Archive 47 of JWL Blog

Home

Index of All Blogs

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

RSS Feed

Scottish Place Names

Our Blog 467 began an account of the third volume of reprints entitled English Phonetics in the 20th Century by dealing with its forty pages of the LJ (A Lloyd James) BBC “Recommendations to Announcers Regarding the Pronunciation” of ‘Doubtful’ (general) Words and its seventy pages of English Place-Names.’’ Continuing from these we now look at the thirty-eight pages containing about 800 Scottish Place-Names. We find that the large numbers of Celtic-derived items these contain seem to make for many more departures from ordinary spelling patterns than do place-names in England.

In considering this list we shall take the opportunity to compare some of LJ’s entries with the publications of his two successors in this field. The first of these was by the late G M ('Elizabeth') Miller who headed the BBC Pronunciation Unit set up in 1939 to replace the Committee of which LJ had been secretary. She edited the 171-page BBC Pronouncing Dictionary of British Names published by OUP in 1971 the year she retired. Incident·ly your faithfull bloggist advised her on behalf of the OUP on its preparation and also enjoyed carrying out the reading of its proofs. Her longest-serving successor Graham Pointon, who was at the helm of the Unit for twenty-three years, in 1983 produced a revised edition of it extended by more than a hundred pages. Unfortunately OUP have not seen fit to republish this valuable work now for three decades. This has been a great pity because it had far more names than the general pronunciation dictionaries cou·d spare the space for. It’s to be hoped that some more enlightened regime at the Beeb and/or OUP may realise how very worthwhile wd be the production of an updated edition of the book.

For the LJ 1932 booklet of Scottish place-names fourteen correspondents were consulted in addition to the twenty members of the regular committee. The sometimes oddly exprest advice from LJ to the BBC’s mostly London-based announcing staff was in certain respects less than realistic. One imagines that he was influenced by an inclination to avoid criticism from the predominantly phonetically unsophisticated committee members etc consulted and his far from tolerant Scottish employer John Reith. I’m referring to remarks like “The Scottish pronunciation of English always differentiates between w and wh…” and “It must be remembered that r is invariably pronounced in Scotland”. In my observation both of these expressions are more applicable to aspiration rather than performance, certn·ly as regards the latter. Most curiously the transcriptions displayed “the symbols [eː] and [oː], representing the sounds largely used in Scotland in place of the Southern English diphthongs [ei] and [ou]”. One wonders what, if anything at all, he can’ve possibly expected his London announcers to do about words transcribed with these.

The most striking diff·rence from English names is due to the circumstance many of the Scottish ones involve a phoneme not even present in the speech of most people in England. This is of course the voiceless velar fricative /x/ which is very often the value of the digraph <ch> in Scottish words. It’s best known from the word loch. The declared policy of the three chief pronouncing dictionaries, to give first for any word its most frequently used form, is currently only properly implemented for the word loch by the CEPD which gives \lɒk\ first. LPD and ODP unrealisticly give \lɒx\ first. From 1917, when Daniel Jones first publisht his EPD, he gave \lɒx\ first. He hadnt reconsidered that judgment until at his final major revision in 1956 he “got real” and actually did put \lɒk\ first. The OED (online)  pronunciation for loch has been unchanged (understandably, taking into account the stupendous amount of data they have to cope with) since Bradley delt with the word (no dou·t in consultation with his Scottish senior co-editor Murray) in 1903. It gives /lɒx/ without any variants! However, for Lochinvar, an item introduced in 1976, it gives a more up-to-date /lɒkɪnˈvɑː(r)/. Other 1976 entries are Lochlann (Scandinavia) as /ˈlɒxlæn/, Loch Fyne (type of fishing boat) as /lɒx faɪn/ with no stress marking supplied and Loch Ness with no pronunciation at all.

Evidence that [x] is a sound most English-speakers with no Celtic background are not at home with is reflected in the fact that LPD3 and CEPD18 give first \bɑk\ for the name of the German composers called Bach. Most of us are familiar with the markedly Scottish or Irish exclamation Och more or less equivalent to Oh. Many know [x] from one or two well-known Scottish names such as that of the Glasgow Sauchiehall /sɒxi`hɔl/ street or of the town Pitlochry /pɪt`lɒxri/. Some have heard of names like Rannoch /`ranəx/ or even Auchtermuchty /ɒxtər`mᴧxtɪ/. It’s known that quite a few people other than just seafarers tend to lissen to the BBC weather forecasts. They may hear any day Ardnamurchan Point (the most westerly spot on the island of Great Britain) which LJ gave as \ɑːdnəˈmərxən\. It’s heard today only as /ˈɑdnə(ˈ)mɜkən/ from most forecasters. Unfortunately, as LJ’s lists extensively revealed, the Scottish /x/ isnt only spelt <ch> and the spelling <ch> at times represents other sounds. The other chief spelling for /x/ is <gh> as we see in Brough /brɒx/, Redheugh /red`hjux/ and Drumsheugh /drᴧm`ʃ(j)ux/. Unhelpfully, the spelling <ch> may also stand for /k/ as in /`baŋkəri/ Banchory, /kɪl`katən/ Kilchattan, /lə`maŋkə/ Lamancha, /`mɜkɪstən/ Merchiston, /rɒθi`mɜkəs/ Rothiemurchus, /abə`kɔldə/ Aberchallder and /abə`kɜdə/ Aberchirder. These last two ‘Elizabeth’ Miller, a Scot herself by the way, repeated in 1971 but evidently Pointon’s researches led him in 1983 to convert them from /k/ to /x/. LJ’s /`baŋkəri/ Banchory with only /k/ was given an alternant with /x/ in 1971. No alternant has been given for the equally unguessable /bən`kru/ for Bunchrew. For the <ch>of Ballachulish /balə`hulɪʃ/ only /h/ has been given. The name Buchan only had /x/ given: non-Scots almost always say /`bᴧkən/ for it. A common name LJ didnt include was Buchanan for which CEPD and LPD rightly give /bju`kanən/ both adding the variant /bə`kanən/ with a comment by LPD, as in Miller 1971, that the latter is “normal Scottish usage”.

The LJ list contains no word beginning <ch> with the value /x/ though it did have Acharacle /ə`xarəkl/, with /x/ beginning its accented syllable, which must’ve been rather intimidating for London announcers to face. He gave Champany as /`ʧampəni/ and Chryston as /`kraɪstən/.

Except for OED quotes which are given as shown in its text, versions between forward slashes are to be taken as strictly phonemic representations of current GB pronunciations. They're given in a transcription which makes no use of lengthmarks. They're not quoted as in the LJ original text in order to avoid certain possible misinterpretations. Back slashes \-\ have been used to enclose transcriptions that are not necessarily strictly phonemic.

(To be continued)

Ash and Bush Keepers

Many English-speakers have accents very close to General British in most respects except that their types of English pronunciation didnt undergo, roundabout the late eighteenth century, the change in a hundred or so common words from an ‘ash’ type vowel to a pref·rence for a so-called ‘broad’ /ɑ/ that (the very most general) GB has in words like ‘pass’. These were words in which an earlier ‘ash’ /æ/  immediately preceded either one of the voiceless fricatives /f, θ, s / or /ð/ or a preconsonantal /n/ (within the same syllable).

Apart from the retention of word-final and preconsonantal /r/, this retention of an ash vowel before fricatives is perhaps the most striking feature of General American types of accent distinguishing them from the usual General British type. This is also a chief characteristic of what Cruttenden 2014 in his Chapter 7 on ‘Standard and regional accents’ categorises as ‘GNE’ ie ‘General Northern English’ (ie of England) a slightly regional General British type of accent. This has been effectively represented since 1990 in all editions of the Wells Longman Pronunciation Dictionary identified  by a “Non-RP” (equivalent to ‘not most general GB’) indicator (from 2000 using the section sign “§”), in his words "Pronunciations widespread in England among educated speakers, but .. judged to fall outside RP...". This term RP (equivalent to General British) he still retained in 2008 in the most recent third edition of his Longman Pronunciation Dictionary (p. xix) Cruttenden in his 2014 edition of Gimson's Pronunciation of English has now replaced it with 'General British'. Roach and his fellow editors of the CEPD (Cambridge English Pronouncing Dictionary) have referred to the term 'RP' since 1997 as 'archaic'. In 2011 in their 18th edition they repeat the comment that, other than with certain placenames "words are given only in the standard [sic] accents chosen for British and American English". They had always exhibited such a policy until in 2006, in the 17th edition, they strayed from their past practice with certain words including one and nothing which suddenly appear with alternant versions that replace /ʌ/ with /ɒ/. (They didnt treat any of our 'ash-keeping' words similarly.)  One and nothing etc continued with their new alternants in the 2011 edition we've mentioned but this time with a worryingly crude inserted comment deplored in our Blog 419.

In groups of words such as those in the lists we give below there are something like twice as many basic (non-compound) words which show agreement between the ash-retaining and ash-retracting varieties as display contrast. On the other hand, when we consider words from the point of view of occurrence in non-specialist daily use, we find that GB items that exhibit the ‘ash-type to pass-type’ change are so common that stretches of speech of a minute or so by any speaker will almost always contain one or more of them. Often listeners to unfamiliar broadcasters will be unaware that the speakers arnt  most-general-type GB users until, only after a minute or so or less, it’s noticed that they have an ‘ash-keeping’ type of pronunciation.

Below we list the ash-retracted (to /ɑ/) common words of General British in relation to ash-retaining United States varieties conveniently termed (ordinary) General American:

GA /æf/ corresponds to GB /ɑf/ in: after, (after)math, autograph, behalf, calf (plurals GA /kævz/ GB /kɑvz/), cenotaph, chaff, craft, crafty, daft, draft, draught, epitaph, giraffe, graph, half /hɑf/ (plurals GA /hævz/ GB /hɑvz/), laugh, monograph, photograph, raft, rafter, shaft, staff, telegraph. Cf the verbs ‘calve’ GA /kæv/ and ‘halve’ GA /hæv/.

Contrast ‘ash’ for both varieties in: affable, baffle, daffodil, gaffe, gaffer, graphic, graphite, Jaffa,  photographic, raffle, snaffle, riffraff, naphtha, scaffolding, Stafford, traffic.

GA /æθ/ corresponds to GB /ɑθ/ in: aftermath, bath, lath, path. Contrast ‘ash’ in both for: Bathsheba, Bathurst, bathysphere, Catherine, hath, mathematics, math(s), osteopath, pathological, polymath, psychopath, Strathclyde,  Kath,  Kathleen.

GA /æð/ corresponds to GB /ɑð /in: baths, lather, paths, rather. Both have ‘ash’ in blather and gather and /ɑ/ in father. (It’s noticeable with quite a few Scottish speakers who have very nearly entirely GB accents that they prefer /ɑ/ in the word ‘gather’. Much less often Scots speakers may be encountered with /ɑ/ in one or two other words such as ‘aspect’ and ‘asthma’. Another such word is ‘mass’ which was also formerly similarly favoured by some Roman Catholics in England).

GA /æs/ corresponds to GB /ɑs/ in: alabaster, aghast, ask, ass (as term of contempt only), bask, basket, bastard, blast, brass, broadcast, casket, casque, cast, caste, caster, castle, castor, clasp, class, contrast, dastardly, disaster, fasten, flabbergast, flask, forecast, gasp, ghastly, glass, grasp, grass, hasp, impasse, last, mask, masque, mast, master, nasty, outcast, outlast, overcast, pass, past, pastime, pastor, pasture, rascal, repast, steadfast, vast.

Names with GB /ɑ/ include: Aldermaston, Belfast, Castleford, Damascus, Glastonbury, Grasmere, Grassington, Madras, Newcastle, Plaistow, Prendergast. The widely known British trade term (equivalent to US Band-Aid) Elastoplast is GB /ɪ`lastəplɑst/ evidently because its final element is an abbreviation of the word 'plaster'. Also /ɪ`lastəplast, ɪ`lɑstəplɑst/.

GA /ænd/ corresponds to GB /ɑnd/ in: Alexander, Alexandra Alexandria, Alexandrine, command, commando, countermand, demand, Flanders, reprimand, slander and some other names including Sanders and Sandra.

Both have ‘ash’ in: Amanda, Andrew, android, band, bandy, bland, brand, candid, Candida, candle, candour, candy, Cassandra, coriander, dander, dandy, expand, gander, Gandhi, germander, gerrymander, gland, grand, hand, land, Mandalay, Mandy, meander, Menander, oleander, pander, panda, philander, Randall, Randolph, random, salamander, Samarcand, sand, sandwich, sandal, Sandy, Scandinavian, shandy, stand, Tandy, vandal.

GA /ænt/ corresponds to GB /ɑnt/ in: advantage, aren’t (when ‘am not’), can’t, chant, chantry, enchant, disenchant, grant, implant, plainchant, plant, shan’t, slant, vantage(-point). Cf aunt.

Contrast ‘ash’ in both GA and GB for: adamantine, ant, anti-, antic, antler, Antrim, Atlantic, banter, Bantry, Blanton, cant, Canterbury, cantilever, canto, commandant, corybantic, descant, elephantine, extant, fantasy, frantic, gallant, gallivant, gantry, gigantic, hierophant, Infanta, Levant, mantle, mantra, pant, panties, pantomime, pantry, pedantic, Plantagenet, plantain, plantation, rant, recant, romantic, Santa, scanty, shanty, sycophantic, tantamount.

GA /æns/ corresponds to GB /ɑns/ in: advance, answer, chance, chancel, chancellor, dance, enhance, glance, lance, prance, trance and names including France, Frances, Francis and Vance.

Both GA and GB usually have ‘ash’ in: askance, ancestor, cancel, cancer, circumstance, expanse, fancy, finance, manse, rancid, romance, stance, sycophancy, trans- and numerous derivatives.

GA /ænʃ/ corresponds to GB /ɑːnʃ/ in avalanche, blanch, Blanche, branch, ranch, tranche.

From the point of view of their historical origins, altho the pre-fricative /ɑ/ words exhibit what was perfectly possibly a phonetic gradual developmental process, the existence of the /an/ versus /ɑn/ groupings is praps more likely t’ve  come about from adoption of one rather than the other of pairs of original French loanwords all of which exhibit earlier spellings with ‘aun’ alongside ‘an’ which suggests that the /ɑ/ type may’ve simply been preferred to the ‘daunt, flaunt, gaunt, haunt, saunter, taunt, vaunt’ type most of which have early 'an' variants. Historically speaking the ‘chance’ type finally settled into their present distribution only in the earlier nineteenth century. John Walker’s Critical Pronouncing Dictionary and Expositor of the English Language in 1791, praps rather conservatively, showed /ɑ/ before /n/ only in ‘-mand’ compounds and ‘aunt, chandler, salamander’ and ‘prance’ but /æ/ in ‘advance, advantage, chance, dance, lance, slander, plant, slant’ etc.

Many of the ash-changed items listed above have ‘unchanged’ alternative pronunciations. For more detailed information than is given above, see §3.1 of the main ‘home page’ division of this website. That article may be of particular interest to quite advanced users of English as an additional language. Cruttenden 2014 is in full Gimson’s Pronunciation of English by Alan Cruttenden.

One other feature that may often accompany ash keeping and is exclusively associated with northerly counties of England is the retaining of the ''bush" vowel /ʊ/ in words like cut, cup, come, does, flood, courage, tough and young which the rest of England, Scotland and in all Englishes beyond the British Isles have converted to an opener unrounded vowel of the type /ᴧ/ or to a schwa. It tends to be felt to characterise northern speech markedly more strongly than ash keeping so that it wd be hardly likely that one wd come across a speaker who exhibited it while not also exhibiting ash-retaining. Speakers moving away from a Northern milieu who tend to adapt their speech seem more likely to convert from bush-retaining than from ash-retaining. 

20th Century English Phonetics (ii)

The third volume of this monumental set of reprints lists BBC “Recommendations to Announcers Regarding the Pronunciation” of seven groups of words. Here’s a list of them with the number of pages for each entry: English Doubtful Words 40, English Place-Names 70, Scottish Place-Names 38, Welsh Place-Names 32, Northern-Irish Place-Names 12, Foreign Place-Names 51 and British Family Names 92. It was too gratifying to LJ’s (Lloyd James’s) BBC boss John (later Lord) Reith to have secured the Poet Laureate Robert Bridges as the Committee’s chairman for LJ to dare suggest getting rid of him. Bridges flattered himself that he was an authority on the English language. This inevitably made him an acute embarrassment to LJ and prob·bly to most of the other committee members. C. T. Onions, the OED editor, served on the BBC Advisory Committee from 1930 to 1934. He exprest to LJ in 1932 his dismay at their ‘insufficiently rigorous procedure’ and indicated that he me·nt to bow out ‘silently without giving you any trouble’. He later revealed his disillusionment with Reith’s committee saying ‘It is odd that in no other department than language ... would the distinguished amateur be tolerated’ according to Asa Briggs’s The Golden Age of Wireless (1965).

The inclusions in the first booklet on words of “doubtful pronunciation” had all “caused difficulty to announcers” or “given rise to criticism from listeners”. Plenty of the recommendations regarding them have long ago needed to be withdrawn. Some, even so, survive into the twenty-first century eg /ˈɔːld ˈlæŋ ˈsain/ for Auld lang syne, a pedantic interference unfortunately included in the OBG (the 2006 Oxford BBC Guide to Pronunciation) tho people usually these days say 'Old Lang Zyne' assimilating the initial /s/ of the Scottish word 'syne' (for 'since') to the last (voiced) consonant of the Scottish version of 'long' in a way that irritates a few intolerant Scottish people. Others now of advanced obsolescence or completely obsolete included /əˈkaustik/ acoustic, /anˈtʃouvi/ anchovy, /ˈberit/ beret, /broˈʃuə/ brochure, /kaiˈrɔpədi/ chiropody, /ˈkʌmbət/ combat, /ˈkᴧndit/ conduit, /kouˈinsidəntli/ coincidentally, /ˈdekəd/ decade, /diˈkɔːrəs/ decorous, /ˈdispjuːtəbl/ disputable, /etiˈket/ etiquette, /ˈfɔːkən/ falcon, /ˈfɔred/ forehead, /ˈfjuːzilidʒ/ fuselage, /ˈpiːʤərətɪv/ pejorative, /inˈhiərənt/ inherent, /iˈrefjuːtəbl/ irrefutable, /ˈproufiːl/ profile, /ˈpristain/ pristine, /ˈrestəroŋ/ restaurant, /ʃiː/ ski, /səˈnɔːrəs/ sonorous, /spontəˈniːiti/ spontaneity and /ˈziːbrə/ zebra.

A notable oddity of the LJ phonetic transcriptions of English words is the inclusion of all the r’s of the orthography never actually he·rd until decades afterwards from any BBC announcer and even then from very few indeed of the numerous official voices featuring in BBC broadcasts. The explanation given was the quaintly ambitious-sounding “.. this booklet is intended not only for South-Eastern speakers but for the whole English-speaking world..” LJ remarked at page 7 oddly “Announcers are not required to use Northern English”. He sed this by way of pointing out that Newcastle was recommended to be pronounced /ˈnjuːkɑːsl/ tho, adding with continued ponderous jocularity, “if he happens to be a native of the place and pronounces it in the native way, he will be forgiven”! “He” because there were, of course, no women announcers until decades later!

The placenames are full of spellings people continue preserving centuries after they’ve become absurdly inappropriate as any guide to how they now say them. Althorp, the ancestral home of the late Princess Diana is a famous example. Announcers were enjoined to pronounce it as /ˈɔːltrəp/ and that edict held until “In 2000 the estate released a press statement saying that henceforth it would be known as awl-thorp” whereupon the BBC “changed our recommendation accordingly” as was noted in the OBG. This was partly prompted by complaints in the press at the patent absurdity of the Spencer family’s saying it one way and spelling it another. In many cases the BBC sanctioned two spellings as for instance with Barnoldswick with a first recommendation /bɑːrˈnouldzwik/ and the alternative /ˈbɑːrlik/. Extreme cases were Slaithwaite with its three alternatives to /ˈsleiθweit/ namely /ˈslæθweit, ˈslouit & ˈslauit/ and Uttoxeter with /ᴧtˈɔksitər, juˈtɔksitər,ˈᴧksitər, ˈᴧkstər & ˈᴧtʃitər/.

Only half-a-dozen (uncompounded) non-proper words in the English language have the letter combination gh representing the sound /f/ but in this very limited list about three times as many of the entries have that feature. One name /ˈbəːrfəm/, which actually has at times had the spelling Burgham, here appears only as Burpham bizarrely spelling its /f/ not with gh but, as if it were derived from Greek, with ph. Philleigh /ˈfili/ even has ph initially in a non-Greek-derived word. Only one word has /θ/ for its gh: Keighley /ˈkiːθli/. Other curiosities include such items as as Ardingly, Bellingly and Chiddingly which look as if one might treat them like the adverbs with unstrest final syllables but they turn out to be strest /ɑːrdiŋˈlai, beliŋˈlai & tʃɪdɪŋˈlai/.  Strange too is Irlams o’ th’ Height required to be /ˈəːrləmz o ˈðait/ in which the initial aitch of Height was forbidden to be sounded in an amazing apparent rare concession to some local sensitivities. His phonetic symbol /o/ represents rather allophonicly the pronunciation of an /ou/ so speeded up that it lost its diphthongality, as it was suggested cd happen at the first syllable of molest.

Besides orthographical freaks like Cholmondely for /ˈtʃᴧmli/ (on which see also our Blog 018),  Clapworthy for /ˈklæpəri/, Happisburgh for /ˈheizbərə/, Scafell for /ˈskɔːˈfel/, Wrotham for /ˈruːtəm/ and Wyrardisbury for /ˈreizbəri/ there are phonological freaks like /buːldʒ/ for Boulge and the doubly odd (orthographically and phonologically) Ebbw /ˈebu/ with its w representing a vowel (that of put ) and one which ends no English nouns in current inconspicuous GB at least. A large number of the greatest oddities are to be found where the linguistic and political borders with Wales and Scotland dont coincide and in the far southwest mainly in Cornwall. The other booklets must await a later blog.

Weakforms (xiii) He

The pronoun he was included in 1885 by Henry Sweet in his Elementarbuch des Gesprochenen Englisch first-ever list of sixty-odd weakform-possessing words he identified under the heading ‘gradation’. This referred to the alternation between phonemicly, as we shd say now, contrasted variant forms of a certain class of words reflecting their differing degrees of prominence. The more prominent variants he termed ‘strong forms’ and the less prominent (normally unaccentable) ones ‘weak forms’.

There was a certain discrepancy between Daniel Jones’s comments on he in his EPD’s (ie English Pronouncing Dictionary’s) editions from 1917 to 1963, and those of his Outline of English Phonetics from 1932 to 1956. In the former the entry always re·d “hiː (normal form), iː, hi , i (frequent weak forms)”. It must be remembered that Jones used /iː& i/ for what it has become, after Gimson, more often preferred latterly to represent as /iː& ɪ/. In his Outline §487 he listed the same forms but added a cross reference to its section §472 which re·d “In the case of he and we, the strong forms hiː, wiː are commonly used in unstressed positions; weak forms hi, wi exist but are not often used”.

The fact is that the effects of occurrences of these alternating weakforms can’t be satisfactorily described so briefly. The selection of one or the other of them depends on matters like the degree of colloquiality and prosodic and segmental contexts. For example /hɪ/ used prepausally in eg /`ˏɪz hɪ/ Is he? now tends to sound Conspicuous and/or oldfashioned GB. On the other hand /hɪ `ɪz/ for He is wd be much less conspicuous or pass unnoticed becoz an anticipative assimilation may convert the vowel of he to a value at least close to tho not necessarily exac·ly the same as that of the following is. And /hɪ `dᴧz/ for He `does, in which the he wd usually be of minimal length, tho no dou·t less common than [hi `dᴧz] wd sound more or less inconspicuous. It seems that qite a lot of speakers tend to use /(h)ɪl/ for he’ll.

Within that Outline §472 which we have just quoted no mention occurs of the lengths of the /hiː, wiː/ strongforms but earlier in that paragraph, with regard to me and she, we find the comment  ..the use of the strong forms miː, ʃiː (with vowel shortened owing to lack of stress) is also quite frequent in unstressed positions. An example “she said so” is added remarking of it “though usually said with ʃi, may also be said with ʃiː, even when the word is quite unstressed”. These shortened prosodic variants were exactly comp·rable with what in the latter  twentieth century came about in the conversion of the earlier GB (RP) mainstream value for the final unstrest vowel of words like happy from /-ɪ/ to /-i/.

When Gimson in the first edition of his Introduction to the Pronunciation of English (1962) §10.04 came to list the “most common” words with strongform/weakform alternation he gave unaccented he as “/hɪ, iː, ɪ/ ([hi])”. This last square-bracketed notation acknowledged the existence of an aitch-keeping weakform that had the close-vowel character of the strongform without the length that’s normally taken to be a characterising feature of its phoneme’s identity. What it failed to do was make clear exactly what its addition designated — whether [hi] was a norm or only a possible variant. The fact is that this type of close articulation with weak value is exhibited in a variety of circumstances in tokens of both the /i/ and /u/ ie the street and root vowels, notably word-finally and word-internally before vowels. Representations of this feature of GB phonology have produced a number of disharmonies between the transcriptions of various writers. The cat was thrown amongst the pigeons when the Longman publishing company put its great weight behind the practic·ly universal transfer, in the General British EFL sphere, from the Jonesian intrinsically length-marked vowels to the Gimsonian redundantly length-marked transcriptions of them in 1978.

My impression has been that, once trapped into the gen·ral lengthmark-keeping style, users much preferred the lengthmark-dropping /`hapi/ to the strictly phonemic /`hapiː/ partly at least becoz this latter version was perceived as a socially dubious — if not dialectal — variant pronunciation. This was not because of the phonemic identity of the final vowel but because of the prosodic value it suggested. (In forms of you, notably in an expression like Thank you, the version /`θaŋk jʊ/, at least when firmly spoken rather than muttered etc, has tended to parallel the development [`hapɪ→`hapi(:)] in becoming CGB (socially conspicuous GB). Compare Cruttenden-Gimson 7ed p. 268 with 8ed p.275.)

I’m afraid I’m not happy with the new treatment in the Cruttenden-Gimson eighth edition of the vowel of the aitch-keeping variant of he \hi\ and the final vowel of \hapi\ happy as examples of an allophone of /ɪ/ rather than “an allophone of /iː/ conditioned by accent and position”. I perceive it as just that. So (in lengthmark-free phonemic transcription) hungry as /`hᴧngri/ and pedigree as /`pedəˌgri/ seem more satisfactory transcriptional solutions than resort to such a counter-intuitive expedient as accepting [i] as an allophone of /ɪ/.

PS: I use forward slashes /.. / around strictly phonemic transcriptions but backslashes \ ..\ around sequences not phonemic thruout.

BBC English PS 25

Please go to §4.1 in the main part of this website when you wish to access the sound file of this dialog.
The text is:
1. The English language has gone to the dogs these days on the BBC.
2.  I wouldn’t say that.
3. You used to be able to count on good English clearly spoken.
     Not now you can’t, though.
4.  There’s a slightly wider spectrum of accents as you’d expect.
     Fewer posh ones. More classless ones.
5.  Half of them sound downright uneducated.
6.  That sounds like prejudice to me.

This short dialog, of only 60 words, gives us plenty of things to discuss. It has a speaker expressing  sentiments that might be he·rd forty or fifty years ago from many middle-aged people and might be encountered even today from a few. Looking at our transcription you may notice that, unlike the usual one for General British pronunciation in EFL circles today, it omits the lengthmarks that regularly accompany the five original simple-vowel phonemes /iː, ɑː, ɔː, uː, ɜː/. Cruttenden’s latest eighth edition of the Gimson book has made a timely updating of that transcription by substituting /ɛː/ for the former /eə/ and /a/ for the previous /æ/. I rather deplore the retention of Jones’s lengthmarks by Gimson in his replacement for the original EPD transcription in 1978. The reasons he gave for this move were that it made for better legibility than the lengthmark-free versions and preserved a degree of continuity. This is perfec·ly true for the native speaker of English and the highly expert user of English as an additional language. For them it must make recognition of those phonemes much quicker and easier and so speed up reading of the transcription. However, this was at a very unfortunate cost to those learners who are less than fully aware of the actually very variable lengths that occur in English pronunciation. Learners will be no dou·t by far the largest number of those seeing and using the Gimsonian converted transcription for the EPD and almost the only users of the great variety of textbooks that very soon converted their transcriptions to the new style introduced by the justly most highly respected and admired authority of the day on the British pronunciation of the English language. However, the fact is that, to the learner, these lengthmarks convey, if anything, very false ideas of the realities concerning what lengths those five or six vowels may have in ord·nry spoken English whether conversational or formal. The fact is that they are truly regularly pronounced fully long only when occurring in the accented syllables of words uttered in their lexical or isolate pronunciations when any following sound occurring in their syllable is not /p, t, k, ʧ, f, θ, s / or/ ʃ/ and/or no further, weak syllable follows in the word in which they appear. Otherwise they may be of any length mainly according to the inclination of the speaker and indeed do vary very greatly.

The Gimson replacement transcription was not questioned or objected to at all much but in fact remarkably readily embraced in almost all quarters. However there was one sign of dissatisfaction with the misrepresentation of length in one very numerous set of words — those like happy which required under the new regime to be transcribed /hæpiː/ at the time when mainstream GB was increasingly observable as using [hapi] while the traditional (RP) /hæpɪ/ with its diff·rent final phoneme was rapidly becoming perceived as oldfashioned and/or socially conspicuous. This minor revolt came rather paradoxically from the source which had been at the forefront of those abandoning the old Jonesian transcription in favour of the new Gimsonian style. Its first steps were taken with the excuse (praps one might say disguise), in the Longman flagship EFL publication, its LDCE (Longman Dictionary of Contemporary English 1978), in the form of a suggested space-saving common symbolisation to cover simultaneously British and American pronunciations of that final vowel of happy etc. In subsequent editions the device came to be explained in more than just that way. In 1990, twelve years later, when the LPD (Longman Pronunciation Dictionary) appeared, Wells remarked that he “judged it best to stick with the system which in recent years has at last provided a de facto standard in EFL work..”.  He, however, acknowledged making some minor deviations from it saying at its Introduction §3.4 (i) “The symbols i, u are used for iː, -ɪ, uː, -ʊ” in what he termed “positions of neutralization”. Whatever else, what this did do was remove numerous cases of glaringly inappropriate length marking which had characterised the unmodified adoption of the Gimson transcription.

The following is a strictly phonemic transcription with no lengthmarks accompanied by very approximate tone indications. The bracketing of the symbol /d/ at Turn 5 indicates uncertainty or ambiguity:

1 A  ði ˈɪŋlɪʃ ˏlaŋɡwɪʤ ⎪ əz ˈɡɒn tə ðə `dɒɡz ðiz ˏdeɪz⎪  ɒn ðə ˌbi bi ˏsi
2 B  ˈaɪ ⎪ wʊdn seɪ ˎˏ ðat
3 A  ju ˈ jus tə bi ˏ eɪbl ⎪ tə ˈkaʊnt ⎪ ɒn ˈɡʊd ˏ ɪŋɡlɪʃ ⎪  ˈklɪəli ˎspəʊkən ⎪⎪  ˈnɒt `naʊ ju ˏkɑnt ðəʊ
4 B  `ðəz ə slaɪtli waɪdə spektrəm əv `ˏaksnz ⎪ az jud ɪk`spekt ⎪⎪ `fjuə ˎpɒʃ ˏaksnz ⎪ ˈmɔ `klɑsləs wʌnz
5 A  `hɑf ˏɒv ðəm ⎪ saʊn(d) daʊnraɪt ʌnˊ`edjukeɪtɪd
6 B  ˈðat⎪ saʊnz laɪk ˊ`preʤudɪs ⎪⎪ tə ˎmi

Comments:

Turn 1 has the contrastive accent one wou·d normally have expected to occur on these delayed to the following word — tho not with an effect that a liss·ner wd positively find strange. BBC in strict Gimsonian style wou·dve had to have the rhythm /ˌbiː biː ˈsiː/, which is the sole version given in CEPD (the Cambridge English Pronouncing Dictionary) but current predominant usage is most usually what LPD supplies, tho only as an alternative, viz \ ˌbiː bi ˈsiː\. Our speaker here has the very normal three approximate lengths — respectively medium, short and long.

Turn 2 shows elision of the /t/ of the negative suffix -n’t which from any word so ending is very common even when a vowel is the next sound. (It’s not normal tho, in GB, at the end of a prososdic phrase). The vowel /a/ of that receives no lengthmark in the Gimsonian style but it’s very offen pronounced as long as any other vowel gets to be. Rather long here.

Turn 3 begins with two monosyllables both with the “long” vowel /u/ with perfec·ly normal very short values — the first from being unstressed, the second from occurring in a syllable closed by the voiceless consonant /s/. The next word, /biː/ which is lengthmarked in  Gimsonian is dou·tless as short as the speaker can make it. Another very short /u/ occurs in the you before can’t which word has an /ɑ/ of only medium length. We have at the very end of this turn at the word though a shortening of the diphthong so extreme that out of this context one might not recognise it at all and be inclined to transcribe it as [ʊ]. Not that people wd hear that as strange. The late Roger Kingdon preferred to transcribe many words such as follow with a final /ʊ/.

Turn 4 at its first word it has an accented schwa whose common existence I’m rather weary of pointing out yet agen in the face of repeated assertions by so many highly respected authors that such a thing doesnt exist in GB.  In its last phrase it has a quite short /ɔ/ at more and an only medium length /ɑ/ in classless. The version of accents, which occurs twice, we shd normally expect to see as /aksn(t)s/ but in fact we clearly hear something that rhymes with Saxons. Even tho it’s such an unusual idiosyncratic form of the word, this’d prob·bly pass unnoticed by most lisseners. I’ve noted the famous broadcaster Lord Melvin Bragg (whose speech reflects his Cumbrian upbringing to some extent but hardly in this respect) repeatedly pronounce the word science exactly as scions /saɪənz/ rhyming it with lions. Incident·ly the /n/ of accents is notably long both times. This suggests asking Gimsonian transcribers, completely reasonably if slightly mischievously, why the Gimsonian transcription shd not’ve included (comparably non-phonemic and redundant) lengthmarks on cert·n consonants.

Turn 5 has a hardly more than medium length /ɑ/ at half tho it’s markedly accented. It sh·d be noted that becoz the intonations supplied are tonetic rather than tonemic, the Rise tonemark is placed where the rising movement actually occurs. Had the notation been tonemic, the speaker’s choice, not of two simple tones but a single falling-rising complex tone wou·dve been indicated at the accented word half only, leaving the unaccented word of unmarked. The fact that the speaker used the strongform /ɒv/ of of didnt mean that he was accenting the word, but merely that he chose the slightly less rapid of the two alternative rhythms available to him. The faster one wou·dve been selected by choosing to use the weakform /əv/ of the word of. The tonemic notation `ˏhalf of them (or `´half of them), if ord·nry (non-phonetic) spelling had been used, wou·dnt’ve revealed which of the two possible speeds the speaker had selected.

Turn 6 has a slightly ‘detached’ ending. The speaker’s prosody suggests that his remark has ended at prejudice but the notably less vigorously uttered following words to me sound like a pensive afterthought.

In the main division of this Website at §5.5 the reader may see my 1975 article on 'The undesirability of colons in EFL transcription' first publisht in the Journal of the International Phonetic Association

20th Century English Phonetics (i)

Last year, saw the publication by Routledge of a set of six large, weighty (about a kilogram each), solidly bound, handsome (and correspondingly expensive, suggest your library gets’m!) volumes entitled ENGLISH PHONETICS: TWENTIETH CENTURY DEVELOPMENTS. Its editors were Beverley Collins, Inger Mees and Paul Carley. This set belonged to a series by Collins & Mees of which the previous item was a seven-volume set PHONETICS OF ENGLISH IN THE NINETEENTH CENTURY (2007) which also belonged to a uniformly styled valuable series begun with their DANIEL JONES: SELECTED WORKS of 2003. The present new set of these photographically reproduced copies of historic texts are labelled I to VI thus “I: J. A. Afzelius’s pronouncing dictionary, II: Arthur Lloyd James: Broadcasting and Spoken English, III: Arthur Lloyd James: Broadcast English, IV: English phonetics, V: Landmarks in the study of English intonation and VI: Phonetics of English as a foreign language. Except for items by two Continental precursors, they all consist of works with some degree of association with the Daniel Jones UCL school.

The first volume of this set is devoted entirely to republishing for the first time in over a century the truly pioneering but long forgotten lexicographical achievement, some years before anything of the kind from Jones, of the 1909 Concise Pronouncing Dictionary of Modern English by Jon Arvid Afzelius. He was a Swede who taught, and wrote textbooks on, English at the then Gothenburg Business Institute. The Collins-&-co prefatory notes observe that its 472 pages (smallish originally but reprinted at double size with improved clarity) with “around 24,000 headwords” were “by any standard a remarkably high total for a so-called concise dictionary” and “all the more notable for being the work of just one man.” I was unaware in the 1970s of even the existence of the Afzelius book so it was something of a coincidence that, when I came to compile my Concise Pronouncing Dictionary of 1972 for OUP my aim was that it shd be approximately half the size of the Jones EPD of the day and so it came, by something of a coincidence with the Afzelius, to contain about 24,000 entries (in my case each with at least one GB and one GA transcription).

Afzelius used an adaptation of Henry Sweet’s phonemic-equivalent ‘Broad Romic’ notation. He checked his impressions with the Oxford Dictionary, which by then was about half completed, as well as with the works of various other English phoneticians of the day. He curiously mentions no de·t to the Dane Otto Jespersen whose extensive contributions to the Dictionary of the English and Dano-Norwegian Languages by John Brynildsen (1902-07) constituted the most complete body of English pronunciations in modern phonetic transcriptions before Jones’s works and can hardly have not been known to Afzelius. He had in the 1880s belonged to the International Phonetic Association for some years during which Jespersen had been a leading figure. When one considers the immediate enthusiastic reception Jones’s EPD seemed to’ve received on its publication, it’s rather surprising, as the editors remark, that such a useful book wasnt immediately taken up by the EFL world beyond Sweden. The only dedicated complete pronunciation dictionaries of English available at the time had been the grossly out of date (little changed since 1791) Walker’s Pronouncing Dictionary and the 1836 enlargement of it by Benjamin H. Smarte who called his revision Walker Remodelled. They continued in print into the early twentieth century.

The second volume in this set contains reprints of three sep·rate short books by Arthur Lloyd James. The very short (47 pages) Speech Signals in Telephony came about as a training manual in the matter of “radio-telephonic speech” for members of the Royal Air Force in 1940. Among the people he thanked for their cooperation in his RAF work were Squadron-Leader D. B. Fry who was later to succeed Daniel Jones as University College London Department of Phonetics's second ever Professor of Phonetics. Another was L-G’s son Flying Officer D. Lloyd James who was much later, as a BBC staff member, to invite me to broadcast on the BBC. While still at school aged about fourteen I came across in a public library, as my first-ever ecounter with English phonetics, a modest book of 176 pages Our Spoken Language in which Lloyd James gave first an explanation of the speech mechanism and then a concise and very readable description of (GB) English sounds and prosodies. Re-reading it now, despite admiring the clarity of its style, one’s bound to find its tone in places embarrassingly patronising. Of course its assumptions were for the most part simply those that had been typical of the middle classes since Victorian and earlier times.

The third part of this second volume reprints the 1935 book entitled THE BROADCAST WORD in whose 207 pages were republished (with a joint Index etc) seven “talks, lectures and essays”. Their titles were: I. The Broadcast Word (pp 21-27). II. The B.B.C Advisory Committee on Spoken English (pp 29-73). III. Speech in the Modern World (pp 75-129). IV. Speech and Language in the World today (pp 131-150). V. Standards in Speech (pp 153-172). VI. On Reading Aloud, with Special Reference to the Bible and the Prayer-Book (pp 173-199). VII Some Thoughts about Minority Languages with special reference to Welsh. (pp 193-219).

Among his most frequent topics was his deploring of so-called “intrusive r”. This nineteenth-century term was unwisely adopted by Daniel Jones: A. J. Ellis had called it more sensibly “euphonic”. See the discussion on this website at Division 6 §8. Lloyd James struggled to dissuade the BBC announcers under his supervision from using what is now so universally accepted as to cause surprise that it cou·d ever have been objected to. He acknowledged use of pronunciations like “sonatar in A” (page 117) to be an ‘ingrained’ habit and (at page 107) an “established feature of so-called Standard English” yet condemned them as “unseemly”(page 118) and “deplorably common” (page 183). He indulged his prejudices in other pronunciation matters as well. He never fully acquired the toleration and discretion and thoro·ly scientific outlook that characterised the mature writings of Daniel Jones, but he was a notable phonetic figure whose work deserved to have its availability restored. The editors provide excellent informative comment on him in their gen·ral introduction to the volume (pp1-8).

Intonation/Accentuation Points

An old fre·nd of this Blog in Japan, Emeritus Professor Tami Date, recently enquired about a recording he heard from an audio CD that accompanies an American EFL textbook. Tho it wou·d be nice to, I don’t think you need to hear the items he’s talking about to follow the discussion and appreciate the points being made.

He as·t questions about the following items.
I’ve lissend to the recordings of them and I’ve added the intonations I perceived.
1 Do ˈyou ˈwatch ￨ TˏV?
2 Do ˈyou ˈlike ⎟ watching TˏV ⎟ in your ˈfree ˏtime?

Tami remarked:
It sounds like, after 'do', the pitch goes up on 'you' with prominence?  So the auditory impression that I get is that 'you' is contrasted with some other person.

I replied:
When you say “it sounds like, after 'do', the pitch goes up on 'you' ”, you’re right but the way that puts it is a bit ambiguous. So I’d’ve been happier if you’d said at ‘you’ or to (instead of on) ‘you’ ”. ‘Do’ has lowish pitch but ‘you’ is accented by having been given contrasting high pitch. There’s no movement upward during this word ‘you’ which in itself stays at a single level pitch. The fact that it’s higher, of course, me·nt that a rise in pitch had occurred. That it is accented does tend to suggest a contrast of some sort tho not at all necessarily strong or with reference to any specific person.

In some other situation a speaker could accent ‘you’ for a contrast with another person.
For example one might say:
ˈI’m ˈhaving a cup of `tea. Do ˈyou preˈfer ˎcoffee?
Tami added:
Most Japanese teachers like me would probably pronounce 'you' with a relatively low pitch till we get to the main verb or the adverb.
My reply was:
That’s just simply a perfectly okay alternative.
He also sed:
…[P]lease explain why the pronoun is apparently accented.
I replied:
A word is usually accented by a speaker merely to give it a bit more attention or prominence ― and that’s no big deal. Sometimes people even accent a word simply to make what they’re saying more lively in general ― what for many years I’ve been describing as using ‘animation stresses’.

A colleague James Kirchner commented that the prosody used wou·dnt sound natural in ord·n·ry conversation. I agreed and mentioned that I thaut the tempo is too slow in particular. Of course most learners wd be fairly well aware that such a style is partly adopted to be extra clear in order to help them to imitate it properly. In ord·n·ry conversation most naturally the first two words wd usually be replaced by the single contraction “D’you” which most often wou·dnt be accented as for instance like this:
D’you ˈwatch ˈT´V?
or, a little faster, not bothering to accent the T of TV:
D’you ˈwatch T´V?

And if a strong contrast between two persons was intended, as James suggested, the speaker wd be most likely to use a different intonation, usually a falling tone eg
Do `you watch TˏV?  or   `D’you watch TˏV?

Here are the questions with intonations marked as they’re to be he·rd on the recording:

1 Do ˈyou ˈwatch ￨ TˏV?
2 Do ˈyou ˈlike ⎟ watching TˏV ⎟ in your ˈfree ˏtime?
3 ˈDo you ˈgo to ˏclubs?
4 Do ˈyou enˈjoy ˏclubs?

5 Do ˈyou ˈeat ˎdinner ⎟ in ´restau[´]rants?
6 Do ˈyou ˈeat ˏout some[´]times?

7 Do ˈyou ˈever  ˈplay ˏsports ⎟ on ˏweekends?
8 Do ˈyou ˈplay a ˏsport ⎟ on ˏ[ˈ]weekends?
9 Do ˈyou ˈvisit ´relatives ˏoften?
10  Do you ˏoften ⏐ visit ´relatives?

For anyone who doesnt find that the simple intonation markings dont make things fully clear I offer detailed explanations below.

 In what follows I use “m” to stand for any syllable.

If the first word or words in a tone phrase aren’t marked they are to be taken as uttered on a lowish pitch.

ˈm = high (level) pitch.
Any subsequent ˈm in a tone phrase will indicate a step down to a slightly lower pitch as happens to     ‘watch’ in the first line

This vertical bar ⏐ signals (a change to) a new tone phrase within a sentence (so the next unmarked word will have lowish pitch).

ˏm = lowish rise. ´m = highish rise.

[´m] I’ve enclosed in square brackets extra tone movements that wdnt be classified as accents coz they merely smoothly renew a movement begun at an earlier really accented word eg the word ‘out’ in #6 which in itself doesnt move but does constitute the beginning of a rising tone ― as we gather from hearing the next syllable a little higher.

In 5 the notation in ´restau´rants? means that the speaker had a second higher rise on the last syllable that one wou·dnt classify as employing a diff·rent new tone but as either as renewing the rise begun earlier or constituting a complex single tone (a Rise-Climb in my terminology).

In 8 the notation ˏˈweekends conveys that the speaker didnt use a new tone but levelled out at the top

of her rise thus using a single complex tone (in my terminology a Rise-Alt)

My full system of tone marks is expounded at §8.3 and §8.5 on this website.

Festschrift for Professor Hyun Bok Lee

Alan Cruttenden's Gimson 8th Edition

      

Hangover Lodge PS 24

This is the title of #24 of the dialogs of my book People Speaking for which with this present item I continue to, as was promised, provide further phonemic-tonetic transcriptions, with comments, for the use of advanced students of spoken English.  If possible, it’s best to transfer the contents of the sound file into Audacity (the freeware audio facility) to be able to select convenient slices for repeated playback. Section 4.1 provides descriptions of the tones to be found in the avowedly broad ie 'unsu·ttle’tonetic transcription.

1. /ˌsəʊ ju av `-klɑsɪz | frm ˈnaɪn | tə `-wᴧn | ˈsɪks ˈdeɪz | ə ˎwik /
2. /ˎ jes | ɪkˈsep fə ðə wik | wen ðɛz ən ˈɔl deɪ trɪp | tə ˎstratfəd /
3. / ˈwɒt də ðeɪ ˈdu | wɪ ðə `rest əv ðə deɪ /
4. / `ˈjuʒ(l)i | ðɛ ˈfri tə du wɒt ðeɪ `laɪk |
5. / ðə(r)ə `lɒts əv | vɒləntri ak`tɪvətiz leɪd ˏɒn |
6. / `ðeɪ kŋ ˈgəʊ fə `wɔks | ɪf ðə ˌweðəz ˏfaɪn |
7. / ɔ ˎʤᴧs sɪt ɪn | ð(ə) ˏgraʊnz | ə (ð)ə `hɒstl |
8. / ðə `siriəs ˏwᴧnz | raɪt `letəz ˎhəʊm |
9. / m̩ `praktɪs | fənetɪk trn̩`skrɪpʃn /
10. /an(d) ðə ˏfrɪvələs  ́wᴧnz /
11. /ðeɪ ˈkɒŋgrɪgeɪt `-naɪtli | ət ðə ˈred `haʊs / ɑ ˈləʊkl `pᴧb/
12. / ˈwɒʧu ˈkɔl `(ð)ɪs pleɪs/
13. /ˈhanəʊvə `lɒʤ /
14. [ɦə ɦə] /aɪm nɒt səˎpraɪz | ɪf ˈɔl ˈðat | ˎbuzɪŋ | gəʊz ɒn ˏðɛ /

On this occasion no indications of vowel length are offer·d such as appear redundantly in the common Gimson-style transcriptions of the main pronouncing dictionaries etc. The symbol /ɛ/ represents the phoneme more traditionally shown as /ɛə/ or /eə/ but which is now increasingly recognised as not typically diphthongal in current GB but most offen [ɛː]. The vertical bar ( | ) signals a break in the rhythmic flow which will usually may be short or very slight indeed. The pitch, unless markt otherwise, drops after a bar to the lowish prehead value of the first syllable of any new utterance not markt otherwise.

At line 1 the aitch-dropping at the word 'have' is not unusual in informal conversation where, as here, the word is minimally strest in very fluent utterance. The Fall-Mid tones `- at 'classes' and 'one' indicate a restricted descent from the top third but on·y to the middle third of the speaker's ord·nry voice range.

In line 2 the elision of the final /t/ of 'except' is completely ord·nry before the following consonant.

At line 3 the form /wɪ/ of 'with' doesnt create the impression of its being a casual weakform becoz the mere simplification of the sequence of the two occurrences /-ð ð-/ by eliding the first of them sounds completely ord·nry. Before most other consonants that were beginning a following word, /wɪ/ wd prob·bly sound quite casual.

At line 4 the bracketed (l) is used to indicate a 'ghostly' (or even dou·tfully present) rather than really firmly articulated segment. In this case an /l/ appears to be used in 'usually'. The tone mark `ˈ at the word 'usually' is me·nt to convey a narrow fall in the uppermost third of the speaker's voice ran. I call it a Fall-Alt. (More at my website main section 8.5.8).

At line 5, the asterisked letter, in this case (*r), at the first word 'there’re' is used with the same sort of meaning as in line 4.

At line 6 the assimilation of the weakform /kn/ of 'can' to /kŋ/ before the closely following /g/ in 'go' is completely ord·nry.

At line 7 the asterisks at the occurrences of the indefinite article 'the' and the word ‘of ’ are also used as in line 4.

In line 8 the strest syllable of the word 'serious' is represented as having the vowel phoneme of words like 'street'. This p·onunciation is increasingly commonly he·rd from Gen·ral British speakers but the p·onouncing dictionaries are rather slow to catch up with the fact, currently on·y giving /ɪə/ for it.

At line 9 the syllabic /m/ with which it begins is converted from the normal syllabic /n/ which is a very common weakform of ‘and’ by a perfec·ly ord·nry assimilation to the bilabial consonant which begins the following word. Note the extra vigour of the humorously truculent 'explosive' (praps contemptuous) manner in which this word 'practise' is uttered. The syllabic /n/ in the prefix of 'transcription' is a variant pronunciation much less offen he·rd than /a/ or /ɑ/ plus /n/ in this word.

At line 10 the use of the weakform /an/ of 'and ', if that's what we have, wou·dnt sound in the least casual. The two rising tones neednt be taken as at all implying any pitch or rhythmic discontinuity between the two words, just a smooth transition from low to high.

At line 11 the form /ɑ/ of our is a weakform for most GB speakers but many have it as their only pronunciation for the word. Anything stronger such as /ɑʊə/ wd be inclined to sound unnaturally careful or formal.

At line 12, a much more usual stressing of this sentence wd be as ˈWhat d'you `call this place?’ Which isn· to suggest that there’s anything unnatural about the way she did say it, even in her use of /wɒʧu/ given the fluency and the informality of the occasion.

At line 13 she ovvisly thinks she's he·rd  /ˈhaŋəʊvə `lɒʤ /. Or jokingly pretends so.

At line 14 the lack of the past-tense marker /d/ at the end of 'surprised’ isnt surprising. The two initial voiced-aitch symbols etc are used just to give some vague impression of the indistinct vocalising she produces.

BlogRoll

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