Thursday 7 April 2011

Avoch aye

Here is an interesting graphic created by Jim Scobbie at QMU Edinburgh. It shows ‘a typical Scottish vowel system, single tokens from one speaker’, as revealed by ultrasound tongue imaging.
We know that the vowel system of Scottish English is rather different from the vowel systems of all other kinds of English (except northern Irish, which has historical links with it). Most striking is the absence of the FOOT-GOOSE opposition that is found in all other varieties of English. In Scottish English, good rhymes with mood, look is a homophone of Luke, and put rhymes with shoot. All have the same close vowel, phonetically somewhere in the range u - ʉ - y.

(For some reason the Co-op decided to use a Scottish actor to utter the payoff line good with food at the end of each of their current series of television ads. For example, right at the end here.)

There is also typically no opposition between LOT and THOUGHT or between TRAP and PALM. So cot and caught are likely to be homophones kɔt (as in many other kinds of English, though not in England), and so are Sam and psalm sam.

This gives the vowel system i ɪ e ɛ a ʌ u o ɔ shown in Scobbie’s plot. It is striking that it shows ɪ as lower than ɛ, with u just as front as i. (NB: the lips are to the right, differently from our usual vowel charts.)

There are no intrinsically long vowels in the Scottish system, although some vowels undergo lengthening when morpheme-final or when followed by v, ð, z, r (the ‘Scottish Vowel Length Rule’ or Aitken’s Law). In particular, vowels are short before d, which means that the vowel in Scottish need sounds shorter than that in English need, though the quality is still tense i. If we take the verb to knee, however, its past tense kneed has a long vowel because of the morpheme boundary. Thus we have a minimal pair nid – niːd. Other comparable pairs are brood – brewed and tide –tied. Try side and sighed on your Scottish friends: they will typically be amazed that non-Scots do not distinguish between them.

Scottish English varies from place to place and class to class, and Scots (language/dialect) even more. Here’s a clip of fishermen in Cromarty talking about the loss of their special vocabulary. (Thanks, Amy Stoller, for this.)

This clip is also notable for the placename Avoch. As you can hear, the speaker calls it ɔx. I didn’t put the name of this tiny village in LPD, though EPD has it, as ɔːx, ɔːk. Wikipedia transcribes the name ɒx, and has a sound clip. Since there is no LOT-THOUGHT opposition in Scottish English, it is pointless to argue which of the two is correct.
_ _ _

I shall be busy at a conference over the next few days. Next posting: 12 April.

95 comments:

  1. Wow... u as front as i. Does that mean that u is actually [e] and e is actually [ɪ] or [ʏ]? It's fascinating how ‘all-inclusive’ English is and how it considers Scottish pronunciation of English as an accent of the language. In comparison, no Russian scholar would ever include the Tajik pronunciation of Russian as one of the ‘official’ accents of Russian. I think.

    ReplyDelete
  2. Actually the tide-tied opposition is not just the Scottish Vowel Length Rule; these are completely different phonemes, even without the morpheme boundary. I would transcribe tide as tɐɪd, and tied as taed, where ɐ is the Scottish KIT vowel, which is definitely not pronounced ɪ, even though it's usually transcribed that way.

    Lie as in tell a lie is lɐɪ, lie as in lie down is lae.

    Similarly eye and i (the letter) are ɐɪ, while aye and I (pronoun) are ae...although in Edinburgh I've also heard eyes as aez.

    I'd call this the FLY-TRY split: flɐɪ, trae. So tide, (tell a) lie, eye and (the letter) i all have the FLY vowel, while tied, lie (down), aye and I have TRY.

    There are other minimal pairs too but I've noticed a lot of variation.

    ReplyDelete
  3. Pete, I think you’re one of those cases which a few weeks ago one drama school speech instructor described. Namely, how difficult it is today to teach people to say instead of ɐɪ or ɑɪ.

    ReplyDelete
  4. @ Pete: I remember a distinction like this being written into a book of Scots and Gaelic words, which used respelling rather than IPA. How do you pronounce "ceilidh"? I have always heard it (in England) with the FACE vowel in the first syllable, but I recall that this book listed either FLY or TRY. My limited knowledge of Scots and Gaelic prevented me from understanding the difference.

    When I asked this to a native speaker of Gaelic in a Kingussie hotel, he replied by simply saying what a pretentious word "ceilidh" is.

    ReplyDelete
  5. Since there is no LOT-THOUGHT opposition in Scottish English, it is pointless to argue which of the two is correct.
    Only if no English speaker from (say) England ever mentions that village. (Which doesn't sound *that* unlikely to me, given its population, but the transcription Wikipedia uses between slashes are supposed to be diaphonemic.)

    ReplyDelete
  6. @Mary A - I'm not sure what you mean. I'm one of those people...how difficult it is to teach people to say ???

    I'm quite capable of saying when I want to - I can speak RP pretty fluently. But in my native accent this vowel comes in two varieties, as described above.

    @Ed - I pronounce ceili(dh) with FACE and then happY: keːlɪ.

    This is the only way I've ever heard it pronounced in the Scottish Lowlands (where admittedly Gaelic isn't widely known). It's also how it's pronounced in Northern Ireland, where Irish Gaelic is widely known, although it's a bit different from the Scottish variety. It may well be pronounced very differently in the Scottish Highlands...perhaps daːns!

    ReplyDelete
  7. As a side note, how should the Scottish vowels be transcribed? In particular the usual transcriptions ɪ and ʌ for the KIT and CUT vowels seem very inappropriate.

    For KIT I'd suggest ɐ or maybe ʌ. Not sure about CUT though...ɤ? Or ɞ?

    Any suggestions?

    ReplyDelete
  8. "they will typically be amazed that non-Scots do not distinguish between them." - I do

    ReplyDelete
  9. Pete,
    « I've also heard 'eyes' as aez.»
    That's what you'd expect, but keep the flag flying for 'yeen'.

    Only a few days ago, John confirmed my understanding of a comment by Michael Everson as follows:

    «Yes, and as I expect you know Welsh has two (or in some varieties more than two) diphthongs of this type, ei~əi and ai. So we have the nice minimal pair tei = English tie vs. tai 'houses', pl. of tŷ.»
    http://phonetic-blog.blogspot.com/2011/04/un-pet-de-fromage.html?showComment=1301666536495#c8839531224037554354

    And I raised the Ulster correlate of the question you are now raising:

    «Yes, but I don't know if any form of Welsh English has the Londonderry opposition between ei and aɪ eye~I, with no morphophonological reason for it like teid~taed tide~tied.»
    http://phonetic-blog.blogspot.com/2011/04/un-pet-de-fromage.html?showComment=1301667194700#c2282376233742879119

    And way back I made a similar point about it to the one you are now making:

    «Daniel Jones gave the examples 'tide' and 'tied' for Sc., proposing to distinguish them as [təid] and [taed], and although there are other phonological reasons for that opposition in Sc, the phenomenon is too widespread to have phonological oppositions to fall back on in most cases.»
    http://phonetic-blog.blogspot.com/2009/10/wholly-holy.html?showComment=1257531292140#c1350034733464279822 (excerpt)

    I was just mentioning DJ's examples from memory, and I haven’t checked, but I think I got 'tide' wrong. I think he had [tʌid] for that. Anyway he obviously thought they were different phonemes, whatever he thought phonemes were.

    But although I was thereby saying that like you and DJ I think they are different phonemes, and am now saying that the Ulster opposition is parallel, but the realizations are a bit different (ei~aɪ for eye~I), what I meant by "the phenomenon is too widespread to have phonological oppositions to fall back on in most cases" was that need~kneed brood~brewed etc have to be entered into the equation. You need something like ɐɪd~aed for eyed~I'd to demonstrate that the opposition between those different phonemes is not neutralized in the analogous contexts. Your Edinburgh example aez doesn't do that.

    ReplyDelete
  10. we have a minimal pair nid – niːd

    Why is is that such minimal pairs are not usually taken to prove the existence of separate phonemes /i/, /i:/ in Scottish English, whereas near-minimal pairs such as "clearing-keyring" are taken to establish the existence of separate phonemes for FLEECE and NEAR in RP and similar accents?

    If morpheme boundaries are allowed in our phonemic representation, then NEAR, SQUARE and CURE could be /niːr/, /skweɪr/ /kjuːr/, with synchronic rules of Breaking, Laxing and R-dropping leading to surface realizations as [nɪə], [skwɛə], [kjʊə]. "Keyring" and "clearing" would be distinguished as underlying /kiː#rɪŋ/ vs. /kliːrɪŋ/

    ReplyDelete
  11. Mary A.: The Tajik accent, no, but the Ukrainian accent? The relationship of English to Scots is very similar to that of Russian to Ukrainian, complete with "It's not really a different language, they just talk funny" attitudes among the first group.

    Pete:

    The tied-tide distinction is usually considered part of the SVLR, since it is predictable from the following consonant plus morpheme juncture. Close variants of it are widespread in North American English under the name of Canadian Raising. Particularly interesting is the case of writer with a raised vowel and rider with a lowered one, which shows that the underlying distinction of voicing in the following consonant, which is neutralized to /ɾ/ in production, is still present at an underlying level. The specifically Canadian version of CR extends the same principle to the /aʊ/ diphthong; in both cases it is thought to be a partial survival of an earlier stage of the Great Vowel Shift when Middle English /iː/, /uː/ had become central rising but not yet low rising diphthongs.

    Canadians are widespread in the U.S. media, and usually alter their specifically Canadian lexis: supposedly the former anchor of ABC News, Peter Jennings, had lieutenant spelled "lootenant" on his teleprompter to remind him to pronounce the word "properly". Their phonology, however, remains distinctive if subtle, and I find the game of "Spot the Hidden Canadian" quite enjoyable. Here are the rules, for any American who cares to try it:

    1) If your suspect says aboat for about, congratulations! You've spotted a Hidden Canadian.

    2) If your suspect pronounces rider [morpheme boundary] and spider [no boundary] differently, you may have spotted a Hidden Canadian, but it may just be a Western Pennsylvanian [a region with CR in /aɪ/ only] after all.

    3) If your suspect says "eh?" at the end of every other sentence, you've spotted a Hidden Canadian who's mocking you.

    ReplyDelete
  12. mallamb said
    ...the Ulster opposition is parallel, but the realizations are a bit different (ei~aɪ for eye~I)...

    I've seen those Ulster phonemes narrowly transcribed in several interesting ways. I've seen eɪ(or ɛɪ)~ɑe, əi(ː)~a(ː)e and əi~ɑe. I get the impression that there are a wide variety of pronunciations of these phonemes in Ulster English. Either that or different people hear them differently.

    Also, while we're on the topic of Ulster English, which happens to be related to today's topic of Scottish English (as John pointed out), do you know if Belfast has this opposition between eye and I?

    ReplyDelete
  13. I don't think Belfast has the FLY-TRY split, no. Belfast speakers pronounce both as ae.

    So come on people - how do you transcribe the Scottish/Northern Irish CUT vowel?

    ReplyDelete
  14. If I may beat my own drum a moment, I see that Scobie's profiles show what I have reported many times from from X-rayed vowel profiles from a number of languages, that high ɪ is lower than e, and low a is higher than ɔ.

    ReplyDelete
  15. Tide-tied is not predictable from SVLR and morpheme boundaries. A better example is died dɐɪd versus dyed daed.

    Clearly these are two completely separate phonemes.

    ReplyDelete
  16. I've noticed the very open quality of /ɪ/ that Professor Wells notes; does anyone know if KIT and DRESS are merged for any Scottish English speakers?

    ReplyDelete
  17. @ Pete: I've found the book I was referring to earlier: "The Pocket Guide to Scottish Words" by Iseabail Macloud. I had remembered wrongly as regards the word "ceilidh": it indicates the FACE vowel. However, I think that the FLY-TRY split may be indicated on page 9, which prescribes separate vowels for "bite" and "by".

    The same page, whilst discussing the -ch sound, seems to think that Germans use the same sounds at the end of "Bach" and "ich", so this book may not be reliable for phonetics. "Bach" is the appropriate one to indicate the Scots sound.

    Scots may be far away from other forms of English, but still not as far away as Swiss German is from other forms of German. I've seen Swiss German speakers be given subtitles on television in Germany.

    ReplyDelete
  18. "For some reason the Co-op decided to use a Scottish actor to utter the payoff line good with food." Surely for the obvious reason that the slogan only works in a Scottish accent.

    The irony is that in other Co-op adverts the actor in question sometimes does make a distinction between the vowels of "good" and "food", like many, many Scots who talk less "broad" than others. If memory serves, it's more like "[gʏd] with [fʉ:d]". For what it's worth, based on impressions over many years of living in Scotland, it seems to me that the lack of /ʊ~u(:)/ distinction is far from universal among Scots, and much less widespread than the lack of /ɒ~ɔ(:)/.

    ReplyDelete
  19. BTW Ukrainian is a different language from Russian, maybe those who said 'it's not a different language they only speak funny' heard Ukrainians speaking Russian with an Ukrainian accent. I have seen cases of Germans hearing Swiss folks speak Standard German with a Swiss accent and saying 'O but Swiss German is not so difficult as I used to believe'.

    In this context it would interest me whether there is a continuous transition between say real Scottish (Lallans) and English with a Scottish accent or not. In case of both Ukrainian and Russian and Swiss German and Standard German there is no continuous transition (unless, of course, someone intentionally or lacking competence speaks a mixture of both languages, but that would be a unique case).

    ReplyDelete
  20. Lazar: No, KIT and DRESS are definitely not merged in Scots.

    KIT is very lax and open as you say. I'm not sure how to transcribe it but something like ɐ comes close. It's also the first component in the FLY vowel ɐɪ.

    DRESS is then very close: e (rather than the ɛ found in RP), with a long allophone conditioned by the SVLR. So there's no merger.

    Funnily enough, in Derry (Northern Ireland but speaking an accent closely related to Scots) the FACE vowel is usually e but has an allophone ɪ. But again there's no merger with KIT because of the open quality of the latter.

    Come on though - is no-one going to have a go at transcribing the Scottish realisation of CUT? It's also the first component in MOUTH: ɤʊ in Scotland and ɤʏ in most of Northern Ireland.

    ReplyDelete
  21. @ Pete: If it were up to me, I'd use ʌ for Scottish STRUT and ɐ for southern English/RP STRUT, but that's never going to catch on.

    The Northern Irish government employs some staff to translate between Ulster Scots and English, considering them different languages. However, Westminster's Parliament doesn't seem to consider Ulster Scots a language, as the rule that nobody can speak in a language other than English in Parliament is not used against Ulster Scots MPs.

    @ Wojciech: I recall something in Peter Trudgill's book that a Scots phonology was forced onto Scottish English, which is why Scots have a FOOT-STRUT split unlike the Northern English. A lot of Scots words have been lost in Scottish English, but then that is also the case in many traditional dialects of England.

    The reasons why dialects are made into languages are political. If Switzerland had been invaded by Germany at some point, they would've not considered their largest language to be "Swiss German" any further.

    ReplyDelete
  22. Ed, they don't consider it a language. Among the official langauges of Switzerland, you'll find "German", not "Swiss German".

    Wojciech, there's a continuum on the dialect level in both cases. The dialects in Switzerland, mostly Alemannic, just enjoy a different social status than in Germany, and even that was much the same as the status of the Alemannic (and other) dialects of Germany until about a hundred years ago.

    ReplyDelete
  23. @ Lipman: Yes, you're right. That was very poor use of quote marks by me. What I was getting at is how political disputes transfer into language, so that we now have Serbian, Croatian, Bosnian and Montenegrin, and the differences amongst these are no greater than amongst the different dialects of German.

    Where are you from, Lipman? What's your native dialect of German?

    ReplyDelete
  24. Too complex and off-topic for a comment, and none.

    ReplyDelete
  25. For aught I know, Swiss German Alemannic dialects are different from other Alemannic dialects, and while there are continuous transitions from Alemannic dialects in say Germany or Austria (Vorarlberg) to Standard German, there is no such transition in Switzerland.

    Re Serbian, Croatian, Montenegrin and Bosnian to my knowledge again they're rather like say English English and American English or Germany's German and Austrian German and (Standard) Swiss German, so they're not really dialects but variants of the same standard language. Slovenian and Macedonian are a different case, these ARE different languages, and so are Slovak and Czech. So while it may be true that 'dialects' are made 'languages' for political reasons, S, C, M and B are not the best example for it.

    Speakers of different variants of German, both standard and dialectal, describe their language as 'deutsch' or German, they usually don't carve out separate languages from it. Yiddish and Luxemburghish may be an exception, but then, they are very peculiar even measured against other German dialects, which are pretty variegated amongst themselves.

    ReplyDelete
  26. @ Wojciech: I'm not sure if you "got" what I was saying. S, C, M and B used to be considered one language: Serbo-Croat. Slovene and Macedonian were recognised as separate languages in Communist Yugoslavia. Political independence often becomes linguistic independence.

    Linguisphere counts the following Alemannic dialects as languages:

    gct – Alemán Coloniero
    gsw – Swiss German
    swg – Swabian German
    wae – Walser German

    ReplyDelete
  27. Alemannic non-continuum: where do you get this?

    SCMB: why? It would even be a better example if they were just variants of the same standard language, but they represent dialects all right.

    Macedonian: disputable, Slovenian and Czech vs Slovak: obvious, but Ed didn't talk about any of these anyway, I think.

    Yiddish: ignoring parts of the vocabulary (and the use of Hebrew/Aramaic words in everyday language is commonly overestimated), the average German will understand much more than of a Highest Alemannic dialect

    Luxembourghish: what's so peculiar about it?

    ReplyDelete
  28. Ad Ed

    I seem to have got(ten) what you were saying but my point was that S, C, B and M are really different (to a degree) forms of the _same_ standard language, like British English and American English, say. This to my knowledge, which may --- needless to say --- be wrong. And again to my knowledge: there are dialects in the SCMB area, yes, but they cut across the political borders and are not subsumable under Serbian etc. as decreed by politicians.

    Ad Lipman

    Macedonian is a Bulgarian variant, some say, I dunno, do you? Slovenian could be lumped together with SCBM but as distinct from these 'four' (or one, as I would say), it's really different. Czech and Slovak are really close, but not in the sense in which British English and American English, or Serbian and Croatian are 'really close'. In the sense, rather, in which Swedish and Norwegian, or Spanish and Portuguese are really close, without ceasing to be different languages.

    Yiddish contains not just Semitic but also various Slavic words, afaik, plus a number of developments unparallelled in other German dialects. I don't deny that it's to some extent intelligible to German speakers, but I sort of agree with those who 'hear' it as a different language Luxembourghish teems with Gallicisms, at least if you go by its written version. "Digniteit" for "dignity" and such... . Not "Wuerde" or any phonetic variant thereof.

    Alemannic non-continuum, where I got it from. From observation. I was saying: for Switzerland only. I have observed Alemannic continuum in Vorarlberg, but not in the canton of St. Gallen, where I stayed for a time. You cross the CH-A border and you hear the difference straightaway. There is, to be true, a kind of transition form between Swiss German and STandard German: when a Swiss German person tries to speak standard German but can't (due to deficient school education), for istance: 'das isch eine niet zu verachtende Verbesserig unsere (or: oisere) wirtschaftleche Situation'. But that's just broken Standard German, which no-one speaks _ungezwungen_ except on special occasions when one is supposed to speak 'pure' Standard German. In Vorarlberg, Suedbaden there are real transition forms, that is right.

    ReplyDelete
  29. On a choir visit to Alsace, a choir member of Swiss German descent heard a lady speaking Alsatian dialect and said it was nearly identical to the Swiss German he knew. He tried to speak to her in dialect, but she insisted in speaking to this foreigner in French!

    ReplyDelete
  30. Ad Steve,

    yes, Alsatian dialects are Allemanic too, and they are all very close to one another (though on the other hand some dialect speakers from Freiburg i. Br., Germany) say they can't understand dialect speakers from Freiburg i. Ue., Switzerland.) I do not know, though, whether in Alsace they ever spoke a transition form from their dialect to Standard German; I'd think if they ever did, they do not do nowadays. In Switzerland, transition forms between a local Alemannic dialect and Standard German do not exist (at least not in the speech of native Swiss persons, maybe in that of German guest-workers, so numerous there today).

    ReplyDelete
  31. Wojciech

    I'm no expert — just an Englishman living in Scotland who takes an interest. I can give you my sense of a relatively recent relatively stable status quo and the ambitions of those who are working to change it.

    Within living memory, children in school were punished for using local vernacular. This did not necessarily destroy the vernaculars but it kept them as disparate and localised L varieties in a diglossia relationship with English as the H language.

    At the same time, there was a healthy Scots literary language. This was not just a minority interest. The literature included the immensely popular and respected vernacular poems of the national poet, Robert Burns. In the Middle Ages and Early Renaissance, the prestige language of Scotland differed from that of England — although, confusingly, the language of Scotland was called 'Inglis'.

    Clearly, 'Inglis', literary Scots and many of the local vernaculars are varieties of a single whole. It's less obvious that the vernaculars of Scotland's cities are varieties of Scots. To an outsider they seem more like dialects of Scottish English with a sprinkling of Scots vocabulary and a hint of Scots grammar. In fact, I don't believe there's diglossia in Scottish cities, but rather incremental code-switiching between a cline of sociolects.

    Others might give you a more expert picture of the status quo which many are trying to change. But I think the goals (some of which may, arguably, have been achieved in part) are reasonably clear:

    • to establish a Scotland-wide spoken standard that can be shared by speakers of different localised vernaculars

    • much wider use of Scots in non-literary prose

    • elevation of spoken Scots to the status of H language used in publish discourse: politics, administration, law, education, religious worship etc.

    Already, Scots is an official language of the Scottish Parliament, but I'm pretty sure there are no Scots-English interpreters employed and not speeches made in Scots. The number of Parliamentary and Scottish Governmental tests published in Scots is not zero but not big.

    Serious writing and public discourse in Scots is out there somewhere, but you need to know where to look for it.

    The 2011 Census (Scottish version) had a question asking us whether we could understand/speak/read/write English, Gaelic and Scots. The replies are intended to inform future Government policy. I wonder what the figures will be for use of Scots, and what the Government will do about them.

    ReplyDelete
  32. @Ed:

    If Switzerland had been invaded by Germany at some point, they would've not considered their largest language to be "Swiss German" any further.

    Switzerland was invaded and occupied by France between 1798 and 1803, but the version of French spoken in Switzerland is still called "French", is it not?

    ReplyDelete
  33. Ad Steve,

    thank you for this extensive explanation. 'It gars me greet' to speak with Burns. There was a poet, I recall, Hugh McDiarmid, who wrote beautiful Marxist poems in Scots. Given the deplorable states of affairs as it is now, I doubt whether Scots (as distinct from 'Scottish English', boiling down to Scottish accent in Standard English) can be resuscitated. It truly 'gars me greet', though I understand that it can't be help, nor, perhaps, ought to be helped given that English is threatened in its world-language status.

    ReplyDelete
  34. vp, I get your point, of course, and you're right that it's all called French, but in fact, the situation of the Gallo-Romance dialects in Switzerland is pretty colourful. (The local Italian dialects aren't Italian either but Lombardic.)

    ReplyDelete
  35. Wojciech

    The standard languages in Serbia, Croatia and Bosnia are being forced apart not just by the will of politicians but by the implementation of educators, publishers, editors and compilers of dictionaries. In these independent states the process is surely irreversible. In Montenegro it may not be too late for politicians to change their minds or for the implementers to withdraw cooperation.

    ReplyDelete
  36. @ Pete 8 April 2011 09:22 (I feel bad that I can't contribute anything to the discussion now going on between Lipman, David, Wojciech, etc., but I'm learning a thing or two by reading it): Is there a possibility of a DRESS-FACE merger in Scottish English then? Because you mention that DRESS is [e] or [eː] depending on environment, but isn't that also what FACE is in this variety? If there is such a merger, I can't think of anywhere else in the English-speaking world where it happens. Thank you.

    ReplyDelete
  37. Swiss French does not sound like the standard (Parisian) French.

    ReplyDelete
  38. @ vp: I didn't know that. My theory is thus refuted.

    ReplyDelete
  39. It should have occurred to me sooner, but here's a rich source of information on Scottish and related accents: sound comparisons.

    It's currently not working properly with the Safari browser. Firefox is recommended.

    @Phil

    No sign of a DRESS-FACE merger in the accents sampled here.

    ReplyDelete
  40. @Ed:

    I did some further research (aka Wikipedia browsing) that may unrefute your theory!

    Of the current French-majority cantons of Switzerland, only Vaud and Fribourg were part of Switzerland in 1798. Vaud was treated as a quasi-colony of Berne at the time, and its inhabitants seem to have treated the invading French as liberators.

    ReplyDelete
  41. Ad David Crosbie

    but do you think those educators etc. can really make different languages out of what essentially was and is the same language? Such is the power of these ...tors? Imagine in the UK they 'implemented' a new spelling in which you must henceforth write 'hensfawth' and 'wuhk ov aht' and the like, and you must never-never say 'checkers' and 'you are welcome' but always only 'draughts' and 'don't mention it' and such... . That would be called the 'Royal British Languij' or some such... . Obviously, it would be a farcical masquerade and no-one would think the British have a _different_ language, different from that of the North Americans.

    ReplyDelete
  42. @Wojciech: I'm afraid you'll find that there isn't a universal linguistic criterion to unambiguously decide what is a "separate language" and what is a "dialect". The only clear cases are where there is no mutual intelligibility whatsoever.* Within a family/grouping/language, there will always be some doubt because there will always be some mutual intelligibility, even if it's only at a level of, say, 3%. Hence Weinreich's definition of "a language": a dialect with an army and a navy. The problem is, what is the cut-off value? 60%? 80%? 95%?

    At the other end of the spectrum, no two people have exactly the same grammar. There will always be small differences -- even if really very small, they're still there. That's why people talk of idiolects. The grouping of idiolects into a "dialect" is essentially arbitrary, and -- to a considerable extent -- done on extralinguistic grounds. Again, at what level of similarity do you decide that two speakers are not from the same dialect? 99%? 95%? 80%?

    (*) BTW, is the mutual intelligibility between, say, English and French really 0%? And if not, what does that mean?

    BTW 2: The labels "American English" and "British English" are not particularly meaningful from a strictly linguistic point of view, in particular in terms of phonetics and phonology. (Unless you specify "Standard" in both cases, which still leaves quite a bit of room for uncertainty.) Glasgow has probably more in common with Maritime Canada than with London, and London has more in common with Sydney, Australia.

    ReplyDelete
  43. @Wojciech: BTW 3, re your last comment. If there was a political will/need, I don't think what you describe would be at all that bizarre. Cf. South Africa (Afrikaans), or SCBM, for that matter.

    ReplyDelete
  44. Wojciech

    but do you think those educators etc. can really make different languages out of what essentially was and is the same language?

    Of course I do. It was done with Hindi and Urdu. These were once subsumed under 'Hindustani', and there are still published teaching materials for 'Hindi-Urdu'. I remember hearing on the radio an account of the efforts under the forgotten Prime Minister Lal Bahdur Shastri (1964-66) to push Hindi further away from Hindustani than the mass of speakers could accept. It got so bad that significant numbers of Indians were listening to Radio Pakistan because the Urdu spoken there was closer to everyday language than the Sanskritised language of Radio India.

    Nobody speaks of 'Hindustani' today, and pretty soon nobody will speak of 'Serbo-Croat'.

    It's not just the spelling. Nor just a different alphabet — although that undoubtedly helps. What language creators wisely focus on is lexis. Where possible, they promote words with cultural resonance in the history of the ethnic and/or religious group that has decided to become a nation-state. Shastri overdid it, but wiser heads decided to slow down. The aim is to differentiate from the preceding language, so if they can't find word-elements from ethnic/religious sources, they can make different use of international word-elements.

    When a language is split, people will continue to speak the same vernaculars as before. But those vernaculars were affected by the official language of the time. And their children's vernaculars will be influenced by the new official national languages. The new languages will be the medium of school education, so their grandchildren will have no experience of the old undivided official language.

    In short, the grandchildren of Serb and Croat speakers of Serbo-Croatian will be speaking quite like their grandparents in intimate or domestic settings with only a few national differences, but their language of writing and public life will be one or other of the well-established official languages, each crafted to be as distinct as possible without falling into the madness of Shastri's Hindi.

    ReplyDelete
  45. @ vp: I have thought of a nearby problem for my theory, which is that you might consider the Anschluss 1936 to have been a German invasion of Austria, yet the Austrians still call their language "German". Going into whether this counts as an "occupation" or not is likely to make enemies for me, so I'll leave it at that :)

    On the subject of educators' trying to change the linguistic makeup, the Republic of Ireland tried to supplant English with Irish as the main language of the nation for many decades, but it didn't work. Educators on their own cannot have power to change language. With B, C, M and S it is a much easier job though, as you're not asking the pupils to speak any differently: just to know what to call their language. I expect that some of them will be more than happy to embrace linguistic independence.

    ReplyDelete
  46. wjarek

    The only clear cases are where there is no mutual intelligibility whatsoever.

    Even this may be unclear. I wish I'd made a note of a paper I once read on language, history and society in parts of West and central Africa. Among the remarkable case histories was a region where two ethnic groups lived together in a situation of massive power imbalance, having evolved bizarre rules of speaking to member of the other group. As I remember it, the inferior group had to gibber almost incoherently, while the superior group could bark orders with complete fluency. I may have got some of the details wrong.

    The result was that one group said 'Yes, we speak the same language' while the other said ''They speak a different language, which we have to learn'

    ReplyDelete
  47. To all my partners-in-discussion: I have to suspend my activities till Monday. Thank you.

    ReplyDelete
  48. ed

    Surely the idea of 'Germany' included Austria right up to the day in 1871 when a political state called 'Germany' was created. Of course, the words used were Deutsch and Deutschland, but the words in other languages (Germans, Allemands, Tedeschi, Немцы) had long been exact translations.


    It would be different if the army of Romania were to march into Moldova. Under the right political conditions, Moldovans might well agree to accept that their language is really Romanian.

    I understand that the only reason Switzerland recognised a fourth official language was a fear that Mussolini would invade on the grounds that Roumansch was really a dialect of Italian.

    Irish didn't work in Ireland for two reasons:

    1. It was too difficult for the English speaking majority to master quickly and easy.

    2. They made a total mess of school-teaching and motivation. Children grudgingly learned the minimum necessary for Civil Service entry.

    With less ambitious goals and much better teaching and motivation, Irish is now doing very well.

    A striking example of forcing a different language on a nation-state is the imposition of Modern Hebrew on a population of Yiddish, Arabic and Ladino speakers.

    ReplyDelete
  49. @David: That would probably be the Nembe and Kalabari -- see How many languages are there in the world by S. Anderson.

    But this was my point exactly -- maybe I wasn't clear enough. In purely linguistic terms, differences between language forms are quite difficult to measure, and if you do measure them successfully, the decision to recognise one speech form as the same or different must be arbitrary, in essence. (Plus, of course, different things happen at different levels of language structure.) But on top of all that, there are extralinguistic factors -- (geo)politics, social factors (including e.g. status and attitude, like here), and so on, and so on. And, often enough, these tend to trump the linguistic factors when deciding whether something is a "language" or a "dialect" or whatever.

    ReplyDelete
  50. John Cowan: "Their phonology, however, remains distinctive if subtle, and I find the game of 'Spot the Hidden Canadian' quite enjoyable."

    Well I'm right over here so come get me! ;o)

    "Particularly interesting is the case of writer with a raised vowel and rider with a lowered one, which shows that the underlying distinction of voicing in the following consonant, which is neutralized to /ɾ/ in production, is still present at an underlying level."

    Yes, I'm one of those that does that. It's neat how historical distinctions can still stay intact but nonetheless be altered or obscured like this.

    ReplyDelete
  51. @David Crosbie:

    Nobody speaks of 'Hindustani' today

    Yet the language Hindustani lives on, most robustly perhaps in the dialogue of Bollywood movies, which is generally fully comprehensible to both "Hindi" and "Urdu" speakers. Now combined with a healthy admixture of English, or course.

    ReplyDelete
  52. @David Crosbie: I'm curious; how would a Scottish English speaker, with DRESS as [e(ː)], maintain the contrast with FACE? It's pretty clear that they do maintain it, I just don't know how.

    ReplyDelete
  53. Wjarek

    Yes it might well have been the Nembe and Kalabari.

    However, the paper described a much starker contrast in perception of intelligibility than accounts of these languages today. It was in an anthology published in the seventies, but the paper itself was, I think, earlier — and the period described earlier still. My impression is that it described societies as they were a hundred years ago, if not before.

    I'd love to track it down. It also described the linguistic interaction of Tutsi and Hutu when the imbalance was feudal rather than political. And there was a pleasing tale of families arguing over the genealogies and land ownership recorded in oral tradition, but politely declining any interest in British Colonial written records of the events and people in question.

    I take your point about the difficulty in establishing objective measurements of linguistic difference. Funnily enough the research project behind sound comparisons is an attempt to do just that: to establish measurable and quantifiable phonetic distances between Scottish accent — and further the distances separating other accents of English, other Germanic languages and earlier reconstructed pronunciations.

    ReplyDelete
  54. Lazar Taxon

    Do go to sound comparisons. (Dont' use the Safari browser, though.)

    There you can see and hear that the various Scottish accents sampled all use variations on ɛ for better and variations on e: for day.

    Eight, naked and name are similar, though not identical to day for what would be a single FACE vowel for me.

    Head, read and ten are likewise similar but not identical to better in what would be a single DRESS vowel sound for me.

    None of the speakers sampled there have anything like e: for any of the DRESS words.

    ReplyDelete
  55. @ Lazar Taxon: If you look at the link to Sound Comparisons that David posted earlier, the DRESS vowel is usually ɛ(:) and the FACE vowel is usually e(:). Incidence in each set may differ from in England though: for example, the traditional Edinburgh accent on that site has the KIT vowel in "red".

    I was interested to see that the site uses ɛ: to represent both FACE and SQUARE in Morley (West Yorkshire). FACE is usually represented as /e:/ here, but is true that our FACE vowel has moved away from [e:]. If you watch the old film "Kes", you'll notice a FACE vowel closer than is now common in Yorkshire. FACE is still slightly closer than SQUARE though.

    ReplyDelete
  56. Lazar Taxon, Ed

    I posted a reply to you yesterday, Lazar, but it disappeared into the ether. Ed expresses it more succinctly than I did.

    What I'd add, though, is that the traditional vowel in red in Edinburgh, and in head in several accents suggests a phonological rather than allophonic difference. In several accents head is not in the DRESS lexical set, and in the accent of at least one traditional Edinburgh speaker, not is red. However, all the Scottish accents sampled have some sort of ɛ in better and ten.

    ReplyDelete
  57. Wjarek

    I posted a reply yesterday, but it disappeared into the ether — twice, I think.

    Yes, it might well have been the Nembe and Kalabari that i read about. But the paper described a more extreme situation than referred to in current descriptions of those languages. The contrast in perception of intelligibility was much, much starker. This was no doubt because this was the paper was already old when included in the anthology (in the seventies, I believe) and described societies as they had been observed earlier still. My impression is that the observations were made a hundred years ago or more.

    I wish I could track it down. The paper also had an interesting account of linguistic relations between Tutsi and Hutu when the imbalance was feudal rather than political. And there was a pleasing account of families disputing genealogy and historical land ownership on the basis of oral tradition, while politely ignoring offers of help from British Colonial administrators who had written records of the people and land holding in question.

    I take your point about the difficulty of establishing objective measures of linguistic difference. Coincidentally, that's exactly what the team behind 'Sound Comparisons' is trying to do. I lack the expertise to follow and judge, but the aim is to measure how different Scottish accents from each other, from other English accents, from other Germanic languages and from reconstructed historical pronunciations.

    ReplyDelete
  58. @David: Yes, the Sound Comparisons project is rather sophisticated -- in fact, probably among the most sophisticated in the field, even though other methods do exist. Googling "dialectometry" gives a good feel of what has been/is being done; in particular, work by John Nerbonne is worth a look. But it's still a bit arbitrary at points, and limited in e.g. only looking at a limited database of isolated words, and only at phonetics and lexical semantics.

    And then, if two varieties show up with a similarity index of 0.93 (as e.g. "Traditional Tyneside" and "Typical Tyneside" do), is that the same dialect/accent? How about, say, 0.56?

    What I find attractive about such approaches is in fact their being excplicitly gradient rather than enforcing binary same/different categorisations...

    ReplyDelete
  59. As far as I know, there is no such thing as a "Moldavian" (or "Moldovan") "language": it is only a variety of Romanian.

    There is an old paper by Carlo Tagliavini about that.

    ReplyDelete
  60. Pete, don't ask me. Trust Prunella

    http://www.telegraph.co.uk/news/1582128/Acting-advice-from-those-who-know-best.html

    or listen to younger actors.

    BTW, since this off-topic is flourishing, may I ask about the equivalence of diacritics. I.e. if I add a lowered and centralized diacritic to ʌ, do I get an ɐ? And is then a lowered and less rounded ɜ equal to ɐ too?

    ReplyDelete
  61. Mickey

    However similar the vernaculars on either side of the border, official Standard Moldavian is a different language because Stalin made it so.

    According to Bernard Comrie in The Languages of the Soviet Union (1981):

    "The main difference between Moldavian and Romanian at present, apart from the different alphabets, lies in the fact that Moldavian uses a number of Russian loans that are not used in Romanian, including stump-compounds of the type rajkóm (in full in Russian rajonny komitet 'regional committee'). The procedure of forming stump-compounds is even transferred to native morphemes, e.g. korsát 'village correspondent' (cf. korespondént 'correspondent' sat 'village'), although the native order with the modifier after the head is retained (cf. Russian sel'kor). Although Moldavian is currently little different from Romanian, it is likely that continued political separation will serve to increase these differences with time."

    Of course, the political situation has changed radically since 1985. Moldavian could be abolished as a language at the stroke of a pen if the governments of Romania and Moldova wished. But it hasn't happened yet.

    ReplyDelete
  62. Sorry! I should have written:

    The political situation has changed radically since 1981.

    That was the publication date of Comrie's book.

    ReplyDelete
  63. Phil Smith

    The speaker represented on the 'Sound Comparison' site as having a 'typical' Belfast accent pronounces eye and I identically. The site transcribes it as [a⸱e].

    In the examples given for 'traditional' Antrim, 'typical' Antrim and Tyrone accents, the two words are not homophones.

    ReplyDelete
  64. David Crosbie

    Caesar non supra grammaticam! ;)

    ReplyDelete
  65. Ad wiarek et alii

    I of course know there are no hard-and-fast criteria for distinguishing 'languages' from 'dialects'; yet the cases in which is it really difficult to say whether a given speech form is a dialect or language are, at least in Europe, not so numerous. All things political (ambitions, correctness..) put aside, we are in most cases quite clear about the status of a speech form in most cases, are we not? I am currently in Sardinia, so Sardinian is perhaps one of those rare cases where there is no such universal or nearly-universal certainty. So seldom is there any need to investigate whether the given community has an army and a navy.

    Yes, I was thinking about Standard British English and Standard American English. I don't think there are many serious persons competent in 'either' who would claim they are different languages. And even if the British were forced to spell 'forrid' (forehead) 'wuhk ov aht' (work of art) and what have you. Or punished for saying 'you are welcome' or 'I insist that he be here on time'. Afaik, Serbian, Croatian and so on are more or less like Standard British English, Standard American English and so on to one another... . Afrikaans and Dutch are not like that at all, Dutch (Netherlands) and Flemish (Belgium) are a better example. Two forms of the same standard. Roumanian and Moldavian or Moldovan are like that, despite Russian loan-words, this is at least what I seem to know.

    Of course after generations of saying 'braces' instead of 'suspenders' and 'wireless' instead of 'radio' you may come to believe you speak a diffent language than the guy who habitually says the other thing.

    ReplyDelete
  66. Mary A.:
    No, diacritics do not make a symbol they're applied to equivalent to another symbol, because they indicate a smaller difference than the distance between two adjacent symbols. So a lowered ɜ is not the same as an ɐ, and it's not even the same as a raised ɐ. They're actually four phases of a gradual quality change: ɜ, lowered ɜ, raised ɐ, ɐ. The change from ʌ to ɐ is similar: ʌ, fronter ʌ, backer ɐ, ɐ.

    ReplyDelete
  67. This comment has been removed by the author.

    ReplyDelete
  68. This comment has been removed by the author.

    ReplyDelete
  69. Wojciech

    There are no institutional rules against writing 'suspenders' in Britain or 'braces' in America. We abstain for the purely practical reason that each word has a more common meaning in one language or the other.

    By contrast, take this anecdote from when I was involved in English language exams for the armed forces of the Czech Republic. When the advanced Level Three exams was set up, it was decided that a measure of English-Czech interpreting skill was desirable, so a little simulation was included in the Oral Paper.

    But the Czech Armed Forces still contain officers from the old Czechoslovakian forces, some of whom speak Slovak. This had not been foreseen when the marking criteria were drawn up. They called for markers to consider the quality of the Czech interpretation. I was in a markers' meeting where some members were seriously considering deducting marks from the Slovaks.

    In the modern world, languages are defined, assessed and enforced by officialdom. There are societies where we can ask linguists to identify languages, but none of them are in Europe or the rest of the developed world.

    No speaker of American English in Britain can be denied employment, justice, political rights, freedom of speech or any other limitations on simple linguistic grounds. The worst that can happen is the editors may change their spellings, or that they may not be invited to speak a second time in public fora.

    Moreover, English speakers the world over lay collective claim to the English language culture. Many of our great writers were Irish, but many of us don't know and wouldn't care. Popular culture is largely American and our soap operas are Australian. English is too big to be the possession of any one political unit.

    English attained this significance before the age of modern nationalism. Indeed, it was before the age of single national languages. English was of secondary importance compared to Latin for many purposes, and then for a time of tertiary importance compared to Latin and French. In the modern world, nation states feel the need to have single national languages. And with the modern definition of nation that means an exclusive.

    Modern national languages are codified the way that the great languages of the past were codified. As I understand it, Portuguese was codified with emphasis on the differences with Spanish. In Former Yugoslavia, the new nations have fewer differences to emphasise. In Bessarabia, the colonial power relied more on a change name (to Moldavia and Moldavian) and a change of alphabet than anything grammatical or lexical.

    The right of the de jure governments of Croatia, Bosnia, Montenegro and the de facto government of the Soviet Republic of Moldavia to codify distinct languages is unassailable. Mutual intelligibility is of no relevance whatsoever.

    This works both ways. By mutual the intelligibility criterion Chinese would not be a language

    As to what in English we call Flemish and Dutch, my understanding is that they are the same language because the political establishments say so.

    ReplyDelete
  70. Wojciech

    There are no institutional rules against writing 'suspenders' in Britain or 'braces' in America. We abstain for the purely practical reason that each word has a more common meaning in one country or the other.

    By contrast, take this anecdote from when I was involved in English language exams for the armed forces of the Czech Republic. When the advanced Level Three exams was set up, it was decided that a measure of English-Czech interpreting skill was desirable, so a little simulation was included in the Oral Paper.

    But the Czech Armed Forces still contain officers from the old Czechoslovakian forces, some of whom speak Slovak. This had not been foreseen when the marking criteria were drawn up. They called for markers to consider the quality of the Czech interpretation. I was in a markers' meeting where some members were seriously considering deducting marks from the Slovaks.

    In the modern world, languages are defined, assessed and enforced by officialdom. There are societies where we can ask linguists to identify languages, but none of them are in Europe or the rest of the developed world.

    No speaker of American English in Britain can be denied employment, justice, political rights, freedom of speech or any other limitations on simple linguistic grounds. The worst that can happen is the editors may change their spellings, or that they may not be invited to speak a second time in public fora.

    Moreover, English speakers the world over lay collective claim to the English language culture. Many of our great writers were Irish, but many of us don't know and wouldn't care. Popular culture is largely American and our soap operas are Australian. English is too big to be the possession of any one political unit.

    English attained this significance before the age of modern nationalism. Indeed, it was before the age of single national languages. English was of secondary importance compared to Latin for many purposes, and then for a time of tertiary importance compared to Latin and French. In the modern world, nation states feel the need to have single national languages. And with the modern definition of nation that means an exclusive.

    Modern national languages are codified the way that the great languages of the past were codified. As I understand it, Portuguese was codified with emphasis on the differences with Spanish. In Former Yugoslavia, the new nations have fewer differences to emphasise. In Bessarabia, the colonial power relied more on a change name (to Moldavia and Moldavian) and a change of alphabet than anything grammatical or lexical.

    The right of the de jure governments of Croatia, Bosnia, Montenegro and the de facto government of the Soviet Republic of Moldavia to codify distinct languages is unassailable. Mutual intelligibility is of no relevance whatsoever.

    This works both ways. By mutual the intelligibility criterion Chinese would not be a language

    As to what in English we call Flemish and Dutch, my understanding is that they are the same language because the political establishments say so.

    ReplyDelete
  71. Ad David Crosbie

    Yes I know there is no law against 'suspenders' or 'I insist that he not say ''suspenders''' in Britain, but even if there had been any, I claim, no-one would (political correctness and stuff aside) seriously think British is a different language from American.

    British (stadard) and American(standard) are two standards of the same language.

    Czech and Slovak are not, though. Neither are Swedish and Norwegian, neither Spanish and Portuguese. They are all very similar, pairwise, but different.

    C, S, B, M, by contrast, are (methinketh); so are Rumanian and Moldovian (ditto), so are Dutch and Flemish (ditto).

    Flemish and Dutch are two variants of the same standard, called Algemeen Beschaafd Nederlands, or General Literare (or: Cultured) Dutch, which provides for local variants in the Netherlands and in Belgium.

    I do not set much store by mutual intelligibility. It is subjective and gradeable and generally unreliable. Chinese is a language and is not at the same time: putonghua is _the_ Chinese language, and then there are various Chinese languageS.

    ReplyDelete
  72. Wojciech

    British and American English are not codified standards. But Norwegian does have two standards — varieties of Dano-Norwegian that became national codified standards overnight in 1917.

    A similar standard to Dano-Norwegian was Serbo-Croatian — a codified standard made official at the stroke of a pen in 1850.

    Just as Norwegians in a newly independent Norway chose to have codified national standards, so do the Croats, Bosnians, and (possibly) Montenegrians in their newly independent states.

    Everyone agrees that Dano-Norwegian outlived its usefulness. I believe we'll eventually agree that the same is true of Serbo-Croatian.

    The dialect with an army and a navy definition is robust — at least for the tongues of modern states that are not landlocked. And it works for many languages in the past. Scotland used to have an army, a navy and a language distinct form the English of London.

    ReplyDelete
  73. Ad David Crosbie

    I did not claim that British and American English are two codified standards. What I claimed was a counterfactual (maybe too risky to claim): Should ever (however unlikely that is) the differences between the two varieties be codified, speakers of both of them would for still a long time agree that they speak the same language. Would they not?

    As for Norwegian, I am not sure which of the many standards you are alluding to, but nynorsk in any case is not a variant of Dano-Norwegian. Dano-Norwegian did not, methinks, outlive its usefulness, for it continues being the standard form, but the name 'Dano-Norwegian' did. It's 'bokmaal' now, a rather misleading name. But the Norwegian situation is extremely tricky, and in flux. Nynorsk and bokmaal are not two variants of the SAME language, they are different (albeit close) languages called the same name. Bokmaal could be considered a variant of Danish, but for one thing it has considerably developed away from its mother tongue, and then it is politically highly inopportune to say a thing like that... . One very important factor is that the Norwegian phonetics is wildly different from the Danish, far more than the American from the RP.

    Serbo-Croatian was a somewhat artificial creation of Vuk Karadzic, but Italian was a somewhat artificial creation of Dante... . The dialects Karadzic subsumed under his standard were not those we know as 'Croatian', 'Serbian' or what not --- their borders cut across today's political borders.

    Against the adage by Weinreich I would adduce two counterexamples: Frisian, and Catalan. Would it be unfair to mention Welsh, or Faroese as well? I also sort of sympathise with Sardinian's aspiration to languagehood: sa limba sarda. No navy and no army, except the Italian ones. Not without reason do they sometimes write on their walls: A foras sus colonisadores itaglianos.

    Re Scotland and Scots, as I wrote somewhere else it really 'gars me greet' (makes me weep) that real Scots is so moribund to-day as it is, or seems to be. It was close enough to English, though, to produce a number or intermediary forms, including various Scots or Scottish accents in what in writing is standard British English. Sunt lacrimae rerum, or, in this case, linguarum.

    ReplyDelete
  74. @Wojciech: Should ever (...) the differences (...) be codified, speakers of both of them would for still a long time agree that they speak the same language. Would they not?

    Well, that was my Afrikaans example. The differences between Afrikaans and Dutch were codified quite recently. But you say yourself that they are "not like that at all" (i.e. unlike SBE vs. SAmE). Do they consider each other speakers of the same language?

    BTW, there has been dialectometric work showing that some Flanders dialects (in the west, of course) are linguistically more divergent from Standard Dutch than Afrikaans is... Hmmm...

    ReplyDelete
  75. Wojciech

    Your latest post is full of unsupported assertions that I can't accept for a moment. I find I can't argue with you any more; it's wearying and gets us nowhere. What you call 'a language' is for me a nonsense.

    And Scots is not moribund.

    ReplyDelete
  76. Ad wiarek

    Re Afrikaans and Dutch I am not sure. May be one of those really tricky cases I said are rather rare in the European linguistic space. One thing that strikes me is that, if you compare the same text, such as the Lord's Prayer or the Universal Declaration of Human Rights, or such, in both of them, you find some differences in literally every word, and quite often you find outright different words. This is not the case with British and American English. Codifying the r-lessness of the British ('wuhk of aht') and the d-quality of the American intervocalic 't' ('love-ledders', 'twendy', 'liddle' --- not just intervocalic) would increase the number of discrepancies dramatically but it would still be not exactly the way it is between Afrikaans and Dutch. OK, you can codify lots of things, such as the characteristic diphthongal pronunciation of the short 'a' in AmE ('come arn, mayn!' or '.. mairn' or some such) or you-all, or youse or lots of others. But then, not all Americans say 'thayt bayd mayn' for 'that bad man' or 'you all' or even 'you guys' for 'you' in plural. It would be a political folly to legislate that all should.

    I don't know those Flemish dialects, but one characteristic feature of all dialects of a language (German, Polish, Italian --- these are I have some idea about) is that despite their sometimes very deviant (from the standard language) phonetics they copy the standard language's idiomatics, phraseology, learned and not-so-learned words, and even grammar. (For instance, when I studied Swiss German, the books said: 'never use the active participle in -nd, it does not exist in Swiss German'. But I heard them quite often in everyday Swiss German speech...) Luxemburghish, for instance, has emancipated itself at least in part from Standard German in this respect, borrows words from French rather than German ... that is why, I feel, its claim to independent languagehood looks credible (though still disputable, I agree). Whether this be or not the case with Afrikaans and Dutch --- I don't know and can't check very quickly, lacking any mentionable competence in either.

    Ad David Crosbie,

    I am not a lingust, I must modestly admit, so I don't know what language 'really' is, i.e. what science claims (with good reasons, supportedly, and so on) it is. And in addition, I am quite insensitive to political and/or nationalist ideas about language, and decrees on languages. I flatter myself to have a 'robust sense of reality' (as sir Bertrand would have said) of various languages, but that's, admittedly, very naive and bound to seem nonsense to a skilled linguist. I wish I had understood exactly why it is so naive, but ... let the matter rest at this.

    If Scots is not moribund --- all the better, this 'gars' me rejoice, by contrast. I by all means like linguistic variety (where there really is some...)

    Re unsupported assertions (such as for instance?..) come on (or 'cuhm ahn' as a future American language writer would write) --- we are not discussing on the pages of a learned Journal (founded 18-something on the incentive of some Imperial Academy or another) so we don't have to build in learned footnotes into our posts, I'd think...

    ReplyDelete
  77. Ad David Crosbie

    Lest I should be misunderstood: I am NOT against anyone's declarations of (political) independence, at least not _a priori_, nor against anyone's wish to call his/her/its/their language anything he/she/it/they please(s) --- if the chaps from Grevelfingen decide to call their language 'Grevelfingisch' rather than just 'German' -- why, let'em.

    Such strivings are by in themselves innocuous and deserve some (tongue-in-cheek) respect: after all, the Grevelfingians may have a eighty-something words for local realities, and pronounce the word 'Boden' ('floor') with a particular inflexion of voice... What I have a certain problem with, rather, is the expectation that such Grevelfingisches should really be, and in all seriousness, considered languages different from those from which they separated (in name, and maybe in the political status of the community of their speakers).

    ReplyDelete
  78. The Ausbausprache – Abstandsprache – Dachsprache distinction which John has directed us to allows clearer discussion of the various 'languages' that popped up in the thread. But the distinctions are perhaps too precise to the 'language' that prompted us — Scots.

    In some ways, Scots fits the description of pseudo-dialectized abstand language. If Scots literature had ended with Robert Burns, that might now be accepted. But there is a modern literature and a modern aspiration to a Scots of public discourse. And in the changed political dispensation, Scots has the status of one of three official 'languages' — albeit more limited in use even than Gaelic. In short, those who use and advocate Scots as an Ausbausprache are becoming more numerous and less marginalised.

    The existence of urban vernaculars that resemble real non-abstand dialects of English complicates the picture. These are sometimes described as 'Scots' — even believed by some to be typical of Scots.

    In Accents of English I understand John to imply a clear distinction between abstand Scots and non-abstand Scottish English. If I understand John correctly, each has its own phonology, but the phonology and phonetics of Scottish English have been affected by Scots.

    ReplyDelete
  79. Ad David Crosbie

    not knowing much about Scots, I must confess I have always believed things are the way the you are describing them, so in this case here is perfect agreement between us. Pseudo-dialectised Abstand, plus non-Abstand English dialects, shading off little by little into just a Scottish accents in standard English. Plus benighted people believing Scots is 'nothing but' those non-abstand dialects and accents... . But after Burns there was one Hugh MacDiarmad, a Scots poet, authors of 'Hymns to Lenin' and other beautiful poems -- was he not (in his poems) Scots?

    Sardinian and certain variants of Frisian are in a somewhat similar situation, even if not exactly the same. Kashubian in Poland may be a case in point too, though opinions differ. In the Netherlands there is a thing called 'stadfrysk' or town Frisian, a Frisian-influenced Dutch dialect, in Germany they call 'Ost-Friesisch' (Eastern Frisian) certain Low German dialects spoken where Eastern Frisian had been spoken centuries ago, and, too, to some extent influenced by it in phonetics and phonology.

    ReplyDelete
  80. Standard Serbo-Croat was never a single standard; rather, it was a fusion of two existing standards, an agreement that Standard Croatian and Standard Serbian (both of which already existed) would be treated as equally acceptable for all purposes. In this way it is like Standard Bokmål-Nynorsk, and like what would be the case if British society decided to accept American English as a written standard with a status equal to British English, or vice versa. It is that agreement which came apart when Yugoslavia did, and it has been followed by the creation of a third standard for Bosnian and a nascent fourth one for Montenegrin.

    All four standard languages are founded on the historic dialect of Eastern Hercegovina, a neo-Shtokavian form which is now the most widely spoken dialect variety of naš jezik 'our language', as it is politely called, in the whole of former Yugoslavia. They differ roughly as follows: Standard Croatian is exclusively Ijekavian, admits influences from the Chakavian and Kajkavian macro-dialects, is relatively hostile to Western loanwords and does not normally respell the ones it accepts, and is written exclusively in the Latin script; Standard Serbian is mixed Ijekavian and Ekavian, has no such influences from the other macro-dialects, is relatively friendly to Western loanwords and respells the ones it accepts to match Serbian pronunciation conventions, and is written with equal acceptability in the Latin and Cyrillic scripts. Standard Bosnian is close to Standard Serbian, has some influences from palaeo-Shtokavian macro-dialect, is exclusively Ijekavian, and uses the Latin script only; Standard Montenegrin will probably wind up using the Latin script only and being exclusively Ekavian. There are of course many differences in vocabulary, on about the same scale as BrE-AmE differences.

    ReplyDelete
  81. David Crosbie:

    Although it's true that there was a hard break in the official spelling of Romanian in Moldova between 1918 (when Romanian Cyrillic went away) and 1924 (when Moldovan Cyrillic was promulgated) due to the political union between Romania and Moldova at that time, six years is a short time in the life of an orthography, and it is plausible to see Moldovan Cyrillic as a reformed spelling in the same sense that Soviet Russian was a reformed spelling of Tsarist Russian (though somewhat more drastically so; more letters were abolished). That being the case, it is plausible to say that the imperial power in Moldova retained rather than imposed a Cyrillic orthography as a way of distinguishing the Romanian of Moldova from the Romanian of Romania.

    There certainly are differences between Moldovan Romanian and Romanian Romanian, but it is the official position of the Moldovan government that the official language of Moldova is Romanian, though Moldovan is its official name in Moldova. This is not equivalent to the situation in former Yugoslavia.

    Lastly, I'd say the point at which Austria was definitively divided from the German lands was either 1805 or 1866/67, but by no means 1871.

    ReplyDelete
  82. ad John Cowan

    since you seem to know so much about 'our language': do Croatian and Serbian, say, differ at least as much as do British and American English? To my knowledge, bokmaal and nynorsk differ to a far larger degree than the said BrE and AmE.

    ReplyDelete
  83. infinitum scribam12 April 2011 at 18:10

    @John Cowan: good summary for Serbo-Croatian

    @Wojciech: how does Norwegian come in here?

    ReplyDelete
  84. Ad infinitum scribam

    Norwegian comes in through the initial fragment of J. Cowan's posting:

    "Standard Serbo-Croat was never a single standard; rather, it was a fusion of two existing standards, an agreement that Standard Croatian and Standard Serbian (both of which already existed) would be treated as equally acceptable for all purposes. In this way it is like Standard Bokmål-Nynorsk, and like what would be the case if British society decided to accept American English as a written standard with a status equal to British English, or vice versa."

    He forgot to explain that Bokmål and Nynorsk are both Norwegian in a sense. The difference between them is, to my knowledge, far larger than that between British English and American English or, again to the best of my knowledge, Serbian and Croatian.

    Example:

    Article 1 of the Universal Declaration of Human Rights, Lord's Prayer, in:

    Nynorsk:

    Alle menneske er fødde til fridom og med same menneskeverd og menneskerettar. Dei har fått fornuft og samvit og skal leve med kvarandre som brør.

    Fader vår, du som er i himmelen!
    Lat namnet ditt helgast.
    Lat riket ditt koma.
    Lat viljen råda på jorda
    så som i himmelen.
    Gje oss i dag vårt daglege brød.
    Forlat oss vår skuld
    som me òg forlet våre skuldmenn.
    Før oss ikkje inn i freisting,
    men frels oss frå det onde.

    Bokmål:

    Alle mennesker er født frie og med samme menneskeverd og menneskerettigheter. De er utstyrt med fornuft og samvittighet og bør handle mot hverandre i brorskapets ånd.

    Fader vår, du som er i himmelen!
    La ditt navn holdes hellig.
    La ditt rike komme.
    La din vilje skje på jorden
    som i himmelen.
    Gi oss i dag vårt daglige brød.
    Forlat oss vår skyld,
    som vi òg forlater våre skyldnere.
    Led oss ikke inn i fristelse,
    men frels oss fra det onde.

    ReplyDelete
  85. John Cowan

    Thanks for clarification and details.

    Do you see any chance of Moldova reverting to the Roman alphabet? Or do your italics in 'official position' imply that the Romanian government is more attached to separate language status than it admits?

    Why did the German-speakers of Austria not consider themselves part of the German Lands? I can see why foreigners saw 'Germany' as a geographical entity before it was a state, but why should German-speaking Austrians think so? Or did they believe in a multi-ethnic multilingual Empire?

    ReplyDelete
  86. Andrej Bjelaković12 April 2011 at 22:48

    As a speaker of Serbian, I'd say John Cowan has pretty much nailed it. Just to clarify, "Standard Serbian is mixed Ijekavian and Ekavian" is true, but only in the sense that both are considered standard in Serbian, not that they are literally mixed in SS.
    Basically, people in Serbia use Ekavian, Serbs in Republika Srpska part of Bosnia use Ijekavian (and they'll probably call their language Serbian rather than Bosnian).

    ReplyDelete
  87. Wojciech: At least as much as BrE and AmE, probably much more. In truth, Nynorsk, Bokmål, Danish, and Swedish are likewise four standardizations of a single dialect continuum, though more diverse than the four standardizations of "our language". What is more, the spoken dialect isoglosses often run east and west in Norway-Sweden, whereas the political boundary runs north and south, with the result that central Norwegians may well understand their Swedish neighbors better than their compatriots in the south. (Norwegian, unlike the other two but like English, does not have a prestige accent/dialect.)

    Infinitum, Andrej: I should note that my understanding is completely dependent on the work of Miro Kačić, the Croatian linguist (in both senses of that term). While highly respected, Kačić's work is of course controversial, like everything else about the language he worked on.

    David: I don't understand. Moldovan is written in the Latin script and has been since 1989 — except in Transnistria, the Slavic-dominated region east of the Dniestr which the Moldovan government does not control. Indeed, all parties outside Transnistria agree that Moldovan and Romanian are the same language, use the same orthography, and so on, except that it is called Romanian in Romania and Moldovan in Moldova, as if in America we insisted on calling English American (we sometimes do, of course, but not often, certainly not for all purposes).

    The Austrian version of German nationalism would be Grossdeutschland, a state uniting all the German lands under Austrian hegemony. Once the 1866 Austro-Prussian War made it clear that Kleindeutschland was the order of the day, it's hard to see Austrian Germans as feeling national (as opposed to cultural and linguistic) identity with the rest of the Germans.

    Digression: The original intent of German's much-derided national anthem, Deutschland, Deutschland über alles / Über alles in die Welt, was not world conquest, though that's the Nazis made of it. It was essentially civic-nationalist, and proclaimed loyalty to Germany-the-shared-language-culture-ideology over the particularist German microstates that were all that existed in 1841, to "Einigkeit und Recht und Freiheit", as the third stanza (now the only official stanza) has it. Brecht's unofficial new lyrics, "Und nicht über und nicht unter / Andern Völkern wolln wir sein / [...] / Und das Liebste mag's uns scheinen / So wie anderen Völkern ihr's" set the record straight.

    Andrej: Thanks for the correction: mixed was an ill-chosen word on my part.

    ReplyDelete
  88. John Cowan

    Thanks for that. No puzzle about my comments of the alphabet for Moldovan — it was just ignorance. Everything I knew about Moldavian/Moldovan was learned before 1989, and the detailed book I had to hand was also out of date. I did know about Transnistria, but assumed that they didn't bother with the language.

    I'm sort-of glad I put out some inaccurate comments. As a result I'm now much better informed!

    ReplyDelete
  89. Ad John Cowan

    "Wojciech: At least as much as BrE and AmE, probably much more. In truth, Nynorsk, Bokmål, Danish, and Swedish are likewise four standardizations of a single dialect continuum, though more diverse than the four standardizations of "our language"."

    When one compares texts (the same chapters of the Bible, say, the same Universal Declaration of Rights, the same use instruction of your motor-saw and such like) in languages grown out from the same dialect-continuums, such as: Danish-Swedish-Bokmaal-Nynorsk, or Czech-Slovak, or Russian-Ukrainian-Byelorussian, or Spanish-Portuguese, or German-Dutch-Luxemburghish-Yiddish, or Dutch-Afrikaans, or maybe various standarised or semi-standarised versions of Frisian or Rumantsch, one is immediately struck by two things:

    1. The texts are composed of recognisably similar words, here and there a completely different word, but rather seldom, in recognisably identical or nearly-identical order;

    2. 50 percent or more of these recognisably similar words display subtle differences, though -- here and there a different vowel or consonant or none where the other language has one, or some different prefix or suffix ... or such-like. This is seen from the texts in Bokmaal and Nynorsk I quoted above: for instance #1 of the Declaration of Rights contains 8 (subtly) different words in both languages, 6 identical (I am counting types, not tokens) plus a number of quite different words that can be discounted as stylistical variety of the translations.

    Now the above is not true of BrE and AmE (in their standard variants); the same text written in BrE and AmE looks, for the most part of it, exactly the same. Differences like 'colour'-'color' etc. are rare; they would increase drammatically in number if the r-lessness of BrE ('wuhk of aht') and the American 'd' (I am wriding a ledder) were codified in spelling, and perhaps the quality of the American short 'o' (oh my Gahd!), but even then they would remain regular and predictable, in way in which the differences between Czech and Slovak are not (the contrast between Bokmaal and Danish is a special case).

    A similar situation as in BrE and AmE obtains, methinks, in Germany's German and Austrian German, and Swiss (Standard) German, or France's French and Belgium's French, or Dutch and Flemish (standard, not dialects).

    Now, it'd interest me what your opinion is on Cro-Ser-Bos-Mont in this respect. Are these languages more like the former group (the Scandinavian ones, Czech-Slovak etc.) or are they rather like the various Englishes, or Germans, or Frenches, or Dutches? My impression is that they are decidedly more like the latter. Contrast Slovenian, which relates to 'our language' more or less like the languages of the former group: close, very close, very similar, yet still, clearly different.

    ReplyDelete
  90. A postscript to my previous posting: at http://www.omniglot.com/writing/serbo-croat.htm

    you find the same §1 of the Universal Declaration of Human Rights:

    Croatian:

    Sva ljudska bića rađaju se slobodna i jednaka u dostojanstvu i pravima. Ona su obdarena razumom i sviješću i trebaju jedna prema drugima postupati u duhu bratstva.

    Serbian (in transliteration)

    Sva ljudska bića rađaju se slobodna i jednaka u dostojanstvu i pravima. Ona su obdarena razumom i svešću i treba jedni prema drugima da postupaju u duhu bratstva.

    Bosnian (in Latin alphabet):

    Sva ljudska bića rađaju se slobodna i jednaka u dostojanstvu i pravima. Ona su obdarena razumom i sviješću i treba da jedno prema drugome postupaju u duhu bratstva.

    Need I count the differences between (pairwise) the three? I can't having the impression that Cro, Ser, and Bos of this example differ far less than Nynorsk and Bokmaal of the example above.

    Compare Slovene: Vsi ljudje se rodijo svobodni in imajo enako dostojanstvo in enake pravice. Obdarjeni so z razumom in vestjo in bi morali ravnati drug z drugim kakor bratje.

    It remains to be seen, tho', to what extent the omniglot site is reliable in the relevant respect.

    ReplyDelete
  91. Andrej Bjelaković14 April 2011 at 13:41

    Hm, my comment seems to be disappearing, I'll try again....

    There's no reason for Omniglot to call the Serbian bit transliterated, Latin and Cyrillic are both commonly used to write Serbian.

    Anyway, these translations seem okay, however if you take say the Lord's Prayer as an example, you'll find a few more differences. But yeah, your point still holds.

    ReplyDelete
  92. David Marjanović18 April 2011 at 20:58

    All four standard languages are founded on the historic dialect of Eastern Hercegovina

    ...so that Standard Croatian is much, much closer to Standard Serbian than to the dialect of Zagreb (the capital of Croatia), and Standard Serbian is AFAIK closer to Standard Croatian than to the dialect of Niš in eastern Serbia.

    but do you think those educators etc. can really make different languages out of what essentially was and is the same language? Such is the power of these ...tors? Imagine in the UK they 'implemented' [...] Obviously, it would be a farcical masquerade and no-one would think the British have a _different_ language, different from that of the North Americans.

    Replace "think" with "want to believe", and there you go.

    An army, a navy, and an ideology.

    I have thought of a nearby problem for my theory, which is that you might consider the Anschluss 1936 to have been a German invasion of Austria, yet the Austrians still call their language "German".

    There was a short period in the 1960s when school reports said Unterrichtssprache ("language of teaching") rather than Deutsch, but Austrian Standard German is almost identical to the Standard German varieties of Germany and quite different from any dialect spoken in or near Austria, and it's not going to go away (or be greatly modified) anytime soon – if anything, it's becoming more similar to the standards of Germany.

    Most Austrians seem to have considered themselves Germans till the early 1950s or so.

    ReplyDelete
  93. I was speaking of 19th-century Austrians of the Empire: after 1918, everything changed, with Austria becoming the smallest possible Small Germany.

    ReplyDelete

Note: only a member of this blog may post a comment.