Simon Akam in the New Yorker has one of the best essays I’ve read (archived) about the struggle to standardize a language. Here’s the start:
Ask him how it all began, and he remembers the ice. It was a bitter morning in January, 1982, when Bernard Cathomas, aged thirty-six, carefully picked his way up a slippery, sloping Zurich street. His destination was No. 33, an ochre house with green shutters—the home of Heinrich Schmid, a linguist at the University of Zurich. Inside, the décor suggested that “professor” was an encompassing identity: old wooden floors, a faded carpet, a living room seemingly untouched since the nineteen-thirties, when Schmid had grown up in the house. Schmid’s wife served Rüeblitorte, a Swiss carrot cake that manages bourgeois indulgence with a vegetable alibi.
Cathomas had already written from Chur, in the canton of the Grisons, having recently become the general secretary of the Lia Rumantscha, a small association charged with protecting Switzerland’s least known national language, Romansh. Spoken by less than one per cent of the Swiss population, the language was itself splintered into five major “idioms,” not always readily intelligible to one another, each with its own spelling conventions. Earlier attempts at unification had collapsed in rivalries. In his letter, Cathomas said that Schmid’s authority would be valuable in standardizing the language. Cathomas wrote in German but started and ended in his native Sursilvan, the biggest of the Romansh idioms: “Jeu engraziel cordialmein per Vies interess e Vossa attenziun per quest problem.” Translation: “I thank you very much for your interest and attention to this problem.”
Schmid, the man he was counting on, hadn’t grown up speaking Romansh; he first learned it in high school, and later worked on the “Dicziunari Rumantsch Grischun,” a Romansh dictionary begun in 1904 and still lumbering toward completion. But the depth of his expertise was formidable. By the time Cathomas knocked on his door, Schmid had already sketched a plan for standardizing Romansh: a “majority principle” in which the most widely shared spellings across the idioms would win out.
“He really already had everything,” Cathomas recalled. “He had worked it all out in his head.”
What Cathomas hadn’t reckoned with was how quickly the tidy scheme, once loosed into the valleys, would ignite quarrels that engulfed Swiss classrooms, newspapers, and eventually cantonal politics—a parable of how an attempt to secure a language’s survival can feel, to those being standardized, like an assault on what makes them distinct.
I hope that lead-in entices you to read the rest; it’s gripping, sad, and funny, with sympathy for all sides. (And specially for J.W. Brewer, I cite the band name Liricas Analas ‘Anal Lyrics.’) Here’s a bit near the end:
The Lia Rumantscha is also asking the International Organization for Standardization to classify each idiom as a separate language. Some people doubt that this hyperlocalism will pay off. One member of the Zurich team told me about a Swiss firm that sold a G.P.S. device with directions spoken in Swiss German. “No one bought it,” he noted. “People said, ‘That’s not my Swiss German.’ ” You can give the machine a voice, he suggested, but people still want it to sound like their cousin.
I might add that I am the proud owner of a Lia Rumantscha publication, the Dicziunari rumantsch ladin-tudais-ch. Thanks for the link go to my dear old friend Dave, to whom I tip all my hats.
Attempts to create standard languages artificially, as opposed to picking one dialect and promoting it at the expense of others, seem rarely to work out well, not really pleasing anyone but their creators. (The one-dialect solution does work, so long as the speakers of that dialect are dominant anyway, usually politically, but sometimes culturally. If that’s not the case, you’re probably doomed whatever you do.)
Even if the new creation isn’t stillborn, you end up with with XKCD’s standards problem:
https://xkcd.com/927/
With Kusaal, on the Ghana side, the literacy people have picked Agolle Kusaal as the written standard, which makes sense inasmuch as there are substantially more Agolle Kusaal speakers in Ghana than Toende speakers; on the Burkina side, Toende Kusaal is the written standard, on the reasonable enough grounds that there aren’t any Agolle speakers there. Subdialects within Agolle and Toende exist, but their differences don’t seem to be big enough to cause any major issues.
One of the example texts in my Agolle Kusaal grammar (which I got from a newspaper) was evidently written by a Toende speaker, and their Agolle Kusaal has some giveaway spelling mistakes and incorrectly written particles. On the other hand, it’s perfectly comprehensible.
very excited to read this!
especially because at my gut level “a ‘majority principle’ in which the most widely shared spellings across the idioms would win out” seems such an odd choice, since it doesn’t do anything to make the writing system coherent for any of the lects, much less make it easier for readers who are speakers of different lects to grasp how they’d each render a word. (it is a very useful thing about YIVO transliteration that it isn’t the primary writing system for yiddish, so it doesn’t interfere too much with (for example) ײַ being spoken as /ɑ:/ [the historical majority realization], /aj/ [the version implicit in YIVO transliteration], or even, if i’m remembering the small dialectological corners, /ɛj/)
Jeffrey Heath’s Dogon grammars flag this up as a problem for literacy work. The French administration regarded all the Dogon languages as “dialects” of one mega-Dogon “language”, despite their mutual incomprehensibility, and the post-independence regimes continued the mistake. So schoolchildren are presented with texts in what they are told is their mother tongue, but is in fact incomprehensible to them. (The language that was chosen as “standard Dogon” is not even the largest in terms of speakers, though apparently it did already have some cultural cachet.)
In Welsh, after the failure of the artificial Cymraeg Byw, it now seems to be accepted that you can basically write in your own dialect, so long as you avoid some of the most distinctive features. It helps that the traditional orthography is conservative enough that it’s pretty dialect-neutral, so speakers can happily write e.g. siopau “shops”, whether they actually say /ʃɔppɛ/ or /ʃɔppa/.
This is a or perhaps the metal band (pretty tonal) with lyrics in Romansh: https://www.metal-archives.com/albums/Prefix/Sgratta%27m/158440
“Sgratta’m” seems like it should be a perfectly cromulent name for a metal band, so long as it doesn’t mean something like “kiss me quick.”
Sadly, it seems that the band name is the somewhat dull “Prefix.”
An early comment on Rumantsch Grischun was indeed that its creation had simply added a sixth standard to the five existing ones.
Like in German, you have to learn the written standard as a lect even though it didn’t use to be one, it seems. I think that was the idea.
And the lyrics of at least the first one are very much not in Rumantsch Grischun. Fa lö a prümavaira!
I think the only real solution is to have already produced a written standard a few hundred years ago.
Impractical, of course, but possibly true – English spelling, which people vaguely grumble about, doesn’t represent any actual form of the language, and probably didn’t in 1700 or whenever modern spelling started to settle, but it also doesn’t really force anyone to learn *someone else’s* specific form of English in order to write, so we’ve mostly just got used to it.
(RP and its modern descendents are really quite badly represented by standard spelling, which, for example, is rhotic. Actually, when I think about it, ‘English’ spelling is quite Scottish – it’s rhotic, it distinguishes wh- from w-, it doesn’t distinguish the two ‘ah’ sounds… I’m ok with that, we suffer in other ways!)
@jen
Don’t forget the bisyllabic diphthongs, e.g., great, straight.
… and fir–fur if you’re a Yes True Scotsman.
The Scots and Irish are anecdotally better spellers than the English. I confess to smiling with lofty pity whenever I see another poor fool has written “peninsular” instead of “peninsula”.
Previously on Resuscitating Romansh.
“Jeu engraziel cordialmein per Vies interess e Vossa attenziun per quest problem.”
I guess that from now on I’ll be mentally translating “thank you for your attention to this matter” into Romansh, which may improve my mood for a second.
I think the only real solution is to have already produced a written standard a few hundred years ago.
i’m not sure – i think that the first wave of yiddish standardizers’ approach was actually a reasonably good one (and i think john m. tait’s suggestions about scots have been somewhat similar): create a standard orthography that (as DE put it) “is conservative enough that it’s pretty dialect-neutral”, and stop right there. in yiddish, the early-20thC standardized spelling moved significantly away from most of the then-current spelling systems (which were largely germanizing, which is even sillier with a different alphabet as well as language), and didn’t really empower any single topolect.
multiple-standards problems did emerge, for sure, but mostly about particular details (e.g. melupm-vov ⟨װוּ⟩ vs. mekhitse-alef ⟨װאו⟩ to clarify where the vowel is in ⟨װו⟩, or whether to use a ⟨בּ/בֿ⟩ contrast, or ⟨ב/בֿ⟩). but without the nazi attempted genocide i think there would’ve been a general settling-out to a few variants of a single core standard. the meaningful problems have come in the post-khurbn period, when YIVO (and others) have tried to teach the language based on a new lect (that they don’t acknowledge as such) built off their variant of the standardized orthography, which corresponds to no previously existing lect. YIVOish now does exist (it’s more or less what i speak, though i push it towards a southeastern topolect), but most non-hasidic yiddish teaching neither respects YIVOish as a living lect nor acknowledges its artificiality – and it’s now verrrry divergent from other (larger) living yiddish lects, which have informally converged towards a different orthographic standard.
rozele: create a standard orthography that (as DE put it) “is conservative enough that it’s pretty dialect-neutral”, and stop right there.
Ivar Aasen’s Norwegian Landsmaal – and successfully so, if I may controversially say so. It may have lost out in the end, but that outcome was far from certain in the early 20th century.
More conservative than dialect neutral, though. Dialect neutrality wouldn’t really have been possible without losing all visible connection to the spoken language. It’s probably not possible in any language of some size. My take on that is that a standard shouldn’t be too narrow. Exposure to natural variation is a good thing. That means that the standard must allow it.
I’ve been told that some of the weirder features of the Kusaal orthographic reform of 2016 were attempts to be dialect-neutral between Agolle and Toende Kusaal: for example, writing duoe for /duɪ/ “rise” (previously written due), because Agolle /uo/ usually corresponds to Toende /ɔ:/ before a consonant, and the Toende Kusaal for “raise” is dɔɔe /dɔ:j/; the idea being that Toende speakers would just read uo as /ɔ:/ throughout.
The dialects are just too different for this laudable aim to really succeed though: all it’s done in practice is to complicate Agolle orthography.
(It’s a pity, if they were going to consider Toende Kusaal, that they didn’t adopt the Burkina-side Toende Kusaal use of a tilde for vowel nasalisation: no good reason why not, in this post-typewriter age. Instead, they continued the prior system of writing n after the vowel, but made it more ambiguous. I mean, it’s their language, and they can write it how they like, but really?)
Any claim that X is the “only” workable solution is highly vulnerable to historical counterexamples, but it does seem broadly plausible that successful standardization approaches were easier to pull off successfully in prior centuries via in many cases the sort of imposition-by-brute-force methods that are now quite difficult to employ in liberalish post-modern societies due to features of those societies that are in many respects praiseworthy.
OTOH, it seems like orthographic standardization in English happened without too much obvious oppression of rival approaches, but maybe that’s because for whatever reasons you didn’t have competing standards strong enough to look viable (and stable within their own user base if left alone) but weak enough to be squelchable, with a limited number of trans-Atlantic-type variants being found tolerable in practice. (In practise?) And of course there was concerted squelching of written Scots, but that brings us back to the thorny issue of whether it’s a competing standardized form of “the same language” versus a different language, with I guess the complaints about squelching seeming to some to have more strength if it’s the latter. But is Scots as distant from Scottish Standard English as the two official standard brands of Norwegian are from each other?
Standardization of Yiddish is an interesting historical example because there wasn’t (except maybe in the USSR?) that much in the way of obvious state power that any faction could hope to enlist on behalf of its own preferred orthographic scheme, which must have created different incentives for how promoters of any given proposal could hope to succeed.
A Romansh aside: I was wondering about the surname Caviezel*, and it turns out to be Romansh, but I couldn’t find its etymology. Von Planta’s Rätisches Namenbuch might have an answer, but I don’t have a copy handy. Michel-Vincent Caviezel, who introduced the potato to Brittany in the 1700s, has been claimed to be of Irish origin, but that may be false.
* American movie actor Jim Caviezel, of a right-wing bent, has recently been announced to be playing the lead in a biopic of poor ol’ Jair Bolsonaro.
Now you’ve got me curious!
there wasn’t (except maybe in the USSR?) that much in the way of obvious state power that any faction could hope to enlist
the u.s.s.r. is indeed the big exception – and oktybrish, which i have a soft spot for*, is the distant outlier in the standardization schemes*, and the big gap in what i wrote earlier.
but it’s a bit fuzzier than that: among the main drivers of standardized orthography were the yiddish-medium school systems that emerged after WWI, with, i believe, some degree of state support in the polish republic. so institutional authority was still in the mix elsewhere, though not with the power to enforce one version. the YIVO system comes out of that polish sphere; its main, if uncodified, competitor, the informal hasidic standard**, is as i understand it descended from the usage of the religious schools and newspapers of the former habsburg lands.
someday i’ll get around to reading mordkhe schaechter’s detailed history of yiddish spelling reform.
.
* partly because it’s the only one devised with expanding literacy as a stated goal.
** which i think has somewhat converged with the oxford system developed by dovid katz and others with a less strictly prescriptivist intent than the YIVO system.
Re Caviezel,
https://www.srf.ch/radio-srf-1/auf-den-spuren-eures-namens-caviezel
has
Casa = house +
(Rhaetian form of) Wetzel = (nickname for) Werner
Although I find the speaker’s Swiss German comprehensible, there is no transcript, and I am not skilled in transcribing dialect.
Thanks!
The variants Cavieziel and Cabieziel also exist. Do they still work with that etymology?
Thanks for that, and now I know Caviezel is (in German) four syllables: /kaviˈetsel/.
“Wetzel” I know as a (German-American) surname. Or I thought I did. Although in googling up one specific example I had in mind it turned out to actually be a one-letter-different German-American surname, although then wikipedia has a list of plenty of notable Wetzels, including two Medal of Honor recipients (from two different overseas military ventures).
@Y
I would guess your variants are the same name:
Cavieziel has a second palatalised e, whereas in Caviezel the second e is unpalatalised (or realised as schwa).
Cabieziel is similar, but the b is due to transcription by a Romance speaker for whom b is the best fit for Rhaetian v, eg., the Romance speaker has only [BETA] and writes “b”.
No state involvement in German until the orthography of 1901. Rather, the archaizing lect and the spelling conventions used by the chancellery of Upper Saxony at Meißen somehow became popular among neighboring chancelleries and spread from bureaucrat to bureaucrat in all geographic directions until Luther took it as the base for his translation, making sure to pick words that were as widely understood as possible. For a while there was a separate Catholic standard, but it was derived from Luther’s and slowly but surely converged with it…
X-D Well, he’s already played the other Messias, so he’s the logical choice…!!!
(He was Jesus in Mel Gibson’s Greek-free passion play.)
I’m sure they do.
The speaker tries hard to make it three syllables with a diphthong that Romansh or part thereof probably has but German, certainly Swiss German, does not. He tries to get close to [jɛ] but also to not actually reach it.
In other words, he regards it as Very Much Not German and therefore tries to pronounce it à la romande; it must not have an established German pronunciation.
Attempted transcription-and-translation into Standard German, ? = my uncertainties, articles with personal names preserved:
The laughter at the end covers too much.
* Amazing. He’s taking etwas “something”, applies *[tw] >: [pː] as seen e.g. in Yiddish, further extends the rounding to the preceding vowel, and slaps -er on it so it can refer to people. I’ve never encountered that before; it took me a while to figure out!
** wo, invariant relative pronoun throughout Alemannic. The second occurrence is endowed with a linking -[n]-; compare zune Tür farther north and rosane Rosen farther east.
Maybe we’re talking past each other but I’m confused as to why David M. doesn’t think of the standardizing practices of the chanceries of various, you know, states as constituting “state involvement.” If you want to say there was no “state involvement” at the Germany-wide level before there was a unified German state to be involved, sure. And the process by which multiple political/state actors* ended up converging rather than adopting separate standards (either a multiplicity or one Protestant v. one Catholic) is an interesting one to study. And then there are separate interesting issues as to how the standard adopted by the chancery of the Grand Duchy of Wotsit came to also be used by the Wotsitinian local schools and newspapers etc., possibly without explicit direction from the Grand Duke.
*Of course one big factor was Luthers Bibel, and the relevant German Protestant churches that uniformly adopted it were themselves state actors, not private voluntary associations relying solely on their powers of persuasion. That the non-Lutheran ones played along rather than promote a rival translation or a multiplicity thereof is important though, so again you have the question of how multiple state actors ended up working in harmony when there was no single sovereign power above them to impose uniformity.
Because, before Luther, it was literally just the scribes and literally just in the chanceries. Hardly anyone else wrote in German at the time; schools and newspapers were far in the future. The princes didn’t care and don’t seem to have made any decisions related to this.
…or, in other words, in addition to et-was he has et-wer.
You could say my dialect has both of these without the et-…
Wetzel occurs as a given name in West Virginia; probably it’s originally a tribute to Lewis Wetzel, the beloved genocidalist.
So casa in Romansh is ca` same as in Venetian?
@Rodger C: Apparently, another Lewis Wetzel was one of the founders of West Virginia during the Civil War. Born in the 1820s, he was probably named after the previous Lewis Wetzel (who I don’t know if he was actually related to), but I wonder if there is some conflation of the two in more recent naming patterns.
It looks like an abbreviated form of casa, usually ca, stretches from Catalan through Occitan (co) and all of Gallo-Italic. Wiktionary only gives extended forms in Rhaeto-Romance (Romansch chasa) and offers no Franco-Provencal etymon.
You could say my dialect has both of these without the et-…
Is that analogous to something I hear here in New Mexico, while standing in line: “I can help who’s next”? That always sounds odd to me.
I think that’s pretty much universal by now — I hear it so often here in Massachusetts it barely registers any more.
I think in Massachusetts it was more usually, “Can I help who’s next?” when I lived there, although whether it really had the intonation of a question varied from speaker to speaker.
About 35 years ago I was driving on a minor road in Catalonia, and every 100 m or so there was a sign in the neighbouring woodland that said, I thought, Caca Prohibida. Reasonable enough, I thought, though why it was necessary to say it so often was unclear. Then I realized that Catalan, unlike modern Castilian, has a letter ç, and that it actually said Caça Prohibida, obviously a Catalan spelling of Caza Prohibida
I think in Massachusetts it was more usually, “Can I help who’s next?”
Oh! Yeah, that’s what I meant — I didn’t even notice the reversed words in JF’s quote. That way it is a bit weird, but not because of “who’s next.”
No, it’s “who” used to mean “someone/anyone”, not “whoever”.
@dm 09.12:19.14
Re your missed word zw??, I understood this as “zwei davon” [Zw? dra], where dra (= daran) means “davon”. This would refer to de domo et casa, where the “house” word is recorded in the name twice. However, I cannot say why the “zwei” part sounds like zwie (maybe compare the word zwiebäck?).
Ah, that makes sense. (I didn’t know daran could mean that, but why not.) Other than in Zwieback (…now I want some…), zwie- occurs in a number of obsolescent words (Zwietracht “quarrel, lack of unity”, Zwiegespräch “conversation of two people”, Zwiespalt “dilemma”…), and… I can’t find out what its etymological vowel is and therefore how it would come out in Switzerland. Looking up the etymology of Zwietracht, it’s confusion all the way down. Hm.
@DM: Thanks, got it.
@Hat and Brett: So the part that I find strange is widespread in the U.S., and the part that seems normal to me may be regional. Yes, what I remember from childhood is “Can I help…” or “May I help…” but “I can” makes more sense at the pharmacy. The technician is telling you that they’ve finished the clerical after the previous customer and can give you the help that you unquestionably came for. “Can I help…” and “May I help…” make sense when a stranger looks lost, for instance, but if you’ve been standing in the line for people picking up prescriptions, I take the question versions as polite non-compositional idioms.
On the other hand, using “who” to mean “whoever” or “the person” would seem archaic to me in any other context.
In 2006 Microsoft decided that the thing the world needed most urgently was a Mapudungun version of Windows. What they failed to take into account was that different groups of Mapuches are passionately convinced, to the point of violence, that only one way of writing Mapudungun is tolerable, and that others must be suppressed, by force if necessary. Microsoft had used what seemed to be the most widely used orthography, but not the one the activists wanted.
I happened to be in Valdivia when the story blew up — not in the area with many Mapudungun speakers, but within shouting distance — so I could read about it in the Diario Austral, the principal (or maybe only) local newspaper. Although accounts you can find on the web today emphasize the theft of intellectual property claim, my impression is that they were primarily angry that Microsoft had chosen the “wrong” version of the language.
One of the few times in my life when I have sympathized with Micro$oft.
Athel Cornish-Bowden : that’s a very elaborate way to state your opinion on Vista. Kudos. What prompted that, though?
Don’t get me wrong, I hate Vista as much as the next guy, it’s the reason I finally switched to Linux, after initially trying FreeBSD and Plan 9.
Plan 9
Respect!
I wish Plan 9 was actually usable in Real Life. Love the idea of it …
Ah, for what might have been.
I grew up in New England and to me “I can help who’s next!” strikes me as far more common than “Can I help who’s next?” In fact, the latter phrase seems more likely to me to produce a “whoever” instead of a “who”.
I wonder if you can analyze this “who” as essentially a contraction, a kind of retail jargon caused by the pressure of working in a retail environment and the need to deliver a clear, concise, sometimes loud, message.
Plan 9? Not the one from outer space, I suppose?
(I did watch that one. Hilarious.)
I should mention that “who” meaning “whoever” also exists – and is standard.
Wer zuletzt lacht, lacht am besten. “S/he who laughs last laughs best.”
Wer zuletzt lacht, hat den Witz nicht verstanden. “…didn’t get the joke.”
Wer misst, misst Mist. “If you measure, you measure trash.” (physicist saying)
Rette sich, wer kann! “Sauve qui peut !” (optative use of Konjunktiv I)
Wer das glaubt… “If you believe that…”
that’s a very elaborate way to state your opinion on Vista
I miss Vista, with its surprisingly useful widgets, though I miss XP and Win 7 much more.
And I sufficiently dislike the idea of using Win 11 that I have been considering switching… not sure if Linux or Mac OS would be less impractical, and maybe I’ll have to grudgingly go back to Windows after all.
I grew up in New England and to me “I can help who’s next!” strikes me as far more common than “Can I help who’s next?”
Sure, as a statement I’ve heard it that way (no idea which is more common). I was just taken aback by the idea of it as a question.
I sufficiently dislike the idea of using Win 11
I dislike it too and held off as long as I could, but now that they’ve stopped supporting 10 (why? is it really so vital to force everyone to buy the latest and greatest?) I’ve bought a new computer (good Black Friday price) — my old one was too old to upgrade even if I wanted to. I’ve turned it on and connected it to the network but haven’t done anything with it, and probably won’t until the old one actually stops working. Bah.
I think the “Can I help?” version is at least in some contexts phrased as a question for the sake of politeness but is really an instruction to whoever’s next in line to come up to the counter asap so as not to create additional delay for those behind them in line.
I’ve been waiting for Windows 12 ever since Windows 8 came out. Bah indeed.
Pullum on “can I help who’s next”, in 2005: “it’s a fused relative construction with human denotation, headed by the relative pronoun lexeme who. And that is a possibility that has mostly been extinct for some fifty to a hundred years.”
Lynne Murphy, in 2007, first noticed it from young English shop assistants: “this just sounds weird to me, and I don’t recall ever hearing it in my native land.” (She hasn’t lived in the US since 2000.)
I haven’t noticed the “can I help”/“I can help” distinction, but I’m sure either way the “… help who’s next” phrase is new (or at least newly widespread) within my lifetime; I don’t think I ever heard it before the 1990s at the earliest.
Mark Liberman followed up on Fused relative clauses with who in 2013, finding that they’re not as extinct at Pullum thought: “Iago clauses [“who steals my purse steals trash”] are definitely still Out There these days, though they’re clearly informal and maybe to some extent regional. There are plenty of examples in COCA …”
That Pullum post is interesting: I hadn’t noticed that the “subordinate interrogative” construction is syntactically (as opposed to just semantically) distinct from a clause with a fused relative, but CGEL is all over this (pp1070ff.)
In Kusaal, both types are expressed by (internally-headed) relative clauses, but there’s a syntactic difference too: within a clause that represents a subordinate interrogative, you can’t prepose the object to the beginning of the clause (which is otherwise common in relative clauses.)