and that you get a reading wherever you point the detector into space. That doesn’t make it ‘uniform’ — as later more accurate detectors have established.
Central America ceased to be a peninsula three million years ago when its tip became connected to South America. Did you mean “isthmus”?
The tectonic autonomy was within Europe all along. The Adriatic Plate continues to drift north, and the Alps keep rising and becoming narrower. Sometimes there are magnitude-3 earthquakes in Vienna.
I think this comes simply from pronouncing the acronym instead of unfolding it in speech (or, for that matter, pronouncing it as a word – compare “a URL” and contrast “an earl”).
In the thread Brett linked to, I casually mentioned the Bullet Cluster but didn’t explain what that is. Behold one of the places where you can see dark matter: the bright matter of two galaxies collided and slowed down; dark matter does not interact electromagnetically, so it does not collide, it passes through and simply continues on, so that by now there are two gravitational lenses on either side of the directly visible collision. As the article says, a few other examples have been found since. Gravitational lensing is something you can see with your own two eyes (and a very big telescope).
But it remains unknown if any of these are genuinely unknowable and thus examples of Gödel’s incompleteness, or if they’re going to be proved or disproved tomorrow like Fermat’s Last.
Victor Mair has made this profound misunderstanding, too.
If AA spread with agriculture, as language families often do, it may well have originated somewhere far from its current center of diversity. The Natufian culture of 15,000–11,500 BP is very popular in this school of thought.
But it does. What it doesn’t pass is the Michelson-Morley experiment.
You seem to have quantified the quality of these observations. Please show your work.
@Hans, I think about it similarly to JWB but.
“any competent linguist looking at the linguistic facts”
@DE, what facts? The spatial distibution, what else?
I tried WP and was disappointed. It says “…is seen as being well-supported by the linguistic data.[101]”
[101] is Frajzyngier and Frajzyngier in the cited work doesn’t express any opinion (this is not what his article is about) but refers the reader to “Ehret, Christopher 2015. Africa from 48,000 to 9500 bce.” I would say that this “is seen” is precisely THE reason to hate passives and other impersonal constructions, because their frequent use in sceince does make people write such things. But Frajzyngier doesn’t even speak of “well-supported”, he says Ehret thinks “that the location of the split was in or near what is now Southeast Ethiopia“
“so that by now there are two gravitational lenses on either side”
@DM, beautiful. Never heard about it!
Central America ceased to be a peninsula… Did you mean “isthmus”
no, i meant what i said, which was “subcontinent” (admittedly i was pushing it a bit to include it as a type specimen, since the only landmass that usually gets called that is the south asian one).
Surely Central America’s peninsular credibility was restored just over a century ago when U.S. capital and engineering know-how restored a water barrier between it and South America.
The canal has locks, so there is always at least one “land” route across the isthmus, although that route changes as the locks are opened and closed.
Europe is my metaphor for the diversity of India. (speaking of Indian continenthood)
From A Passage to India:
Bullet Cluster … Behold one of the places where you can see dark matter: the bright matter of two galaxies collided and slowed down; dark matter does not interact electromagnetically, so it does not collide, it passes through and simply continues on, so that by now there are two gravitational lenses.
I’m not convinced two lumps of our-sums-don’t-work explains any more than one lump. AFAIUI, the Bullet Cluster observations are trying to pick between competing ‘explanations’ ** of apparent gravitational phenomena. They’re not throwing light (ha!) on the nature of what causes the lensing. These two (claimed) masses are separated by some 2 million light years, so it seems … curious … to describe this as colliding. They mightn’t interact electromagnetically, but they have mass so should interact gravitationally(?) Like deviating in their trajectory, even getting captured into orbit around each other?
Anyhoo none of this shows it’s a new form of matter, only that it continues to be invisible/does not interact.
** @Hat I think the charge of ‘my five-year-old could do better’ doesn’t hold: I’m not proposing or defending any other theory. I’m looking for rather more “we don’t know”/”so far we’re guessing” — like what five-year-olds aren’t known for.
@Y I’m tortured by an enquiring mind, and perhaps an over-developed iconoclasm. In what way was trying to get to the bottom of Tocharian peninitial stress or “left edge” ‘picking an argument’? Thank you to @DM for providing concrete information.
But it remains unknown if any of these are genuinely unknowable and thus examples of Gödel’s incompleteness, …
That’s the bugger with Gödel: if a conjecture is incomplete (in the required sense), its contradiction is also incomplete. So we can never know it’s incomplete unless we happen to (dis)prove it, as you say, at which point pfff! it magically turns out it wasn’t incomplete all along.
Re the Aether, before M-M already:
So the contradiction I see is wanting to both claim the Universe is uniform (85% Dark Matter), and claim the local Universe (Solar System) isn’t uniform (< 0.1% Dark Matter — giving the convenient inability to produce any).
Please show your work.
I think I did already (also in the earlier thread Brett linked to). We now know neither Cosmic Rays nor the CMB are as uniform as thought when first discovered — because we now have more accurate instruments. Estimates for the mass of conventional matter/that is, ways for detecting it are getting revised steadily upwards.
Now I regret having changed “her continenthood” to “Indian continenthood” before pressing “post comment”:)
The solar system is, by the standards or our universe and even our galaxy, an extremely dense clump of matter. Ordinary/baryonic/luminous matter clumps a lot more easily than dark matter. Because it interacts readily via electromagnetism, ordinary matter is subject to dynamical friction, heating, and radiation away of energy. This causes clouds of ordinary matter to collapse more readily than dark matter can, and that is what is responsible for the formation of stars. There’s really nothing mysterious, therefore, in why a star like the sun represents a gigantic overdensity or baryonic matter but only a modest overdensity of dark matter; a proton hitting the sun gets caught up on the stellar plasma, while a weakly interacting massive particle just passes through ballistically.
From a particle physics standpoint, there is no reason to believe that there are not lots of heavier weakly interacting particles that are too heavy to produce in colliders on Earth. On the contrary, there are very good reasons to believe that those kinds of particles; for example, if right-handed neutrinos are very heavy, there is a natural explanation (known as the “see-saw mechanism”) for why the left-handed neutrinos we observe are very light (so light they were long though to be massless).
Dark energy is another story. It really is weird, and most physicists probably believe that it isn’t really just the cosmological constant that we stick into Einstein’s equations. The dark energy density just seems like an absurdly small number; if you guessed its value from quantum field theory, a naive estimate would be about 10¹²¹ times larger than what we see (the hierarchy problem). It also, for no apparent reason, happens to be of roughly the magnitude as the amount of matter mass-energy in the universe (the coincidence problem). These are recognized as the biggest problems in cosmology, and people would love to have a deeper understanding that would explain them. However, no more complicated model does a better job of explaining various cosmological measurements so far, although a lot of people have tried to come up with more natural-looking models of what our experiments perceive as dark energy.
and perhaps an over-developed iconoclasm.
You think?
@DE, I tried another link in WP, that is Blench.
As I understand the “linguistic facts” you mean is the spatial distribution of languages.
It is a good argument, I believe.
And speaking of ideology, I do think that the perception of Africa as a shithole is a bad thing. But I can think about a plenty of reasons for someone to believe or want to believe that they came from the east other than ideological.
(Besides depending on arguments both theories don’t even contradict each other. They could have come from the east and then branch in Africa. Then Semites move to Arabia and from there conquer both the shore and Sumer)
* In what way is Dark Matter “explaining” anything rather than a label for ‘our sums don’t otherwise work’?
It’s a hypothesis that accounts for various things, including some like the Bullet Cluster that were unknown when the hypothesis was formulated.
And “our calculations don’t otherwise work” had had some outstanding successes in physics, such as the prediction of neutrinos and that of the charm quark.
* “observations” from a speck of flyshit orbiting an unremarkable star in an unremarkable galaxy;
Why quotation marks?
Cosmology would be out of luck if we were orbiting a remarkable star in a remarkable galaxy, since we’d have a biased view.
* using puny instruments, and only over a period of a blink of an eye in cosmological terms; and
Puny but effective, and the period accessible to astronomical observation is about 98% of the history of the universe (astronomers see the oldest galaxy known as it was 280 million years after the Big Bang, which was about 13.8 billion years), plus the CMB, which was emitted a few hundred thousand years after the Big Bang, according to current models.
* certainly not based on even beginning to understand all the weird shit that was going on in the first microseconds after the Big Bang.
Irrelevant to the question of whether there’s dark matter now.
* Observer bias much?
Evidence for bias much?
The few astrophysicists who doubt the existence of dark matter don’t seem to doubt the observations.
* wrt “uniformity”, Dark Matter is 85% of the mass of the universe, allegedly; yet
* less than 0.1% of the mass of the Solar System.
That’s because the Solar System is far denser than the the universe as a whole.
I’m 20% of the human population of the building I’m in now, but if I were in the [*searches*] Regency International Center in Hangzhou, I’d be more like 0.005% of the population.
* We lack an explanation of the nature of Dark Matter, and why/how it can be spectacularly absent locally.
It is true that the nature of dark matter is unknown. There’s no reason to think it’s absent locally.
* (Not that we’ve actually detected it locally, that <0.1% claim is equally dubious.)
There’s nothing dubious about the “less than” claim, which is consistent with 0.
* Then I don’t see “a very ordinary hypothesis”; I see an unfalsifiable hypothesis.
If the rotation curves of galaxies weren’t so similar, the hypothesis would have a much harder time.
A comparison to what I regard as sensible science:
* Logic/Mathematics has Gödel’s incompleteness theorems.
That’s not science. There’s a big difference between making models based on observations of nature and studying the properties of formal systems, not that there’s anything wrong with either.
* These aren’t just some obscure purely-theoretical corner of algebra, we can point to familiar examples:
* It’s unknown if there are infinitely many prime pairs;
* It’s unknown if there are infinitely many Perfect Numbers, or any that are odd;
* The Kollatz Conjecture seems intractable.
Those things were unknown before Gödel. (Though Collatz published his conjecture later, he had apparently started working on it before Gödel’s publication.) The most I see that you can say about their connection to Gödel’s Theorem is a guess about the sociology of mathematics: Probably after Gödel, many more mathematicians have thought those questions might be unanswerable.
* And that’s without going into all the ramifications of Riemann’s Zeta function, which might be built on sand.
I don’t think so, unless you’re talking about the mathematicians who, I’ve read, just assume that Riemann’s Hypothesis on the zeta function is true.
So a Mathematician can honestly/humbly say ‘we just don’t know, we might never know’.
So can physicists, and I hope some do—about why the universe exists, whether there are other universes, etc. (I say such things to my students.)
It’s when cosmologists use seemingly-confident language like “which explains … quite well” (and yes Science journos might be failing to pass on the caveats),
The Wikipedia article on dark matter includes a statement you might like: “The prevailing opinion among most astrophysicists is that while modifications to general relativity can conceivably explain part of the observational evidence, there is probably enough data to conclude there must be some form of dark matter present in the universe.[21]”
There was a good quotation from a physicist at Wikip, about three alternatives including that the dark-matter hypothesis for galaxy formation was wrong, but I can’t find it now. It’s not in the article on dark matter or the one on the Bullet Cluster.
that I can practically smell the ectoplasm.
I think your sniff test is less well founded than dark matter.
@AntC: If you’re wondering why my answer to your point about 85% versus <0.1% differs from Brett's, I think it was because we understood your tersely expressed argument in different ways. I think he thought you expected the distribution of dark matter to roughly match the distribution of ordinary matter, maybe since they feel the same gravitational forces. I thought you were confusing amounts with percentages. Now I don't know which of us understood you correctly—maybe neither.
Also, I said astronomers can see galaxies from when the universe was 2% of its present age, but I can’t say whether they can measure the rotation curves of those galaxies, for example.
Thank you again Brett, thank you Jerry. ‘fraid I have a plane to catch, but a couple of quick thoughts. (I’m not espousing any of these ideas, just pointing out that to confirm or reject them needs closer observations than we currently have.)
— that is, large, stable particles of ordinary matter. I understand we couldn’t/wouldn’t want to create those in a particle accelerator. If they’re a thing, they’d have been formed in well less than the first 2% of the age of the Universe. So not knowing conditions then is confounding.
If they’re a thing, they might have subtle effects tunneling through/collapsing white dwarfs. How do we observe that?
What I meant by “blink of an eye in cosmological terms” is (for example) we haven’t studied the rotation of a single galaxy over extended time. How can we be sure we’re accounting for its angular momentum and consequent mass? — the ‘Winding Problem’.
a proton hitting the sun gets caught up on the stellar plasma, while a weakly interacting massive particle just passes through ballistically.
The chief requirement for dark matter particles is to have mass. Never the less, their mass would be tiny compared to a star. Then why don’t they get stopped/caught in the star’s gravitational field? And/or get gravitationally attracted to the black hole at the centre of a galaxy. Ok, you’re claiming they don’t ‘clump’ as readily as ordinary matter; they don’t need to at first: over the lifetime of the Universe, the gravity of the formed clumps will pull them in. (As I understand it, they have to be co-located in the region of ordinary matter, to get the observed lensing effects.)
How would getting stopped in the gravitational field of a sun (not black hole) work? Without electrostatic friction, wouldn’t the dark matter either (1) escape, (2) orbit, or (3) tunnel through the centre of the sun in an oscillatory manner? In the case of (3), how likely is it that improbable interactions with ordinary matter would have stopped the dark matter over time? In the case of (2), wouldn’t most of the orbits be very eccentric (since no clumps of dark matter closer to the Sun)?
There’s no hypothesis as to the mass of individual particles, nor their speed (can it vary?). If they approach a sun tangentially, gravitational capture would give an initially eccentric orbit.
If they zoom straight through or past/through the whole galaxy, in what sense are they contributing to its mass in order to explain lending or rotation? If there’s equal numbers of particles zooming in random/opposite directions, how does this form the hypothesised ‘halos’ around galaxies?
peninsulas can be somewhat ill-defined
Ontario Peninsula
the spatial distribution of languages
Partly, but also the diversity, which leads to hypotheses about the likely time depth of the protolanguages for each AA branch. Cushitic (for example) is so diverse that it has been seriously suggested that all of AA should really be regarded as branches of Cushitic.
I note that the major proponents of an Asian origin for AA who are mentioned in the WP articles are a Nostraticist, the fabulist Jared Diamond, and an archaeologist specialising in the Pacific region … and the argument that you can track language origins from archaeological evidence for agriculture strikes me as a good example of imposing your wonderful Theory on the messy Data (Diamond certainly has form in this respect.)
And Blench is surely right in pointing to an ideological aspect to the Asian-origin view.
I don’t myself have any axe to grind in this. It would not cause me any personal grief if the idea of the Natufians as the speakers of proto-AA panned out. I’m not a bit bothered by the fact that the great majority of current speakers of AA languages in Africa do in fact speak AA languages which have origins in Asia (Arabic and Ethiosemitic.) So what?
I am allergic to confusion of human genetic origins with language family origins, and (especially) to the toxic myths of African cultural history summed up in the term “Hamitic.” The Natufian thing need not rely on such errors, though.
peninsulas can be somewhat ill-defined
The
https://en.wikipedia.org/wiki/Taymyr_Peninsula
is one of those where I went: “Peninsula? Yeah, well, I suppose it is, once you start looking at it.”
This morining’s “All Threads Are One” etymological update: the “pen” in peninsula is from the same Latin etymon as the “pen” in “peninitial.”
AntC: If you’d been around a century ago, you’d have been saying Einstein’s theories didn’t pass the “smell test.” I admire your chutzpah, though.
@DE, by spatial distribution I mean diversity too. I won’t call a simple contour “distribution” (of what?!) and conversely, knowledge of diversity within the treee without knowledge of how its leaves are distributed in space can’t tell you anything about the “homeland” (whcih is located in space, not on the tree).
And as soon as you have subscribed names of “languages” on a map you’re speaking about diversity.
For me distribution is a way to connect two values. Mapping of one structure (linguistical distances, the tree) to another (the geography). So for me it is one criterion – but I agree that it is complex.
The quotation I couldn’t find (which appears to contain further quotations) is from https://en.wikipedia.org/wiki/Cuspy_halo_problem:
The frustratingly repetitive and uninformative article is about an annoyingly named but real problem with dark-matter models.
The current spatial distribution of Cushitic languages is meaningless on its own. People will plug it into Urheimat-location theories because they have added an assumption (whether explicitly stated or not) that migrations over the last X thousand years have been sufficiently modest in scope that the current spatial distribution gives you meaningful insight into where the predecessor was spoken X thousand years ago. Historical/archeological evidence could in principle tell you one way or another, but a default assumption about whether people are more likely to have stayed put or moved over the last X thousand years if you don’t have enough actual non-linguistic evidence to tell you one way or another is not the sort of assumption that historical linguists, as such, have any particular competence to assess the strength of.
No: this is about the diversity of the Cushitic languages, not just their current location.
Either all the Cushitic languages coincidentally spread, one after the other, from the homeland of Cushitic in Asia to Africa, over the six millennia or so needed to account for this diversity, or proto-Cushitic itself was spoken in Africa.
The former is logically possible, but it’s hardly reasonable to take its thumping improbability as a mere “assumption” by historical linguists amateurishly intruding on the rarefied realm of Proper Science.
Sure, absolute certainty is unattainable. But it’s unattainable in archaeology too. (And in cosmology …)
@lamguagehat: Einstein, of course, later believed that (the interpretation of) quantum mechanics didn’t pass the sniff test.
I unfortunately managed to get myself muddled between different levels of generality. The plausibility of the Proto-Cushitic Urheimat being in Africa doesn’t establish that the (earlier) Proto-AA Urheimat was also there, just as locating a plausible Urheimat for Proto-Germanic doesn’t mean that the PIE Urheimat was in the same place or even particularly close.
Here’s a piece speculating about the proto-Kartvelian Urheimat, which I suppose may be of interest to certain nationalists but can perhaps be viewed dispassionately from a distance. https://pmc.ncbi.nlm.nih.gov/articles/PMC10689496/
It addresses actual linguistic data of a lexical nature. E.g. “Cattle-breeding vocabulary and terms for wine are common to Svan and the other Kartvelian languages, whereas terms for crop cultivation, sheep-breeding and metallurgy, which are common to the other Kartvelian languages, are absent in Svan.”
DE, their diversity COMBINED with their location in Africa.
(And “came from the east” does not mean “branched in the east”
And “first branching took place in the east” does not mean “all branchings took place in the east”.)
@DE we know the the east mediterranean is a place of numerous important inventions made precisely within the time range that you need that certainly led to fast growth of the number of locals, and could give them competetive advantage (if not then their numbers gave it).
I think you’ll agree with this. Our interest to this place and period is excessive, but it has little to do with our views on other places. It is not “Africa” that is worse than, say, “Russia”, it is the east mediterranean that fascinates everyone.
In my head, I will mehanically keep in mind possible mechanisms of migration from there… and mechanisms of migration to there because didn’t Akkadians move into Mesopotamia from elsewhere?
And if you can choose between two options, there will be scholars who prefer both. Because even “everyone speaks of A but not B!” is a good reason to take interest in B, and having taken interest is a reason to like it.
“Competent” linguists are not competent in evaluating spatial distributions. They are competent in calculating linguistical distances. But then once you’re looking at a map, you can’t precisely tell what exact numerical weight this distribution in space gives to the likelihood. It’s impressionistic to you.
And for this reason someone who has a reason to prefer eastern homeland may assign a lesser weight – they don’t know any exact method for this.
And such reasons (to prefer the eastern versions) are plenty and may have nothing to do with one’s view of Africa. You may like such mechanisms (spread of languages with the spread of numerous peoples) or you may like agriculture, or you can be a specialist in East Mediterranean (like Semitologists and Egyptologists) and want to expand your area of interest.
Speakign of ideology (and thinking “racism!”) whenever anyone disagrees is a bad idea. Ideologies TOO may play role, but they’re not the only reason someone may disagree with you.
>Either all the Cushitic languages coincidentally spread, one after the other, from the homeland of Cushitic in Asia to Africa, over the six millennia or so needed to account for this diversity, or proto-Cushitic itself was spoken in Africa.
A lot seems to hinge on the accuracy of six millennia. I’m less certain than you that we know much about the processes and timelines of diversification in pre-state societies, especially in what might have been settings of population movement without subsequent regular long-distance communication, across areas with (maybe) wildly varying substrates. In other contexts, I thought you disparaged the idea of glottochronology being so definitive.
A lot seems to hinge on the accuracy of six millennia.
No, I plucked that number out of the air (though it’s probably not unreasonable.) I just meant to convey some idea of the sheer diversity we’re talking about. Much greater than Indo-European …
The internal linguistic diversity of the non-Semitic parts of AA is vastly greater than that within Semitic. This is surely significant no matter what reservations one has about glottochronology (I have many.) You don’t need any putative measurements of time in years to say this. Lexicostatistics is easily abused or misinterpreted, but its value (such as it is) is independent of the mug’s game of trying to make it yield anything like firm dates.
Lexicon is not the only relevant criterion for divergence, either: we’re looking at major differences in phonology, morphology and syntax too.
It’s absolutely true that language change doesn’t happen everywhere and always at the same rate, or anything like it. I have noticed, though, that this point tends to be coupled with an assumption that change in (for example) Africa is likely to be faster than elsewhere, an assumption that itself may come from some rather dubious ideological ideas about Africa, along with the perennial tendency to underestimate the enormous diversity of the continent and its peoples, cultures, histories and languages.
No doubt language change really has been rapid in some places and at some times in Africa. But I can also produce clear instances of the opposite. Take the Kusaal verb tʋm “send.” This has clear cognates all across Oti-Volta, which is unsurprising given that the whole group is much less diverse than Indo-European (though more diverse than Germanic, Celtic or Italic.) But cf also proto-Bantu *tʊ́m- “send”, as seen in e.g. Swahili tuma “send” (where the -a is a flexion.) In the case of Bantu, there actually is relevant archaeological work suggesting that the expansion began about the same time as that of Indo-European; proto-Oti-Volta was very different from proto-Bantu, and at the very least you need to add another millennium to get to a halfway plausible date for a common ancestor of Kusaal and Swahili. But the cognate verb stems are still virtually identical after all that time.
I’ve cherry-picked here, of course, but this example is not isolated. I have actually been rather surprised that Oti-Volta and Bantu don’t seem different enough, given the admittedly rather equivocal archaeological evidence – but much of that is very much open to interpretation too.
Speaking of ideology (and thinking “racism!”) whenever anyone disagrees is a bad idea. Ideologies TOO may play role, but they’re not the only reason someone may disagree with you.
Indeed not – and I said as much. But both the racism and the colonialist ideologies in question were (alas) quite real and (also alas) demonstrably misled linguistic scholarship on Africa. It’s a thing that actually happened, not a flailing-out accusation I’m making about scholars whose conclusions I happen to disagree with.
I also think that these things have cast a long shadow, and still affect the preconceptions of scholars who are in no way personally racist or prone to colonialist attitudes. Blench is right.
A cause of bias which is by no means racist or colonialist is that people inevitably interpret new data in terms of what they already know. If you have a solid background in Semitic languages, you will be extremely likely to see AA as a whole through a kind of Semitic lens, unless you are aware of this danger as a source of misinterpretation of the data and consciously strive to avoid it. It’s easy to slip into a mindset where Semitic becomes the norm from which the rest of AA has deviated historically …
I’ve had problems with this kind of thing myself in Oti-Volta work. I know Kusaal very much better than any other Oti-Volta language, and have quite often misinterpreted things by falling into an unduly Kusaalocentric viewpoint. (Unfortunately, Western Oti-Volta as a whole turns out to be a pretty aberrant Oti-Volta branch in many ways, despite accounting for the great majority of all actual speakers of Oti-Volta languages.)
But I am really by no means xenophobic about the peoples of northern Togo and Benin … even though they don’t speak proper Western Oti-Volta.
It seems to me that much of the discussion around the age of “Cushitic” is wildly premature, inasmuch as, to my knowledge, nobody has even tried to make a purely linguistic case in favor of there ever having existed a Proto-Cushitic language (distinct from Proto-Afro-asiatic) in the first place.
Doubts have been expressed about Beja (“Northern Cushitic”, or a separate branch of Afro-asiatic altogether?) and about Omotic (Most scholars no longer think it is part of Cushitic, but it is unclear whether it is part of Afro-asiatic at all), but even the remaining languages (“Core Cushitic”?) might well constitute more than a single branch of Afro-asiatic.
As an outsider to Afro-asiatic historical linguistics, I am struck by the fact that the “shared common features” of Cushitic and Chadic (in sharp contrast to those of Berber, Egyptian and Semitic, each of which makes up a clearly-defined branch on purely linguistic evidence) seem to be areal rather than genetic in nature. This certain belief in the existence of a Proto-Cushitic and/or Proto-Chadic language seems all the more unwarranted considering how under-described, or indeed wholly undescribed, many Cushitic and Chadic languages are.
DE, as I said, I do like the argument (diverstity).
I’m objecting (and very strongly) to your words that it’s hard to see how any competent linguist etc. will question the African version without an ideological bias. This does mean “everyone who disagrees likely does so because of an ideology”.
When we speak of ideology in this context, very ugly things come to mind – but I guess Jared Diamond is interested in agriculture and the way it affected people, which you can call ideology maybe but isn’t what one will think about when you say “ideology” – and similarly Semiticists all take particular interest in their geographical area of study but may differ in views on Africa.
I say “very strongly” because I think such interpretation of scientific disagreement destroys science.
But I agree that certain ugly ideas may affect opinions of SOME people.
Jared Diamond certainly has form when it comes to ideologically-driven misinterpretation of the facts, but the ideology in question there is not racism or colonialism.
https://www.imediaethics.org/jared-diamonds-factual-collapse/
(Interesting sidelights on New Yorker fact-checking, incidentally. All Threads are One.)
I was intending “ideology” in a much more neutral sense – it’s not in itself a boo-word for me: there are good as well as bad ideologies.*
The problem is not in just having an ideology (everyone does, including those who vehemently deny it) but in letting your ideology cloud your view of reality. Avoiding this is often easier said than done, and it’s a lot easier to recognise the phenomenon in others than in oneself. I find.
Those who never reflect on the sources of their own ideology (or deny that they have one at all, which is much the same thing) are in particular danger, though.
* https://www.songlyrics.com/fela-kuti/opposite-people-lyrics/
Some mixing up of truly ugly thigns with simple excessive excitement about the East Mediterranean and everything that people invented there (which is oversimplification but not necessary “ugly”) is contained in the very word “ideology”. When Blench uses it he may mean both.
But once it is said, it is difficult not to think about you know what. Besides, as I said, some people make like the idea of migration for different reasons, and there is not even a contradiction: wherever the branching happened, speakers could have moved to and from Africa 20 times before it.
@DE, thanks. Now I understand you better.
You spoke about ideological biases of teachers and some of such biases come from 19th century when racism and other ideologies placing humans, cultures, languages and places one above other were the norm.
It is difficult not to think about this all – and it MAY play a role here. For SOME people.
Besides, Semiticists and some others will have a bias which I won’t call “ideology”
Всяк кулик своё болото хвалит, “every cook praises his own broth” or even the bias of people who either know a lot about place A and little about place B whcih is not ideological.
I gave an example recently of a strong (even crazy) bias based on knowing too little (precisely nothing) about something. The view on Arab women frequently heard in Russia (it is not “everyone’s view”, though). The reality is not that many of them don’t have problems specific to Arab culture. But they’re talented people in a rich (intellectually too) word who atop of that have those problems. A Russian instead thinks “I don’t see them, I never hear about them – perhaps all they do is obey men” and a view meant to support (unless it is simple islamophoby) becomes an insult.
Всяк кулик своё болото хвалит, “every cook praises his own broth”
Exactly.
Nowadays, at least in linguistics, I’m sure that this is a vastly more important source of bias than actively held toxic political ideologies. Chomskyans see the world as data for yet further confirmation of the One True Path … Bantuists mistake proto-Bantu for proto-Niger-Congo …
Fortunately, there are still people like you and me to tell it as it really is …
@DE: the fabulist Jared Diamond
I know some criticisms of Diamond and can imagine a few more, but what are you referring to in particular?
(Ed.: I had missed your earlier comment.)
>The internal linguistic diversity of the non-Semitic parts of AA is vastly greater than that within Semitic.
To say nothing of the internal linguistic diversity of the non-Egyptian parts versus that within Egyptian!
Which should tell you that something is going on other than “more diversity = deeper point of divergence”. It’s clear that there has been a massive erasure of diversity in the last 6,000 years, stemming from cultural innovations and population expansions out of the Middle East. The question is whether AA was a similar but more ancient expansion. I’m definitely not saying you’re wrong. Various parts of Africa do seem like plausible homelands.
But claims based on diversity when we know little about the dynamics that created diversity prehistorically and nothing about the pattern of Semitic diversity that may have been overwritten by known cultural expansions seems weak. It certainly doesn’t convince me that “any competent linguist” would have to agree with you.
And I find this somewhat surprising as well:
>Incidentally, I was struck by the claim here (https://en.wikipedia.org/wiki/Prehistoric_Egypt) that the goat was only introduced to Africa as late as 6000 BCE, given that the word for “goat” is clearly reconstructable to proto-Volta-Congo. It’s all pretty much guesswork, but I’d say proto-VC has to be quite a bit older than, say PIE. Seems a tight timescale to me …
That again seems like a startling reliance on your instinctive glottochronology over evidence.
I believe the most recent estimates from Green Sahara researchers have goats spreading through Africa a tiny bit earlier than that — “after 7,000 years ago” (as in Brierley 2018, which seems to be up to date). Why not accept that date for sake of argument as terminus post quem for proto-VC. We know little about pre-state divergence dynamics, and the possibility of dating proto-Volta Congo seems like an interesting opportunity.
What other dates might flow out from a ~6,500 ya estimate for proto-VC?
@ryan, de
Who is to say that the word for “goat” in Proto O-V did not mean “some wild horned grazing animal” and was repurposed or restricted by daughter-language speakers to goats. I think there are parallels for PIE plants and animals.
NB that 6000 BCE = 7949 BP*, which is earlier than 6500 ya (= 6423 BP). This piece FWIW is broadly consistent with Ryan’s statement and says “Archaeological data suggests that goats entered Africa from the Middle East 7,000 years ago [as of 2019] with humans.” https://sciafmag.com/2019/06/26/genetics-help-solve-puzzle-of-how-goats-spread-through-africa/
*Here in the Eternal Present it is always Jan. 1, 1950. Which creates some ontological problems for those of us whose parents have not yet met each other.
On PlasticPaddy’s point consider the difficulty in translating the Year of the 羊 in the traditional Chinese zodiac sequence. Sometimes it’s rendered as the Year of the Goat, other times the Year of the Sheep, or of the Ram. The Sinitic word apparently occupies a higher-level of taxonomic generality, but no one wants the accurately-vague Year of the Caprid/Caprine.
That again seems like a startling reliance on your instinctive glottochronology over evidence
Did you miss the part where I said “It’s all pretty much guesswork”?
So, guessing (yup): archaeological evidence for the beginning of the Bantu expansion suggests something like 4000 BCE. The relevant etymon is found in both proto-Bantu and proto-Oti-Volta. These languages, though certainly related, were very different:
https://www.academia.edu/37373745/Bantu_grammatical_reconstructions
https://zenodo.org/records/14787434
(Not in the same league as Meeussen by any stretch of the imagination, but the best thing yet available)
If you can point me to a case where two related languages anywhere got to be as different as this in less than two thousand years, I shall be grateful.
Once again, I reiterate that this is guesswork, and I said so explicitly. No glottochronology has been harmed in the creation of these comments.
One point that is relevant here is that although proto-Bantu and proto-Oti-Volta are indeed very different, the actual words involved here were quite similar, viz pB *nɪ̀-bʊ́dì, pOV *bʊ́d-ká; the different position and identity of the singular class affixes is characteristic, but the actual stems are entirely close enough that borrowing looks like a real possibility, rather than shared descent: that would make the dating issue moot.
This is not such a stretch: borrowing within Volta-Congo has pretty certainly happened with “cow”, pOV *nâg-wʊ́, an etymon found right across West Africa, including not only Grassfields “Bantu” but languages which are not Volta-Congo at all, like Fulfulde and Dogon; but it’s absent in yer actual Bantu.
Skepticism about Omotic has since reached the next level. I think the progression has gone like this:
1. “North Cushitic”.
2. Not Cushitic; new name Omotic. But still AA.
3. Perhaps not AA.
4. Probably not AA.
5. Maybe Omotic doesn’t real; some of its very diverse “branches” may be AA after all.
I thought of that; but shifting in the same direction in all (?) branches is not very parsimonious.
Unless, of course, the meaning shift was itself borrowed (“check out this new domestic type of [some wild horned grazing animal]!”); a bit like firewater.
Who is to say that the word for “goat” in Proto O-V did not mean “some wild horned grazing animal” and was repurposed or restricted by daughter-language speakers to goats.
Well, it’s “goat” specifically in every language …
But Oti-Volta is a lot less diverse than PIE say, so in that case there just isn’t a problem: there’s no need whatever to project pOV back to 6000 BCE.
But you probably really meant “proto-Volta-Congo”, in which case, yes, maybe.
The other thing that occurs to me is that proto-Oti-Volta (as reconstructed by me) might not go back more than three millenia or so. which means that there is quite a time gap from the presumed common ancestor of Oti-Volta and Bantu during which it could have been enweirdifying in situ and getting less like proto-Bantu. It’s not legitimate to compare it with proto-Bantu as if they were spoken contemporarily.
Against this is the fact that Oti-Volta has several much closer relatives than Bantu, including the Grūsi languages, along with Miyobe and Koromfe. The proto-Central-Gur language ancestral to all these has pretty much got to be getting on for as old as PIE, unless the whole area broke all records for rapidity of language change. Of course, it might have done; but it would be nice in that case to have some non-linguistic evidence that the conditions were so abnormal in world terms as to cause this, not once, but over and over again.
I don’t think people who’ve not actually tried to do historical linguistics on these languages themselves quite appreciate what “diversity” actually entails here. You’re talking about relationships which are often near the limits of what proper comparative methods are even capable of. (Greenberg confused the situation greatly by using methods which led many to imagine much closer relationships between these languages than have actually been demonstrated properly, or in some cases, even can be demonstrated properly.)
It’s always possible airily to assert that Africa is so unstable that African languages change much faster than elsewhere*, but in the absence of any actual evidence for this proposition it seems more, erm, parsimonious to suppose that the difficulties might actually have something to do with great time depths.
* I’m reminded of R M W Dixon’s view that comparative methods basically don’t work with Australian languages Because Culture. This has not exactly persuaded many Australianists. Seems to me that uniformitarianism is a good position, unless you’ve got pretty solid evidence to point to for doubting it in some really specific circumstances.
Might be useful to know what sort of undomesticated horned critters were common in the postulated proto-Volta-Congo urheimat before goats got there and whether there are one or more reliably reconstructed PVC words for them that definitely don’t mean “goat” in the daughter languages.
a higher-level of taxonomic generality, but no one wants the accurately-vague Year of the Caprid/Caprine
Got me thinking about мелкий рогатый скот. It’s an economic/bureaucratic expression encompassing domesticated Caprinae. Seemingly untranslatable to English.
@de, jwb
Sorry, I must have shortened a longer comment. As de says, the issue is whether the Proto Volta-Congo word had to have had the specific referent “goat”.
You seem to be trying to do physics without maths.
Yes, they should interact gravitationally; the two invisible masses should therefore slow down slightly – but much, much, much less so than the visible matter. Like a neutrino. In other words, they should continue on the original paths of the two galaxies, at nearly the original speeds, while the visible matter that traveled with them has also been interacting in other ways and has therefore slowed down much more.
The people who’ve done the math say neutrinos alone – at least left-handed ones, the only ones yet found – cannot account for this.
@PP:
I’m still wondering about the damned Volta-Congo tortoise … why the devil should “tortoise” be reconstructable to proto-V-C when not even “man” or “woman” are?
Proto-Bantu borrowed its words for “cow” …
Kusaal kukur “pig” is a loanword from Portuguese porco. Anything is possible …
In the nature of the case, it’s difficult to come up with clearcut examples of an existing etymon that has essentially become a calque in a whole language family. You need extralinguistic evidence that the meaning as found in every daughter language can’t possibly be original, on historical grounds. Easy to end up in a circular argument …
I’ve often wondered about “cowry” in West Africa. The word is apparently reconstructable in proto-Oti-Volta, but that has just got to be an illusion. (There’s actually good internal linguistic evidence for borrowing from Western Oti-Volta by other Oti-Volta languages, which helps, but doesn’t solve the problem: the WOV form has nothing about it phonologically to suggest a loan. What did it mean before it meant “cowry”?)
@de
This is not too hard. They used peanut shells, but they biodegraded (or the peanut crop failed for too many years, or there was a bumper crop which caused inflation, or a very influential person had a nut allergy).
“Year of the Caprid/Caprine” – The Day of the Triffids
Maybe I overreacted, since you (DE) do say it’s all guesswork.
It’s worth noting there are other lines of evidence, like research on goats that suggests “a single major expansion event of goats into central (Cameroon) and west Africa regions.” It’s seductive to think that a single expansion of a primary economic differentiator might relate to the expansion of populations who share the word for that commodity.
PP, it’s always possible that several different populations who spoke related VC languages independently adopted the same cognate that they had used for some African bovid (noting there are no caprids in the relevant parts of Africa) as their goat word.
It’s worth considering whether the range of languages in Africa, even allowing for DE’s reasonable caveat that too much lumping has gone on, is as diverse as that in New Guinea. To me, that’s evidence for sweeps that eliminated African linguistic diversity, and we would need to ask what factors prior to pastoralism and agriculture could have inflicted that on a territory whose languages show some strong centrifugal tendencies.
David, you ask for an example of linguistic diversification on a 2,000 year timescale. But that’s part of my point. We don’t have a lot of analogous situations where an agro-pastoralist population expanded and intermarried across a vast diversity of hunter-gatherer languages from different language families without wheeled technology to sustain the centripetal linguistic effects of large scale long-distance trade, possibly with some cross-mapping expansions in the time since. My hypothesis is that diversification can proceed very quickly in that setting. I do think there is evidence that linguistic diversification is more like punctuated equilibrium than slow, steady evolution. What are the events that punctuate the equilibrium?
I would also hypothesize that the full Bantu economic package was more self-reliant than the initial pastoral expansion. Self-reliance would lead to lesser interaction with the HG populations, and weaker substrate effects.
“It’s an economic/bureaucratic expression” –
lit. “small horned cattle”
And the weird thing is that I heard or read a thousand times about similarly “economic” крупный рогатый скот, “large horned cattle” (cows) but NEVER about “small cattle”.
D.O. мелкое быдло и большое быдло* would sound less clumsy)
___
*I think – based on its sound – that it could once be a word for cattle. (checked the dictionary… yes)
But today it is only used as a self-applied insult.
“liberals [so called “elite”, Muscovites, the rich, intelligents…] think everone else is б.!”
(I’m not sure if I EVER heard anyone call someone else so)
Isn’t the point (as far as the Europe classification question goes) that the idea of a peninsula is pretty naturally linked to lack of definition in one part of the boundary?
I agree with Ant’s high-level point that in practice we don’t usually know whether a conjecture is unproven or unprovable, but it’s worth keeping in mind that the thing that’s incomplete in the incompleteness theorem is a formal system, not a conjecture. I suppose there’s room to label statements exemplifying incompleteness that way (by analogy with NP-complete, perhaps?), but in talking about such statements without reference to the system in which they are unprovable is the first step to a whole lot of takes on incompleteness that shouldn’t pass your sniff test.
In b4 “turtles all the way down”… you’ve suggested previously that the word is a sound-symbolic Pan-Africanism; maybe it just, like, speaks to how the human mind works, so it’s adopted everywhere it becomes known…
It has been suggested that the spread of Pama-Nyungan languages across most of Australia, and that of Austronesian across Indonesia & surroundings, happened as part of the spread of a culture & religion. (The spread of agriculture across this Austronesian-speaking region would have happened earlier, together with a spread of Austroasiatic, according to Blench.)
Bokelji (people of Bay of Kotor) will disagree, they say that their bay is the “southernmost fjord in the world”…
Residents of New Zealand and Chile would disagree, and if there were residents of the Antarctic Peninsula, they would too.
factors prior to pastoralism and agriculture
Linguistic evidence is consistent with agriculture and pastoralism having got to the Gur area about when proto-Central Gur was around, which my best guess (if Ryan will forgive my glottomancy) would put possibly as late as 3000 BCE; that seems consistent with the archaeologists’ notions, and also with the beginnings of the Bantu expansion to the southeast.
Certainly proto-Volta-Congo must predate this; on the other hand, the homelands of both Central Gur and Bantu may well have been a lot closer together at that point. A previously relatively compact Volta-Congo area was then able to expand both southeast and west when the speakers adopted agriculture.
I’ve long found it both notable and surprising that it’s easier to find cognates between Oti-Volta and Bantu than between Oti-Volta and what were traditionally regarded as peripheral “Gur” languages to the west of Oti-Volta and Grūsi, like the Senufo group, and Kulango and Tiefo. I woudn’t be surprised if at the end of the day Central Gur actually turned out to be more closely related to its eastern neighbours than its western neighbours, and represents the outcome of a comparatively late Drang nach Westen.
sweeps that eliminated African linguistic diversity
While Africa is certainly less diverse than New Guinea (where isn’t?) I think that it is actually very considerably more diverse than the accepted Greenberg-lumpery suggests. Most people think Mande is a separate group now; in my (minority) view Atlantic is neither a unity nor demonstrably related to Volta-Congo (I will be proved right one day); there is really no good evidence that Dogon is related to V-C; even Greenberg admitted that his reasons for including Kordofanian were really just typological similarities; “Ubangian” is not a real grouping, and much of it is not Volta-Congo … Ijoid is not Volta-Congo … Greenberg’s Nilo-Saharan is a methodological catastrophe, in reality consisting of about a dozen language families which have not been shown to be related to one another; Khoisan is not a single family …
AA is a thing, but epically diverse internally, and Omotic is probably not a real group and is mostly (or all) not AA.
Even Blench has happily identified several presumed isolates in Africa, where he’s not blinded by his stare decisis attitude to Greenberg’s scheme.
> It has been suggested that the spread of Pama-Nyungan languages across most of Australia, and that of Austronesian across Indonesia & surroundings, happened as part of the spread of a culture & religion.
I think you’ve mentioned that before so maybe I should try to remember it this time.
> his stare decisis attitude to Greenberg’s scheme.
LOL.
I couldn’t find papers suggesting a consensus on the Bantu expansion date, though maybe Ehret has some seigneurial rights over the question, and maybe the consensus has developed recently enough that you can’t find it on academia and researchgate.
And you’re probably right about the rest too. But I do wonder how the pVC goat word came about. And why tortoise domestication happened so early in the homelands of the Volta-Congo languages. Maybe that was the factor behind their slow, steady expansion.
“and we would need to ask what factors prior to pastoralism and agriculture…”
Ryan, but why “prior to”?
drasvi: an economic/bureaucratic expression encompassing domesticated Caprinae
Norw. småfe “small (farm) animals” as opposed to storfe “large (farm) animals” and even fjørfe “feathered (farm) animals”.
It just struck me that “All creatures great and small” might reflect a similar technical expression in the English part of the European calquosphere.
@Ryan, they had to wait for the tortoises to catch up? Sorry for ruining the joke, but I though cultural expansion had a speed of like a village or two per generation*, so single digit kilometers per year. I see claims of 200m/h (meters, not miles) as the lower limit for giant tortoises, so they could easily beat that even with sleeping, eating and fornication breaks figured in.
________
(*) Even with horses/chariots, isn’t the usual story is that you defeated the next village over that didn’t have them, raped the local girls (or boys) and settled down to farm the land? Then your kids got to use the chariots for the next village.
that of Austronesian across Indonesia & surroundings
@DM, do you happen to know where this fits, timing-wise, into the emergence of the lowland grain-based states in the region? (obviously my scott-fixation carries on apace)
i’m finding myself fascinated by the idea of that kind of expansion being enabled by the absence of the state form (rather than imposed by its practices), and wondering whether there would be a parallel to the language/culture refugia that james scott argues were enabled by the un-taxably-farmable terrain that defines the uplands. what provides that “friction of terrain” when the terrain itself doesn’t? or is it still based on the landscape, but in other ways? (if the most important deity is the sea-god, can you missionize effectively inland?)
but why “prior to”?
Because, if Greenberg’s scheme is valid, not only does all of Africa contain a mere four language families, but this would have to have already been the case long before the introduction of agriculture and pastoralism, however vague and disputable the linguistic evidence is when it comes to actual dating. So you’re left with a mystery about why these huge language families overran Africa, much like the mystery of Pama-Nyungan (though that problem is worse, because the expansion there seems to have been relatively recent and oddly rapid.)
That’s why I was saying (a) that the Greenberg scheme is in fact poorly evidenced, and there are many more independent language groups in Africa than four, and (b) the core Volta-Congo area may actually have been a lot more compact even in West Africa at the time that farming got going. (As far as the expansion into Central and Southern Africa goes, this is just the accepted wisdom already.)
I think that although this does explain (away) a lot, there is an issue there even so. Part of this could be fitted into Johanna Nichols’ spread zones versus residual zones scheme (there aren’t a lot of natural barriers to expansion in the savanna, for example.)
Also, it’s too easy to slip into thinking of Africa before the European invasions as a kind of timeless stasis, so that where the languages are now is where they’ve been for millennia. This is certainly wrong, in West Africa as in central and southern. Hausa is gobbling up small languages now, and the fact that it is remarkably uniform over its wide area suggests that this has been going on for centuries: no telling how many isolates or small language families have been overwritten. Similarly with the expansion of Mande, or Songhay, or Akan, or indeed of Western Oti-Volta: all this has been going on for centuries, but long after the introduction of agriculture (long after the introduction of ironworking, for that matter.)
This certainly has a bearing on why Africa is now less linguistically diverse than New Guinea: Africa was home to sizeable kingdoms long before the Europeans came, and West Africa has not been like New Guinea, politically or culturally, for millennia.
Incidentally, just how linguistically diverse is New Guinea? Papuan has its lumpers, too … (We’ve been talking about numbers of language families rather than individual languages here, after all. Nigeria alone has hundreds of distinct languages. Even tiny Togo has about fifty.)
Wiki claims 42 language families in New Guinea. I’m sure there are classification issues there too, but even accounting for David’s concern by estimating that Greenberg was off by a factor of ten in Africa, the different scales of geography make the disparity notable.
In addition to David’s explanations, which I can’t disprove, I’m just suggesting there’s a third — that pastoralism reached the proto-VC area earlier than agriculture (but later than DE believes plausible for VC), gave a lesser advantage than the Bantu pioneers got, leading to a different ratio of language blend between VC languages and the likely myriad, wildly diverse hunter-gatherer languages (and language families, a la New Guinea) that they mingled with, resulting in rapid language diversification across the band of central Africa where they were successful.
Since we have few analogies for such a situation, we don’t know what such a glottochronology should look like, so we don’t have the basis for saying it would have to be surprisingly fast.
We do know that there are situations of language contact that seem to accelerate diversification. In a situation in which incoming speakers of a language or dialect continuum settled among dozens of different language communities whose languages were more or less unrelated, with relative population parity and some degree of local economic interdependence, but without long distance trade (which might have better sustained the integrity of the incoming tongue), initial language diversification might be very fast. And then add another 5,500 years of “normal” evolution along with areal effects to confuse us.
DE is still likely right. Just trying to pose the alternative more succinctly.
pastoralism reached the proto-VC area earlier than agriculture
FWIW, “sheep” is reconstructable for proto-Central Gur. (No connection with any proto-Bantu words as far as I can see, though, unlike “goat.”)
The pOV form is *pê-kʊ. (Kusaal has pɛ’og.)
Proto-Grusi reconstruction has not really progressed very far, but e.g. Kassem piə and Kabiye héu do correspond regularly, and suggest *pe- for proto-Grusi too.
Koromfe pes- doesn’t help, as it is very probably a loan from Mooré pésgò. (Koromfe has a lot of those; nowadays the Koromba mostly speak Mooré.)
But I think the Oti-Volta and Grusi forms justify attributing the etymon to proto-Central Gur.
Sheep eat grass and goats eat forbs and liddle lambzy divey? It might make sense to see a first wave of goats advancing into previously forested areas as the climate dried, just to continue my wild speculation.
The pOV form is *pê-kʊ. (Kusaal has pɛ’og.)
You’re sure that OV is not a far-flung branch of PIE?
The relationship is, of course, the other way round.
KONGO!
One of them is Bangime, about which you’ve remarked: “they say they’re Dogon and their language is Dogon; nobody else believes them”.
(…and I just found the original paper from 2007. Here’s a later Blench paper on African isolates more generally.)
Kleinvieh, Großvieh, Federvieh.
Much earlier than any states, AFAIK.
Maybe start with this presentation.
Here’s the paper on the Austronesian expansion that I’m not sure I’ve actually ever read through, though I think I’ve read the presentation version. The earlier Austroasiatic presence hypothesized for Borneo & Sumatra is presented here. There may also be something in Reconstructing Austroasiatic prehistory.
Also by Blench: ARGUMENTS FOR THE COHERENCE OF NILO-SAHARAN, and the presentation version here. I haven’t read them yet.
Does West Africa have any typological isolates that stick out, presumably genetic isolates that were also not subject to much areal infuence?
I suppose “sticking out” is a question of degree; for example, the typologically unusual word order of Mande (which has spread to Songhay and Senufo) probably doen’t qualify, especially as it’s by no means confined to one language, or even a few languages.
Some of the most striking typological oddities turn up in languages within known large families; Chadic is good at throwing such things up. And even Kusaal’s rococo external sandhi system probably counts as a typological oddity (Kusaal, of course, being about as far from an isolate as a language well can be.)
Not (of course) that I know about more than a tiny fraction of West African languages: but my off-the-top-of-the-head answer to this question would be “not as far as I know.” In many ways, West Africa is a huge Sprachbund,
https://www.eva.mpg.de/lingua/conference/08_springschool/pdf/course_materials/gueldemann.pdf
and I’ve often been struck especially by semantic resemblances even between languages that are genetically quite unrelated. This extends into syntax and even morphology: languages tend to mark similar categories in flexion even if the actual markers are quite different, and I’ve found enough parallels between e.g. Hausa and Kusaal syntax that I thought it helpful to refer to them several times in my grammar.
Even exotics like the English-lexifier creoles have been moulded to fit regional patterns, and I expect any isolates have been similarly adapted to their linguistic neighbourhood long since.
I must say that Bangime doesn’t strike me personally as particularly aberrant. I don’t think it’s actually any less likely to be related to Volta-Congo than Dogon is (and I suspect that even Blench himself would regard Dogon as an isolate family if he’d just discovered it himself.)
Oh gosh, one of the researchers in the Blench paper on Bangime that David linked died of an intestinal ailment at 42 while conducting the research.
Also, the AI summary:
>the language is being transmitted to children but it’s being lost above ten
The paper:
>the language is being transmitted to children but there appears to be a loss of vocabulary. For example, the numbers above ten have been replaced…
Pere (spoken in Côte d’Ivoire) has been suggested as an isolate, but it actually looks pretty Volta-Congo-ish to me. Probably no close relatives within Volta-Congo though.
https://zenodo.org/records/3346581/files/Heath%20Tiote%20Pere%20grammar%202019%20BW.pdf
Typologically, it’s pretty bog-standard for West Africa.
Oh gosh, one of the researchers in the Blench paper on Bangime that David linked died of an intestinal ailment at 42 while conducting the research.
Stefan Elders:
https://www.afrikanistik.uni-bayreuth.de/en/Fol03—Team/Fol03—in-memoriam/Stefan-Leonard-Elders1/
He wrote a really excellent grammar of Kulango (which was published postumously.) He worked a lot with Jeffrey Heath, too.
the language is being transmitted to children but it’s being lost above ten
I’ll have to keep that around as a cudgel to use on people who tout the benefits of “AI.”
@DE, thanx, but you repeated Ryan’s “prior to” rather than told why… Germanic, Italic and Slavic speakers formed one cultural and technological complex and were taking each other’s example in colonisation despite 6ky.
My comment was in fact an extended attempt to explain why Ryan said “prior to”: the issue was that widespead language family spreading appeared to have preceded the introduction of agriculture and pastoralism, the very things which have often been invoked to explain this kind of spreading. Obviously such explanations are a touch implausible if the spread happened “prior to” the introduction of farming.
It’s not about the 6000 yrs particularly, or whether cultures can affect one another with no farming going on (obviously, they can): the issue was specifically about the date of the introduction of farming.
I actually agree with what I take to be your underlying assumption. I suspect that the focus on farming as a presumed driver of the expansion of (presumed homogeneous) language groups is at least partly due to the fact that archaeologists can actually dig up evidence of farming, and people tend to assume that what data they can actually get hold of must be the key factors, rather than other things that have left no archaeological traces.
This seems pretty inevitable methodologically, and you obviously have to use whatever evidence happens to be actually available; but just because it’s available to us now, it doesn’t follow that it was in reality the most important thing at the time.
It’s like reconstructing a protolanguage: no matter how rigorous your methodology and how copious your source material, you will never be able to capture more than a fraction of what the language was like when actually spoken; moreover, what you can reconstruct will actually be highly skewed towards certain aspects of the language at the expense of others which would rank as equally or more important in a decent description of a living (or well-documented dead) language. You’re limited by what the methods of comparative work are actually capable of.
It’s still a valuable thing to do – so long as you don’t start getting delusions about what you’ve managed to achieve – or what can be achieved.
“I’m 20% of the human population of the building I’m in now, but if I were in the [*searches*] Regency International Center in Hangzhou, I’d be more like 0.005% of the population.”
Stand on on Zanzibar? But that was a metaphor.
the numbers above ten have been replaced…
Even when extricated from the AI slop, it’s not a great example. The numbers above ten have been lost in Hausa, too: probably the least threatened language in West Africa. Japanese, too …
I’d also point out that (assuming that Bangime is indeed not “Niger-Congo”) the numbers “three, four, five” have been borrowed from Volta-Congo (as also, indeed, in Dogon.) Numbers are actually very borrowable out there in the world beyond Indo-European and Semitic. Even quite small ones. (Hausa appears to have borrowed “two” from Volta-Congo …)
@DE, I don’t understand why you and Ryan think that spreading occured before the arrival or invention of agriculture and pastoralism.
This means:
1. splits – say, distinct Cushitic and, say, Berbero-Semitic – can’t be dated forward in time and agriculture can’t be dated back in time for the second to come before the first
2. a group of several branches in same area and maintaining some unity couldn’t together acquire a technology that made them all spread from that area.
Europe is an example of (2) – Slavic and Romance were already very different when Russians and other Europeans were exchanging technologies (some of them unknown or vaguely known without adoption in the Arab Semitic-speaking world) and not only them but even the idea “everyone must colonise!”
But I don’t understand (1) too.
For example, we don’t question the link between Berber, Semitic, Egyptian, Chadic and Cushitic, it is not “Greenberg”, it is everyone.
And they together have eaten an Enourmous (but largely Sahel-Sahara which could be somewhat different back then when Sahara was humid) chunk of the continent.
All you can do with them is either speak of a mechanism (2) or decide that the split is not as old as you thought and that the technologies were adopted or invented very early (1).
I don’t understand why you and Ryan think that spreading occured before the arrival or invention of agriculture and pastoralism.
I don’t, myself, in this particular case, but a lot of people do.* (My take is that the supposed problem is largely – though not entirely – an illusion, caused by the assumption that Greenberg’s Niger-Kordofanian and Nilo-Saharan are valid linguistic families.)
But spreading certainly can happen without agriculture and/or pastoralism as the explanation (cf Pama-Nyungan, again.)
* The idea that pastoralism explained the spread of “Hamitic” was deeply embedded in the ideology, and still survives in remnants. In the case of some particular subsets of the Non-Family Previously Known as Hamitic it’s very probably actually true. Just as the adoption of agriculture probably really did kick-start the Bantu expansion: but Bantu is a subgroup of a subgroup of a subgroup of Volta-Congo, and to call this an expansion of “Volta-Congo” (or worse, “Niger-Congo”) is a confusion and anachronism. It’s like calling the spread of English to north America part of the spread of Indo-European. I mean, technically, yes …
It’s also noteworthy that the exchange between Slavic and Italic speakers is a continuation of something old.
Namely,
0. (6kya) branching of IE
1. (4000) Italics (as predators) come from the steppe to Europe, and then to Italy where adopt some of the Mediterran cultrue and ways. [Germanics come to Scandinavia?]
2. Slavs, Iranics and Germanics communicate.
3. (2000-1500) Italics build an empire, Slavs and Germanics become predators on its limes.
Meanwhile in the south of the empire
– Canaanite, Egyptian and [maybe Berber and something else] are spoken
– Aramaic, Canaanite, Egyptian and Berber [and here and there maybe something else*] substrates with Greek and Roman superstrates are spoken.
– a religion comes from Arabia and makes all these shift to Arabic
– the religious barrier does not let Arabs be a part of technological Europe.
___
* I find Moroccan Arabic and Berber phonologies very peculiar and wonder if it is an other-than Berber arealism and substrate. If so, then perhaps somewhere in the Atlas some other langauge was spoken in Roman times?
@V: Stand on on Zanzibar? But that was a metaphor.
I liked that book a lot, and I remember the image of the size of the world’s population, but otherwise I don’t know what you mean.
Following the links DM provided above got me to this Dimmendaal/Ahlund article which isn’t a full treatment of Nilo-Saharan but might be described as an eclectic clean-up and update. It has a really interesting section on Gomuz and Koman (within Nilo-Saharan), which they propose to pair as the Komuz group. The sound correspondences look decent.
I’d been looking at Ahland’s work on Gomuz, so that section was more accessible but the article treats many aspects of Nilo-Saharan in more detail than the Blench slides.
Their reasoning for pulling Saharan and some other languages formerly considered to be on separate branches into “Northeastern Sudanic” is more complicated and I’d look to someone else to evaluate it.
Also, what do you think of the proposal that some people have been quietly passing around that the Fut or Bufut people have a more conservative Bantu lect that should be used in considering affinities within Volta-Congo?
As is, I don’t think much of this Futovoltaic hypothesis. But maybe they’d get some constructive critiques that would help them make it work, if they’d just bring their proposal out into the sunlight.
Sorry. I’ll see myself out now.
In a longer but less recent version, Dimmendaal’s A Typological Perspective on the Morphology of Nilo-Saharan Languages (2019), he refers to Koman/Gomuz, Songhay, Kadu and Kuliak as potentially not part of NS, and gives them less attention. By the time of the more recent paper I linked above, he had brought Kuliak into Northeastern Sudanic, and seemed to be more convinced the Komuz (=Koman/Gomuz, the latter now being called B’aga) was part of NS.
Sometimes I think “why wouldn’t you focus on a single grouping, completing the work there, rather than grazing languages in different groups and families” but then I see a footnote in an article saying “recent military action in the Gomuz area…” and realize I might not put all my eggs in one basket either.
I also wish our recent discussions here had fallen wherever our discussion of Meroitic, et al. had fallen two or three years ago, rather than in “Roman English”. We kind of need a main thread for African languages.
If an expansion of a language occurs through discovery of pastoralism / agriculture, it is presumably accompanied or even driven by a population explosion in the earliest (or most successful) adaptors. If this is true, it would be reflected in the genetics. Would expansions among hunter gatherer populations be expected to play out differently (marginilisation of non-expanding populations but no massive disproportion in population sizes)? I suppose it would also depend on whether the expanders thought it would be best to kill all the non-expanding ones, rather than just marginalise them and demand the odd slave or tribute in kind.
DE, but WHY do they think and why do you and Ryan think that IF (!!!!) we accept Greenberg’s ideas, THEN the spread happened before agriculture?
How confident we are for example that argriculture was not practiced by NS speakers in such and such millenium?
I disliked Stand on Zanzibar, but I won’t derange the thread to explain why.
Also, what do you think of the proposal that some people have been quietly passing around that the Fut or Bufut people have a more conservative Bantu lect that should be used in considering affinities within Volta-Congo?
I haven’t come across that. But in principle, that sounds like a basic methodological error: cherrypicking.
I suspect this is not about Bafut specifically, though. Fut is not a Bantu language: it’s what’s called a “Grassfields” language. Formerly, these were called “Grassfields Bantu”, but it now seems to be generally accepted that the relationship between Grassfields and Bantu is more like the relationship between Baltic and Slavonic. It’s not that Grassfields is more conservative than Bantu (those Grassfield languages I know anything about are anything but conservative) but that it’s a separate branch, so comparing it with Bantu can get you farther up the reconstructive tree.
Grassfields as a family does lack some of the specifically Bantu innovations. One lexical one that caught my eye when looking through some of Larry Hyman’s work on Grassfields is that proto-Eastern Grassfields had *nàkˊ “cow”, the etymon found across West Africa, e.g. Fulfulde nagge, Toro Tegu Dogon nàŋá, Bangime nnàà, Waama nako, Kusaal naaf (plural niigi, stem *naag-), Kulango nà, Samba Leko naà … this etymon was lost in proto-Bantu (as I mentioned above); or, if it’s an ancient Wanderwort, never made it to proto-Bantu in the first place.
However, Grassfields is unequivocally much more closely related to Bantu than to Oti-Volta. Grassfields and Bantu are part of the same V-C subgroup, and it is that which should be compared with Central Gur etc.*
A complicating factor is that many people are now beginning to question where the boundary between Bantu and the neighbouring Bantoid languages really is, or even whether there really is such a boundary at all. Northwestern Bantu is much more internally diverse than the rest of Bantu, and there have been (so far inconclusive) suggestions that these languages can’t be straightforwardly descended from traditional proto-Bantu (which is indeed heavily based on the more homogeneous Eastern and Southern parts of Bantu.)
I can’t find a lot about intra-Fut lects, but Pius Tamanji’s grammar has a brief table of lexical diffferences between the court’s and the commoners’ dialects. It looks very much like it’s the commoners’ dialect which is more conservative, in fact: e.g. posh mù’û “eat” versus plebspeak dʒɨ̂, where the latter is from the pan-Volta-Congo form (e.g. Kusaal di “eat.”)
* I have sinned methodologically in my proto-Oti-Volta treatise in comparing Oti-Volta, a mere subgroup of Central Gur, which is itself probably but a subgroup of Gur-Adamawa, directly with Bantu, which is but a distant twig on the tree of the Benue-Congo subgroup of V-C. (I do own up to my transgression in that section, though.) The proper thing to do would be to reconstruct proto-Gur-Adamawa and proto-Benue-Congo first and compare those. But we are far indeed from being able to do that, and I got impatient.
Apologies — “Futo-voltaic” was merely an extended photovoltaic pun prompted by reading a treatise that described one of the hyphenated Volta groups as Voltaique in French. I was hoping the idea that they should bring their proposal out into the sunlight would expose the joke.
But at least it prompted an interesting comment on Grassfields.
The rest of what I wrote in those two comments was serious.
this etymon [*nàk “cow”] was lost in proto-Bantu
Hmm … (Northwest Bantu) Eton ɲàg “ox” (can’t find “cow.”) I thought I remembered something like that.
But I suspect that this is just the Awesome Power of Sheer Coincidence. Moreover, the initial consonant seems wrong; contrast Eton -nà “four”, beside e.g. Kusaal (a)naasi.
a treatise that described one of the hyphenated Volta groups as Voltaique in French
“Voltaïque” is just the French word for “Gur” (in linguistic contexts.) I like “Voltaic” better than “Gur” myself, but I think it got avoided in English because of potential confusion with the adjective for Haute-Volta/Upper Volta in the days before Thomas Sankara came up with a better name for his country. Mind you, that’s never stopped people talking about Chadic languages. Or Germanic languages, come to that.
Lameen is the person to tell us what the state of play really is with “Nilo-Saharan.”
One problem is that Greenberg’s ideas are questionable not as much for “African languages” as for a narrow belt in Africa.
You can’t do anything about undiversity above the belt and you have Bantu below the belt.
https://en.wikipedia.org/wiki/File:Koppen-Geiger_Map_Af_present.svg
https://en.wikipedia.org/wiki/File:Tropical_Forests_2000_by_Major_Ecological_Domains.tif
Weirldy “rainforests” in the first map (as in PNG) is Bantu.
But the tortoise in the second is a perfect match for
https://en.wikipedia.org/wiki/File:Map_of_the_Atlantic%E2%80%93Congo_languages.svg
which you can interpret either as “the tortoise leads to diversification” (much more diverse than Bantu) or “the tortoise leads to unification” (peoples with similar cultures and technologies are there)
You can’t do anything about undiversity above the belt
Oh, yes I can. Afroasiatic is comfortably the most internally diverse language family accepted by all mainstream historical linguists. (The fact that it is accepted is in part due to the sheer accident that AA includes some of the earliest-documented of all languages, along with hundreds of modern languages. There’s a lot to work with, along with some daughters almost as old as PIE before you even get started.)
Linguistic diversity is a thing within accepted language families too. And insofar as it’s an argument for great time depth, it makes no odds whether the languages involved are ultimately all related or not.
The Bantu languages are much less diverse linguistically than the Afroasiatic languages: there’s no mystery about this, as we have actual extralinguistic evidence that the greater part of the Bantu expansion only started a mere 3000 YBP or so. It is eminently reasonable to conclude that proto-AA was spoken (a lot) earlier than proto-Bantu on purely linguistic grounds, and what extralinguistic evidence we have amply supports this.
[The relative linguistic undiversity of Bantu is also a good argument against the common assumption that there’s something about African languages, or about Africa, that predisposes to unusually rapid linguistic change.]
Wiki needs a splitter map of African language families, to put the alternatives in easier conversation. David, does this listing capture most of your suggested splits? Are you a maximal splitter, or are there others who go further? (with caveats about the thinness of some of the documentation.)
1 AfroAsiatic = Berber, Semitic, Egyptian, Cushitic
Chadic
Omotic
2 Nilo-Saharan = Northeastern Sudanic, Central Sudanic
Songhay
Komuz
3 Niger-Congo = Bantu, Voltaique et al.
Ubangi
Atlantic
Kordofanian A
Kordofanian B
Mande
Ijoid
Dogon
4 Khoe sans san
San
(Trying a different visual, where the 4 lumper groups are numbered and bolded, with new families split out of them listed below them.)
DE, internal diversity is a problem because you and Ryan say then you need to explain the spread before agriculture.
Ryan, I like it.
https://en.wikipedia.org/wiki/File:Nilo-Saharan.png is weird
Also if one of us actually makes such a map, it WILL have an effect on Greenbergism in education.
@Ryan:
Afroasiatic:
Not sure if your formatting is meant to imply that Berber, Semitic, Egyptian, Cushitic belong in a group over against Chadic (definely AA) and Omotic (probably not AA, though some “Omotic” languages might be.) If so, I don’t think that’s justifiable. Cushitic is very diverse indeed, and I know that Lameen feels that Berber and Chadic may actually belong together. But there is no consensus on top-level subgrouping within AA.
Niger-Kordofanian:
Neither Atlantic nor Ubangian is an actual unity.
In the case of Atlantic, scholars of these languages mostly agree on this point, and actually have done for a long while; the prevalent view is that its various components are individually each at about the same level as all of Volta-Congo, but that all are ultimately related. I am more cautious. If there is a real genetic relationship there, you’re really talking about something like Altaic, not Indo-European. I think Kordofanian is much the same, at best.
Ubangian is partly Volta-Congo (e.g. Gbeya), but Greenberg also included a lot of languages for which there is no evidence of V-C affiliation at all. There’s a nice table at
https://en.wikipedia.org/wiki/Ubangian_languages
where you can eyeball the words in various “Ubangian” subgroups for yourself. (People will add numerals to such lists. The message about the great borrowability of number words has not yet really cut through. Cf what I said above about Bangime.)
Khoisan:
Khoisan is really five groups, but some experts in that difficult field think that Sandawe may be distantly related to Khoe, and Tuu and Kx’a to each other. I don’t think any of this is particularly controversial, at least among actual Khoisanists. Greenberg just made his characteristic error of confusing typological similarity (phonological similarity in this case) with genetic relatedness here.
Nilo-Saharan:
Lameen is your man for “Nilo-Saharan”, but I think it comprises many more groups than four. What arguments I’ve seen for groupings like Northeastern Sudanic and Central Sudanic give me indigestion by their near Victor-Mair-like* methodological impropriety. However. Lameen knows very much more about this than I do, and it’s a pretty active area, I think. I may easily be behind the times or just plain wrong.
There are a good few presumed isolates out there too, as we’ve been discussing.
* OK, too harsh. But they give me the same feeling of seeing undoubtedly eminent scholars messing in stuff they aren’t properly equipped for. Just as some perfectly intelligent non-linguist writers make fools of themselves when pronouncing on linguistics because they don’t even know that there are basics of which they are ignorant, so too some truly excellent descriptive linguists seem to have no idea about even the commonest pitfalls of comparative work.
Skinner’s etymological dictionary of Hausa comes to mind as a particularly horrific example … there’s a review of it by Paul Newman which magnimously tries to find some nice things to say about it, but the strain is evident:
https://scholarworks.iu.edu/iuswrrest/api/core/bitstreams/807e532e-40fd-492f-a081-d18dce435c1d/content
My formatting was merely meant to suggest that for instance, Semitic, Egyptian, Berber and Cushitic, because they’re next to the heading AfroAsiatic, all still belong in AA, but nothing about grouping. And that Chadic, because it’s on its own line, might be perceived as an independent family by some splitters.
I will probably try to come back later, refer to your suggestions and recreate the list, with the idea that maybe we could perfect it, perhaps with citations, as an alternative map that showed where inclusion is disputed. And then maybe someone (drasvi?) could make a map. I don’t think I could do it, but much of the geographic information would already be available at wiki. One would just need to recombine and recolor a lot of units. (He laughs. It would be a lot of work actually, especially for those new isolates that are currently just names across a general region of the larger family.)
Or alternately, simply post the list with a paragraph of description in the wiki for Languages of Africa. At minimum, I would then refer to it every time this stuff came up here.
We really should pick a thread here for this too, the way Son of Yamnaya works. Options with some related discussions include:
Meroitic Inscriptions Found
Timbuktu Manuscripts
We also had one that talked about Subsaharan genetic data.
And I’ll take note here of Guldemann as seeming like an academic splitter worth looking into.
I’m not aware of anyone splitting Chadic from AA. The only questions are Omotic and its parts. Without Omotic, AA is after all just plain obvious. (…Admittedly, no Chadic or Egyptian examples in that post.)
Blench himself, in his otherwise maximalist works on Nilo-Saharan linked to above, insists that Cabu/Shabo, always included in previous treatments, is an isolate and does not belong to Nilo-Saharan.
Shabo was not known to Greenberg. As far as I know it’s never been considered anything but an isolate. Same with Laal.
And I’ll take note here of Guldemann as seeming like an academic splitter worth looking into
I’m a Güldemann Fann; I don’t think it’s just because he’s a fellow-splitter …
(After all, I’m not a fan of Lyle Campbell … and I do rate Gerrit Dimmendaal, despite his Lumpist deviationism in re Nilo-Saharan. Come to that, I approve of Roger Blench, unless I’m feeling particularly Calvinist that day.)
Are not there dubious “Cushitic” languages?
Oh yes. A particularly interesting case is
https://en.wikipedia.org/wiki/Mbugu_language
There are also a great many African languages which are just too poorly documented for assignment to a family to be more than wild guesswork.
In West Africa (and I imagine elsewhere), a problem is that languages often converge typologically (and borrow extensively), so that a language which is probably a real isolate ends up looking so much like its neighbours that only a really intensive study will reveal that it’s actually not genetically related. And very few African languages have been studied intensively.
Conversely, a perfectly respectable member of one of the big groups can end up so divergent through historical changes that the resemblance is no longer apparent without (again) intensive study. I’ve no doubt some “isolates” are of this kind (and we will probably never know, with most of them.)
In the area I actually know about, there is Miyobe, which is clearly Volta-Congo, but whose particular closer relationships are said to be unknown in the reference books. I think it is actually the closest outside relation of Oti-Volta, and try to demonstrate as much in my proto-Oti-Volta thing. The evidence is pretty copious (I reckon): it’s not that I’m particularly perceptive, it’s just that nobody with extensive knowledge of Oti-Volta as a whole has ever looked at Miyobe properly before (not least because the relevant documentation is even now scanty and up until a decade or so ago was confined to extremely brief word lists.)
Blench is a syncretic lumper-splitter.
He is one of the first linguists to have worked in Arunachal Pradesh after it opened up. He claims that several languages there are objectively isolates, though it had been presumed that everything there is ST.
Search term for his Academia page: “Declassifying Arunachal”.
Blench’s own site seems to be inoperative, which is a pity; I hope this is a temporary thing. Most of his stuff seems also to be on academia.edu, but I find navigating that to be a real pain.
Per archive.org, Blench’s website has been on the fritz since some time in December. His Facebook account is hoppin’, however.
Ah dinna dae Facetube. Yon Zuckerberg’s a heidbanger.
Me neither, but if I can peek in without paying, then why not. Anyway the posts I could read were repostings of internet drolleries.
It’s exciting to see that some of the uncertainties of Nilo-Saharan are being rapidly explored. I talked above about the work of Ahland on Gomuz; Otero’s dissertation on Koman can easily be googled. He says his data came from native speakers of “all the living Koman languages, including from previously undocumented varieties”.
He reconstructs proto-Koman based on reconstructions of “subnodes” — Gwama, proto-Central Koman, proto-Komo-Uduk, proto-Dana-Opo, etc. It seems painstaking. As an example, his cognate set for PKMN *ɓ includes the words for neck, strong and fog, which in all five of his subnodes start with ɓ or p and look plausibly related to me.
He summarizes his work this way:
>One main issue in high-level classifications is the lack of low-level reconstructions of families established with verifiable sound correspondences coupled with morphological evidence to support the internal structure of a given family. This dissertation addresses this issue by reconstructing the basic phonology, including segmental and suprasegmental domains, and tracing the evolution from Proto-Koman down through the nodes to the modern-day sound systems.
Further in, he criticizes previous scholars for assuming NS relatedness and cherry-picking cognates.
As I mentioned above, in the slideshow DM linked, Blench rely on Otero and Ahland for a “Komuz” cluster, and he believes this strengthens Nilo-Saharan. Ahland also seems to believe in the Komuz cluster. Otero is more circumspect in the dissertation, writing “While Koman’s affiliation to the purported Nilo-Saharan super family is still under debate, the overarching aim of this dissertation is to provide a conservative reconstruction of Proto-Koman which will hopefully serve future Koman scholars as well as those interested in higher-level genetic classifications”.
Meanwhile, Blench has posted on Academia what seems to be a draft of a joint paper with Lameen. It’s in Not To Be Quoted status, and the main author-info page doesn’t mention Lameen, but the p. 2 ff. header carries both names. The paper states that, in my not-quoting paraphrase,
a) Songhay is undoubtedly Nilo-Saharan;
b) the membership of Saharan within Nilo-Saharan has not been seriously questioned since Greenberg; and
c) Songhay and Saharan belong in a clade together. The paper suggests that Songhay is more closely related to West Saharan, while Beria and Sagato (otherwise often considered part of Saharan) are in a different clade.
You can read the definitely not quotable conclusions on p. 39.
“undoubtedly Nilo-Saharan”
means undoubtedly same clade as what? If Saharan is doubted, then is it the “Nilo” part (Nilotic? Nubian?) or Fur or what? Where do we (they) anchor the term?
I’m realizing that the simplest way of outlining a coherent splitter language family set may be to follow Guldemann’s “exhaustive survey“. He finds 40 to 45 genetic lineages, believes the evidence is yet insufficient to believe strongly in higher clades other than these, and shows his work.
I’m going to read that more deeply and then think about whether it’s worth putting a summary list on wiki.
One disappointment is that he refers to a complete map at the end, but it’s not found in the linked version.
drasvi, I combined (non-)quotations from the intro and conclusion there perhaps inartfully, but they describe a Songhay/Sahara cluster with the two clades I mentioned, as part of Nilo-Saharan. I don’t think the -Saharan in NS refers directly to the Saharan clade. It’s just a geographical term. But you can read or skim it. Their cladogram is near the end.
The title of Blench (and crypto-Souag)’s paper is misleading. Theres nothing Nilo- at all in the paper. It’s all about Songhay-Saharan.
(Which at first glance looks like a good working hypothesis. The semantic matches are tight, and the phonological matches are not absurd.)
The language family they assign it to is Nilo-Saharan.
Also noting here that in addition to the 40-45 basic classificatory units of Guldemann, he also lists 43 “unclassified” languages, often because there is too little documentation and most of them were “already virtually extinct” when the paucity of material was collected.
Huh. Both Laal and Shabo have separate male and female first person singular pronouns. Maybe an ancient areal feature?
@Ryan, I will.
When everything is doubted, I think the way is: “let’s call X and everything that undoubtedly is in the same clade ‘Nilo-Saharan’.” Else you can’t know what Nilo-Saharan means.
On gender-marked first person singular: Thomas Berg has surveyed this very subject (“Gender marking in the first-person singular: A case of paradigm (in)consistency”, J. Ling. 60, 527–561, 2024, here, OA). After checking grammars of over 1750 languages (!) (I repeat: !), he found 30 languages with gendered 1sg pronouns. About half are in South America. The rest are spread around the world, and include Tocharian A, but no Austronesian or Bantu languages. In Africa such languages are Laal, Shabo, Hadza, Korana (Khoe-Kwadi), and Tihami Arabic (spoken along the Red Sea coast of Yemen). Korana also has gendered 1sg verbal markers, as does Sidaama (Highland East Cushitic). Shabo was not netted in the original survey, and was added after it was pointed out to the author.
Lameen about that paper 3 years ago.
Lots of other discussion of Nilo-Saharan, comparative reconstruction in general, and also Natufian genetics in that long thread.
After checking grammars of over 1750 languages (!) (I repeat: !), he found 30 languages with gendered 1sg pronouns.
This seems odd. Since gender is such a huge feature of languages in general, why so rare?
Presumably because when you say “I”, the referent is unambiguous.
Third person singular is probably where gender provides the most disambiguating power, and first person singular is where it provides the least.
but no Austronesian or Bantu languages
Dunno about Austronesian, but as Bantu (like all of Volta-Congo) is sublimely indifferent to biological sex throughout its grammatical gender system, that’s hardly surprising.
I notice that Kokama-Kokamilla features in the list. It’s worth pointing out that this language has pervasive differences between male and female speech: it’s not just 1sg pronouns by any means. Rosa Vallejos Yopán’s grammar cites e.g.
uri tsenu ikian yawara=kana=uy tana ku=kuara
3SG.L.M hear DEM.M dog=PL.M=PAS1 1PL.M farm=INE
“She heard the dogs in our farm‟ (male speaker)
ay tsenu ajan yawara=nu=uy penu ku=kuara
3SG.L.F hear DEM.F dog=PL.F=PAS1 1PL.F farm=INE
“She heard the dogs in our farm‟ (female speaker)
There are distinct male/female-speaker forms for 1st and 3rd person pronouns, for the plural clitic, for demonstrative and indefinite pronouns, for “like this” and “like that”, for “also”, “but”, “there”, “then” and “after that.”
I think there may be a danger of comparing apples with oranges in just counting up languages with a sex distinction in first person pronouns.
The paper does indeed invoke Grice’s Maxim of Quantity to account for the cross-linguistic rarity of gender-marking in first person pronouns.
Berg does look at male and female di separately (table 2).
(Ed.: I take it back, I misread it.)
Presumably because when you say “I”, the referent is unambiguous.
I understand the logic, but if languages were logical they would be very different. Why are there languages with dozens of cases or noun categories? Why does Chinese require you to have a classifier between a number and the noun being counted? I could go on.
When you say “you”, the referent is also unambiguous…
Exactly!
I understand the logic, but if languages were logical they would be very different.
Sure, and you’d really need to look at the diachrony of personal pronoun systems and see what story comes up. Statistically, though, I’d expect gendered first person pronouns to be unstable as they do no semantic work that I can think of.
When you say “you”, the referent is also unambiguous…
Usually yes, but there are special circumstances. For example, when there are several people present and you are directing your speech at only one of them. It would be interesting to see if and when and how gendered 2sg pronouns got added to languages which have a gender category but didn’t have it marked in that circumstance.
>Lameen three years ago
Sigh. That was somewhere deep in my brain. I even wondered about saying Lameen may not have signed off yet.
Given what he says in the other thread, Blench must be pretty oblivious to write Songhay is obviously NS in a paper he posted with Lameen’s name.
Rule of twice heard — this 2nd time I’ll surely remember Lameen’s disavowal.
Einmal heisst keinmal und zweimal heisst immer!
In my youth, there was a saying among junior hospital doctors:
If a consultant* has seen a case of some disease once before, he says “In my experience …”
If he has seen two cases, he says “In my series …”
If he has seen three cases, he says “Time after time after time …”
* Most senior grade of hospital doctor. Not sure how you say it in USian. “He” (rather than “she” or “they”) is used for historical period authenticity.
I also note that polite or honorific second-person pronouns are common, but that in the first person (e.g. the royal “We”) they are rare or restricted.
I wonder how common it is cross-linguistically for 2pl to be used as an honorific singular (as opposed to some other way of expressing an honorific meaning)? Obviously there are some extremely familiar European examples, but that may well be an areal thing.
WALS has a map of honorific pronouns, but not for this specific point:
https://wals.info/feature/45A#1/23/149
As usual, Africa is not well documented. They’ve missed Mooré, but picked up the much-obscurerer Koromfe, presumably because this is one of the questions in the Lingua questionnaire that the Croom Helm grammars were organised around. I’m pretty sure the Koromfe system is in fact based on the Mooré (honorific plural for both 2nd and 3rd persons.)
I was thinking about this particularly, because I was wondering if this particular* technique of honorification might not actually correlate very well with how hierarchical your society is. It seems to have taken off in a big way relatively late in bougeoisifying Europe, and although traditional Mossi society is very stratified, so is Mamprussi society, and Mampruli doesn’t use honorific plurals. (Neither does Kusaal, but you wouldn’t expect it to: the Kusaasi are not into hierarchies at all.)
* As opposed to e.g. the Japanese system, which surely is tied up with how the traditional culture operates. (But doesn’t work by honorific pluralisation.)
Classical Nahuatl did it with reflexive causatives: a tlatoani doesn’t sleep or eat (like a mere commoner), he causes himself to sleep or eat.
Thai has personal pronouns for everything. I read Thai personal pronouns described as an open word class.
My favourite one is (literally) “mouse”, used as an anti-self-honorific by young women (some of whom stick with it even as they grow older.)
Vietnamese apparently likes using kinship words as quasi-pronouns.
Kusaal does that in vocatives, but doesn’t use kinship words instead of pronouns. If you’re a man, you might politely ask a lady stallkeeper of about your own age
M diemma, bɔ ka fʋ kuosida?
“My mother-in-law, what are you selling?”
Now I think about it, “Your obedient servant” is the 1sg mirror image of honorifics. Surely some such got grammaticalized opaquely in some language somewhere, similarly to Vuestra merced.
Japanese boku originated as an actual calque of Diener/servus* in student slang, and then became a familiar-register male-speak 1sg pronoun Because Japanese.
* At a certain level of ridiculous abstraction, boku is thus “the same word” as ciao.
Reading Guldemann on Nilo-Saharan was revealing and depressing. Ehret in particular comes off horribly, reminding me a bit of the toxicologist in this week’s New Yorker, who diagnosed one baby, almost certainly in error, as having died of codeine poisoning from breastfeeding after his mother took tylenol; and then convinced a generation of mothers either not to take tylenol with codeine or not to breastfeed, on the basis of his sheer senior authority and indeed celebrity in a narrow field deterring anyone from challenging his ridiculous methods and faked data.
That’s probably going too far with Ehret, but it’s hard to believe you can get by with such dipsy doodle methodology.
For me it really helped to see Guldemann’s comprehensive takedown in a way that a dozen anecdotes didn’t get across.
>your obedient servant
LOL that’s a fun moment of recognition.
Jay Jasanoff, “Language and gender in the Tarim Basin: the Tocharian 1 sg. pronoun,” Tocharian and Indo-European Studies 3 (1989), 125-148, here, concludes (§12) that Tocharian A nom./obl. näṣ ‘1sg. masc.’ and ñuk ‘1sg. fem.’ go back to PIE accusative *me and nominative *eg̑oH respectively (or however you’d reconstruct them nowadays). The reason for this split, he suggests, is conservative women’s speech vs. innovative men’s speech, a phenomenon which he illustrates with a similar example in English:
Sherman’s memoirs reproduce some humorous passive aggressive examples of enemy commanders writing to one-another as “your obedient servant.”
I guess the “we” is honorific for “I”; I think many of his soldiers would not have been prepared for this effusion of blood — after all, it was their blood, not his.
And was the tortured syntactical inversion of “Your communication…I acknowledge receipt of” normal for this kind of communication?
Why does Chinese require you to have a classifier between a number and the noun being counted?
I always assumed it’s because of the large number of homophones or near homophones in Middle Chinese. And indeed in Mandarin where polysyllabic words are the rule and there is less ambiguity in speech, the amount of classifiers actually used in colloquial speech has sharply reduced.
Or it could also be that most people never knew or used a lot of the more obscure classifiers and what we see in grammars/textbooks is pedantry from the past. I assume someone has looked into this.
distinct male/female-speaker forms for 1st and 3rd person pronouns, for the plural clitic, for demonstrative and indefinite pronouns, for “like this” and “like that”, for “also”, “but”, “there”, “then” and “after that.”
not to beat my same old hobby horse to death, but wouldn’t it be nice if we could use “grammatical gender” as the umbrella term for all of this, instead of it being used wildly misleadingly for (e.g.) the distinction in kiswahili between ki/vi and m/wa nouns?
I read Thai personal pronouns described as an open word class.
this is actually, i think, the most sensible way to handle such things. i’ve been part of (anglophone) subcultures where it’s been common practice for varying periods of time, and while i could tell some stories about (otherwise also annoying) people doing annoying things with it, it’s generally a very sweet experience, and not particularly hard to get used to.
it could also be that most people never knew or used a lot of the more obscure classifiers and what we see in grammars/textbooks is pedantry from the past
or the highlights of a neologizing fad – or multiple periods of neologizing for fun layered on each other, like my copy of An Exaltation of Larks.
passive aggressive examples of enemy commanders writing to one-another as “your obedient servant.”
“thank you for your attention to this matter”
wouldn’t it be nice if we could use “grammatical gender” as the umbrella term for all of this, instead of it being used wildly misleadingly for (e.g.) the distinction in kiswahili between ki/vi and m/wa nouns?
In principle, yes. But in Bantu studies, apart from the practical difficulty of overturning over a century of settled usage, any new terminology would have to contend with quite a few complications
The Bantuist convention is to call the group of nouns that take (say) the prefix ki- or vi- a “noun class”, while “gender” is conventionally used for sg/pl pairings of “noun classes.” This is a rather weird convention, quite unlike the way the term is used in traditional SAE grammars; but you do need some different term from “noun class” for this.
In case that wasn’t already complicated enough, there are two further problems.
One is, that grammatical agreement of pronouns etc doesn’t always match the formal classes (or genders.) In Bantu, there is usually a much closer match between these two different concepts of grammatical gender than in other language groups, but even so there is a mismatch (for example, in Swahili, animates take the “human” m-/wa- concords regardless of their actual noun-class/gender.)
The other is that the classes/genders have (complicated and variable) semantic affiliations, though never with sex. But this is of course exactly how (grammatical) gender got specifically sex-related names in SAE, so this too feeds into the whole complex of ideas.
So in Bantu, if you eschew “gender” as a technical linguistic term, you need not one but three new terms. (Not that this would be altogether undesirable, because the current terminology is misleadingly ambiguous. But getting consensus on the new terms would be quite a challenge.)
@DE: Japanese boku originated as an actual calque of Diener/servus* in student slang, and then became a familiar-register male-speak 1sg pronoun Because Japanese.
Are you telling me that European students referred to themselves as Diener or servus?
This sort of thing survives (or servives) in Spanish servidor, though I think it’s at least partly jocular.
And apparently a non-servile “your boy” is not unknown in American English. Originally AAVE? Sort of the way an earlier generation said “yours truly”. Did all the “yours” complimentary closes come from “your servant”?
@rozele:
this is actually, i think, the most sensible way to handle such things. i’ve been part of (anglophone) subcultures where it’s been common practice for varying periods of time, and while i could tell some stories about (otherwise also annoying) people doing annoying things with it, it’s generally a very sweet experience, and not particularly hard to get used to.
Could you give some examples? Is that like calling yourself “your boy” or referring to a co-racialist by means of an anarthrous racial insult, if you get my drift? What second-person pronouns were used?
Are you telling me that European students referred to themselves as Diener or servus?
Yes, I am.
In greeting (and acknowledgment) formulae, specifically (elliptical for “your servant.”) Ciao is of a similar origin (though it is the Venetian cognate of “slave”, not servus.
This is where the Hungarian greeting szervusz came from.
Boku itself is 僕, Sino-Japanese for “manservant.”
https://en.wiktionary.org/wiki/%E5%83%95
The whole complex of words is engagingly polyglot.
Thinking of servus, I was just wondering what the Kusaal and Mooré Bible versions did about the Hebrew עֶ֫בֶד and Greek δοῦλος, given that both languages distinguish sharply between “slave” (Kusaal yammʋg, Mooré yàmbá or yɛmbga) and “servant” (Kusaal tʋmtʋm, Mooré tʋmtʋmda.) Much the same as the English versions, I see: “slave” only for for those implied to be literal chattel slaves, not for e.g. Paul calling himself δοῦλος Χριστοῦ Ἰησοῦ, or Naaman the Syrian general calling himself עַבְדֶּ֑ךָ when talking to Elisha.
I know of one Twitch streamer, Dr4gonBlitz, who calls himself “your boy” on stream quite a bit—especially when he’s doing something like reminding YouTube viewers to, “Like, comment, or subscribe. It’s a great way to help your boy!” He’s a quirky white guy from Florida, around thirty years old.
…and was promptly adopted by their mothers for 2sg.
I think Japanese doesn’t have personal pronouns at all, just titles. Thai probably too.
Just the number of different classifiers, or the number of times ge is interposed between a numeral or pronoun and a noun, too?
My sister says it’s totally true in Austria currently, and brings up another version:
Woman from Italy: “He’s Italian! He’s fibbing! If he’s from the north and says he’s done it often, he’s done it twice, and if he’s from the south, he’s done it once!”
Possibly even with once vs. never.
@DE: Thanks. I knew the origin of ciao but not that Hungarian has szervusz (though a lonely gray cell is whispering that I might have forgotten). This book chapter by Dobrota Pucherová says it’s used with different spelling in Slovak too.
Remains common in Vienna (as a T greeting – all greetings are segregated between T and V). Usually vowel-reduced (u as [ɐ]), sometimes all the way to a single syllable: [sˑɛ̆ɐ̯̆sː].
Just the number of different classifiers?
Yes, I mean the increasing use of ge as a generic replacement for the “correct” classifier. But not being a native speaker my intuition might be faulty.
@DE, (slaves and the Bible)
Once after reading a comment by a Muslim reader to a novel (which I didn’t read) I realised that metaphors of slave, servant and child in the context of human relationships with God are interesting and I know too little about their history.
The novel begins in the first Hijri years, the hero is a boy – Abu Bakr’s slave recently converted in Islam. He is studying his new religion and thinking about God, humans and so on, to which he refers as his children – using an old Slavic word for children, used in the Slavonic Biblical translations of τέκνον and often in Russian translations as well. Because it sounds vaguely religious in modern Russian, the authors of the novel used it in describing pious [Muslim] thoughts. You can guess what the commenter said.
I don’t even know how old is Russian “slave God’s such and such” and how do you say it in English.
Abu Bakr’s own name is Abdullah, it looks monotheistic but he was born “c. 573” WP says (when Muhammad was 3) and so that of his first son, Abdurrahman (which makes me think both about names of god in Islam and bismillah and Rahmanan) said to be “a polytheist during the early years” by WP, while his first wife’s name Qutaylah bint ʿAbd al-ʿUzzā is not.
>I guess the “we” is honorific for “I”; I think many of his soldiers would not have been prepared for this effusion of blood — after all, it was their blood, not his.
ulr, I think that’s an anachronistic understanding of war. Here is a list of 124 general officers killed during the Civil War, or roughly 10%.
https://thomaslegioncherokee.tripod.com/listofuniongeneralskilledormortallywoundedinbattle.html
In fact, Gen. Corse was wounded in the face during the battle. The Union troops were surrounded, so there was no escape but surrender, and it’s really poor form for the general to fight to the second-last man and then surrender.
When Corse was being bandaged, an officer named Rowett took command. “With ammunition running low, Rowett, hoping to conserve the supply, ordered a cease-fire. Many Union soldiers, thinking Rowett wanted to surrender, cried out, “Never,” and “Die first.”
https://warfarehistorynetwork.com/article/the-desperate-battle-of-allatoona-pass/
(It doesn’t say who reported those cries or how many soldiers were hoping or even cheering for a surrender.)
If you’re in suspense, the Union force would have been doomed except that the Confederates left; they were needed or seemed to be needed elsewhere. The Wikiparticle seems to disagree with the above article, and itself, on that need. The Union victory, protecting the railroad, made Sherman’s march to the sea possible.
Language connection: Some say this battle inspired the expression “Hold the fort.” (I usually hear “Hold down the fort,” possibly by influence of “hold down a job”. Ngram result.)
DE –
It’s not completely clear, but it looks like Mandarin nín 您 (polite 2nd person pronoun, earlier written 恁) comes from a contraction of 2pl nǐmen 你們 or something similar.
nín 恁 for some kind of ‘you’ goes back at least to the Ming, but it’s not clear to me when it started being used as a polite pronoun instead of just a plural one, so it may or may not have been influenced by European usage (Chinese pronoun usage in general certainly has European influence since the late 19th century).
More details here, though I’m not sure everything in this entry is correct: https://en.wiktionary.org/wiki/%E6%82%A8
Thanks!
@drasvi:
Abu Bakr is said to have been a
https://en.wikipedia.org/wiki/Hanif
prior to his conversion to Islam. So, yes, monotheist. Whatever one’s view of this, there was a lot of monotheism about in Arabia prior to Islam.
Could you give some examples?
one of the more ostentatiously aesthetic moves was a friend who (at the height of a philly-punk-centered cycle of endless pirate jokes*) used “arrr” for a summer or more. more common were the various “neopronoun” sets then in contention for wider use, which singular “they” has almost entirely displaced: ze/hir/hirs, xe/xer/xers, etc. some other versions from adjacent social worlds (adopted somewhat later than the period i’m thinking of) are “v”, which justin vivian bond uses, and “judy”, which taylor mac uses.
.
* what’s a pirate’s favorite grain? barrrrrley. (claims were made about having exhausted the entire english wordhoard’s supply of words with “ar”)
Sounds unlikely at first glance — shouldn’t “hold the fort” be as old as the word “fort”? — but may actually be true, for the figurative use. Of course “hold the fort” can be found from before 1864, but only with reference to literal forts, as far as I could see in a quick look. A hymn titled “Hold the Fort” appeared in ca. 1870 and quickly became popular (and the phrase jumps more than 10x in the ngram between 1870 and 1885):
The composer, Philip P. Bliss, in a preface to a pamphlet with the words and music, includes a story about Corse receiving a message via signal flags: “Hold the Fort; I am coming. W.T. Sherman.” (Sherman’s memoir says he *received* signal flags during the battle indicating that Corse was there, but not that he *sent* any such message.)
The OED’s original entry for fort in 1897 did not mention “hold the fort”; it was added in the 1972 Supplement, with the hymn as the first citation. Not yet revised since then, but it looks to me like good evidence that the phrase as an idiom was indeed popularized by the hymn.
@David Eddyshaw: There seem to be folkloric claims about practically all the important companions of the prophet (except for the ilk of Abu Sufyan, obviously) having been hunafa before their conversions. However, there are apparently doubts among some scholars whether there had ever been a significant hanif presence in the south-central Hejaz.
On the other hand, though one might have doubts about the traditional hanifness of one or another Companion, it does actually seem to be the case that monotheism of one kind or another had been quite trendy for some time in Arabia at that point.
https://en.wikipedia.org/wiki/Monotheism_in_pre-Islamic_Arabia
In the circumstances it seems quite plausible that the kind of people to whom Islam appealed might already have been monotheists of some kind. (And perhaps less atypical of their time and place than later pious anecdotes suggested.)
The more observations we get of the Universe, the weirder Dark Matter becomes. That is [my take] the more ad-hoc characteristics of DM get introduced to “make it work” — contrive the theory to fit the data. Is there anything left that would make Lambda-CDM falsifiable?
This latest is that there’s two large voids of not-even-Dark-Matter sitting either side of our local galaxy cluster, in a very much not uniform distribution. (With low probability– but then no less probability than seems to be the norm in this Humpty universe.)