Archive 27 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 for this siteRSS Feed

10/05/2010Diphthongs in Dictionaries#270
09/05/2010British Vowels Acoustic Data#269
04/05/2010Rounding of the NURSE vowel#268
02/05/2010Unusual Initial Consonants#267
27/04/2010IPA Postwar Specimens#266
26/04/2010Some Topical Pronunciations#265
25/04/2010IPA Specimens tween the Wars#264
21/04/2010Pre-fortis Whatsitsnames#263
19/04/2010Specimens Agen#262
16/04/2010More on Specimens#261

Blog 270

The 10th of May 2010

Diphthongs in Dictionaries

 John Wells’s blog of the tenth of May 2010 was devoted to the pronunciations of the name “Naomi”. He sed:
I’ve always said ˈneɪəmi. That is also what you will find in Daniel Jones, alongside a less common variant with an unreduced GOAT vowel in the second syllable, ˈneɪəʊmi (notation modernized). As with other -eɪə- sequences, there is also the possibility in RP of compression and smoothing to -eə-, giving ˈneəmi.
It wd’ve praps been safer to avoid possible confusion by specifying which work of the Daniel Jones he had in mind.

Roach & Setter, as editors of the Jones EPD of 2006, gave as the current British forms
ˈneɪ.ə.mi; neɪˈəʊ-
The Gimson 1997 and Ramsaran 1981 editions had
ˈneIəmɪ [ˈneIəʊm-]
The editions Jones himself was responsible for had:
ˈneiəmi [ˈneioum-] 1956, 1963
ˈneiomi [ˈneiəm-] 1926, 1937, 1947

Jones’s notation \o\ was presumably a non-diphthongal allophone of /ou/ the phoneme which EPD showed as /əu/ and /əʊ/ respectively from 1967 and from 1977. His \ei\ it seems may be understood as indicating the possibility of \e̝\ as a non-diphthongal allophone of /ei/ (later /eɪ/). His \eə\ disabbreviated from \eiə\ can hardly represent what he already had the different notation /ɛə/ for. That notation Gimson converted to /eə/ from 1977. It has to be a rarely represented diphthong /eə/ ie [eə] for which he provided the keyword 'they’re' at the EPD 1963 front endpaper and earlier.
Jones never regarded the diphthong /ɔə/ as obsolete “RP” in his era always recognising it as a subvariant at words like 'four', a practice maintained by Gimson and Ramsaran but understandably discontinued by Roach & Setter in 1997.

Gimson’s similar list contained only eight diphthongs by contrast with Jones’s fifteen. However, he gave as the predominant version of 'ruin' /rʊɪn/ not */`rʊɪn/ indicating that the pronunciation was monosyllabic and therefore that /ʊɪ/ must constitute a diphthong. It’s also found in other words. Such a diphthong was regularly recognised as such by Jones in his keyword lists. Similarly Wells in LPD3 etc transcribes 'ruin' as ˈruː ̮ɪn which permits disabbreviation to ˈruɪn. This notation we may set beside the “Weak Vowels” panel in LPD3 which observes “The vowels ə, i, u are always weak” at its §3 adding at the end of the paragraph “If a diphthong is created through the COMPRESSION of weak syllables, it remains weak, as in annual ˈӕn ju  ̮əl.” However, his transcription of 'annuity' as \ə ˈnjuː  ̮ət i\ which disabbreviates to \əˈnjuəti\ seems to contain a diphthong which is strest while not being /əˈnjuːəti/.

 Jones’s list also included, distinct from /ɔi/ (=/ɔɪ), the diphthong /oi/ with keyword 'going'. And he had /oə/ distinct from /ɔə/ with keyword 'Samoa' which he conveyed by /səˈmouə/ with its removable italic /u/. At his explanations of the meaning of italic letters he sed at EPD1963 p. xxi:
In the case of « eiə », « ouə » and « oui » the use of italic « i » and « u » is to be taken to mean not only that the words transcribed with these symbols are said either with the disyllabic sequences  « eiə », « ouə », « oui » or with the diphthongs « eə », « oə », « oi », but also that they may be said with disyllabic sequences  « e-ə », « o-ə », « o-i »”.
His count of fifteen diphthongs also included the rising and falling pairs /iə & ĭə/ and /uə & ŭə/ respective keywords 'here, happier, gourd' and 'influence'.

Blog 269

The 9th of May 2010

British Vowels Acoustic Data

The current issue of JIPA (the Journal of the International Phonetic Association) has in its first 32 pages an article ‘Formant frequencies of vowels in 13 accents of the British Isles’ contributed by Emmanuel Ferragne and François Pellegrino of Université Lyon 2. It presents “F1/F2 graphs ... which could be used as starting points for more thorough analyses”. Its aim was “to obtain an up-to-date picture of within-and-between-accent vowel variation in the British Isles” hoping it may contribute to “a better understanding of phonetic changes”. Speakers were used from the Irish Republic, Ulster, North Wales, the Scottish Highlands, Glasgow and nine locations in England ranging from Truro to Newcastle. The source of the recordings of the vowels is given as “D’Arcy, Shona [and three co-authors]. 2004. ‘The accents of the British Isles (ABI) corpus’. In Modélisation pour l’identification des langues des variétés dialectales. Paris 115-119.” No further information was given but I’m kindly informed by Petr Rösel that ABI is a database produced at Birmingham University and marketed (expensively!) by a commercial company Aurix Ltd and available via something called “The Speech Ark” ( ).


The authors offer their findings as a “compact overview of geographically-induced [sic] phonetic and phonological variation in the British Isles”. They are at pains to admit to the many limitations of their procedures. No information on the participants is available except that they had lived all their lives in the regions in question and so had their parents — at least that was the aim as far as proved practical. Their ages are not known. In some cases invented words are required to be spoken at least one or two of which seemed to cause problems. Other drawbacks are also mentioned. From the “ABI corpus” they excluded female data on two separate grounds. A phonetician, a native speaker of English, was asked by them to listen to the speakers and score them on a five-point scale for “the typicality of the speaker’s accent as well as intra-accent homogeneity”. No explanation is offered why the “expert phonetician” entrusted with this unenviable task shd be anonymous. Their findings are illustrated by 45 well presented formant graphs and spectrograms. They say that “along with the acoustic data, an auditory analysis of all speakers ... was carried out by the authors” but don’t say much about its use.

Their data on accents “not widely mentioned in the acoustic phonetic literature” are of obvious  value. However, their comments on their very small and under-specified sample of General British speech, on which there exists already such a wealth of instrumental and auditory data, will be felt to be relatively less valuable. It was no surprise that their data on the goose and foot vowels indicated values “more front than recent pronouncing dictionaries suggest” especially since we have no idea of the ages of the speakers. Incident'ly it seemed rather uncomf'table to find one of the dictionaries referred to as “Jones 2003” seeing that Jones had been de'd thirty years when the description in question referred to had been entirely re-written in 1997 by Peter Roach.

The author’s claim “we will guard against inferring potential phonetic changes from symbols used in broad transcriptions” seems far from borne out. They say “the vowel plots provided in these dictionaries [LPD 2008 and EPD 2003] explicitly show that the authors keep considering [sic] that the first element in MOUTH is slightly more back than that of PRICE. In contrast to the latter, other recent publications have symbols for [them] that are much more in line with our findings”. These “other recent publications” turn out to refer to the controversial opinions of Clive Upton alone. The reference to Olausson & Sangster (2006) is completely inappropriate. The Oxford BBC Guide to Pronunciation deals solely with the distributions of phonemes within the words it glosses and offers no judgments whatsoever on any particular English phoneme’s articulation. Its authors wdve been required to adopt its notation because that set of symbols had become on Upton’s advice a house style for the non-EFL reference publications of the OUP. Contrary opinions on this topic exprest by sev'ral leading authorities are quoted at our Blog 248 ‘A Notational Heresy’. When they come to discuss the transcription of the vocalic phoneme of square they agen seem to be giving insufficient consideration to the principles on which decisions about choice of transcriptional symbols are to be made.

A completely unwarrantable conclusion in this paper (at p.8) regarding statements made in my 2003 JIPA article ‘IPA vowels for British English in dictionaries’ (a version of which may be found as Section 5 Item 1 on this website) is “he seems to reject any possible phonetic development in the pronunciation of these two vowels’ [sc /aɪ & aʊ/]. By contrast the opinion I exprest was that /aɪ & ɑʊ/ wd still be “totally defensible” as IPA notations in lexicographical applications. Use of “ɑ” is within IPA traditions a perfectly allowable choice for a more-or-less fully open completely central vowel. A more venial misrepresentation of something else I wrote there followed in the comment “he reminds the reader that” Gimson in 1977 “departed from earlier versions of Jones’s dictionary by dropping [ɑʊ] in favour of [aʊ]”. I did not say that because Jones never did use the notation “ɑʊ” for the mouth diphthong in any edition of the EPD. He did do so in all the editions of his The Pronunciation of English.

Another of the candid admissions of problems with the reliability of their “sse” data is the remark “the assessment [of it] by a British phonetician ... yielded typicality and homogeneity scores of 3/5; which suggests that our definition of sse (which we equate with what others label RP) is indeed quite lax”.  Entering into controversy on the basis of a minute amount of data collected from a mere six speakers about whom so very little is known seems ill-advised. I have for years been making daily observations of far more than six hundred speakers especially including the usages of national television and radio newsreaders. I’m able to ascertain the speech background and age of most of them. Consequently I feel very reluctant to set much store by their meagre GB data.

Yet another disclaimer comes when they say “given that our subjects have all been recorded in London, it may well be that their type of Standard English is influenced by local features such as the PRICE-MOUTH Crossover”. This  term was used by Wells at Accents of English p. 310 in referring to “the starting-point of PRICE” in Popular London usage as “very considerably backer than that of MOUTH ... whereas in RP it is fronter or perhaps identical, but definitely not backer...”. I dou't if Wells wd now wish to modify these observations. What I have observed in recent years, from a very limited number of speakers indeed who sound nothing like mainstream GB in this respect, is a value for /aʊ/ which verges on the one that, as Wells sez, characterises “Popular London” speech. This cd well be what the authors have he'rd from some of their six GB London speakers. It’s so unusual among GB speakers as a whole that immediately two and only two well-known persons spring to mind as displaying this value which approaches [ӕʊ]. They both seem very much mainstream in all their other articulations. One of them is the 43-year-old, Eton-educated current leader of the British Conservative Party, David Cameron. The other is the very distinguished principal Channel Four Television News editor-presenter Jon Snow, twenty years his senior. He was educated at the comp'rable School St Edward’s, Oxford.

This article, on which I've felt obliged to comment, can’t fail to win admiration for the vigour and scholarship it displays. I regret having neither the energy nor the erudition to evaluate it adequately. 

Blog 268

The 4th of May 2010

Rounding of the NURSE vowel

The current issue of JIPA (the Journal of the International Phonetic Association) has an impressive article ‘What exactly is a rounded vowel? An acoustic and articulatory investigation of the NURSE vowel in South Wales English’ by Robert Mayr of the Centre for Speech and Language Therapy at the University of Wales Institute at Cardiff. I was naturally rather int'rested to see it having spent all my early life, till I was 18, in Cardiff. I wrote about the vowels of Cardiff and Cymric (by which I mean Welsh-language-influenced) English in my book (of 424 pages) on Glamorgan Spoken English which I worked on on-and-off between 1949 and 1964. I’m afraid I’ve never been able to summon the energy required to put it into a state fit for publication and even less in any position to be able to attempt to bring it up to date. Two extractions from it appeared in the collection English in Wales (1990) edited by Nikolas Coupland. They’re included on this website as the Section 7 Items 7 & 8. It’ll praps be clear from the first of them that I regarded native Cardiff speech, at least as I knew it sixty years ago, to contain no indisputable influences from the Welsh language, perhaps a matter of some interest in the context of this study.

When I came to deal with the nurse vowel in my book I didnt think it appropriate to suggest it had important rounded varieties tho I knew speakers who did round it. At that time I’d observed hardly anyone who rounded it consistently. I had for quite a while as a nei'bour someone with an ordinary Cardiff accent who did regularly round it quite strikingly but I regarded her habit as so exceptional as to be quite idiosyncratic. I’ve noticed over the years that very many people with all kinds of accents from time to time use what I can only describe as paralinguistic expressive rounding. I  offen use it myself, not only when I may be sed to be pouting! I’ve generally been at a loss to figure out what might be triggering it. Anyway I’ve very offen noted rounding that seemed to be only some sort of slight degree of emphasis in many kinds of speakers where one cou'dnt merely attribute it to the influence of adjacent segments as one might when it’s used in a word like urgent where it precedes a rounded consonant. Within the past week I’ve observed a speaker on television (with no regional features I cd notice) say the word “certainly” twice within same bre'thgroup once with /ɜː/ clearly visibly lip-rounded and once clearly without rounding but with no obvious motivation for the differentiation.

It's very obvious that lip rounding and spre'ding dont function solely as items among the features which characterise segmental phonemes. Simultaneously with that function speakers may employ them very much as they may or may not make use of various bodily gestures. In their most obvious occurrences for example open lip rounding may serve towards expressing attitudes like judiciousness and spre'ding sometimes can be somewhat analagous to wincing. Smiling and frowning obviously constantly affect lip contours. In just the way that a speech community may contain numerous speakers who are given to frequent use of certain particular gestures there may be large numbers of speakers who habitually employ lip gestures which are even so not completely universal and essential elements in their community’s communicative repertory. Deciding in characterising a phoneme whether one is contemplating only one or both of these phenomena can be a problem for the observer. I can’t help wondering how far these two possibilities apply to the very considerable numbers of speakers in South Wales who these days have marked rounding accompanying their nurse vowel.

 You get very little indication in the works of Sweet, Jones and later writers that /ɜː/ ever occurs in General British pronunciation other than as a “spread” articulation. A trivial exception was Daniel Jones’s comment regarding G. Nöel-Armfield, his first assistant when, with Jones’s concurrence he transcribed his pronunciation of the o of innocent as front-rounded. This was (at p.110) in one  of a set of illustrations of individuals’ usages that was included in Jones’s Pronunciation of English of 1914. It seems to’ve been a product of trying to blend /ɜː/ & /ɒ/ in a way that must’ve sounded pretty precious!

    Regarding that nurse rounding so common these days at least among younger South Wales speakers, I've noted it as particularly marked with the now elderly comic actor Windsor Davies  observable in the once popular sitcom "It Ain't Half Hot Mum" (YouTube has lots of clips of him). On the other hand it’s also been my impression over the years that quite a lot of Cymric speakers have not had frequent rounding of nurse. One well-known speaker it’s offen been possible to observe, Neil Kinnock, cd be seen not at all to have regular rounding. On the other hand it’s clear that changes of the kind have been taking place quite rapidly to South Wales English. Inger Mees’s analysis of recordings made in the later seventies for her 1983 study of The Speech Cardiff Schoolchildren clearly evidenced nurse rounding by them. One neednt be too surprised: nurse rounding at the other end of Wales at Liverpool is well attested. See eg Cruttenden’s Gimson’s Pronunciation of English 2008 p. 130. At any rate the well illustrated twenty-page article that touched off these remarks is to be welcomed as a substantial new contribution to this subject.

Blog 267

The 2nd of May 2010

Unusual Initial Consonants

Cruttenden’s Gimson’s Pronunciation of English lists at §10.10.1 Word-initial etc  Phoneme Sequences but, presumably because of its extremely limited distribution, omits /zl-/. There’s in fact only one word in the OED which is listed with a pronunciation beginning with initial /zl-/. Its spelling begins <zl-> and it’s the Polish monetary term złoty. Our alphabet wont run to that spelling so we write zloty and pronounce it, prompted by that spelling, /`zlɒti/. OED2 (in contrast with the policy of the ongoing OED3 which has yet to get to the treatment of this particular word) gives a second pronunciation which it indicates by the sign “||” to be an “unnaturalized/alien” ie forren version [zwoti].  

There are three other OED items that have the initial spelling <zl> two of which are “Zlead(s” [sic] and “Zlid” and (thirdly) “Z’life”: all three of these are de'lt with at one and the same single alphabetical position. They’re indicated as obsolete and consequently given no pronunciations. They raise the question of whether it’s completely appropriate to regard words as obsolete which are still to be heard in performances of dramas surviving from a bygone age. At any rate there is the question of how they are pronounced in current dramatic productions. It seems certain that in the times when these expressions were to be heard in daily use or were well known in writing the speakers wd’ve in general been aware that the initial <z> was a reduction of God’s and wd’ve accordingly made them /zliːd(z), zlɪd & zlaɪf/. Incident'ly the first two me'nt “God’s eyelid(s)”. The /iː/ of the first was no dou't a euphemistic “mincing” distortion. A well-informed director wd  presumably expect a present-day actor to use /z-/ versions but in one production I’ve noted, a BBC tv presentation of Shakespeare’s Merry Wives of Windsor, at III. iv. 24 where Slender has to say “Ile make a shaft or bolt on’t, slid, tis but venturing” what the actor sez is unfortunately /slɪd/. Dale Coye’s Pronouncing Shakespeare’s Words (2002), a practical guide to 21 plays for teachers and actors who he takes it wd quail before IPA symbols, usefully at his p. 315 recommends “’slight” in Twelfth Night to be pronounced /zlaɪt/ .

Another similar item when spelt swounds, swouns or swoones etc (OED lists these spellings at the first of them) of course stands for “God’s wounds” and suggests using another non-present-day word-initial sequence. A reader or actor is surely best advised to begin them /zw-/. Finally there will be many people who’ve heard another archaic such sequence when plays by Shakespeare are performed namely /zbl-/. The first quarto of Henry IV Part 1 (Act I Scene ii line 82) has “Zbloud I am as melancholy as a gyb Cat” which gives a good indication of how it and Sblood etc were spoken in Elizabethan usage.  Henry the Fifth (first Folio) has (IV. viii. 10) Sblud. Hamlet (second quarto) at II/ii.384 has Sbloud. This last Coye (op. cit. p. 80) recommends to be /zblᴧd/ and Helge Kökeritz in his superb scholarly volume Shakespeare’s Pronunciation p. 318 points out that this and zouns in Othello I.i. p.108 “preserve the voiced s of God’s”.
Let’s put in a plea that the current editorial staff who’re doing such a fine job of OED3 shd regard themselves as having the responsibility to provide pronunciations, where feasible, for words like these that may truly be archaisms but are certainly not stone de'd.


Blog 266

The 27th of April 2010

IPA Postwar Specimens

IPA Postwar Specimens in mf (Le Maître Phonétique) numbered about 55. Of the 52 issues 19 contained none. After JIPA took over in 1971 the category as a regular feature  faded out. Fortunately it resumed fully after the 1999 publication of the IPA Handbook with greatly renewed strength.

By the way, the lists I’ve been giving have taken no account of the kinds of relatively simple passages that for a long time featured in the pages of mf in the section called the “parti dez elɛːv” ie Students’ Corner (or however you prefer to render it into English) but these samples of English, French and two or three other widely tau't languages. These consisted chiefly of short literary excerpts transcribed by UCL Phonetics staff members. For more on these see this website Section 7.1.6-8. They can perhaps still be not without a certain amount of interest for many language teachers. By the postwar period it had become a long time since mf had been in some considerable part the province of the kinds of EFL teachers who originally were a major element in its readership and indeed had been largely the founders of the IPA.

Here follows a record of the Specimens by year from 1945 to the time when JIPA replaced the mf. They were still almost all of NWS but an occasional item was different. Notably 67/1 Jamaican Creole transcribed an unscripted telling of the Cinderella story in about 380 words. This was a valuable innovation that deserved to be emulated in a way we may yet hope to see.

45/1 Arabic | 45/2 Oriya; Frisian
46/1 Norwegian; Danish | 46/2 Armenian; Swedish
47/1 Persian; Zürich German; Chinese [niŋpɔ] | 47/2 Greek of Cyprus; Polish
48/1 Dutch | 48/2 Zapoteco; Latvian
49/1 Slovak; Guaraní of Paraguay | 49/2 Chinese [hakˈka]
50/1 Lusatian | 50/2 Italian of Naples; Gaelic
51/1 None| 51/2 None
52/1 None| 52/2 Faroese; New York City English; Southern American English
53/1 Japanese; Macedonian Greek; Guaraní of Asunción | 53/2 None
54/1 Norwegian of Trondheim; Nigerian English  | 54/2 None
55/1 None| 55/2 Spanish of Llanes; Bokmål Norwegian
56/1 None| 56/2 Ukrainian; Pennsylvanian German
57/1Scottish English; Thuringian German; Indonesian; Stavanger Norwegian;                        Hainanese |57/2 Catalan
58/1 Syriac of Jerusalem ; Slovene | 58/2 None
59/1 Rumanian | 59/2 None
60/1 None | 60/2 Chang Ting Nien Chuang
61/1 None | 61/2 None
62/1 None | 62/2 Hindi
63/1 None  63/2 Black Country English
64/1 Cardiff English | 64/2 None
65/1 Gateshead-on-Tyne English | 65/2 RP English; Modern Greek
66/1 None | 66/2 Southern Mountain American; RP English
67/1 Jamaican Creole; Korean | 67/2 None
68/1 None | 68/2 Cockney English
69/1 None | 69/2 Phillipines Tausug & Ilokano; Cockney English
70/1 None | 70/2 None

Blog 265

The 26th of April 2010

Some Topical Pronunciations

Three speakers in particular, of course, are making us aware of them in the UK at the moment. None of them has displayed any notable features that we werent aware of already. Gordon Brown’s most idiosyncratic habit is his well-known inclination to say Al-Qaeda in a way which I shd think he shares with no other public figure, namely /alki`eɪdə/. Something that has not been noticed much by me before is versions of the verb “create” which dont have the /i(ː)/ one expects in the first syllable but seem to be  /kr̩`eɪt/ or at times /`kreɪt/. I hear these from both Brown and Nick Clegg tho not from Cameron. Clegg has a very inconspicuous type of General British accent. Cameron, the Etonian, is slightly more markedly southeastern tho not really posh. I have on occasion he'rd 'im say /`wɒʔ aɪ `ӕm/. His feature that strikes me most often (not that it’s by any means really unusual) is the frontness of the first element of his /aʊ/ diphthong. One usage I’ve noticed not very recently and only once from him was leverage with the formerly solely American /e/ in its first syllable and not as a purely banking term. One musnt be too surprised: I’ve even just he'rd it from the outstandingly satisfying news presenter Hugh Edwards, the one with the unpompous gravitas and the Welsh prosodies.

PS I dont usually comment on any usage I’ve only observed once because it might be merely a simple slip of the tongue but since writing the above I’ve agen noticed Cameron say “/ɔː`θɒrətɪv/”. Well we do have preventive as well as preventative so why not. OED has an entry “authorative” but has only one record of its use, in 1645, so labels it “rare” and obsolete. No-one is recorded as using “*authoritive”.

A word whose pronunciation has been recently very widely commented on has of course been the name of that notorious Iceland volcano “Eyjafjallajökull”. One naturally went strai't to Wikipedia which duly told us it was pronounced [ˈeiːjafjatlajœːkʏtl̥] and supplied a soundfile of it spoken by an evidently native speaker of Icelandic Mr Jóhann Heiðar Árnason. Unfortunately this guy’s version of it was hard to believe to be the same word as what the transcription represented. It wasnt at all clear but it sounded rather like [`eːɪvɛ̈lœvɪk] which is pretty darnd different. The website Language Log came up with it spoken by another Icelander, no less than the Chief Inspector of Iceland's Civil Protection Agency, Mr Rögnvaldur Ólafsson. This was if anything even less helpful being a poor quality recording sounding vaguely like [`eːɪvəlɜːvɪtl] in which at any rate the last syllable was a little more like what one expected. The situation was finally resolved by a recording of a member of the Icelandic London Embassy staff who revealed her identity only as Becca. She sed very clearly, and fortunately very deliberately, something more like [ˈeɪja ˈfjatɬa ˎjœːkʊtɬ].

The other word that’s caut my attention in the last month or so I first came across when a regularly very clearly articulating and accurate Radio 4 newsreader (I think it was Susan Rae, the one with the light Scottish accent) seemed to make a startling misreading of the word methadone by inserting an unmistakable /r/ into its final syllable. Of course she did no such thing! But at that moment I’d never before he'rd the word mephedrone (neither has the OED by the way). What’s struck me about it hearing it various times since has been the slight contribution the [θ/f] contrast makes to my perception of one word versus another. If someone were to say the word mephedrone as */`meθədrəʊn/ in my hearing with ordinary conversational delivery I dou't if I’d notice the /θ/. It’s no wonder that very few of the world’s languages have both /f/ and /θ/ phonemes or that Londoners in great numbers make little use of /θ/. And it’s not only Londoners. I’ve noticed Yorkshire folk, especially youngsters, who produce an /f/ inste’d of a /θ/ and quite a variety of other non-Cockneys on television. Cf our very first blog 001.

Blog 264

The 25th of April 2010

IPA Specimens tween the Wars

As a supplement to my Blogs 259, 261 and 262 I append this list of IPA specimens of languages which appeared in mf (Le Maître Phonétique) in the inter-war period incorporating into it the 51 items of the 1949 final edition of the IPA Principles. Not very long after the 1912 appeal for NWS specimens by Jones and Passy the First World War interrupted publication of the mf. The following were extracted from mf issues which began to appear in 1923 and continued until the latter thirties. In some cases it may be found that the specimens in the Principles of 1949 repeated earlier NWS transcriptions.
Afrikaans Principles 1949 p. 51
Albanian Le Maître Phonétique 1930 p. 37
Amharic Le Maître Phonétique 1936 p. 16
Arabic Principles 1949 p. 34
Arabic of Algiers Le Maître Phonétique 1928 p.56
Arabic of Bagdad Le Maître Phonétique 1930 p. 4
Annamese Principles 1949 p. 41; Le Maître Phonétique 1936 p. 52
Armenian Principles 1949 p. 34
Assamese Le Maître Phonétique 1933 p. 78
Basque Le Maître Phonétique 1937 p. 63
Bengali Principles 1949 p. 37
Biscayan Le Maître Phonétique 1925 p. 9
Breton Le Maître Phonétique 1937 p. 76 Le Maître Phonétique 1937 p. 76; Le Maître     Phonétique 1937 p. 47
Bulgarian Le Maître Phonétique 1937 p. 24
Burmese Principles 1949 p. 40; Le Maître Phonétique 1924 p. 4

Catalan Valencia Le Maître Phonétique1936 p. 28  
Chinese: Amoy Le Maître Phonétique 1930 p. 38
Chinese: Cantonese in Principles 1949 p. 43; Le Maître Phonétique 1937 p. 69
Chinese: Fuchou Le Maître Phonétique 1933 p. 11
Chinese:Hwa Miao Le Maître Phonétique 1923 p. 4
Chinese: Pekingese Principles 1949 p. 42; Le Maître Phonétique 1927 p. 45
Corsican Le Maître Phonétique 1937 p. 48
Czech Principles 1949 p. 30
Danish Principles 1949 p. 26;Le Maître Phonétique1924 p. 18;
    Le Maître Phonétique 1928 p.56; Le Maître Phonétique 1930 p. 37
Dutch Principles 1949 p. 25;  Le Maître Phonétique 1928 p. 39
Efik Le Maître Phonétique 1929 p. 17
English: American Principles 1949 p. 20; Le Maître Phonétique 1933 p. 54
    Le Maître Phonétique p. 79; Le Maître Phonétique 1934 p. 15
    Le Maître Phonétique 1934 p. 16; Le Maître Phonétique 1935 p. 14
    Le Maître Phonétique 1937 p. 71
English: British Principles 1949 p. 20
English:Cornish Dialect Le Maître Phonétique 1923 p. 7
English: Lancashire Dialect Le Maître Phonétique 1929 p. 19
English: New York Le Maître Phonétique 1933 p. 35; 
    Le Maître Phonétique
1937 p. 46
English: Pennsylvania Le Maître Phonétique 1936 p. 28
English: Scottish Principles 1949 p. 21
English: Scottish of Morebattle Le Maître Phonétique 1935 p. 13
English: Tyneside Le Maître Phonétique 1927 p. 48
English: West Riding of Yorkshire Le Maître Phonétique 1923 p. 6
Eskimo Le Maître Phonétique 1934 p. 76
Estonian Principles 1949 p. 31
Finnish  Principles 1949 p. 31
French Principles 1949 p. 21
French: Poutort Dialect Le Maître Phonétique 1925 p. 21
French: Vaudoise Le Maître Phonétique 1924 p. 7
French: patwa d l onis [sic] Le Maître Phonétique 1927 p. 47
French: Pays de CauxLe Maître Phonétique 1936 p. 14
Gã Le Maître Phonétiquee 1925 p. 6
Gaelic: Scottish Le Maître Phonétique 1924 p. 5
Ganda  Principles 1949 p. 47
Georgian  Principles 1949 p. 33
German  Principles 1949 p. 24
Greek  Principles 1949 p. 32;  Le Maître Phonétique1928 p. 38;
    Le Maître Phonétique 1937 p. 71
Gujerati Le Maître Phonétique 1926 p. 18
Hebrew Le Maître Phonétique 1937 p. 22
Herero Le Maître Phonétique 1932 p. 58
Hindi  Principles 1949 p. 36
Hungarian  Principles 1949 p. 32; Le Maître Phonétique 1937 p. 48
Ibo Le Maître Phonétique 1937 p. 20
Icelandic  Principles 1949 p. 28; Le Maître Phonétique 1936 p. 51
Igbo Principles 1949 p. 45
Italian  Principles 1949 p. 22
Italian Dialect of Servigliano Le Maître Phonétique 1926 p. 40
Japanese Principles 1949 p. 44;Le Maître Phonétique 1936 p. 28
Japanese English Le Maître Phonétique 1925 p. 20
Kanarese Le Maître Phonétique 1928 p. 74  
Korean  Principles 1949 p. 44; Le Maître Phonétique 1924 p. 14
    Le Maître Phonétique 1937 p. 21
kətəlɑ [sic] Le Maître Phonétique 1935 p. 32 & 67
Luchu Le Maître Phonétique 1936 p. 63
Maidu Le Maître Phonétique 1931 p.
Marathi Le Maître Phonétique 1934 p. 102; Le Maître Phonétique 1935 p. 30
Malay Principles 1949 p. 39
Martinique Creole Le Maître Phonétique 1932 p. 4
Norwegian  Principles 1949 p. 26; Le Maître Phonétique 1924 p. 19 &
    Le Maître Phonétique 1929 p. 30
Norwegian: West Le Maître Phonétique 1929 p. 20
Oriya  Principles 1949 p. 38
Panjabi  Principles 1949 p. 37
Pashto Le Maître Phonétique 1927 p. 20
Persian Principles 1949 p. 35
Polish Principles 1949 p.29; Le Maître Phonétique 1923 p. 24; Le Maître Phonétique     1927 p. 9
pɔːmɔ [sic] Le Maître Phonétique 1932 p. 5
Portuguese  Principles 1949 p. 23; Le Maître Phonétique 1937 p. 68
Provençal  Principles 1949 p. 23; Le Maître Phonétique 1931 p. 26
Provençal: Béarn Le Maître Phonétique 1936 p. 62
Provençal: Landes Le Maître Phonétique 1936 p. 62  
Punjabi Le Maître Phonétique 1925 p. 28
Roumanian  Principles 1949 p. 30; Le Maître Phonétique 1926 p. 10
Russian  Principles 1949 p. 28; Le Maître Phonétique 1924 p. 6 ;
    Le Maître Phonétique 1929 p. 47; Le Maître Phonétique 1933 p. 78
Sesutu  Principles 1949 p. 49
Shan  Principles 1949 p. 40
Sindhi Le Maître Phonétique 1933 p. 76
Sinhalese  Principles 1949 p. 39
Slovak Le Maître Phonétique 1936 p. 15
Slovenian Le Maître Phonétique 1932 p. 57
Somali  Principles 1949 p. 34; Le Maître Phonétique 1923 p. 15
    Le Maître Phonétique 1933 p. 72
Spanish  Principles 1949 p. 22; Le Maître Phonétique1927 p. 46
Swahili Principles 1949 p. 46
Swedish Principles 1949 p. 27; Le Maître Phonétique 1927 p. 20
Swiss dialect of Val d’Illiez Le Maître Phonétique1923 p. 5
Tamil  Principles 1949 p. 38; Le Maître Phonétique 1937 p. 66
Thai  Principles 1949 p. 41; Tswana  Principles 1949 p. 48
Turkish  Principles 1949 p. 35; Le Maître Phonétique 1937 p. 45
Twi Principles 1949 p. 46
Ukrainian Le Maître Phonétique 1937 p. 62
Urdu (cf Hindi) Le Maître Phonétique1928 p. 7
Vietnamese (see Annamese)
Wallon Le Maître Phonétique 1934 p. 102
Welsh  Principles 1949 p. 24
Welsh of Merionethshire Le Maître Phonétique 1926 p. 19
Welsh of Cardiganshire Le Maître Phonétique1926 p. 19
Xhosa  Principles 1949 p. 50
Yoruba  Principles 1949 p. 45; Le Maître Phonétique 1923
Zulu Le Maître Phonétique 1927 p. 30; 1929 p. 47

Blog 263

The 21st of April 2010

Pre-fortis Whatsitsnames

In his blog of today John Wells explains why in the last two decades he has always referred to the shortening effect, in most varieties of English, on vowels (and diphthongs), nasals /m, n & ŋ/ and the lateral approximant /l/, when they’re followed immediately in the same syllable by one of the voiceless/fortis consonants (p, t, k, ʧ, f, θ, s & ʃ), as “pre-fortis clipping”. He explained the matter very clearly like this:
The /f/’s in self, selfish /ˈself.ɪʃ/, and dolphin /ˈdɒlf.ɪn/ trigger clipping, but not those in shellfish /ˈʃel.fɪʃ/ or funfair /ˈfʌn.feə/. So do the /t/ in feet and the /ʧ/ in feature, but not the /p/ in fee-paying or the /k/ in tea-kettle. The vowel /æ/ undergoes pre-fortis clipping in lap, lamp, happy /ˈhæp.ɪ/, and hamper /ˈhæmp.ə/, but not in slab or clamber.

I have to admit that I’ve never taken to the term “pre-fortis clipping” mostly because it suggests cutting where I feel a more appropriate metaphor would be something more like “squeezing” tho I’m not actually very much inclined to call it “pre-fortis squeezing” or for that matter "pre-fortis compression" or "pre-fortis contraction”. But I should prefer either of the latter two to “pre-fortis clipping”. I’m glad the reasoning behind what has seemed to me to be a slightly eccentric usage has now been made known to us but personally I’m perfectly happy to go on using the simple term “pre-fortis shortening”. I dont see why one shd worry that students need be confused when contemplating this phenomenon beside that of /ɪ & ɒ/ as mainly shortened forms of /iː  & ɔː/ in historical terms. They can surely easily discriminate between synchronic and diachronic processes of these kinds. Of course John and his UCL colleagues may have encountered student problems that I’m unaware of. I suppose I cou'd imagine students being confused about how it might be related to the unscientific expression “clipped speech”.

Anyway the Wellsian lead has now been followed in the two outstanding books for students beginning a grounding in English phonology, the new 2009 fourth edition of Peter Roach’s English Phonetics and Phonology and the 2003 first edition of the Beverley Collins and Inger Mees Practical Phonetics and Phonology in neither of which it previously featured so far as I can remember. This was understandable of them because students using these books in conjunction with LPD (the Wells Longman Pronunciation Dictionary) will find the terminology employed there and the matter explained at the information panel entitled Clipping. (There’s no corresponding panel in EPD, the Roach-et-al English Pronouncing Dictionary).

Not only do I not greatly favour the term “clipping” but I’ve also for many years preferred to avoid “fortis” as far as possible. I’ve latterly been inclined to go back to Sweet’s easily remember'd “Saxon” term “sharp” and to use with it the correlative “soft”. My reason has been that both the terminological pairs voiceless-versus-voiced and fortis-versus-lenis are quite unsatisfactory by themselves. English phonemes in the voiced/lenis category are constantly used without vocal-cord vibration and may well be articulated quite strongly and so on. The slightly clumsy solution I used in my textbook A Guide to English Pronunciation (1969) was to pair them as “voiced-lenis” and “voiceless-fortis”.

Altho I prefer to avoid the term “voiceless” I do find the less often needed combined term “pre-fortis” attractively explicit and recognisable. Something I fail to understand about the thinking of those who’ve adopted “pre-fortis” is how they have been content with the unexplicit term “rhythmic clipping” for the corresponding shortening process to be found when, to use the LPD examples, the /iː/ in leader and leadership is usually observably shorter than the /iː/ in lead. Another such set out of many is chair, chairman and chairmanship. I have for many years prefer'd for this the more explicit term “pre-enclitic shortening”. If they’d been content to be equally less explicit before they decided to adopt “pre-fortis clipping” they might well have called it “combinatory clipping”.

Those who’d be int’rested in a comment on how variation between General British speakers in their pre-enclitic articulations of various occurrences of the ash vowel may indicate generally-quite-unnoticed regional differences within GB may care to look at §3.9.8 on this website. An example is that, tho most GB speakers are likely to say bad as [baːd], what numbers who say badly as [baːdli] relative to those who say [badli] is hard to judge.

In connection with pre-fortis shortening, there’s something that Daniel Jones was meticulous in recording in the EPD of his day (as is LPD today) which the current EPD seems to regard as not worth notating. This is the fact that in small numbers of words like peacock and teapot the first syllable has captured the sharp first consonant of the second orthographical syllable contrary to appearances. They’re accordingly most faithfully to be transcribed as /`piːk.ɒk/ and /`tiːp.ɒt/.

Blog 262

The 19th of April 2010

Specimens Agen

When practic'ly a century ago Passy & Jones invited readers of their phonetic journal to contribute to it versions of NWS (see Blog 259) they didnt explicitly ban other texts. However, hardly any alternative to NWS has ever appeared in an International Phonetic Association publication since. Of course, there’s certainly something to be sed for ev'ryone using the same text but the advantages are related to the purpose for which it’s employed. As we noted in Blog 261 Palmer and Abercrombie used other materials. One of the 29 illustrations in the 1999 Handbook did actually use a different text of about the same length. This, for the Austronesian language Taba, was a passage entitled ‘A riddle about being sick’. An orthographic version of it was provided but no translation — which might’ve been int'resting.

The matter of the choice of specimen texts wz raised in a valu'ble article in JIPA Volume 36 Number 2 of December 2006 by David Deterding /`detədɪŋ/ with the title ‘The North Wind versus a Wolf: short texts for the description and measurement of English pronunciation’. In it he listed “many shortcomings” of the passage for English. Some of these we covered in Blog 261. Others included positional allophones and consonant and vowel clusters. He put forward for consideration an alternative passage ‘The Boy who Cried Wolf’ which he’d adapted from another Aesop fable. It’s almost twice as long as NWS (195 words) but still a very useful length for many purposes. He devised it in a form that minimised repetitions of words, contained sets of mimimal pairs and had clear instances of the English monophthongs in contexts facilitating their measurement. It’s a distinct improovment on NWS and thus to be welcomed as an addition to the repertoire of such passages to be used in place of NWS or alongside it. It’s this: The Boy who Cried Wolf

There was once a poor shepherd boy who used to watch his flocks in the fields next to a dark forest near the foot of a mountain. One hot afternoon, he thought up a good plan to get some company for himself and also have a little fun. Raising his fist in the air, he ran down to the village shouting "Wolf, Wolf." As soon as they heard him, the villagers all rushed from their homes, full of concern for his safety, and two of his cousins even stayed with him for a short while. This gave the boy so much pleasure that a few days later he tried exactly the same trick again, and once more he was successful. However, not long after, a wolf that had just escaped from the zoo was looking for a change from its usual diet of chicken and duck. So, overcoming its fear of being shot, it actually did come out from the forest and began to threaten the sheep. Racing down to the village, the boy of course cried out even louder than before. Unfortunately, as all the villagers were convinced that he was trying to fool them a third time, they told him, "Go away and don’t bother us again." And so the wolf had a feast.

Other texts have been used for various purposes. Notable ones in the field of phonetics include a 240-word one John Wells devised in 1982 for twenty specimens he arranged to be recorded to illustrate his Accents of English. It was this:

One day last year, when I was driving back to work after I’d had lunch, I had an amazing and unforgettable experience. It must have been two o’clock or perhaps a quarter of an hour later — a quarter past two. It was an incredible thing really. I was sitting there, at the steering wheel of my new car, waiting for the lights to change, when all of a sudden the car started to shake this way and that, rocking from side to side throwing me backwards and forwards, up and down. I felt as if I was riding a bucking horse. Worse than that, some mysterious spirit or hostile force seemed to be venting its vast fury upon the earth. And the noise! There were a kind of deep groaning and awsome grinding which seemed to fill the air. And then, a short while after, the whole paroxysm had stopped just as suddenly. Everything was calm and smooth again. Quiet and peaceful once more. I put my foot down. Just a gentle pressure on the accelerator — or the gas-pedal as it’s known in America — and drove off. Everything was utterly normal once more. So, then. Was this some local and momentary earth tremor which had struck us? Or, I ask myself, “Was it a supernatural visitation, a fiery storm of diabolical wrath — or was it rather that I’d drunk a double vodka or two during my lunch?”.

John Laver to illustrate another epoch-making book, his 1980 The Phonetic Description of Voice Quality, used besides EWS a passage (of 175 words) called ‘The Rainbow’. A longer version of this (332 words) has been used at the Finnish University of Tampere for English Public Speaking Courses. Another passage of 373 words from that stable called ‘Comma Gets A Cure’ constitutes a ‘diagnostic passage for accent study’ that contrived to include the very keywords of the Wells standard lexical set. Were they thinking they had some special value in themselves?

It seems strange that NWS has been used by so many who might’ve been better served if it hadnt been purely narrative “spoken  prose” but at least contained an amount of direct speech. It’s not very suitable for investigating prosodic matters. Accordingly some years ago I prepared an elaboration of NWS of 200 words to be more than half direct speech and to include types of exclamation, command, contradiction, question (question-word, yes/no, alternative and tag), hesitations, vocatives and leave-takings. It also contained diagnostic items for the bath lexical set and others. It was this:

One day, the North Wind and the Sun were having an argument.  “You can’t be nearly as strong as I am” the Wind was saying.  “Oh yes I am,” rejoined the other.  Just then a traveller appeared.  “What a fine cloak!” exclaimed the Wind.  “Yes, it is  nice, isn’t it?” added his friend. “Is it warm?”  “Oh! Unusually so! And ... er... it could solve your dispute if you’d give me that pleasure.”  “What did you say?” they chorused. “I suggest, gentlemen, that this very garment can provide the decision you require.”  “What d’you mean?” snapped the Wind.  “My idea is,” continued the stranger calmly,  “that, whichever of you can make me take it off soonest, he’ll be recognised as the stronger.”  “All right.” answered the Wind, “Shall I try first or will you?”  “You start” replied his companion. Then the Wind blew absolutely furiously but the wayfarer only wrapped his mantle tighter around him. At last he simply had to give up the struggle. Then old Sol shone out really powerfully. Immediately the poor fellow was forced to take the thing off. “There!” he sighed. “That settles the matter.”  “Well! I suppose I’d better be getting along. Goodbye, then!”

Blog 261

The 16th of April 2010

More on Specimens

The Jones-and-Passy choice of NWS (see Blog 259) didnt adopt the exact text Sweet had used. His version of the fable was 130 words long and contained one or two oddities such as the sentence “Then broke out the sun” and the phrase “cold and fierce as a Thracian storm”. They eliminated these and, choozing rather more ev'ryday expressions, reduced it to 113 words. Their text sounds totally unconversational and at one point has what is perhaps a very slightly dated grammatical choice of wording but it has worn fairly well for a century-old text.

        The North Wind and the Sun were disputing which was the stronger, when a traveller came along wrapped in a warm cloak. They agreed that the one who first succeeded in making the traveller take his cloak off should be considered stronger than the other. Then the North Wind blew as hard as he could, but the more he blew the more closely did the traveller fold his cloak around him, and at last the North Wind gave up the attempt. Then the Sun shone out warmly, and immediately the traveller took off his cloak. And so the North Wind was obliged to confess that the Sun was the stronger of the two.

A passage which has had quite an amount of use, especially in America in connection with DARE (the Dictionary of American Regional English), is another fable-type piece which like NWS first appeared in a work by Henry Sweet. It’s not clear whether he took it from some unacknowledged source or concocted himself. (But one wonders how come there’s a soft toy on sale with just such a name and a children’s DVD called “Arthur: The Rat Who Came to Dinner” too. Influence of DARE?) Its earliest version appeared in 1890 at pages 66-68 of his Primer of Spoken English with the title The Young Rat. At something like six hundred words this was longer than most users wished for. We find David Abercrombie in 1964 in his EPT (English Phonetic Texts) using his version of it of about 360 words, of which about a third were direct speech, ascribed as “After Henry Sweet” and re-titled Arthur the Rat. He had modified it in 1950 for use in recording specimens of English at Edinburgh University. By contrast, the DARE version, using that title, extended it making it not very natural-sounding in the process of cramming in various dialect shibboleths into its 583 words.
        Once upon a time there was a rat who couldn't make up his mind. Whenever the other rats asked him if he would like to come out hunting with them, he would answer in a hoarse voice, "I don't know." And when they said, "Would you rather stay inside?" he wouldn't say yes, or no either. He'd always shirk making a choice. One fine day his aunt Josephine said to him, "Now look here! No one will ever care for you if you carry on like this. You have no more mind of your own than a greasy old blade of grass!" The young rat coughed and looked wise, as usual, but said nothing. "Don't you think so?" said his aunt stamping with her foot, for she couldn't bear to see the young rat so coldblooded. "I don't know," was all he ever answered, and then he'd walk off to think for an hour or more, whether he would stay in his hole in the ground or go out into the loft. One night the rats heard a loud noise in the loft. It was a very dreary old place. The roof let the rain come washing in, the beams and rafters had all rotted through, so that the whole thing was quite unsafe. At last one of the joists gave way, and the beams fell with one edge on the floor. The walls shook, and the cupola fell off, and all the rats' hair stood on end with fear and horror. "This won't do," said their leader. "We can't stay cooped up here any longer." So they sent out scouts to search for a new home. A little later on that evening the scouts came back and said they had found an old-fashioned horse-barn where there would be room and board for all of them. The leader gave the order at once, "Company fall in!" and the rats crawled out of their holes right away and stood on the floor in a long line. Just then the old rat caught sight of young Arthur — that was the name of the shirker. He wasn't in the line, and he wasn't exactly outside it — he stood just by it. "Come on, get in line!" growled the old rat coarsely. "Of course you're coming too?" "I don't know," said Arthur calmly. "Why, the idea of it! You don't think it's safe here any more, do you?" "I'm not certain," said Arthur undaunted. "The roof may not fall down yet." "Well," said the old rat, "we can't wait for you to join us." Then he turned to the others and shouted, "Right about face! March!" and the long line marched out of the barn while the young rat watched them. "I think I'll go tomorrow," he said to himself, "but then again, perhaps I won't - it's so nice and snug here. I guess I'll go back to my hole under the log for a while just to make up my mind." But during the night there was a big crash. Down came beams, rafters, joists — the whole business. Next morning — it was a foggy day — some men came to look over the damage. It seemed odd that the old building was not haunted by rats. But at last one of them happened to move a board, and he caught sight of a young rat, quite dead, half in and half out of his hole. Thus the shirker got his due, and there was no mourning for him.

Harold Palmer in 1925 in A Few Documents on English Phonetic Notation showed plainly that he was not satisfied with NWS for his purposes because it didnt contain all the symbols he wanted (it even lacked words providing ʧ, ʒ, ɪə, ɛə & ʊə) and failed to adequately exemplify various connected speech phenomena such as “weakening, shortening, stress, word-linking etc”. The untitled passage of 116 words he devised was preferred, especially for its brevity, by Abercrombie for the eight specimens of different types of transcription he gave in Appendix I of his EPT. Some of its usages are perhaps now slightly out of date but it has the advantage of being all direct speech. 

        At what time are you going to the exhibition? I thought I heard you tell your brother this morning that you expected to meet him there at about two.  Yes. Would you like to join us there?  I would, with pleasure, but I am not sure whether I can. In any case I must leave early to catch the four train. I do not live here now; I live in the suburbs and I want to get home before it is dark.  Are you really in such a hurry to get home?  Must you? If it is solely on that account, we can take you back in our car.  Can you? That will be splendid!  All right.