Archive 26 of JWL Blog

Readers who need explanations of any of the abbreviations used may find them at Section 1 of the Home Page.

RSS Feed

Not Really a Mis-Hearing

If my mem'ry serves me right, some years ago it was usual on my favourite BBC Radio-4 early-morning Today news programme to repeat unchanged, from an earlier session to a later one, rather more items than is usual these days. At any rate I remember tuning in just at the end of what was probably the 6-to-7-am first transmission of the five-minutes-or-so slot that was entitled Thaut for Today. This item, which still continues to be scheduled, was in fact an opportunity for some representative of a religious group to say something often intendedly uplifting at an hour when many of us were struggling to get ourselves up from our beds to face the ardours of preparing for the journey to work. On this morning, as I tuned in, I only cau't, so it was clear, the very last word or at most couple of words uttered by the speaker and I felt stunned to seemingly be hearing what sounded like some sort of remarkably indecent dismissal.

Naturally I was exceedingly anxious to catch the repeat of this item when it came round agen one hour later. The speaker turned out to be, as far as I can recall, an Anglican clergyman, someone no dou't like the Bishop of Mbawawa in Barbara Pym’s endearing novel Some Tame Gazelle, who was making a moralistic point out of telling the story of an experience he’d had when he’d been a missionary in Africa. I think he described a day-long safari-type journey he’d undertaken with native bearers, “good souls,” who for many hours had cheerfully manhandled their backbreaking burdens mile after mile to enable him to walk unencumbered thru the jungle. As they reached a point in sight of their destination with less than half an hour to sunset, to his dismay they all suddenly stopt and sat down.

The speaker sed that he begged them to carry on, reasoning with them that it'd take such a relatively short time compared with the earlier part of their journey and be so much better to arrive at their destination before the disadvantages of nightfall set in. However, they were adamant that they shd rest because, as they explained, according to their religion, it was essential that they should stop at that juncture to “allow their souls to catch up with them”.

“I’ve many times thaut since”, sed he, “that it wd so often give so many of us such a lot of spiritual benefit if we cou'd, like those good fellows, from time to time stop and ponder on our past lives and, as it were, wait to be cau't up with by /`ɑː səʊlz/”. This of course was “our souls”.

American readers may be int'rested to be apprised that the majority of General-British speakers pronounce our most offen as /ɑː/ and that the great Oxford English Dictionary at its entry ass n(oun)2 “Now chiefly U.S.” added “vulgar and dial[ectal] sp[elling] and pronunc[iation] of ARSE.] appending also “(Webster 1961 ‘often considered vulgar’.)” This is not the only such retention to some degree on the other side of the Atlantic of variant forms of words where an /r/ has been lost over the centuries before a following /s/. Others include cuss, hoss and passel which are more elegantly curse, horse and parcel. And we both like to use a form of burst converted to bust if only informally. Oh, and people who'd stoop to using such a coarse expression as I seemed to've he'rd wd have no troubled conscience about dropping the /h/ of a word like holes.

IPA Specimens

The International Phonetic Association was inaugurated in 1886 and in 1897 its title became the present one. In 1888 the name of its journal (in the first three years called Dhi fonètik tîtcer) became settled as Le Maître Phonétique. That is what it remained until 1971 when, on its ceasing to be printed entirely in phonetic script and assuming its present style, it was re-titled The Journal of the International Phonetic Association (well known as JIPA /ʤaɪpə/). From early in its history a very regular feature of its contents had been what were called ‘Specimens’ ie of various languages from anywhere in the world. Initially a variety of passages were tried out for the purpose all shorter than a single page. Then in the fifth of ten pamphlets expounding and exemplifying the principles of the IPA, published mainly as supplements to issues of Le Maître Phonétique, a particular text was recommended to provide ‘specimens of phonetic transcription’, the well known Aesop fable of ‘The North Wind and the Sun’ (hereafter NWS). This fable had been first used with such a function by Henry Sweet at p. 209 of his Handbook of Phonetics in 1877.

The footnote to page 19 of the 1912 version of The Principles of the International Phonetic Association said “The editors [Paul Passy and Daniel Jones] will be pleased to receive versions of this fable in languages or dialects not included here, for publication either in subsequent editions of this pamphlet or in the Maître Phonétique.” Whatever may be thought of the limitations of the usefulness of the NWS passage, it was at least concise and employed lexical matter that could reasonably well be accommodated to languages with a wide variety of cultures and geographical backgrounds. Compared with what had been tried out previously as illustrative materials it was quite an improvement. What is more it produced a remarkable response from a wide range of people interested in the aims of the IPA. 

In the pages of that 1912 edition of the IPA Principles which followed, NWS, in the form of a paragraph of about ten lines containing not a lot over a hundred words, then appeared in twenty-three versions in most cases preceded by a few lines of comment. In 1949 the booklet that became the last edition of the Principles was published. It contained fifty-one versions of NWS about half of them of European languages the rest being mainly Asian and African items. It was to remain in print as the Association’s publication in most demand until in 1986 its printing was discontinued in anticipation that a new version would very soon be produced. As the present writer warned the meeting of members at Oxford when the decision was discussed, this discontinuation was distinctly premature. It took another 13 years for the Handbook to appear, in 1999, with a smaller number of illustrations (29) but much improved and containing a more worldwide range of languages to which further items have steadily been added in subsequent issues of JIPA with more than ever valuable accompanying matter. Latterly they have usually contained information on where audio files of the readings analysed can be accessed and downloaded via the internet and in various cases formant charts, which are especially welcomed, and occasionally spectrographic materials etc. The IPA authorised vowel diagram has generally been used tho the editors have sometimes been perhaps too indulgent in accepting contributions that have rather departed from its proper shape which was the only one used in the Handbook. There were excellent instructions on how to produce new specimens given at pp 89-91of JIPA Volume 32 Number 1 of June 2002.

Besides the earliest specimens, when in 1923 Le Maître Phonétique finally resumed publication after the break occasioned by the First World War, there began a steady trickle of versions until 1971. When mf was replaced by JIPA they seemed to dry up. Until then only two or three issues of mf had ever appeared without some specimens. The peak was in 1937 when there were seventeen. The next twenty fallow years ended in 1990 when there were seven. From 1991 to 1997 there were twenty-one. Since 1998 there have been almost sixty more to be added to the Handbook set. These have been helpfully listed by the editors in JIPA most recently at page 129 of the current Volume 40 Number 1 issue of the 1st of April 2010, which proudly declares on its front cover “40 Years of JIPA”.

Beniowski (ii)

No-one seems to know Beniowski’s year of birth but it was probably about 1800. He died in 1867. In the year before he published his Dictionary he issued an 84-page booklet entitled The Anti-absurd or phrenotypic alphabet and orthography for the English language .. invented by Major Beniowski .. Author of the system of artificial memory designated Phrenotypics .. London .. published by the author, 8 Bow Street, two doors from Covent Garden Theatre. In it he gave information about his education etc mentioning: “A course of seven years of mathematical, literary, and medical studies, at the University of Wilno [ie Vilnius]. About ten years ... of authorized, official, medical practice. A course of military studies in the ... École d'État-Major, Paris. About twelve years of travelling in various countries of Europe, Asia, and Africa. Study and practice of the principal modern and ancient languages. [Involvement] in the principal political movements of Europe, from 1816 to 1841, at St. Petersburgh, 'Wilno', Warsaw, Cairo, Paris, and London. Seven years of a studious residence in London ... assisted in my studies, observations, and reflections, by a memory greater than which is not upon record”.

He seems to have supported himself largely by lecturing on "the art of memory": his system of mnemonics, like various previous ones, included devices like assigning numbers to small groups of phonetically similar letters. After participating in an uprising agenst Russian rule, along with other Polish insurgents he came to the UK for political sanctuary. He’d at first received a Government allowance of £40 per annum but lost it when he became obscurely involved with an abortive chartist uprising at Newport in Monmouthshire in 1839. He called himself a “cosmopolitical Chartist”.

His Dictionary, about 30,000 words each delt with in a single line at almost all entries, was more or less a reduced Walker. Even so it showed something of the way things were moving in the British accent he le·rnt from the educated Londoners he mixed with. When quoting him, to make it easier for the reader, his indications’ll be interpreted as they’d be represented today rather than by displaying his actual notations. This includes ignoring the way he, with the lack of realism of all English orthoepists before A. J. Ellis, represented vowels of unstrest syllables with the values they’d be accorded if spoken with full stress. Also many of the r’s he shows are omitted because of being very unlikely to have been sounded.

He may well have deferred to Walker’s judgments too often but he didnt follow him slavishly. He reflected various changes from Walker’s day in eg not giving /baθ/ but /bɑθ/ for bath and likewise with ask, asp, basket, bastard, blast, brass, cast, castle, class, last, mast, plaster etc. Yet he agreed with Walker’s /a/ for branch, casket, chaff, chance, chandler, clasp, craft, gasp, glance, glass, grass, grant, lather, pass, past, raft, rafter, slant, staff etc. For rather he recorded /ɑ/ or /eɪ/ compared with Walker’s /a/ or /eɪ/. He had like Walker the first syllable of envelope as /ɒn/ but unlike him the modern tonic stress first not last. He also had the more modern GB first-syllable stress on premier. Some Walker omissions that he included were avoirdupois, (as /avədjuː`pɔɪz/), Bethlehem (as /`beθlɪhem/), children (as /ʧɪldən/) and Jesus (as /ʤizəs/). For Jerusalem he had /ʤɪ`ruzələm/ with /z/ not Walker’s /s/. It’s curious that he didnt include the word Polish nor his coinage phrenotypic. Even children wasnt in the body of the dictionary but occurred twice in some prefixed exemplificatory continuous texts. The value he showed for it may be compared with Walker’s indication that he considered the ordinary conversational version of hundred to be /hᴧndəd/.

He sometimes differed only slightly from Walker eg at `laboratory where his penult vowel was /ɒ/ unlike Walker’s schwa type. Similarly at `corollary his penult had /ӕ/ rather than Walker’s /e/ and at predecessor he showed /priːdɪ`sesə/ rather than /pre-/. He chimes with Walker in his treatment of the -ile suffix as /-ɪl/ as it mostly is in GA. Agen like Walker he has virile as /`vaɪrɪl/ which one notes is imagined by the Webster Online editors to be still current in Britain with such a first syllable. His different spelling as opposed to sound in futile as “fiutil” shows a more modern representation of the sound of the name of the letter U by having two phonetic letters rather than Walker’s “u” with  superscript numeral.  He also agreed with Walker on guttural as /`gᴧʧərəl/ which, tho it hasnt done so, one might perhaps have expected to survive in US usage in the way cordial has kept the value /`kɔ(r)ʤəl/.

In various places he differed from Walker only in order of preference between alternatives. He gives only Walker’s second choice viz /stӕlək`taɪtiːz/ for stalactites. Likewise for epoch he put /`iːpɒk/ before /`epɒk/ and for herb /hɜːb/ before /ɜːb/. Finally, regarding the suggestion in my article on the (final) happy vowel (see §3.2 on this website and my Blog 186) that the value [ɪ] didnt become mainstream educated usage in England until the mid nineteenth century, like Walker he identifes it with /iː/. The original texts of the booklet and the dictionary quoted have been digitised in full and are available free on the Internet. Those who’ve found the above of int'rest may like to look at my Blogs 167, 177, 178 and 179.

Major Beniowski (i)

Fifty years ago or more the UK had a great number of secondhand bookshops and during my time as an impecunious student I was glad to ramble round them in various cities and found many long-out-of-print items I still have. The other day I happend to pick up one I thaut praps I shd throw away now. It’s thick but very small (about 12 by 8 by 4 cms ie 5 x 3 x 1½ inches) and the binding’s in sad condition. I’d never properly looked at it — I guess mainly thru being put off by its ridiculously eccentric title

Don’t try looking for phrenotypic in the OED: it’s not there. It was his personal coinage and ruffly me·nt mnemonic. One of his obsessive notions was that the English-speaking nations were being held back by their “absurd” spelling. 

Anyway, contemptuous of John Walker (see Blog 167) for his superscript numbers and plethora of spelling “Principles”, he set about compiling his own pronouncing dictionary with a set of phonemic (not that he·d·ve called them that) symbols of his own concoction. He boasted justifiably that it cd be used by any printer of English. To quite an extent he was successful. He abjured capital letters in any form and aimed at a single letter for each phoneme. This aim he failed to fulfil for /ʧ, ʃ, θ & ð/ for which he used the combinations {ch, sh, ʇh & th} but for ones he cdnt manage to represent with single letters from the English alphabet he simply used “turned” (ie upside-down) versions of such single letters. Some of the results produced what were to become (not with his values for them or necessarily in imitation of him) authorized IPA symbols including his {ɥ} which he used for IPA /ʌ/, {ɐ} for /i/, {ə}, not for schwa but for /əʊ/, and {ɔ} for /ɔ/. As to the ordinary vowel letters {a, e, i, o, u & y} he used for the first one the sound of its name and for the others their commonest sounds in monosyllables viz /eɪ, e, ɪ, ɒ, u & aɪ/. Having adopted {y} for /aɪ/ he, in a way some modern phonologists wd find quite agreeable, effectively enough, used {i}  for yod, eg use was {ius}.

His least satisfactory symbols visually were the awkward-looking turned g he used for /aʊ/ and the turned {l} he used for /ʒ/. His choice of a turned t {ʇ} in the combination {ʇh} which he adopted for /θ/ in order to distinguish it from {th} which he assigned to /ð/ was also not happy. The really astonishing deficiency in his system was his failure to differentiate /ʊ/ and /u/. He did seem to distinguish eg fir and fur as {fer} and {fɥr} similarly to the way Walker had and even Murray provided for in his 1884 notation for the OED. Like so many nineteenth-century orthoepists before Ellis, he recognised no /ɜ/. Besides {ɥ} ie turned h for /ᴧ/, he had turned y {ʎ} for /ɑː/. He used {oi} perfectly satisfactorily for /ɔɪ/. Like Walker, he showed all words with r’s as if they were pronounced with /r/s but without adding superfluous final e’s after his word-final r’s as Walker had. For example air appeared as {ar} (as if /eɪə(r)/ cf Walker’s "a¹re" — that superscript #1 shd surmount the a in a combination, not follow it, but Unicode doesn’t oblige), here as {hɐr}  and cure as {kiur}. He made no attempt to distinguish /ŋ/ from /ŋg/ as in pairs like finger and singer. Perfectly acceptable was his method of indicating word stress by an acute accent placed after the vowel of the syllable bearing tonic stress. He economically gave it to be understood that, unless marking indicated otherwise, the first syllable of each word was to taken to bear the tonic accent, so his {smɔlpox} was very clear. (To be continued.)

H C Wyld and Daniel Jones

 
Henry Cecil Wyld was born on the 27th of March 1870 to a colonial administrator with the East India Company whose wife was from a landed Scottish family. Both he and Daniel Jones were Londoners by birth but Jones was a dozen years his junior. Wyld became one of the most successful pupils of Henry Sweet at Oxford. Before that he’d already attended universities at Bonn and Heidelberg. First from 1899 to 1920 at Liverpool and thereafter in another Chair of English at Merton College Oxford he pursued a distinguisht academic career in the field of the history of the English language mainly concentrating on its phonology. One curious sideline he had for a time was as an inspector of the teaching of phonetics in the training colleges of Scotland.

He was clearly much respected by Jones who in 1914 commended to readers of his new Outline of English Phonetics “the able articles” in Modern Language Teaching of December 1913 and June 1914 in which Wyld proposed the adoption of the term ‘Received Standard’ to refer to what Jones ended by calling ‘Received Pronunciation’. At his third edition of 1932 Jones dropt the footnote recommending those two articles for reasons not hard to guess. In his earliest books and in the two first editions of his Outline he’d used the invidious term ‘Standard Pronunciation’ but by 1932, no dou't because that expression had caused resentment, he added a paragraph (§62) saying “The term ‘Received Pronunciation’ (abbreviation RP) has been suggested for the type of pronunciation described in this book. This term is adopted here for want of a better. I wish it, however, to be clearly understood that other types of pronunciation exist which I consider to be equally ‘good’.” This was of course just a bit disingenuous because it was Jones himself who’d made the suggestion.

In 1913 in his collaboration with Herman Michaelis on their Phonetic Dictionary of the English Language Jones had only remarked that ‘The pronunciation represented is that generally used by persons of culture in the South of England’. For the first edition of his English Pronouncing Dictionary in 1917 at page viii he sed ‘The pronunciation represented in this book is that most usually heard in everyday speech in the families of Southern English persons whose men-folk have been educated at the great public boarding-schools’. After eight further lines enlarging upon this he continued ‘The form of pronunciation recorded in this dictionary may be referred to shortly as “Public School Pronunciation”; it is indicated in what follows by the abbreviation PSP’. Wyld, at page 3 of his HMCE (History of Modern Colloquial English) in 1920 remarked. ‘If we were to say that Received English at the present day is Public School English, we should not be far wrong’. There Wyld was talking about a variety of English not solely a type of pronunciation. For that he most often used the elliptical term ‘Received Standard’.

From the 1926 revised introduction to the first edition of his EPD onwards, Jones was to influence two generations and more of British linguistic scholars into adopting the regrettably question-begging term ‘Received Pronunciation’ tho he was apparently followed by none of them in his use of Wyld’s term ‘Received English’ which he employed in a few places in his Outline eg at §§407, 653, 662fn but occasionally alternated with ‘Southern English’ (eg 1956 p. 348) and ‘educated Southern English’ (§879) when he didnt write simply ‘English’. Jones’s adoption of “RP” is usually attributed to the influence of A. J. Ellis but it was most probably no less prompted by the example of Wyld’s persistence in using the term ‘Received’ despite its by then quite archaic effect in such contexts. The social attitude of Wyld was staggeringly supercilious, defiantly so at HMCE p.2 where he refers to “the type which most well-bred people think of when they speak of ‘English’ ” and  has a good deal more to say in the same vein.

In his remarkable single-handed achievement of the very large single-volume 1440-page Universal Dictionary of the English Language of 1932, Wyld recorded pronunciations for every headword in two forms of transcription, a popular type and a scientific notation in the Henry Sweet tradition. He always remained aloof from the International Phonetic Association. His description of the pronunciations he indicated was “those current in good society” but he sed also  “the sounds that the writer of this Dictionary had in mind are those in use among the majority of persons who speak Southern Standard, or better, Received Standard English. If this description is considered too vague, it must suffice here to say that Received Standard is that type of English which is spoken by those who have been educated at one of the older Public Schools. It is by no means the exclusive property of these, but from them at any rate we may be pretty sure of hearing it”. It was a bit ironic that Wyld had only one year at any public school tho his was one of the big seven namely Charterhouse. Jones attended the minor-league Radley College for only two years before proceeding to University College School which he didnt consider to be a ‘public school’ at all.

His DNB biographer and sometime student at Merton College, Harold Orton, sed justifiably of Wyld’s Dictionary “Its size and comprehensiveness, its methodical arrangement, its unusually full treatment of etymologies, its clear and precise definitions, together with its brief and racy illustrative sentences mostly of his own coining, gave it a lasting value.” One of my favourites among his definitions was his terse bracketed gloss on currant bun “(with few or no currants)”. He died at his home, Alvescot House, Alvescot, Oxfordshire, on 26 January 1945. The foregoing has contained variations on themes also pursued at Blog 079 and our main section §7.3.7 etc. 

More on Restaurant

Triggered by the new account of the pronunciation of restaurant in OED3, we looked in our last blog at the treatment of the word’s pronunciation in a selection of reference works past and present. One excellent thing about the new OED3 entry is that there’s no nonsense about seeming to suggest that any considerable number of British speakers produce a perfectly French version of it. This get-out has been employed in all too many cases in various publications rather than giving a proper account of how English-speakers really say words. This is not to say that one isn’t grateful particularly to John Wells who so courageously and uniquely offen supplies us with information about forren pronunciations. He’s not suggesting that English speakers do or shou'd use such forms in their conversations. There certainly are some of us who can say French words in perfectly French pronunciations but anyone in ordinary circumstances who elects to do so would be felt to be insuff'rably pretentious if their performance werent intended humorously.

Our selection of versions of restaurant unsurprisingly includes none before the 1908 OED1 edited by Sir William Craigie the other Scottish editor of the dictionary. He included a completely French version but after the OED1-&-2 sign || for “not naturalised, alien”. This practice has been very reasonably dropped in OED3. We may notice that the earliest records all contain forms in which the /t/ of the spelling is included. One sees quite a deal of variation even notably when the same lexicographer is making a different decision at a different time, eg Jones’s changes from 1913 to 1917 and thereafter.

I’ve now been provided very kindly by Petr Rösel of the Mainz Gutenberg University with copies of the recordings of the headwords supplied on discs to accompany the dictionaries by OUP, CUP and Longman the publishers of ALD, EPD and LPD. The ALD speaker gives not what has generally proved to be the favourite choice amongst current lexicographers viz the pronunciation /`restrɒnt/ but /`restrɒːnt/ with the same quality but slightly lengthened. [Amended 1Nov17]

The EPD speaker has, as one wd hope, no trace of the “optional” /ŋ/ which might well sound old-fashioned or unfashionable. He also fortunately does not include the (superscript) schwa that the editors tell us “may be omitted” thus producing the more usual di-syllabic variant. He follows the text faithfully in using a nasalised vowel thus / `restrɔ̃ː/. 

The nasalisation he uses is so strong that one’s tempted to wonder if he’s having to make a special effort to produce a form not habitual for him.
The LPD speaker is particularly int'resting in that he produces a version nowhere to be found at the dictionary entry. What he does is especially notable in that one gathers that ev'ry one of the recordings for LPD was monitored by a phonetician and this version seems to have sounded so normal to the monitor that no discrepancy was noted between it and the text. He uses the version that we referred to in Blog 254 as being quite absent from all the accounts known. This must be a relatively recent development tho it will be remembered that I had noted it from a 1999 broadcast by a General British type of speaker. It was /`restrɔːnt/ with a distinctly longer (tho qualitatively similar) vowel type than that of the ALD speaker.

My observation that this type is now common seems to show some sign of being borne out by its use from a reader who was presumably unconsciously replacing the printed version with his personal habitual one. On a personal note, to the best of my recollection, from the 1960s to the 1980s I vacillated between what I perceived to be /`restrɒ̃/ and /`restrɔ̃ː/ but have in recent years found myself inclined to depart from these much of the time in favour of /`restrɔːnt/.

Pronunciations of RESTAURANT

The third edition of the OED, the great Oxford English Dictionary, has just de'lt in draft with the word restaurant. Here are the variants given there exactly as listed:
OED3: 2010 Brit. /ˈrɛst(ə)rɑːnt/, /ˈrɛst(ə)rɒnt/, /ˈrɛst(ə)rənt/, /ˈrɛst(ə)rn̩t/, /ˈrɛst(ə)rɒ̃/, U.S. /ˈrɛst(ə)rənt/, /ˈrɛstəˌrɑnt/, /ˈrɛˌstrɑnt/. (10+4=14 alternants)
The British form listed first is certainly far less common currently than the second. The bracketing of sounds to indicate that they may or may not be he'rd is a useful space saver but unfortunately in a case like this doesnt reveal the fact that three-syllable variants are nowadays much less usual than two-syllable ones in British usage at least.

Previous and present treatments in various dictionaries are given below for comparison:
OED2: 1989 (ˈrɛstərənt, ˈrɛstərɒnt ||ˈrɛstɔrɑ̃)
OED1: 1908 (re·stǫ̆rănt || ręstoraṅ)
Afzelius 1909 ended with a nasalised back rounded vowel and ŋ. See Blog 205.
Jones & Michaelis 1913 ˊrestərɑ̃ŋ [-rɔ̃ŋ]
Funk & Wagnall: 1914 (In rival transcriptions 1 and 2.)
1. res ́to-rənt or res ́ ́to ́ ́rɑ̄ṅ ́ 2. rĕs ́to-rant or rĕs ́ ́to ́ ́räṅ ́
COD (Concise Oxford Dict) 1: 1911 re·staurant (-or, “or as  F[rench]). This was bungled but dou'tless intended to convey /`restərənt/.
A 1951 Appendix added Fr. [rɛstɔrɑ̃].
COD 1990 /ˈrestəˌrɒnt, -ˌrɔ̃/
ODP (Oxford Dict of Pron) 2001 (not disabbreviated)  BR ˈrest(ʃ)(ə)rɑːnt  ˈrest(ʃ)(ə)rɒnt  ˈrest(ʃ)(ə)rɑːnt  ˈrest(ʃ)(ə)rṇt  ˈrest(ʃ)(ə)rɒ̃. 
AM ˈrɛst(ə)rənt  ˈrɛstəˌrɑnt  ˈrɛˌstrɑnt 
(18 + 4 =22) These dubious /-stʃ-/ versions are fortunately not shown in the OED3 draft. They suggest further/-ʃtʃ-/ variants as shown in ODP at Am question if the /ʃ/ versions indeed exist to any noteworthy extent.
EPD17: 2006 disabbreviated but omitting “cutback” vertical bars (21+4 = 25)
ˈres.tr.ɔ̃ːŋ, ˈres.tər.ɔ̃ːŋ; ˈres.tr.ɔ̃ː, ˈres.tər.ɔ̃ː; ˈres.tr.ɑ̃ːŋ, ˈres.tər.ɑ̃ːŋ, ˈres.tr.ɑ̃ː, ˈres.tər.ɑ̃ː; ˈres.tr.ɑːŋ ˈres.tər.ɑːŋ; ˈres.tr.ɒnt, ˈres.tər.ɒnt; ˈres.tr.ənt, ˈres.tər.ənt; ˈres.trɔ̃ːŋ, ˈres.tr.ɔ̃ː; ˈres.trɑ̃ːŋ, ˈres.trɑ̃ː; ˈres.trɑːŋ; ˈres.trɒnt, ˈres.trənt,
The surprising /ŋ/ versions are now actually verging on archaic (unless as non-lexical assimilative forms). US ˈres.tə.rɑːnt, ˈrestɚ.ənt, ˈres.trɑːnt, ˈres.trənt  
EPD1: 1917    ́restərɔ̃ːŋ [-rɑ̃ːŋ, -rɔːŋ, -rɑːŋ, -rɔŋ]
EPD11: 1956  ˈrestərɔ̃ːŋ [-rɑ̃ːŋ, -rɔːŋ, -rɑːŋ, -rɔŋ]
EPD14: 1977  ˈrestərɔ̃ːŋ [-rɑ̃ːŋ, -rɔːŋ, -rɑːŋ, -rɔŋ], ˈrestərɒnt [-rənt]
We see here the Gimson EPD little-used innovation of more than one principal form.
Universal English Dictionary 1932 (H. C. Wyld) [rɛ́stərɑ̃]
ALD1 (Advanced Learner’s Dict): 1948 [réstərɑ̃, réstərənt]
ALD2: 1963  [ˈrestərɔ̃ːŋ, ˈrestərənt]
ALD3: 1974 `restrõ  US -tərənt
ALD4: 1989 /ˈrestrɒnt; US -tərənt/
LPD (Longman Pron Dict) 1-3: 1990, 2000, 2008 disabbreviated but omitting “cutback” vertical bars   (19+8=27)
ˈrest ə rɒnt, ˈrest rɒnt; ˈrest ə rɑːnt, ˈrest rɑːnt;
ˈrest ə rɒŋ, ˈrest rɒŋ; ˈrest ə rɒ̃, 
ˈrest rɒ̃;  ˈrest ə rɑ̃ː, ˈrest rɑ̃ː; ˈrest ə rɔ̃ː, ˈrest rɔ̃ː;
ˈrest ər ənt, ˈrest ər nt; ˈres trɒnt, ˈres trɑːnt, ˈres trɒ̃;  ˈres trənt, ˈres trnt  
|| (ie American) ˈrest ər ənt ˈrest r ənt, ˈrest ər nt; ˈrest ə rɑːnt, ˈrest rɑːnt;
res trənt, ˈres trnt;  ˈres trɑːnt.  
Merriam Webster Online \ˈres-t(ə-)ränt also -t(ə-)rənt, -tərnt\ at 19 Mar 10
Webster 1962 \ˈrest(ə)rənt, -stəˌränt also -ˌstränt or -stərnt\
Dict of English Pron with American Variants ( H E Palmer et al) 1926
́rɛstərɒŋ $  ́rɛstərənt,  ́rɛstərɑnt
Random House Dict (USA) 1966 (res ́tər ənt, tə ränt ́)
Concise Pron Dict (OUP) 1972 `restrɔ̃ -rõ -rənt -rɒnt $ restərənt -rɑnt -tr-
Longman Dict Contemp Eng 1978: /ˈrestərɔ̃, -rɒnt || -rənt, -rɑnt (Fr rɛstɔrɑ̃)/.
Oxford ALD online /ˈrestrɒnt; NAmE trɑːnt; tərɑːnt/  20 March 10
Cambridge ALD online/ˈres.trɒnt/(US)/-tə.rɑːnt/ 20 March 10
Howjsay online (sound only) had /`restrɒnt/ 20 March 10

One int'resting point is the absence of the common British version /`restrɔːnt/. This I’ve noted from various newsreaders etc including Sue McGregor R4 20 May 99, Andrew Marr 10 May 04, Rory Morrison May 07, Joan Bakewell 11 Oct 07, Neil Sleat R4 News Mar 08 and Jan 10.

The two most complete pronunciation dictionaries EPD and LPD can be seen from comparing the disabbreviations provided above with their highly abbreviated entries to have at times made considerable sacrifices in terms of user-frendliness and clarity in their efforts to be ever more inclusive.

Certainly no chicken

In February, Tami Date, a Japanese phonetician who takes great int'rest in English intonation, remarked in regard to the utterance of the expression “I’m certainly no chicken” in the film Good bye, Mr. Chips, that the “perplexing thing to me is the fall-rise tone of  'chicken.' ... I was wondering if it was the usual tone of this idiom. What if it should be uttered with a fall?” This appeal addressed to phonetician colleagues received very prompt attention. Paul Tench, commented “A fall-rise is precisely what I would expect on "chicken". Even if Mr Chips had not used that idiom but a straightforward response “I'm certainly not YOUNG", I would still expect the fall-rise...” This judgment was immediately endorsed by John Wells saying “Paul is spot on”. I completely concur that the statistical likelihood is that the native-English speaker would be inclined to select a falling-rising tone on the word “chicken”.

However, no-one offered a reply to Tami’s other question  “What if it should be uttered with a fall?” The fact is that almost any other ordinary tone might occur but its selection wd depend on the factors of discoursal context and temperament or mood of the speaker. With less than neutral cordiality a fully descending tone might occur: I’m certainly no ˋchicken/ˎchicken... might reflect some degree of resentment, the higher fall more vehemently, the lower more subdued. In a similar context a speaker might use a rising tone eg I’m certainly no ˏchicken (but I'm `not de`ˏcrepit). Again a speaker might possibly in favouring an airy, light-hearted manner, select a high-to-mid descending tone I’m certainly no ˋ-chicken (but there’s `plenty of `life in me `-yet).

This last choice is prob'bly rather more often he'rd from American speakers than others but it’s common enough among General-British speakers as a variant on the Fall-Rise theme. It’s a good example of the way that tones with strikingly different physical forms can be used with very little contrast of semantic effect. On the contrary a narrow low-rising tone, something which hardly features in the literature known to me, with its relatively modest physical difference from the well-documented low-to-mid rising tone may produce a very different semantic effect from the usual low rise. This, in my terminology a Rise-Bass (exemplified on this site, as are others, at §8.5 'English Melodics Beyond Basic Tones'), used in I’m certainly no ˏˌchicken cd sound lacking in enthusiasm or grudging etc. At the other end of the scale of pitch variation a rising-falling-rising tone might be used viz I’m certainly no ˊˋ ˏchicken (Climb-Fall-Rise in my terminology) in a slightly flamboyant style.

Besides these examples one comes across speakers from time to time who frequently idiosyncratically prefer certain tones to the more usual choices for people with their accent. One that comes to mind is the distinguished British actor Edward Fox, particularly famous for the film Day of the Jackal, who at least in one performance on the London stage, for a whole evening constantly avoided Fall-Rises in favour of Falls. Another is Philip Latham whose superb performance as Planty Pal, Duke of Omnium & Gatherum in the Simon Raven BBC tv adaptation of stories from Anthony Trollope novels may fortunately be obtained still in the DVD of The Pallisers of 1974. He, no dou't as a feature of the characterisation, maintained an effective remarkably constant series of low-rising tones, conveying a calm and controlled temperament.
 

Pronunciation for Opera Singers

Recently a request for help was circulated to teachers of English pronunciation by a member of the staff of the University of Minnesota who, inste'd of her usual work “with international graduate students and international faculty on achieving intelligibility in their speaking” of English, had been asked to work with two students of singing who need to employ any of sev'ral languages. She asked if anyone cd offer to estimate how long “it might take for these young women to achieve native like articulation in their operatic singing”. One of them “is a speaker of Mandarin and the other is from Taiwan”.

I admire her evident enthusiasm but I recommend her to relax and remember that managing a few unfamiliar articulations in order to sing with reasonably satisfactory approximations to the sounds of Italian, French, German, and Spanish, the languages she mentions, is hugely easier than learning to use these languages for conversation. Not only is the text provided but the word and sentence rhythms and the intonations are happily prescribed by the composer in ev'ry case. She shd pref'rably have access for each language to a reliable pronunciation dictionary with a good coverage. For Italian there is the Dizionario d’ortografia e di pronunzia by Bruno Migliorini, Carlo Tagliavini & Piero Fiorelli publisht in 1981 by ERI - RAI Radiotelevisione Italiana. For French there are the Dictionnaire de la prononciation by Alain Lerond publisht in 1980 by Larousse and the Dictionnaire de la prononciation française by Léon Warnant publisht in 1987 by Duculot. For German there is the Aussprachewörterbuch by Max Mangold publisht in 2005 by Duden of Mannheim etc. There is no equivalent for Spanish and so far as I know not for Russian either but there are a variety of books which provide singers with help with these languages and even with Czech and with Ecclesiastical and Medieval Latin. They are to be found by going to Google and putting in “Pronunciation for Singers”.

Pronunciation for Singers was the title of what was no dou't the very first phonetically sophisticated book in the field. It was by the first really modern scientific phonetician of the English-speaking world Alexander John Ellis. (Blog 194 has a brief note on him.) It was publisht in 1877 and I shd perhaps not be mentioning it except that it’s of very consid'rable historical int'rest and it’s actually freely available online even to download. Some readers might well find parts of it quite useful. A book I certainly recommend to coachers of opera singers is Phonetic Readings of Songs and Arias by Berton Coffin, Ralph Errolle, Werner Singer, & Pierre Delattre publisht originally in 1982 by The Scarecrow Press N. J. & London. This provides a goodly number of pretty reliable phonetic transcriptions of well-known operatic arias.

Incident’ly this teacher’s appeal for help with opera singers’ pronunciation didnt seem to bring forth much useful relevant comment but it did trigger a certain amount of discussion of kinds of regional American accents that have been imitated by non-American devotees of pop groups, hip-hop artists and vocalists like Madonna. Finally I’d like to mention a little cautionary tale for the benefit of anyone who might want to try to help a singer tackling French. Many years ago when I was working at a Middle-Eastern university a British frend who was an excellent musician confided in me that he was having terrible trouble trying to teach a singer he was working with to articulate a good strong French uvular /r/. I had gently to convey to him that a French-speaking audience, if she had succeeded in doing so for example in rendering one of Fauré’s delicate chansons, would have probably been horrified because traditionally classical singers use only lingual r-sounds and uvular ones are characteristic of music-hall artistes of the coarseness of Édith Piaf.

Death Reports Exaggerated (iii)

John Maidment sed in his blog of the twentieth of January 2010:
At some date in the future, maybe not too distant, GBE will have no centring diphthongs.../ʊə/ is hanging on by its fingertips, I think, but has for most people (1) merged with /ɔ:/ OR (2) merged with /ɜ:/ OR (3) become a new phonological unit which we could symbolise as /ʊ:/.

There’s no question that by the latter part of the last century several very common words including moor, poor, sure, your and you’re had completely transferred from /ʊə/ to /ɔː/ and that many others had acquired variants with /ɔː/ tho few of them had developed from the sequence /jʊə/. The items that have forsaken /ʊə/ for /ɜː/ are now so far from mainstream GB as to be /ʃɜːli/ considered as /kjɜːri`ɒsətiz/. As to [ʊː], one no dou't hears it for the /ʊə/ phoneme more often than [ʊə] but the tokens of /ʊə/ most free from conditioning are surely predominantly [ʊə] and not [ʊː]. It’s axiomatic, one presumes, that in deciding upon the notation for a phoneme we consider its identity independently of any conditioning by contiguous sounds or non-lexical prosodies. Of course it’s not unreasonable to suggest that there may be numbers of GB-speakers who have a phoneme /ʊː/ but I can’t say I’ve particularly noticed any. I dou't if there’s convincing evidence that they yet exist in any numbers leave alone in such a majority as to warrant abandoning the traditional /ʊə/.

Admittedly, common words which are fairly universally agreed to predominantly take /ʊə/ are much thinner on the ground than those having most other diphthongs. However, I suggest that there are sufficient numbers of them for there to be little likelihood of /ʊə/ suffering, in anything like the near future, the same fate as befell /ɔə/. In the list given below, the majority of whose over 200 items will not be totally unfamiliar to educated GB-speakers, it can be confidently predicted that, in works which reliably represent their GB pronunciations, in almost every case mostly the sole and at least the predominant versions will be listed with /ʊə/. I have to acknowledge that a few are listed according to my impressions of what are predominant versions which differ from LPD and/or EPD.  There are many more less familiar such items in the OED and very many compounds containing these items as constituents other than the few I have entered in the list.

I have excluded from this list all words involving smoothing over morphemic boundaries like brewer, cruel, Ewart, ewer, fewer, fluent, heuristic, hewer, jewel, newer, queuer, sewer, steward, Stuart, viewer etc. These are very reasonably represented in the literature as primarily exhibiting /u(ː)ə/ and only subvariantly /ʊə/. Also clearly to be excluded are words with normally /ʊ/ such as augury, estuary, February, January, mercury, obituary, penury, sanctuary, statuary, usury etc. These may occasionally be uttered by the odd speaker with /ʊə/ but for most of them only in artificial, pedantic or stately etc styles.

The following list is not exhaustive. Few of its items can be classified as really abstruse. It’s surely evidence, along with the very numerous compounds that some of them occur in (only a selection of which have been included), of the need for hesitation before starting to toll the death-bell for /ʊə/. Probably the main reason why they’ve hitherto resisted change, and are likely to continue to do so, is that most of them take forms which accord fairly satisfyingly with their spellings:

abjure, acciaccatura, adjure, allure, amateur, amour, angostura, anthurium, appoggiatura, Arcturus, armature, Arthurian, Asturian, azure, Bloomsburian, boor, Bourbon, bourgeois, bourse, bravura, bureau, bureaucracy, bureaucrat, caesura, camera-obscura, Canterburian, caricature, centurion, coiffure, coloratura, con`jure, contour, couture, couturier, craquelure, curacao, curate, curative, cure, curé, curia, Curie, curiosity, curious, curium, de jure, demure, detour, diuretic, dour, dura (mater), durable, Duracell, dura mater, duration, Dürer, durex, during, durum, dour, Douro, Drury (Lane), durer, durex, endure, endurance, entrepreneurial, epicure, epidural, Etrurian, Eurasian, euro, Europe, European, Eurosceptic, Eurostar, Eurovision, fioritura, fluoride, fluorine, (petit) four, führer, furious, Furtwängler, furuncle, fury, futurity, gourd, gourmet, gourmand, gravure, guipure, heuristic, houri, immature, immaturity immure, impure, impurity, incurable, IndoEuropean, infuriate, injurious, inure, ischuria, Jura, jurisdiction, jurisprudence, juorist, juror, jury, juryman, Kommandatura, Kurath, Kurosawa, Latour, Ligurian, liqueur, Lourdes, lure, lurex, lurid, luxuriant, luxuriate, luxurious, McClure, Manchurian, manicure, manure, Marlburian, mature, maturity, mercurial, Missouri, Muir, mural, Mure, Muriel, Murthia, Namurian, neurasthenia, neurological, neuron, neurosis, neurosurgeon, neurotic, Newry, Nuremberg, Nureyev, Nuristan, Neurofen, obscure, obscurity, ordure, overture, pelure, penurious, petit four, pleurisy, plural, plurisyllable, pot-pourri, pourboire, procure, prurient, puerile, pure, purée, purification, purify, purist, puritan, puritanical, puritanism, purity, purulent, Questura, Ruhr, rural, Ruritanian, sakura, Saussure, secure, security, seigneurial, Sgurr, Shaftesburian, Silurian, sinecure, spurious, Stour, Sturm und Drang, sulphuric, sura, Surabaya, surah, sural, Surinam, suture, tambour, tellurium, tempura, tenure, tessitura, thurible, thurifer, Thuringian, tour, Tourette, tourism, tourist, tourmaline, tournament, tournedos, tourniquet, Tours, Truro, Turandot, Turgenev, Turin, Turing, Turku, Uigurian, Uranus, Ure, urial, Uriel, uric, urinary, urine, urology, Uruguay, usurious, velour, Zurich.


BlogRoll

• DCblog (David Crystal)
• John Maidment
• Linguism (Graham Pointon)
• John Wells