Alman is a simplified dialect of German designed to reduce the complexity of the language while maintaining mutual intelligibility with Standard German. It systematically eliminates several complex grammatical features:
• Gender System: Uses a single article form for most contexts
• Case Markings: Removes most case-based inflections
• Noun Declensions: Simplifies noun endings across cases
• Job Titles: Eliminates gender-specific forms in occupational titles
The English home page opens with a deadpan Mark Twain riff:
In 1880, an American author filed a detailed complaint about the German language, together with a proposal for its reform. He had studied it for nine weeks, under great difficulty and annoyance; three of his teachers died in the meantime. The file has remained open since.
Just keep scrolling; it’s presented with all the interactive frills Youth Today are accustomed to, and I find it impressive and mildly amusing. I can’t quite believe someone went to all the trouble to create such an elaborate site in the service of a joke language, and I can’t quite escape the niggling suspicion that they don’t think it’s a joke.
I got to it via MetaFilter, where (as expected) some people don’t understand language and others don’t understand humor; I enjoyed Bruce Ware Allen’s comment quoting his grandfather describing a talk by Robert Millikan (“Introduced with words of strong praise, Millikan spoke a language which I have rarely heard, except from our own children – a German which knows no dative, and in which genders are distributed by lot”) and one by adrienneleigh beginning “When i was 16 or so, i read Samuel R. Delany’s Babel-17 (still one of the true books of my heart) and then, because of a tossed-off bit of trivia in it, set about to teach myself introductory Finnish.” Now, that’s my kind of reader!
Bring back Wede!
https://en.wikipedia.org/wiki/Wede
Good lord:
I had no idea! (I’m afraid I would have been one of the mockers.)
Surely that was intended as a syllabic… wait… acronym, not a syllabic initialism, and therefore [veˈdeː]?
Write Baumann and ask!
What’s wrong with Volapük, then?
(Actually, I vaguely recall that both elements of the name are supposed to be derived from English, but what the hell, it’s got an umlaut and everything!)
If this is deadpan humor, it is good: I strongly suspect (Can L1 Teutonophones confirm/deny?) that the choice of “Alman” as a language name was made chiefly (solely?) to have an “Alman Institut für Sprachvereinfachung” instead of an “Allgemeines Institut für Sprachvereinfachung”.
One aspect shared with some artificial German pidgins/simplified varieties is the use of “die” as the sole form of the definite article (leaving aside the genitive “der”, in Alman). Now, I had always assumed this choice was natural, and was due to the identity of the nominative/accusative plural and the nominative/accusative feminine singular forms, making “die” a natural choice as an unmarked article…except that in the German creole Unserdeutsch (Papua New Guinea) the sole form of the definite article is “de”, seemingly derived from “der”…and while in one German pidgin, Kiche Duits of Namibia, “die” was the sole form of the definite article, its identity with the Afrikaans definite article almost certainly played a role in its selection.
Again, I would like L1 Teutonophones’ opinion on the subject: If you were creating a German pidgin/trying to communicate using German words (same thing, really…), which form of the definite article would you (spontaneously!) use? “der”, “die”, or “das”?
Probably the tossed-off bit of trivia is the first few sentences here about cases (but I’m including the rest of the paragraph for Hattic interest):
I imagine Mark Twain’s rant about German seemed funny to him at the time. (After a paragraph or so, I feel like shouting, “Sam! Sam! I get it, OK? Enough already!”)
i’m sure i’ve mentioned it before, but spoken hasidic yiddish is drifting towards a single article, /dɛ/ or /də/, which i suspect is written די | di (to the extent it exists on paper). i don’t know (and don’t have time to look up right now) whether the accusative and dative classes are being lost, or/and what simplifications are appearing on that side of things – i assume so, to some extent, because the three historic articles all have different case-based forms (adding up to between three and seven (or 8 if i’m misremembering) noun classes, depending on the topolect).
Looking at those lines from Babel-17, it looks like a bunch of disconnected language facts, rather than something an actual linguist would be thinking. I mean, Finnish is related to Hungarian, and also does not have gendered pronouns. French is related to Spanish, and also uses gendered grammar for disparate nouns.
Speaking of a large collection of pronouns used in different status connotations in “oriental languages”, wasn’t there a recent thread about how Thai has that? Do other languages in that family work the same way?
If you’re going to ditch a German case, wouldn’t the genitive be a more obvious choice? “Der Dativ ist dem Genitiv sein Tod” and all that.
de [də], very common in “uneducated” speech.
Which had the effect that now people use nothing but the genitive after prepositions, even when before they had always been used with the dative. You even find this with supposedly educated people. So it’s one case where a prescriptivist was at least partially successful. On the other hand, in everyday speech, everyone prefers constructions with von to actual genitives, while the “dem/der X sein/ihr Y” construction is still limited to “uneducated speech.
@ulr
Jetz kann i dem Hanswurschtn, dem Heimerl, sei Hypothek aa net zruckzahln!
https://scholarsarchive.byu.edu/cgi/viewcontent.cgi?article=1016&context=sophiefiction
P.11
I think this has to be left as is. Or what would you suggest?
Alman is used as a somewhat jocular colloquial designation for ethnic Germans (another designation is Bio-Deutsche); it can also be used for designating stereotypical German behavior (“The neighbors complained because we mowed our lawn during noontime. Typisch Alman“).
A couple of years ago, a book called Deutsch für alle. Das endgültige Lehrbuch. “German for everyone. The final textbook” made some waves*); in the book, author Abbas Khider describes the difficulties foreign learners have with German grammar and proposes simplifications similar to the Alman project. The book is tongue-in-cheek, but I can imagine that it influenced the Alman project.
*) I got the book loaned from my brother; IIRC, he told me that he bought it because he came across it as a radio show; I’m not sure whether the radio show came first and was turned into a book later, or whether the radio show was based on the book.
Right. Alman comes from Turkish. It is often used by “non ethnic” Germans sarcastically to punch up, along the lines of “that’s mighty white of you” in modern American English. Given that the simplifications proposed by the Institute are fairly similar to “immigrant German” as heard on the streets of Berlin or Vienna, my first assumption seeing “Alman” was that the Institute is proposing to replace old German with modern immigrant-influenced German.
A vital piece of information, thanks!
I’m a bit surprised and disappointed that they didn’t get rid of verb conjugations at the same time.
Being liberated from the strictures of subject-verb agreement should be a bare minimum!
Yes, that is surprising. I wonder why?
The received wisdom has long been that people learning German have more problems with case and noun genders than with verb conjugation. At least with the genders, that makes sense, since there are so many words for which, “Der, die, oder das?” just has to be memorized.
Surprising only if “subject-verb agreement” and “verb conjugations” chap your derrière. That might be the case in a language you have not learned to speak without effort. In a language you have learned to speak without effort you don’t notice them, at any rate not enough to grouse about them.
Oh. I have only studied German. Not learned it!
I’m so happy that my native Swedish ditched subject-verb agreement about a century ago.
there are so many words for which, “Der, die, oder das?” just has to be memorized.
Wrong way to think about it. Those are just part of the “nouns”. They help in fitting words together. In assembling an IKEA sentence, you use different types of screw, requiring different types of screwdriver.. Of course you have learned that this type of screw has a Philips head, that type is a flathead or a torx etc.
So, I clicked through to his Git repository. There is more to it than simply a (pseudo-)conlang, such as a sentence corpus and a translation model which is tested against a number of LLMs, as well as more detailed notes on some of the design decisions. Nothing on the verbal system, though. I think maybe the removal of the verbal system caused the LLM success rates to degrade too much?
Maybe one reason is that the biggest group of foreigners currently struggling with German are speakers of modern varieties of Arabic. Their varieties mostly have ditched the case system and have an invariant article, but still have fully-fledged verb inflection.
I have no idea. But I don’t have any kind of [ə] natively; unstressed e is mercilessly [ɛ], and its dialect cognate is mostly zero (sometimes [ɛ], sometimes [ɐ]).
(Speaking of dialect: der, die, das, when actually used as articles and not as demonstrative pronouns, come out as [d̥ɐ], [d̥], [s]. Only one vowel among the three of them; but, unlike in ulr’s experience, they’re never merged. They also result in much fancier consonant clusters than I’d recommend for an of auxiliary language.)
What, they kept the genitive!?!
I think trotz des and dank des were already widespread before Sick wrote his sick book.
These two dem are not articles; they’re demonstrative pronouns. Dative as expected. The sei* isn’t genitive either, it’s the 3sg m. possessive pronoun (with full adjective declension).
“Now I can’t pay back his mortgage to that clown, that ???, either!” (All very dated.)
* [sɛ͠ɪ̯], Standard sein(e) – the -e has disappeared, but the -n- is actually still there.
So you have to memorize them… as part of the nouns.
@dm
Ok. I thought that was just a more extended version of dem X sein(e) Y, to which ulr was objecting. Reformulating that as die Y von X would give vom HW-Heimerl die Hypothek or without inversion die Hypothek vom HW-Heimerl. But I thought both of these would be more awkward than the original.
It can’t be extended like that.* You’d need to reword the whole thing: die Hypothek von {dem [obsolete insult], dem [forgotten insult]}…
* …in Bavarian anyway. It’s much more widespread.
Hans may be right in seeing the lack of case marking and invariable definite article (minus the genitive) versus the persistence of verb agreement as due to Arabic L1 learners being so numerous among people currently acquiring German as an L2. It is a reminder that popular views of “simplification” depend heavily on the linguistic ecology where such popular views arise.
Nothing new under the sun: In discussing Esperanto (he was very sympathetic to the basic principle of an artificial, simple international language, just to be clear) the French linguist Antoine Meillet bemoaned the fact that Zamenhof, being Eastern European, had deliberately chosen to have a feature such as an accusative marker for nouns, something Meillet (correctly) thought was a needless complication (To someone speaking Yiddish, German, and Polish, of course, ONLY having a single accusative-marking morpheme probably seemed like a radical simplification).
“Alman” as a Turkish loanword in German: Well, you learn something new every day. Does any hatter know why Turkish borrowed this word from French “allemand”? For obvious reasons of geography I would have expected a direct loan from “Deutsch” or perhaps a form found in a Balkan language.
Lewis’ 1967 Turkish Grammar mentions only English and French as major modern sources of European loanwords in the Turkish of Turkey; he goes on to say “the borrowing of French words began in the nineteenth century, and there seems to be still a preference for the French forms of words common to French and English.” He gives deterjan, kilosikl and süperprodüksiyon as examples.
I recall reading somewhere that nineteenth century/early twentieth century progressive-minded Iranians regarded France as the quintessential sophisticated modern country.
Was Alman borrowed in the Atatürk years (the Ottpmans had Nemce from South Slavic)? The turks also borrowed words for American and Brazil from French.
https://tamsozluk.com/en/k/turkish-words-french-origin
David Eddyshaw: I would expect the balance has tilted in the direction of English since 1967. And I would expect a word designating “German” (perhaps not distinguished from “Austrian”) to have existed in Turkish well before the nineteenth century. Along with German itself or a Balkan language, I could imagine a reflex of Italian “tedesco” being borrowed. And indeed perhaps “Alman” replaced an earlier form borrowed from Italian/German/A Balkan language (such as “nemce”, which apparently existed in Ottoman Turkish. Which brings up the issue: why was “nemce” replaced?)
Plasticpaddy: that French served as the source language for such “exotic” (from a Turkish point of view) lands as Brazil or other countries in the Americas is unsurprising. It is having French as a source for “German”, when so much direct contact between the Ottoman Empire/Turkey and the German-speaking world existed long before French influence made itself felt in Turkish that I find odd and in need of explanation.
Well, prior to 1871, there was no “Germany” to have a Turkish name.
“Alman” could also be from Spanish, perhaps indirectly. The *original* “Lingua Franca” of medieval (and early modern) Mediterranean sailors/traders was mostly Romance in vocabulary but with a blend of Italian-origin and Iberian-origin lexemes. I expect it was a vector of at least some lexical transmission into Ottoman-era Turkish. What was its word for “German”?
“It is often used by “non ethnic” Germans sarcastically to punch up, along the lines of “that’s mighty white of you” in modern American English.”
For the record, “that’s might white of you” is not used the way you describe it above. It’s an older phrase to declare someone is stand-up, decent etc. it’s not a send up of white people, the way alman is often used.
For that, it could be a whole variety of things, including “that’s so white (of you)” but this emphatically does not mean “that’s mighty white of you.”
I don’t know about Turkish, but you also have almani for “German” in Arabic; the Slavic word for German, e.g. FYLOSC njemac, has come to mean Austrian (nimsi). Current Turkish for Austrian is Avusturyalı, but maybe it was similar to Arabic in Ottoman Turkish (Nemçe seems to exist as an alternative term for “Germany”.)
Edit: I got distracted while writing this and see now after finally posting it that others already have weighed in.
“Germany” is yr Almaen in Welsh, too. Presumably from French; I never previously thought about it.
Turkish has a thick layer of French loanwords from the late 19th/early 20th century. All the usual internationalisms arrived in French form. This leads to interesting consequences, like Turkish üniversitesi (< French) vs. Tatar universitesi (< Russian), even though Tatar is the one with the stricter vowel harmony overall. It also includes pretty much every word with j – the reason why /d͡ʒ/, common in native, Arabic and Persian vocabulary, was assigned the letter c.
The newspaper Hürriyet (Arabic loanword) comes with SÜPER SPOR.
And indeed Avusturya looks like a reading of an Arabic transcription.
Almain is actually attested somewhere in Middle-or-so English.
(Without the j in Ekavian.)
@Étienne, dm
Yasir Yilmaz (Palacky University, Olomouc / ÖAW -Wien), “From Nemçe to Avusturya. Ottoman Appellations for Austria”, DOKTORATSKOLLEG AUSTRIAN STUDIES INNSBRUCK(SIEGLINDE KLETTENHAMMER & KURT SCHARR (HG.)): WAS HEISST OSTERREICH? Überlegungen zum Feld der Austrian Studies im 21. Jahrhundert (Wieser Verlag: 2021), p.80-98.
https://www.academia.edu/67731736/From_Nem%C3%A7e_to_Avusturya_Ottoman_Appellations_for_Austria
This is focussed on the use of various terms by Ottoman writers (Alman, Nemçe, etc.) with reference to Austria. My takeaway is that before the mid-19th century, Ottomans used Nemçe for Austria, which was part of a greater (not politically unified) Almanya, but that the two terms were not consistently distinguished. After that, Avusturya replaced Nemçe. Likewise, for the ethnicity Alman and Nemçe were not consistently distinguished. The article does not say, but maybe the replacement of Nemçe by Alman followed the gradual identification of Nemçe with Austria and then eventual replacement of Nemçe by Avusturya.
I have only seen the phrase used ironically. I cannot imagine it being used straightforwardly except by white supremacists.
There is an O’Henry short story in which two men are kidnapped, treated well by their captors, and promise to say how “white you’ve treated us.” I doubt there is any relation to race, just the usual “white=good” color symbolism that has pervaded many European languages since long before any of them had to absorb complex ideas about racial hierarchies.
EDIT: Oh, I see the original context was “modern” usage. Yes, I doubt anyone still uses this term in a non-racial way in 2026.
I have only seen the phrase used ironically. I cannot imagine it being used straightforwardly except by white supremacists.
Not these days, of course, but it was still so used when I was growing up (not by me!).
I doubt there is any relation to race, just the usual “white=good” color symbolism that has pervaded many European languages since long before any of them had to absorb complex ideas about racial hierarchies.
Nope, it was definitely racial.
Nope, it was definitely racial.
The OED (I.5.c, ‘Honourable; square-dealing’) says it’s originally U.S., dates it back to 1837, and says “In origin probably chiefly reflecting racial and cultural stereotypes formerly associated with European descent (and hence implying contrast with people of other races), although perhaps partly informed also by sense A.II.7a.” (‘Morally or spiritually pure; stainless, spotless, innocent,’ which goes back to OE.)
my understanding is that french was the favored european/colonial-power language in turkey and iran well into the 20th century – partly a holdover of its status as The language of culture; partly historical semi-colonial ties, partly the british and russians often being more active political enemies.
from my own experience, all i know is (1) in istanbul in 2002, my french was far and away more useful than anything else non-english we had (castellano and polish), and consistently better than english with older people (meaning 40ish vs 20ish), (2) my first translation job, as a teenager, was englishing a french introduction to a qajar-era feminist memoir, which had only been translated into french at that point (c. 1993), (3) my iranian highschool physics teacher – who went into exile in the mid-1970s – spoke french as his second language, and i think was the main one he’d studied before doing his graduate work in europe, most likely somewhere between 1955 and 1970.
perhaps partly informed
Perhaps. But I’m pretty sure that ancient semantic trail going back to OE had nothing to do with how Americans of the 19th and 20th centuries used or understood it.
Yes: Vola means World; Pük means Speech. As English as you could wish.
Yes, they’re keeping the genitive. But in the feminine form, which is identical to the feminine dative, so it sticks out less.
However:
Existing German isn’t recognized by the German state either. Like the US (and unlike Austria), Germany doesn’t have an official language.
…as easy to learn as English…
OK, this is a joke.
Quite. English is like Navajo: the elders evidently sit around their campfires at night discussing new ways to make the language even harder for foreigners.
Not totally convinced by the alleged simplicity of Spanish, either.
Really, none of these convoluted forms of speech can really compete with the limpid clarity and extreme simplicity of Welsh. No cases! Just one regular verb conjugation! VSO constituent order, just as God intended! Most straightforward orthography of all the indigenous languages of the British Isles! Even easier than Old Irish!
English and especially Spanish have hyperbolic learning curves: the basics are very easy (if you’re coming from SAE at least); the hard parts come later.
Navajo starts very steep… and I don’t even know how it continues.
So, this morning I’m reading the Grauniad’s thing on Anton Jäger and I get to him quoting Marx on potatoes, “simple addition of isomorphous magnitudes.” Wait, wasn’t it some other kinda-scientific adjective for „einfache Addition gleichnamiger Größen“? And, indeed, this is a minority, though well-represented, versus “homologous.” Both feel a bit over-done to me.
I am, of course, inclined to believe there is some important factional separation in whose English translation you choose.
Bloody Isomorphist splitters. Icepick material, the lot of them, if you ask me.
GT gives “simple addition of quantities of the same kind.” That can’t be right, of course, because the man on the Clapham omnibus can understand it.
It doesn’t help that the term is evidently extinct; for “gleichnamige größen” -ai, Google only finds one occurrence (on the first few pages) that is younger than Marx (and not a discussion of his work), and that one is literal – a complaint that clothes sizes of the same name are quite different in different countries.
But from context it’s clear that it simply means “numbers with the same unit”; don’t add apples and pears. Isomorphists, homologists – icepicks on them all.
Marx there is comparing French farmers with couch potatoes. When one gets up to go pee, another one takes his place (gewinnt so ihr Lebensmaterial mehr im Austausche mit der Natur).
Those old-timey writers were pretty stuck-up. They revelled in the hard-headed condescension of their political views.
#
Jede einzelne Bauernfamilie genügt beinah sich selbst, produziert unmittelbar selbst den größten Teil ihres Konsums und gewinnt so ihr Lebensmaterial mehr im Austausche mit der Natur als im Verkehr mit der Gesellschaft. Die Parzelle, der Bauer und die Familie; daneben eine andre Parzelle, ein andrer Bauer und eine andre Familie. Ein Schock davon macht ein Dorf, und ein Schock Dörfer macht ein Departement. So wird die große Masse der französischen Nation gebildet durch einfache Addition gleichnamiger Größen, wie etwa ein Sack von Kartoffeln einen Kartoffelsack bildet.
#
KM: I say, I say, I say!
FE: What do you say? What do you say? What do you say?
KM: Why is the French nation like a sack of potatoes?
FE: I don’t know. Why is the French nation like a sack of potatoes?
KM: Well, it’s like this …
[FE then says: but what about the Umschlag von Quantität und Qualität? Boom, boom!]
He probably stole the potatoes line from Freddie just like he did “first as tragedy and then as farce.”
Stole? As fellow workers, they held the means of literary production in common.
French nation is potato.
Sakartoffelo.
Rant: Welsh is simple to the point of being simplistic in places. Since there are no real distinctions between subject and object pronouns, and either can usually be dropped, AND many verbs are labile, it is often entirely dependent on context to determine who, IF ANYONE, is filling a certain syntactic niche.
(end rant)
In general, I agree that “X language is easier than Y” is ignorant and nonsensical. But I think there are objective reasons why some people feel that some languages are easier than others. Often it’s because the difficulty isn’t residing in the first few things you learn about the language. If I struggle with tone, I’m not going to get past day one learning Mandarin. But if I struggle with the subjunctive, I can get pretty far into a basic Spanish course before things get hairy. Some languages really do front-load a lot of their complexity into the “See spot run” sentences, while others don’t.
For example, when I started learning Welsh, it was annoying that any time I encountered a new word, I had to guess whether it was mutated or not. If I see a phrase like “Y gadderfolch” (or whatever), it’s either a masculine noun that starts with G, or a feminine noun that starts with C! Imagine what a pain in the ass that would have been before electronic dictionaries. Day one of any Welsh lesson you will run into this ambiguity, and I’ll bet it daunts people who would otherwise find Welsh very “easy” in the long run.
But the mutations are so simple …
I’ve known even small children that manage them just fine …
Yeah, but small children can eat bugs, too.
For the record, I think Welsh earns a spot alongside Spanish among languages that “feel easy” overall for L1 English speakers. But unlike Spanish, you can’t jump in and start composing coherent sentences five minutes into the first lesson. If you learn two words in Spanish and put them together, nothing happens except a successful Spanish phrase. Try this in Welsh, and there is a 90% chance that you will get a disappointed sigh from your instructor. This is what I mean about “front loading.” Welsh forces you to understand every mutation from the beginning. For beginners learning Welsh while not already fluent in Welsh (which if my math is correct is all of them) this also means that any time you encounter a new word, which again is a frequent occurrence for beginners, there will almost always be two possible words that you could be encountering, and you won’t know which because you are not yet fluent in Welsh. You can’t see “Y gaddfolch” and say “ah, yes, a soft mutation on a feminine noun; how simple; even a small child knows about soft mutation on feminine nouns,” because you don’t know what “gaddfolch” is yet! You don’t know that it’s feminine. You’re looking in the dictionary under G like an absolute moron!
I hope this clears up any misunderstanding.
Perhaps I should not be teasing those who have shown their worthiness by aspiring to learn the Language of Heaven …
To be fair, the full horror of the mutation system is not actually something that even small L1 children can manage; a fair bit of it is not present in Welsh as She is Actually Spoke, where maybe only the soft mutation is absolutely indispensible. And things like the “mixed mutation” are really literary stuff.
I would have guessed that it would have been Welsh syntax that was the biggest challenge for L1 Anglophone learners. Perhaps that only becomes properly terrifying once the fear of the mutations has eased up a bit.
I often think that Welsh mutations are somewhat analogous to external tone sandhi in Kusaal: in both cases, the origin was in transparent phonological processes to begin with; then both languages dropped their word-final vowels, so you can no longer actually see the bits that caused the sound changes any more, and they look quite arbitrary and unmotivated; then, in both languages, analogical changes supervened to the point that the once-transparent origin of the changes is no longer any help at all in actually applying them in the contemporary language, and it all just becomes one more burden on the poor learner.
(And Kusaal only has one regular verb conjugation too, but multiple different noun plurals. Just the way a proper language ought to be …)
I like the note in Marjorie Shostak’s introduction to Nisa: The Life and Words of a !Kung Woman:
Presumably Juǀʼhoan grammar is free of consonant mutations.
Presumably Juǀʼhoan grammar is free of consonant mutations
Initial consonant mutations seem surprisingly unpopular in the languages of the world. (I blame alphabetically arranged dictionaries, infantilisation and violent video games.)
Apart from Insular Celtic, I can think only of Mende and its close relatives, Fulfulde (famously) and the Austronesian language Nias; the Welsh system is easily the most complex among all these. Cymru am byth!
Yélî Dnye has it (what doesn’t it have?) but only in a very limited way: IIRC, just in expressing possession by a second person singular. Amateur stuff, really.
Some Bantu languages (and some other Volta-Congo languages) kinda have it, but only by cheating: it’s all just sandhi between class prefixes (sometimes subsequently lost) and following stems. Doesn’t Really Count.
The WP article on consonant mutation has some examples which don’t count (non-initial, or purely phonological), but the Vanuatu and NC examples may answer, though they still aren’t as messy as Celtic.
@Madeline Kalvis: Thank you for your posts about the differences in initial complexity. I wanted to say something similar and didn’t have the time yesterday, but you made all my points already, probably better than I would have been able to.
Yélî Dnye has it (what doesn’t it have?)
i don’t think i’d ever looked at yélî dnye phonological charts before, and any language that can make IPA look like a devanagari script has my undying respect and i hope my life never depends on pronouncing it.
I don’t know about initial complexity. I think attitude is a big part of it. After 51 years I still have trouble with tones in Mandarin, especially when pronouncing things in isolation and when typing. But for many years now I have often passed for a first-language speaker on the phone (without intending to do so). Surprises me every time. I started with an attitude something like, Ok, I’m here in Taiwan, so I’ll learn this language. Can’t be that hard, can it?
That said, Welsh and Yélî Dnye look pretty daunting.
I was going to say if you’re not used to tones, learn some Mandarin first, that’s tone language in easy mode… it introduces the concept without stacking level tones like even Cantonese does.
(I’ve never used Mandarin on the phone, though, and I haven’t tried to learn any other tone language.)
tone language in easy mode
The simplest full-fledged tone system among languages I know is in Hausa. Just high and low (with a few falling tones from fusion of HL on long syllables), no fancy emic downsteps, and virtually no tone sandhi at all (there are a couple of particles which take the opposite tone to whatever precedes them, but that’s all.)
Presumably Ju|ʼhoan grammar is free of consonant mutations
Having just dug out my copy of Patrick Dickens’ brief grammar, I can confirm this. In fact, Ju|ʼhoan looks rather dull from a purely morphophonological point of view; a few suffixes, added without any exciting sandhi changes, but almost “isolating” overall.
(Dickens himself died of AIDS, aged only 39.)
Not these days, of course, but it was still so used when I was growing up (not by me!).
By the late 1960s «white » was already disparaging in some circles. I was listening this morning to Andrew Hickey‘s brilliant podcast series on the history of Rock Music in 500 songs and the Zappa episode actually includes the following dialogue between Frank Zappa and Davy Jones of the Monkees, from the movie Head:
Zappa, shaking his head: “That song was pretty white”
Jones “well, so am I. What can I tell ya?”
Zappa: “You’ve been working on your dancing though.”
That’s simpler, but I’d say Mandarin is easier because Mandarin has only one level tone.
(Systems very similar to Hausa’s seem very widespread – Navajo for one.)
Systems very similar to Hausa’s seem very widespread
True. Lingala is another African one. Proto-Bantu seems to have been like that too, but most modern Bantu languages like to shake things up with tone spreading and tone dissimilation and allsorts.
I think the difficulty is often not so much in the tones themselves, but in tone sandhi and how tones interact with morphology and syntax. Kusaal’s basic tones aren’t particularly hard to distinguish (or pronounce) for example, but that doesn’t help you much when it comes to e.g.
M̀ dá nwɛ̀’ nâ’ab lā.
I FAR.PAST strike chief the
“I struck the chief.”
versus
Kà m̀ dá nwɛ̀’ nà’ab lā.
and I FAR.PAST strike chief the
“And I struck the chief.”
Hausa is
… comparatively simple tonally beside this kind of thing, but it still manages to do a lot of unobvious things with tone in its morphology, which tends to do with tone changes what the Semitic languages do with vowel changes.
Languages like the Sinitic and Athabaskan ones, where tones have arisen relatively recently from segmental features, seem to be a lot less prone to weird morphological or syntactic tone phenomena than e.g. Volta-Congo languages, where tones had nothing much to do with segments from the beginning, but floated serenely above them all.
The late Sixties Zappa-Jones interaction Vanya quotes was echoing an already-longstanding hipster/beatnik trope whose locus classicus might be the famous passage by Kerouac “At lilac evening I walked with every muscle aching among the lights of 27th and Welton in the Denver colored section, wishing I were a Negro, feeling that the best the white world had offered was not enough ecstasy for me, not enough life, joy, kicks, darkness, music, not enough night.” Note FWIW that this is from what is at least nominally a novel even if the first-person narrator is often taken to be a stand-in for the author. The author’s own historical visit to Denver that may underlie this passage is said by Kerouacologists to have been in 1949.
Shostak’s quote struck me in that it took her three months to get the hang of the fiftyish clicks and dozens of other phonemes in the language, yet twenty months to use its four level tones correctly. Given that her linguistic description is imprecise, I can come up with some possibilities: Maybe she just has a hard time with tones, but clicks are easy; or maybe “tones” subsume phonation differences (glottalized, murmured, pharyngealized, epiglottalized) plus nasalization, which make them much harder to learn. And/or maybe it also took her a lot longer than three months to get all the click accompaniments right, but getting them partly right is enough to be understood, even if you sound like a funny foreigner.
Oh. Very interesting potential implications for historical-comparative Afro-Asiatic.
Absolutely. Most Sinitic languages, and other whole families in the region, do have heaps of tone sandhi, but haven’t morphologized it AFAIK; but Mandarin has very little of that.
(Shanghainese, at the other end of the… horseshoe, has turned it up to 11 and ended up as a pitch-accent language, with at most one unpredictable tone per pretty long word.)
Ah, that sounds likely. For that, 20 months is actually pretty impressive.
@David Eddyshaw
Welsh syntax doesn’t seem too bad to me. The obvious differences, like main verbs before subjects and adjectives after nouns, are things that any English speaker will be familiar with in at least some contexts. Sometimes I will encounter “bod” halfway through a sentence and have to work backward to figure out the parallelism (i.e. “wait, was there a main verb waiting for a berfenw back there somewhere?”). But it’s certainly no worse than any IE language that chains pieces together to make a verb phrase.
I think more consistent challenges for English speakers will be:
1. verbs so weathered by the centuries that their stems are hard to spot at first, like cael or mynd.
2. homophones in different parts of speech (“Bwytodd o llawer!” “Hmm? Bwytodd llawer o? Llawer o beth?”)
3. fake forms that only exist in textbooks like “rydw i” for “dw i.” Nothing sabotages a new learner more than well-meaning scholasticism in a textbook.
Yes, the fake forms mostly go back to Cymraeg Byw (of evil memory.)
My favourite Welsh multi-homophone is /i/ (which includes the words written ei, as you will know, of course.)
Pronouncing ei “his/her/its” as written is not only a Horrid Spelling Pronunciation, but is a spelling pronunciation based on a mistaken spelling: this error goes back to
https://en.wikipedia.org/wiki/William_Salesbury
who wrongly supposed that the Welsh 3rd sg possessive pronoun i was derived from the Latin eius.
Admittedly, it’s quite handy to have a distinctive spelling for /i/ in this meaning, even though it is Wrong, Wrong, Wrong. (Even some L1 speakers have been bamboozled into ignoring their actual, perfectly correct, spoken forms and mispronouncing ei in Literary Welsh. I blame infantilisation and Margaret Thatcher. Yma o hyd …)
“Salesbury’s principles of translation
There are three particular aspects which stand out from Salesbury’s work which, in the view of Isaac Thomas, amount to his principles of translation.
Verbal faithfulness
The title page of the 1567 New Testament states that the translation was ‘word for word’ from the original. In practice, this meant having an equivalent word in Welsh for every corresponding word in the original Greek. However, Salesbury did not keep to the word order (syntax) of the underlying Greek when translating into Welsh. Where Salesbury had inserted a word into the text, he enabled the reader to identify such cases clearly by using a different ‘typeface’ or font.
Variation of expression
Another distinctive aspect of Salesbury’s work was the way he varied which Welsh word he used to translate the same word in the original text. This was a reflection of the influence, in particular, of the important Renaissance scholar Erasmus, who had championed the idea of variety and abundance of vocabulary and expression in literature.
Dignity of expression
Salesbury’s desire for dignity of expression can be seen in his use of antiquated Welsh terms. In the case of Welsh words of Latin origin, he also sought to highlight that by the way he spelt them, even if that did not reflect how they were pronounced.
These distinctive features … proved to be controversial
….
For example, when Bishop William Morgan wrote the preface to his complete Welsh Bible published in 1588, while generously acknowledging Salesbury’s prior work, he also described his task in part as ‘cleansing’ the New Testament. This most likely refers to ridding the translation of aspects of Salesbury’s work, such as the unusual pattern of spelling and the inconsistent use of mutation.
Notwithstanding these criticisms, his translation has been hailed by Isaac Thomas as being among the best of the period, as far as faithfulness to the original is concerned. ”
https://www.evangelicalmagazine.com/article/the-welsh-new-testament-of-1567/
Even Bishop Morgan’s version is in language that must have been archaic already by that time. (Like the English KJV, too; though the less notable, inasmuch as Welsh has pretty much always had a much wider gap between the literary language and the spoken language than English.)
At least eight kinds of Interslavic have been proposed since 1826.
https://isv.wikipedia.org/wiki/Med%C5%BEuslovjansky_jezyk
I wouldn’t mind some actual historical spelling in Welsh. How about gh to indicate a g lost to soft mutation? Irish has a lot of this sort of thing, and while some people find it intimidating, I appreciate the etymological information.
Well, Old Welsh orthography generally ignored the mutations, not only between but also within words, so e.g. the name Tudfwlch was written Tutbulc, and Taliesin’s patron Urien appears as Urbgen.
The weirdness of Old (and Middle) Irish orthography is due to the Irish having adopted Brythonic spelling conventions, which, in turn, arose from writing Latin in the traditional way, instead of how Brits actually pronounced it by then.
I was just thinking that, although Old Irish orthography leaves, um, something to be desired when it comes to representing the actual sounds of the language, adopting Brit spelling conventions was perhaps the best solution if you were going to write the language with the Latin alphabet without inventing lots of diacritics or digraphs.
I was just recently having a (friendly) dispute elsewhere online with someone who was likewise suggesting a (modern) Irish-like orthography for Welsh would be more helpful for L2 learners. I said that, although one naturally welcomes people learning the language as an L2, the orthography is for L1 speakers, for whom it largely works fine*. I don’t see any reason to change the orthography to ease the task of L2 learners (English speakers are welcome to lead the way by reforming English orthography if they disagree.)
* Partly because one generally knows where to expect the mutations, so undoing them is not usually a big deal. Admittedly you can be garden-pathed by e.g. disappearing g‘s and by m and b falling together, but it seems a small price to pay for having an orthography infinitely more sensible than English. (It’s not optimal, though, because the writing of unstressed vowels is historical; but undoing that would cause problems because the spoken dialects often differ in their reflexes.)
Yup, I don’t know any Celtic languages in depth, but I understand why Irish/Scottish Gaelic look the way they do. Which is a good start.
And it is, of course, unnecessary to point out at the Hattery that speech is prior to writing: a L2 learner of Welsh should be learning to speak the language, and if they do that they’ll have to cope with the mutations anyhow, like it or not. (Presumably the same is true of Irish, which has plenty of falling-together of mutated consonants of its own, after all.)
And it is, of course, unnecessary to point out at the Hattery that speech is prior to writing
Here writing always precedes speech. That’s why hardly anyone speaks any of the languages they know so much about. It’s hard to get blood from a beet.