Some kind LH reader sent me a copy of Sydney Goodsir Smith’s Collected Poems
By the time the other European vernaculars had broken away from their Teutonic and Romance beginnings and more or less settled down to national norms based on their own particular vernacular translations of the Bible, the Scots (for various reasons, some political) had adopted the English version instead of producing one of their own — though such in part did exist. The language and spelling of this English version (generally the Eycliff), because it was the only general book available to everyone, gradually became the accepted pattern for all formal literary purposes — legal documents, royal proclamations, burgh records and suchlike. Had a Scots version of the Bible been published widely at this time there would have been a different tale to tell in all probability. [..]
He then talks about people spelling as they spoke and “provincialising matters still further with a very plethora of apostrophes to approximate it, as it were, to the accepted standard English”; he continues:
In 1947 or thereabouts a group of poets in Edinburgh, the Makars Club, decided to rationalise this spelling business if they could manage to. We agreed on some rules and agreed to abide by these and by these means hoped to remove the ‘dialect’ stigma (as we thought of it) so often levelled at the often widely different usages of this ancient and respectable literary language. Our problem was to agree upon a standard spelling so that an Aberdonian and a Borderer would spell a word in the same way while pronouncing it sui generis, just as a Lewisman will pronounce a word differently from a Cornishman yet both spell it the same in print. […]
It is difficult to find the middle way. (Take a simple example: say, the word ‘heart’, pronounced anglice ‘hart’ — or ‘haht’, if you like; scottice ‘hairt’. It was decided it should be spelled ‘heart’, as in German ‘Herz’, or French ‘cherche’, and such Scots words as ‘wersh’: but, alas, one found English or American or anglicised Scots reading it by eye as ‘hurt’ — as in English ‘Bert’ or ‘certain’. Quite apart from the sound values, simple meaning may sometimes be distorted.) It is the difficulty of unfamiliarity with the printed word. The same poem spoken would be perfectly understandable to all.
I have been as inconsistent as any. I now spell ‘heart’ as ‘hairt’, for instance. What then of the present collection ? Should I re-spell the whole lot in accord with my present probably only temporary convention or leave them as they were writ at the time, as a kind of historical record of the vagaries of fashion, one man’s fashion? I dillied and I dallied and at last laziness won and I decided, more or less, on the latter alternative. Explicatus est.
I’m not sure whether the misprint at a crucial point of the penultimate paragraph has more of sadness or of humor; in the sentence “It was decided it should be spelled ‘heart’, as in German ‘Herz’ […],” “heart” should read “hert.” Correctus est.
After skimming it once, I have trouble understanding what the author means.
I think I’ve read somewhere that there is a convention to use the same spelling as in English for those words whose Scots pronunciation (in whatever variety is deemed standard) is identical to their Scottish English pronunciation, and for all other words to spell them in a way that is supposed to more immediately reflect their Scots pronunciation(s). This convention struck me as unusual because it bases the orthography of a language on the orthography of another and the pronunciation of a particular variety of that other language.
That may have been on the defunct Wirhoose website of the late 20 century, whose archived remains may be of interest to LH readers. It had, for example, a comparison of what Shetlandic would look like if written using an orthography modeled on various existing orthographies including Scots and Faroese or Icelandic. Good times.
The Scots word for ‘house’ (which also means ‘apartment/flat with a separate street entrance’) is a good example. How should it be spelled? House is correct, but strongly suggests the English pronunciation. Hoos(e) gets the right pronunciation when read as English, but oo is not often used in Scots writing except for such concessions. Hous is also good Scots but will baffle anglophones.
Also of interest to LH readers may be the translations here between three Scottish poets (using standard English or Shetlandic) and three Russian poets. Example: this translation/adaptation by Christine De Luca of the original, following, by Lev Oborin. (The book I’ve linked to has a glossary.)
He waeled fae da list o haandicrafts
da vynd o makkin gless laek jewels
braethed inta hit, bluesteyned hit, a deep draacht
Luckit bi glim an glink o blueness
atween skelfs an haemly sag o sofas
bookcases an coorteens, faemily sagas
da proil, aa but sparely
vimmered, jöst barely,
but da soleed air tippit da scales, gaffin.
выбрал из предложенных ремесел
мастерство фигурного стекла
выдувал стекло и купоросил
синева мерцала и влекла
между полок и фамильных кресел
этажерок занавесей
клад
чуть звучал
но плотный воздух перевесил
It may, but where is it? Google only confirms it’s defunct, and archive.org draws a blank.
Closely analogous to what I’ve read about ideas how to spell Low German. One convention, again involving poets, is to stick graphically as close to Standard German as possible in order to deviate as little as possible from what the potential readers are used to. For example, where ik would suffice (as it does in Dutch), write ick to make it look more like ich.
“The language and spelling of this English version (generally the Eycliff)” – I suppose Eycliff to be a typo for Wycliff(e).
Hous is also good Scots but will baffle anglophones.
Oh, I don’t think anglophones are quite so easily baffled — certainly not anglophones who read Scots poetry.
I suppose Eycliff to be a typo for Wycliff(e)
Ah, you may well be right; I vaguely thought it must be a Scots form as I was reading the foreword, but on reflection that doesn’t really make sense. But Google Books tells us there was once such a place name (“usque Partem australem de Eycliff”; “In another grant from the same, of land at Eycliff”).
You can download Wirhoose as a zipfile here.
The W is probably just turning sideways. Happens to most writing systems every couple thousands of years.
Richard Milhous Nixon wasn’t even noticed by the makers of the Simpsons.
This convention struck me as unusual because it bases the orthography of a language on the orthography of another and the pronunciation of a particular variety of that other language.
Cf. also the two very different ways of spelling Occitan.
DM, JC, RC: Thank you.
By the way, I think what I wrote in my first comment doesn’t make sense. Quite a few orthographies must be modeled on another language’s. I should have expressed more clearly that what I find remarkable is how, if my understanding of whatever I had read was correct, the spelling of each word depends on whether it is pronounced the same in a particular other language (which I assume most speakers of Scots also have a very good command of), so that you end up with sentences where half of the words are spelled in a consistent, contrived way (which way exactly that is is another matter), and the other half is (are?) spelled whichever derived way the cognate happens to be spelled in English. It’s not one orthography but a salad of two, mixed word by word in a seemingly random way – which of the two gets used doesn’t depend on etymology, for example – whose underlying rule must be far from obvious to people until you tell them. And yet it works, I guess. Quite the Japanese.
I can make no sense of that, because I use a completely different vowel for Bert and certain (and hert, although it does tend to slip towards hairt) than I do for hurt and curtain, and can’t really see why you would use a different one.
Well, if you’re in Edinburgh I presume you’re Scottish; he’s talking about non-Scots who see -ert and pronounce it in the way they’re used to.
The official download site for the Wirhoose zip file is now at Scots Thriep.
‘A thro the night i heard her hert
Gang soundin wi my ain’
If it’s that poem against SGS, and since our fuzzy search capabilities are fast approaching those of the Middle Irish, to hell with standardized spelling.
Maybe the above was hasty and insensitive as I am not a Scots speaker (just read the comments under the Scots Threip post), but I do think that if you’re late you can be new and a language with a spelling for each dialect could be revolutionary and a model to others, and speakers writing their dialects as they are is part of keeping alive the most universally collaborative works of oral lit on the grandest scale (dialects) [/soapbox]. Another matter is that I shouldn’t be so eager to volunteer other people’s language for complete fragmentation into poetry. That feels symptomatic.
to hell with standardized spelling
But once we’ve searched for something, we still have to read it.
Standardized spelling is for the benefit of readers throughout the Anglosphere. As an American, I do not want to have to decipher an article about a feft from the House of Lawds, when those spellings have nothing to do with the sounds I use to pronounce those words. Granted, the gh in might has nothing to do with my pronunciation either, but at least it’s a fixed convention; I don’t have to cope with “I might do this but you mite do that”.
anya: a spelling for each dialect could be revolutionary and a model to others
As a subscribing member of the Union for Linguistic Fragmentation I ought to agree. Hovever, the gripe in Scots Threip isn’t about dialect representation as such, but about representing dialects through the reading conventions of a (however slightly) foreign Standard language — and thus enforcing that standard as the reference point also of the spoken dialects — and about how a (re)established native standard might change the sociolinguistic situation.
Not just a standard, but specifically a diaphonematic standard, such that you can (without too much agony) learn to map the standard spelling to your own pronunciation. Thus, instead of writing “what” for Central Scots and “fat” for Northeast Scots, all Scots are to write “what”, and the latter group will be taught to read that as [fat] rather than [ʍat] or [wat]. In order to create such a standard, one must align it to the most conservative possible dialect: in other words, make all the distinctions that are made anywhere. Older Scots spellings serve this purpose pretty well.
The same is true of English, of course, except that it continues to make distinctions that no living dialect makes any more (my stock example is vein and vain, which are universally homophonous), and it has far more irregularities than necessary. That’s why the spelling reform that I support for English, Axel Wijk’s, cleans up irregularities by making them diaphonematic when possible, as in knoe for know, as it is not a homophone of now. When that is not possible, Wijk goes with the greatest number: any, which has to be fixed because it appears to rhyme with zany, becomes enny, not anny (Irish) or inny (U.S. Southern).
The exception to Tait’s scheme is Insular Scots, which not only makes far more distinctions but makes them in a way that can’t readily be mapped. So when Tait is writing Shaetlan, he does write “fat”, because that fits the pronunciation.
Indeed. Ideally, schools would first teach according to the local reading rules and only later teach the reading rules of the regional, or national, or supra-national, standard, if and where such do exist.
But diaphonemic spelling may get to a point where the number of digraphs and mute letters and conditional pronunciations is too big and the system just won’t work anymore. Or where the local and standard reading rules are too different for learners and teachers alike. There’s a golden middle where one allows a degree of (ideally systematic) variation between more shallow regional diaphonemic systems.
I should have mentioned I was only going on about Smith’s conundrum, and hadn’t meant to comment on the Threip articles at all, just the discussion here, which shamed me for drafting a linguistic Utopia for others.
Poetry as a testing ground and all; but as for language in the political sense, I only touched that because I forgot it was there and not for mythologizing.
I don’t see you being shamed, or shaming yourself. Or presenting a Utopia for others, for that matter. Not that there’s anything wrong with that.
Yeah, I don’t see any shame either. We all get to orate from this soapbox, and you did so quite eloquently, as far as I’m concerned.
Indeed. Ideally, schools would first teach according to the local reading rules and only later teach the reading rules of the regional, or national, or supra-national, standard, if and where such do exist.
Indeed. For English there is no such national or supranational standard (with the very limited exception of RP, and less and less so), so learning a single set of reading rules is enough.
But diaphonemic spelling may get to a point where the number of digraphs and mute letters and conditional pronunciations is too big and the system just won’t work anymore.
Again, indeed. Fortunately, English has not undergone any unconditioned splits that I know of, only mergers, since its spelling was last reformed.
The FOOT-STRUT split is rather irregular; there are conditions, but they’re not very strict. (Put is a regular exception, but and butter are exceptions from the exception…)
Does anyone still pronounce the k? If graphic distinction from the homophone no is the point here, the new -e would already do that.
I don’t know about the k, but the toe–tow distinction is still maintained by some speakers in East Anglia and South Wales.
“English has not undergone any unconditioned splits that I know of”
Apparently the CURE set once had a single vowel, but depending on the dialect, different words have different vowels and in mine sometimes vowel, glide, vowel.
A half-sentence from the book I’m currently editing: “The ‘low’ status of Colloquial Arabic can be ascribed to the fact that it is neither codified nor standardized.”
The FOOT-STRUT split is rather irregular; there are conditions, but they’re not very strict.
True. I don’t remember offhand what Wijk says about this; my suspicion is that he leaves the spelling alone.
the toe–tow distinction is still maintained by some speakers in East Anglia and South Wales.
It’s recessive, though, at least in East Anglia: within another generation it will probably be gone. I don’t know about the Walian situation; WP says it is maintained.
Apparently the CURE set once had a single vowel
Yes, that is an unconditioned split across accents. Few words are involved, and Wijk leaves them alone. (My accent is conservative and maintains the /uɚ/ pronunciation except in surely and its clipped form sure; original sure is conservative.)
Anyone knows what happened to my comment? The system says it’s duplicate when I’m trying to post it again, but it’s not showing up anywhere at all…
EDIT: might as well add the original comment…
Again, indeed. Fortunately, English has not undergone any unconditioned splits that I know of, only mergers, since its spelling was last reformed.
Some dialects did (the “bad-lad split” in Australian, and IIRC some others elsewhere).
IIRC, one of the Bulgarian orthography proposals was a (vaguely) diaphonemic system. It supposedly had sixteen vowel letters (some of which had to be invented on the spot). English would probably need even more (not sure exactly how many, but there are 24 lexical sets, so presumably somewhere around that).
I think I’ve read somewhere that Irish spelling is also trying to be diaphonemic, which with the language’s quite divergent dialects meant that the rules for each major dialect are well-defined but crazy. This is presumably what is meant by “just won’t work anymore” (though it does appear to still be used).
Some dialects did (the “bad-lad split” in Australian, and IIRC some others elsewhere).
That one’s not so much unconditioned as partly conditioned, and that by an eyebrow-raising hodgepodge of factors. For example “bad” in particular seems to have a long vowel because of its meaning—it’s one of the Four Emotional Horsemen, the others being “mad,” “glad,” and “sad.” That said, my siblings and I were always amused by mum’s Tasmanian short-a pronunciation of “sad” (but not “bad”, “mad”, or “glad”).
Huh. What is it phonetically?
I think I’ve read somewhere that Irish spelling is also trying to be diaphonemic, which with the language’s quite divergent dialects meant that the rules for each major dialect are well-defined but crazy. This is presumably what is meant by “just won’t work anymore” (though it does appear to still be used).
Yes, I did have Irish in mind. I’ve also seen claims that Tibetan is diaphonemic with even more crazy rules. Which goes to show that the acceptable level of impracticality is a socio-political question. For a nation under pressure of extinction, cultural fragmentation may appear as a bigger threat than incomplete literacy. Or maybe widespread literacy in the national language is not in the interest of the increasingly exclusive bilingual literary class.
Does anyone still pronounce the k [in know]?
Only in Shetland traditional dialect, which is outside the scope of ‘English’ for spelling purposes. However, Wijk leaves silent consonants alone as long as their pronunciation is uniform; similarly, the f/ph written contrast survives, as ph as /f/ is invariable. Wijk English is meant to be easy to read, not necessarily easy to spell (though easier than English traditional orthography).
the “bad-lad split” in Australian
That exists in some Southern English varieties, and in some American cities, but with lengthening replaced by tensing (mere phonetic detail,
intended to give artistic…) Where was I? Oh yes: lad is /læd/ but bad is /beəd/, with slight differences in realization.However, the split has very low functional load, at least in the U.S.. For me when I’m speaking entirely with my native accent, I tense only slightly (more [ɛə] than [eə]) and only before nasals, and not all of those: function words with simple codas like can are exempt. After 35 years in NYC, though, I’ve picked up a few tensed lexical items. Even when the split is fully phonemic, as in Philadelphia, you have to push it to find minimal pairs: the surname Manning is lax, the participle manning is tense; Mass is lax because it is short for Massachusetts, also lax, whereas mass is tense; stressed have is lax, halve is tense.
SImilarly, Wijk wants to have us write dh for the voiced interdental fricative, which I think is just crazy: the distribution of þ and ð is fairly predictable and minimal pairs are hardly to be found. Granted, it would benefit foreign learners to make the distinction clearer, but hardly natives.
there are 24 lexical sets
Only on a simplified RP vs. “General American” (a notional rather than an actual accent nowadays) view of things. The complete Wells-Mills-Rosta-Cowan set is KIT, DRESS, TRAP, BAD, LOT, STRUT, FOOT, BATH, DANCE, CLOTH, NURSE, TERM, DIRT, FLEECE, BEAM, FACE, TRAIL, FREIGHT, PALM, THOUGHT, GOAT, SNOW, GOOSE, THREW, PRICE, CHOICE, MOUTH, NEAR, SQUARE, START, NORTH, FORCE, CURE. Obviously a separate written form for all of these is a non-starter. In particular, BATH and CLOTH are always equal to either TRAP or PALM and LOT or THOUGHT respectively, and in almost all of North America LOT = PALM (even where LOT ≠ THOUGHT), thus mixing historic short a and short o. In my opinion this mess should be left alone.
the Four Emotional Horsemen
Only three in the U.S., where sad is consistently lax.
What is it phonetically?
Historically toe was [toː] and tow was [toʊ], as the spelling suggests, and this pronunciation survives in the Rhondda Valley. In most of English, the two sounds merged to the latter, which then became [təʊ] in England, but not in America. In Norfolk today, toe is [tʊu] or [tɤʊ], and tow is [tɐʉ] under the influence of Estuary English, which has pushed the fronting of [oʊ] one step further along.
If only people wouldn’t write breath almost every time they mean breathe. One silent letter distinguishing the preceding consonant and the vowel before it, and people don’t even notice amid the general chaos.
@John Cowan: What’s the reason “DANCE” is included as a separate set? Australian English? To me, it seems more like a case of the BATH and TRAP sets containing different words in different accents (the same applies to CLOTH and LOT, actually). I agree with you that the current orthography for these sounds (BAD-TRAP-BATH-DANCE-LOT-CLOTH) is already mostly optimal, although I’m biased because I’m an American English speaker with the cot-caught merger and no bad-lad split, so any of the splits would be a pain for me to have to transcribe (and I don’t have to memorize “BATH” vs. “START” spellings, although I do have to memorize THOUGHT vs. LOT/CLOTH vs. PALM/FATHER/BRA spellings).
One area where I think it could maybe be improved is the spelling of certain words that historically had /ɔː/, but currently have shortened /ɒ/ for most speakers of British English. Some of these seem like they could be respelled with “o”, which is a more regular representation of /ɒ/ for British English, and which is an OK representation of /ɔ/ in American English due to the existence of the CLOTH set. E.g. “sausage” > “sossage” and “caustic” > “costic” (with the vowel of “moss” and “acrostic”), “laurel” > “lorel” or “lorrel” (with the vowel of “forest”/”sorrel”), “cauliflower” > “coliflower” or “colliflower” (with the vowel of “alcoholic” for all British, and at least some American English speakers). Wijk seems to mention this class of words.
(It doesn’t work for words that historically had /ɔːlt/ and currently have BrEng /ɒlt/: re-spelling “salt” and “fault” as “solt” and “folt” makes it look like they should be pronounced with GOAT. But there are no words with original /ɒlt/ that these can be confused with, while there are words with original /ɒs/, /ɒlV/ and /ɒrV/.)
… Wouldn’t that create confusion between halt and holt? or cult and colt? malt and molt?
That’s what I was trying to say: respelling “au” as “o” doesn’t work well for words pronounced with /ɒlt/.
What’s the reason “DANCE” is included as a separate set?
To discriminate between the AusE varieties for which -anC words like dance are treated differently from -aC- words like class and those where they are not. The half-Australianness of And Rosta, my cow orker on the post-Wellsian lexical sets, probably had something to do with it.
Just watched a video I thought Languagehattics might enjoy: Christine de Luca speaking and reading some of her poems in Shetlandic. I’m posting here as it appears to be the location of the only other mention of her on Languagehat. I won’t put the url in case it gets flagged as comment spam, but if you put Christine de Luca and Shetlandic and WikiTongues into your favorite search engine, you’ll find it.
Quite wonderful to start out at > 95% comprehension, and feel it slowly drift down to 75% when she gets to her last poem, “Discontinuity”.
I don’t think you should have any trouble posting a single link.
The Case of the Missing Link. That was solved quickly !
I don’t think you should have any trouble posting a single link.
I’ve had plenty of trouble, which is why I adopt Ryan’s tactic of giving search terms.
I think what happens if I put a single link is: it’s accepted at first; but then I then go back and edit the post (because I always stuff something up); then the edit gets flagged as spam. I think the dumb engine thinks I’m posting two links: the original and the edit, even though they’re the same.
I think what happens if I put a single link is: it’s accepted at first; but then I then go back and edit the post (because I always stuff something up); then the edit gets flagged as spam.
This was indeed the case until relatively recently, but should have been fixed by now. Did you still get that problem in the last few weeks?
(I think it’s more that the dumb engine thought that you were editing a link in, even if you weren’t; the same thing was happening when a post without a link was edited to add one.)
That hasn’t happened to me since our gracious host increased the link allowance from one to, what is it, five?, several months ago.
Before that, editing a comment with a link in it always threw it into moderation or worse retroactively. Since then, not once.
For what it’s worth,
https://youtu.be/m0EwquC6wBU
Thanks!
I finally got around to watching that Shetlandic lady, and it’s fascinating how she starts out in slow and careful Standard English and then over a minute or so segues into dialect (and my ESL comprehension plummets). But the Norse influence is obvious. “As I [w]rat mer and mer”.
(I call it a dialect, but I’m not going to commit to what language it’s a dialect of).
Yes, that’s a wonderful video (and of course I approve heartily of her fierce defense of “dialect” and individual ways of speaking). Here’s the poem she ends with, “Discontinuity,” and when I was googling the first one, “Spelling It Out,” I found this essay, which includes that and a couple of others; I very much liked this bit:
.
Re calaphine, there is an Irish word coilbhín meaning “stem or stalk” (of plant).