Archive 23 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

Syllabic etc l's and Anaptyctic schwas

One can commend to EFL students the formulation in John Wells’s blog of today “The general rule in English is that when you form an adverb in -ly from an adjective ending in l you simplify the two l sounds to one. Thus fully is 'fʊli ...  He continues “In addition, in an adverb formed from a stem ending in syllabic l the syllabicity is lost...This means that for wholly we would normally expect only a single l sound.” He  explains that wholly is exceptional because it’s of’en /`həʊl.li/. In fact there are quite a few exceptions including palely /`peɪl.li/, solely /`səʊl.li/, vilely /`vaɪl.li/ and various items formed from adjectives with the suffix -ile like docile, facile, fertile, futile, hostile, juvenile, puerile, senile, servile and versatile. These he seems a bit hesitant about, saying “when we form a nonce -ly adverb from an adjective that is not usually made into an adverb in that way, double l may be preserved”. I shd say that, altho in ordinary speech these may well take weakforms within fluent phrases that simplify the double /l/ to a single one (colloquially I feel like... is extremely offen /aɪ `fiː laɪk... /) in ordinary unhurried conversation I shd expect them to be regularly he'rd with two l’s. This is a tricky area of analysis on which the dictionaries show quite a bit of variation in their treatments. No dou't he’s right to suggest that it’s relatively unusual to use adjectives ending -ilely. Tho I suspect that they’re more avoided by writers, for their awkward appearance, than by speakers. An example in literature comes in the Keats poem La Belle Dame sans Merci in which a knight awakes to find himself on a "cold hill’s side" and waits "palely loitering". In that one can safely expect /`peɪl.li/ to be used.

By the way, The Holy Oral Method, the title of one of my People Speaking dialogs which you can listen to on this website, (§4.1.26) contains a pun on wholly almost as creaking as in the title and picture heading to the Wells blog we’ve  quoted. 

Advanced EFL learners and their teachers may from time to time find they notice from perfectly respected British speakers what I call ‘creeping anaptyxis’ occurring to words that have traditionally not contained schwas but are given them by an increasing number of “careful” speakers. An example is the pronunciation of struggling as /strᴧgəlɪŋ/ which I’ve noted as used by the admirable and very experienced television news presenter Anna Ford. There’s some discussion of that topic on this website at §3.7.I.11 commenting on which in a blog some time in 2006 John Wells sed I think Jack’s observations are correct. I do quite often hear three-syllable schwa-containing pronunciations of words such as threatening, doubling, gathering (though I don’t think I’d ever use such a pronunciation myself), and even of simpler. I’ve never yet heard a trisyllabic version of gently, though it’s not uncommon in subtly.        

Such speakers are obviously aiming to employ “clear speech”. Other examples are assembly as /ə`sembəli/, gently as /`ʤentəli/, simply as /`sɪmpəli/ etc. Of course these go counter to the rule Wells gives but they are not so commonplace as to be desirable models for learners. Where they seem to be at variance with the usual spelling as in the case of §/`bɜːgələ/ for burglar, §/`ɔːdəbəli/ for audibly, §/`dᴧbəlɪ/ for doubly etc they’re no dou't widely disapproved of when noticed. Perhaps some people may frown on /`sᴧtəli/ for subtly too but I shd say the dictionaries are right to record it as a commendable optional alternative to /`sᴧtli/ as is praps more so /`sʌtl.i/ with syllabic /l/ (I regretfully refrain from trying to represent a syllabic /l/ by use of the proper IPA subscript syllabicity stroke under the /l/ symbol because the fonts available to me can’t manage to site it suitably). By the way, people not very familiar with American usages may find /`sᴧdl.i/ a bit mystifying for subtly.

A related oddity of the development of English spelling, which EFL users shdnt allow themselves to be misled by, is the fact which Wells makes clear by saying “The ending -ically... [is] usually... compressed to simple -ɪkli, thus physically ˈfɪzɪkli”. We’ve let spellings like *authenticly, *comicly, *fantasticly, *mechanicly, *patheticly, *romanticly, *specificly and *tragicly become obsolete or nearly so since the eighteenth century tho they exactly represent how we normally today say such words. People perversely persist in spelling them only as authentically etc. If you say them with those extra syllables today you may to some people sound old-fashioned or pedantic. A further point is that you may notice that we don’t, any longer at least, have the words *caustical, *catholical, *frantical or *publical. This explains why we do have the adverbs causticly, catholicly, franticly and publicly which don’t end -ically.

Wells ends his useful blog by saying “With the suffix -less ... guileless is ˈɡaɪlləs, not *ˈɡaɪləs; similarly tailless, soulless”. This is perfectly true but there is one word, goalless, that’s he'rd from British broadcasters probably more often than not as /`gəʊləs/ rather than /`gəʊlləs/. However, EFL users h'd probably better not adopt that version c'z most of the ones who say it that way are sports reporters and not always regarded as speaking the most elegant form of the language.

 


 

The Pronunciation of Enhance

John Wells’s blogs are constantly asking int’resting and/or provocative questions.
Today, noting that the Queen sez /ɪn`hӕns/ and not /ɪn`hɑːns/ for enhance, he sez
“I wonder why enhance for her has not undergone the usual broadening (backing plus lengthening) in the environment _ns.”

I’m afraid that his question contains a presumption about the development of such words that tends to over-simplify matters. There are a lot of words with similar environments and many of them have had diverse hist'ries, coming into English from diff'rent varieties of French and at widely diff'rent periods and also subject to the influence of various analogies. The word answer has /ɑː/ in GB like the French loanwords but it goes back to Old English.

Let’s look at this /-ns/ set anyway. A fairly complete list of non-rare, non-derivational ones is as follows. I quote it from §27h of my article ‘The General American and General British Pronunciations of English’ at Section 3.1 on this website.

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

Both GA and GB usually or always have /æ/ in askance˃, ancestor, cancel, cancer, circumstance˃, expanse, fancy, finance, manse, rancid, romance, stance˃, sycophancy˂, trans˃- and numerous derivatives containing -mancy such as necromancy. Names include Clancy, Hans, Hanse, Nance, Nancy, Rance, Pallance, Penzance, Vance and Yancey.

We see that there are only a dozen non-proper words with /ɑːns/ as their usual GB value but there are also a dozen such which normally have /ӕns/ in GB. The sign ˃ is used above to indicate the existence of a subvariant with /ӕ/ for words predominantly having GB /ɑː/. We see that askance, circumstance and stance have the same alternation as enhance but the reverse GB preference. Also we have Wells's own LPD3 comment that for words with the prefix trans- “RP prefers ӕ , although a substantial minority use ɑː”
The upshot of this is that the Wells presumption of “broadening (backing plus lengthening) in the environment _ns” to be “usual” isnt made on very safe grounds.

More evidence of similar variation to that of enhance can be found at the entries for words containing (stressed) /ӕnd, ӕnt/ and /ӕn(t)ʃ/ etc at §§27f, 27g and 27i of the article I’ve quoted where we find commandant, plantain, plantation, stanchion and substantial all with GB alternation between /ӕ/ and /ɑː/.

The most curiously improbable pronunciation of an -and- word is to be found in the Hertfordshire name of a Domesday Book village near St Albans called Sandridge and indisputably derived from the word sand but pronounced /`sɑːndrɪdʒ/. You never know how unlikely a word’s development may be.

Finally we shd remember that, before the nineteenth century, the pronunciations of these -an- words were very variably attested by the numerous writers who commented on such things. The most famous and respected of them John Walker in his Critical Pronouncing Dictionary of 1791/1797recorded /ɑː/ only for askance showing the other comparable /-ns-/ words all with /ӕ/. In fact he showed /ɑː/ only for the small group of chandler, command, demand, reprimand and stanch, giving all other such words with only /ӕ/. Compare my Blog 178. By the way, I cdnt imagine the Queen ever saying /ɪn`trӕns/ or /fi`ӕnseɪ/. And Jones, unlike LPD1 (tho not LPD2 or 3) and post-Gimson EPD,  always recognised /ɪn`hӕns/ as an "RP" variant.

Times Spelling Bee Pronunciations

I’ve just stumbled across a feature provided by TIMES ONLINE which you can access free here

This is a site with a variety of games you can play to test your spelling. They’re so entertainingly presented that they’re likely to prove amusing for  all ages. The basic game you can play by yourself: it offers sets of 15 words at a time that you can hear pronounced and repeated as often as you like. This of course is why I commend it to anyone, especially teachers, concerned with English pronunciation. It can be quite challenging to recognise words uttered in isolation.

When you get to the site a little legless sexless android figure pops up with sort-of electronic bone-clacking and awaits your action. It pulls down its right antenna to its mouth ready to speak into a microphone at its end. The display invites you to START GAME. If you don’t do anything, it first stares then blinks and after twenty seconds or so, with ostentatious rustling of its page, takes a copy of The Times from an invisible back pocket.

Some of the words you hear from the assortment of male and female voices have homophones which you have to remember to disambiguate by resort to a WORD HELP button. This is provided also often when not needed eg for /`swɒləʊ/.
If you recommend anyone to try the games, be warned that not all the words are to be he’rd completely clearly. I have very adequate hearing and hi-fi sound equipment but, trying a few rounds, I took their brute to be proofed, illuminate to be eliminate (with very front /uː/ a confusion I find in real life too), scholarly to be scullery, serried to be more like sewed than anything. Euphemistic cd equally well have been euphonistic (which, tho rare is recorded by OED). The last vowel of stupor sounded as if its -or were pronounced with a German accent and the stressed vowel of pudgy and of courier as if with a Northern accent. Otherwise the voices sounded fairly neutral General British with some a bit markedly southeastern. Resolve came over with no audible final /v/. Three words were given rather abnormal pronunciations. /e`femərəl/ sounded to me a bit stilted with that initial vowel. Even more so at`tribute with /ӕ/. Viscose was given what one can only call an unsuitable form. I was incensed to be told I was wrong to have written viscous for the /`vɪskəs/ I he’rd. That ending is perfectly common for varicose but there’s good reason for not letting the same happen to viscose. None of the pronouncing dictionaries countenance it with /-əs/. The word monstrous defeated myself and my wife: it came out not sounding like that and indistinguishable from monsters.

You get a cheer if you get all 15 right.
If you want to use the site for competitions they say:

The Head-to-Head game challenges you to spell the same set of 15 words more accurately and faster than your opponent. And the really great thing about this game is that it's turn-based, which means you and your opponent don't need to be online at the same time.

As soon as you challenge an opponent to a new game (using easy, medium or hard words), they will be alerted in their Notifications area on the site or via e-mail if they have opted to receive notifications this way. Both of you can then start spelling the same set of 15 words.

STANLEY ELLIS 1926 - 2009

The funeral took place today of Stanley Ellis, the noted dialectologist, who died on October the 31st at his home in Harrogate in North Yorkshire. He was no doubt the phonetician best known to British popular audiences in the latter half of the twentieth century and also the person whose phonetic transcriptions were most extensively published in that era. He became in 1952 the principal fieldworker of the eleven phoneticians that Harold Orton sent throughout the length and breadth of England to carry out the extensive Leeds University Survey of English Dialects. His share, 118 of the 313 localities all but a few of which were rural, amounted to about 38 percent of the total. Their findings, which generally required about 18 hours of questioning, were all represented exactly according to the International Phonetic Association’s practice as constituted then (in the 1950s) except solely for the use of [ɽ] for an approximant instead of a flap articulation. The twelve substantial volumes of the Basic Material contained all the transcribed responses to over 1300 questions. He only ever once transcribed his own speech, in 1964 in his article Hypercorrect Dialectal Pronunciation in Le Maître Phonétique (pp 2 & 3). For a commentary on it see this website § 7.4.31.   
 
    Stanley, who was a product of a local grammar school, came from a Bradford family involved in the woollen industry. At the Leeds University School of English he completed an undergraduate dissertation in 1951 and proceeded to obtain an MA on Lincolnshire dialect. He was engaged in 1952 to undertake first a decade of extensive fieldwork throughout England for the Survey and afterwards editorial work on its findings. Later he became a lecturer on English language. In 1983 at the age of 57 he took early retirement from his senior lectureship and devoted himself more fully to the forensic voice recognition work he had begun to undertake in the sixties. To this he added various lecturing activities and also notably a series of highly popular broadcasts in which he genially drew from people in many parts of the British Isles fascinating illustrations of their local speech characteristics and verbal lore. He will be remembered with affection by countless people.

Pronunciations of Spouse and SPOUSES

In his blog of the 5th of November 2009 John Wells remarked of the word spouses I think (though without any hard evidence) that most people pronounce this word with a voiceless sibilant at the end of the stem, ˈspaʊsɪz.
Hard evidence isnt easy to come by about a word that isnt anything like daily on everyone’s lips. I’ve been aware for over half a century of a problem in deciding what one can call the usual (General British) pronunciation of the word spouse. In 1972 I think I was justified in playing safe and recommending the EFL user to say /spɑʊz/ if aiming at British usage and /spɑʊs/ if at General American while acknowledging that /spɑʊs/ was a GB common alternative. The Webster dictionaries only say “also” about /z/ variants including in the Online version. The current O(xford)DP has only /spaʊs/ for American usage and for British /z/ only as subvariant.  I had cert'nly heard /s/ from numbers of GB speakers by 1972. As to the plural, it’s never been my impression that any GB speakers were inclined to treat the variant /spɑʊs/ as GB regularly treats the plural of house by having /z/ rather than the “regular” /s/ in its plural.

John remarked Many dictionaries do record spaʊz as a possibility for the singular noun adding I would be pretty confident in saying that in BrE at least [z] is very much a minority preference. I think he may well be right but I’m inclined to think that it’s only become so in the last half century or so. The 1989 second edition of OED added the /s/ variant but the OED editor Bradley in 1914 only seemed to know of /z/. Daniel Jones himself only ever recognised /z/ from 1917 to 1956 whereafter Gimson took over the EPD. He didnt record /s/ even as a subvariant until 1977. Roach and company have by contrast given the word only with /s/ since taking EPD over in 1997.

There’s no dou't that /z/ was the norm before the twentieth century. The word goes back to Middle English and there are sixteenth-century spellings of it on record as “spowze”. Walker’s widely circulated Critical Pronouncing Dictionary of 1791 which often discussed such variations mentioned no alternative to /z/. I guess that the switch to /s/ has been braut about by the influence of the spelling since the arrival of fairly universal literacy. There are only a few nouns ending -ouse: only house, louse, mouse and grouse which all have /s/. The only exception is blouse. This has /z/ as far as GB is concerned but mainly /s/ in GA. Anyway there seems to be universal agreement on /z/ for arousal, carousal and espousal.

John was given to speculate about the possibility of some speakers having a /haʊs ~ haʊz/ type of alternation for spouse on hearing a television discussion between the former Home Secretary, Jacqui Smith, and the brilliant presenter Andrew Marr in which both used the pronunciation /ˈspaʊzɪz/ for the plural spouses.  However, he  didnt make any mention of having he'rd either of them utter the word in the singular and I suspect the odds are that both of them wdve sed that too with /z/.  

The Future of English Spelling (ii)

    We now may be seeing on the horizon the possibility that the transition which has seemed so full of horrors may very probably well before the end of the present century be found to be manageable relatively painlessly. Anyone who contemplates the breathtaking speed with which in the last couple of decades computers have extended the scope of the tasks they can perform must surely be encouraged in such a thought. In particular, in view of the ever-increasing pace at which the immensely complicated task of machine translation from one language into another has developed, it must surely be recognised that conversion of texts from the present traditional orthography into a new rational regularised one would not be beyond the writers of such kinds of software once they found it profitable to take the matter up. After all, for some years now one has been able at the click of a mouse to hear the spellings of any written text converted into sounds that produce a very satisfactorily intelligible audible “spoken” version. Surely this process can hardly be simpler than the conversion of a traditionally spelt text to a more rational set of spellings.

At the moment we are seeing the beginning of what will almost certainly turn out to be a revolution in people’s reading habits. A variety of devices have been appearing on the market during the past decade or so that will no dou't prefigure a huge worldwide transfer from the use of paper-based reading matter to electronic screen-based reading. People can even now read novels on the latest varieties of mobile (aka cell) phones. This kind of transference has already begun to happen extensively in the field of reference works and of journals etc of necessarily limited circulation. Publishers are widely closing down or at least shrinking paper-based reference-book-producing organisations at a rate that has cert'nly alarmed David Crystal for one. See his recent blog of the 30th of September On Chambers.  Already it’s possible to store in a single fairly modest-sized computer the contents of hundreds of thousands of books which would require such hundreds of metres of shelving that it would be utterly unfeasible to attempt to accommodate them in any ordinary dwelling. In the foreseeable future it will before long become not only more convenient and increasingly less expensive for reading to be enabled with its spelling rationalised but it could perfectly possibly come about that readers may, if they prefer, be able to obtain a conversion to phonetic (meaning, strictly, phonemic) spelling in preference to a merely improved spelling.

It cou'd well be that many if not most people might ultimately prefer to read texts translated into a phonetic spelling taking its letters from the well-tried stock of symbols of the International Phonetic Alphabet. Any existing transcriptional version may well not be suitable for the purpose. For example one of the least successful (and writable) features of the IPA its colon-type length mark which is a constant feature of transcriptions of the most widely used current versions of British pronunciation cd easily be eliminated. This wd have one resultant benefit of at least in one way reducing the emphasis on trivial differences to be seen in comparing the commonest transcriptions of General British with those of General American. Compare the versions of GA and GB in the transcription used on this website at Section 3.1.d for American and British pronuncations of English where some trivial contrasts of other kinds are also eliminated. It’s often quite trying to contemplate the way that the transcriptions of the two varieties now appearing online in the third edition of the great Oxford English Dictionary frequently display trivial contrasts of representation that are too often reflections of the personal transcriptional preferences of the respective pronunciation editors rather than significant differences between the two  types of speech.

What’s more it’s perfectly feasible that ultimately the choices available to readers may come to include versions which display the text in their preferred accents. Such texts cou'd in due course also be made audible in particular accents. A reader may in the long run even be able to program a converter to supply the user’s own choices of representations of differently pronounced words. For example, if a user chooses to have texts spoken in the General British accent but prefers to hear neither as /niːðə/ rather than the predominant /naɪðə/, that version could be specified. Similarly if the British reader wd prefer to read bath as /bӕθ/ or /baθ/ rather than /bɑθ/ the preference cd be accommodated for that and the other more than one hundred words that displayed such variation. Of course one cannot predict what might turn out to be the take-up of such offers if they became made but they will undoubtedly become technically feasible in the not very distant future. 

Readers who've found these two blogs of mine at all intresting are recommended to look at Graham Pointon's "Linguism" blog English Spelling Reform of the 26th of October 2008.

The Future of English Spelling (i)

There can be no doubt that the English language is viewed as a desirable acquisition by more people who do not speak it as their mother tongue throughout the world than any other. Its grammar is less discouraging for most learners than that of most languages but its sound system with its relatively large repertoire of vowels and consonants is rather harder to acquire than that of most other numerously-spoken languages. It has at least forty phonemes in its most widely adopted forms — most notably General American and General British. Their vowel systems have eighteen or more units whereas the great majority of their adopters have native languages which operate with either only half-a-dozen vowel phonemes or only two or three more than that. The English normally recognised two dozen consonant phonemes contain /θ, ð, ŋ/ and /ʃ~ʧ, ʒ~ʤ, l~r/ many of which are unusual as units or as opposed pairings over large areas of the world.  

 Unfortunately in addition to these difficulties English is a language that, among extensively used ones, is saddled more than any other with a greater multiplicity of traditional spellings that are frequently anomalous mainly in that they are archaistic rather than representative of current spoken usages. However this may be, there has up to the present been little prospect of the situation being ameliorated to any significant degree. It’s noteworthy that after more than century of propaganda and other efforts to bring about the acceptance of a more rational set of spellings no real progress has ever been made in the matter.

In Britain there is a body, the Spelling Society (for most of its century of existence known as the Simplified Spelling Society), which has had over many years the patronage and participation of considerable numbers of very distinguished scholars and men of letters a number of whom were reasonably well enough informed to be able to discuss the matter sensibly. By the recent hundredth anniversary of its founding it has come to abandon further advocacy of the adoption of the very practicable system proposed as a spelling reform that it had worked out with relative completeness by 1910. The quality of that system was corroborated by the fact that a very similar one was arrived at essentially independently in the USA. Unfortunately, altho such systems wd constitute great improvements on present practice, they wou'd, if adopted, not prove ideal solutions because inevitably they wd still leave us with very many thousands of words in forms none of us wd really be happy using especially when it came to such knotty problems as the representation of the vowels of unstressed syllables. 

This British set of proposals could be seen in the 1910 book New Spelling by Walter Ripman and William Archer which was kept in print, with only fairly slight revisions, much of the time till 1948 but has since not surprisingly been allowed to lapse from publication. This proposed a form of revised spelling for worldwide use but has, again unsurprisingly, not been fully accepted by revisionists elsewhere in the world. Apart from such lack of unanimity about solutions, it is not hard to understand why no progress has been made. The disruption that would occur if spelling was not extremely gradually reformed would be almost unimaginably colossal in both extent and cost. All the vast numbers of existing publications would within a generation or so become difficult to use possibly at times to the point of unreadability for the rising generation brought up on the new spelling.

The alphabetisation of new indexes would become a problem for the older generations and the vast quantity of older ones would be so for the young. The initial letters of perhaps half the words in reference works would be changed. For example there would be no longer any at all beginning with C and many with G would have been moved to listing under J. Items now to be sought out under T might (according to the system adopted) have to be looked for in four different places. And so on. All newspapers and other periodicals, all daily reference sources like telephone directories and dictionaries would if converted to the proposed system become seriously less convenient to use or consult to succeeding generations. The numerous homophonous names of people and places and other words that are now distinguishable would become problematic. In short the chaos that would be consequent upon anything like overnight transfer to the new spelling would make all but the most unrealistic enthusiasts for reform quake at the prospect. 

More Accentual Puzzlements

Tami Date has asked the Supras for comments on the following:
Among many cases of accentual puzzlement that Japanese students come across in conversation class, perhaps the one that ranks uppermost in the list concerns wh-questions as follows:
(1)  A: I haven't time now.
      B: When will you have time, may I ask?
(2)  A: No, that's not Stephen's house.
       B: Well, where does he live then?
(3)  A: It wasn't made with flour.
       B: But how did you make it, then?
(4)  A: Harry's not coming to tea?
       B: Who is coming to tea, then?
(5)  A: I know all about it.
       B: But how can you know?
He adds
I'm sure that Japanese students, or for that matter, even many teachers, would pronounce the wh-questions with the most emphasis on the interrogative words by analogy with their L1 intuition. The other day, I checked the intonation of the questions with my colleagues at work (Canadian, British and American), and they all concurred about the position of the nucleus on the auxiliary/be verbs. They categorically ruled out alternative positions.

I find this last statement very surprising but I suspect it may reflect the danger of putting such questions to people who arent experts in judging such matters. I think they cdve been misled by the paralinguistic etc features eg very strong stress and relative loudness with which Tami may have illustrated the versions with accents on the interrogative words. Or praps equally likely, they thaut how they’d say the sentences th'mselves with such accentuations and considered that they’d most likely use that style if they were being very emphatic possibly because of experiencing irritation. The first item in particular may have triggered an inclination to adopt such an attitude in all of them by having that rather over-formal tag “may I ask” which in current ordinary conversational usage wd probably not be employed if the speaker were not being sarcastic. I’m quite positive that it’d be perfectly possible to accentuate the interrogative words while putting the questions in a completely inoffensive and agreeable manner. This comment, I expect, answers his next question:
But what if students utter "WHEN will you have time?", "WHERE does he live?" etc.  How would you react? What I mean to say is: suppose a non-native speaker has made an error in intonation, but it is not recognized as such by his or interlocutor, who may think the other person is rude, abrupt, aggressive etc. 

These things aren’t merely or even mainly matters of accentuations and melodic values but as much or more of paralinguistic features such as loudness, voice quality, tempo and pitch range. Even body stance and facial expression aren’t necessarily always irrelevant. It wd be very int'resting to at least hear (if not hear and and see) how Tami and his students say these items with the accentuations prompted by the analogy of their own language. I suspect they’d sound fine. I’ve often found myself deploring how some people who’ve written descriptions of English intonation for EFL users have overplayed the danger of giving offence by not using sufficiently idiomatically English intonations. Cf on this Website Section 8.4 §§ 3 etc. None of what’s been said above is me'nt to contradict the generalisation that very, perhaps most, often native speakers are likely to choose to accent the auxiliary or similar word in such situations.

He added a postscript:
Incidentally, Cruttenden has the following to say in Intonation (1997: 85):
(Quote)
Yet another case of focus on old information, but one which is more difficult to explain, is where a wh query is attached to a preceding statement, e.g.
(I went home pronto) Why did you go HOME?
(We went into Manchester this morning) What did you go to MANchester for?
(I only smoke occasionally these days) What do you SMOKE?
Notice that, in all three examples, the nucleus and hence focus falls on previously mentioned items (old information). It is possible to put the nucleus on the new item, the wh word, in such cases, but such a placement seems intuitively a less likely possibility.(unquote)

There’re plenty of examples of exceptions to the “rule” about avoiding re-accenting re-occurrences at my Section 8.1.1 which begins with the  warning that Highlighting of contrasts overrides all other tendencies to give a word stress. Now contrasts may be not only represented by a recent word in a discourse but may simply reflect what is going on in the speaker’s mind but is undeclared. In these first two sentences the contrast is with any destination that the speaker expected his co-locutor might mention. 
    On the other hand the last item is a matter of semantic re-focussing on which a variety of examples are provided at 8.1.12. The first smoke refers only to the habitual action. The second one is focussed on favouring or savouring a particular kind of tobacco. This may seem a bit like loose thinking or clumsy self expression but it’s very much the kind of thing that  most of us are capable of saying.

 
 
    
 

Idiosyncrasy or Tongue Slip?

As a careful systematic observer of people’s pronunciations for half a dozen decades I wasnt surprised by the universal negative replies to the question asked of Supra members whether heiress with initial /h/ cd be said to be a known usage among educated British speakers. This question was asked by the leading Supra member Judy Gilbert who’d heard the America-resident British journalist Tina Brown (age 55, wife of the famous editor Harold Evans) use such a version of the word. I can’t help being mildly surprised that no member of Supras seems to have had the temerity to raise the matter with the lady herself.
 
Having heard Ms Brown’s speech and found it perfectly mainstream General British and knowing of her background of being educated in British "superior" girls’ schools and at the University of Oxford I’ve had little dou't that she wdve been very unlikely to've even heard anyone ever pronounce it so — leave alone have developed a habit of saying it that way herself. The obvious explanation is that it was a mere one-off “slip of the tongue”. After all, the spelling is practically an invitation to make such a slip. Anyway I’ve come over the years to feel that if you’ve heard anyone pronounce a word only once in a particular way it’s very inadvisable to take that to be their habitual way of saying it.

The version under discussion is not attested in any modern dictionary recording educated usages on either side of the Atlantic I’m sure. If it had had any currency in American educated speech one cdve been confident that Edward Artin wdve put it in the 1961 big Webster. Actually there’s slightly less inclination attested in America to re-introduce lost aitches than in the UK. He recorded the very non-GB aitchless humble and homage, as well as the well-known herb. This topic is discussed in Section 18 of my Homepage article General American and General British Pronunciations Item 3.1. where among other matters discussed is the common educated American use in unstressed situations of /ӕv, ӕz/ and /ӕd/ the aitchless forms of have, has and had. I’ve never come across any mention of these elsewhere but they’re strikingly different from British educated usage.
 
In any case one can never be a hundred percent sure in such matters since most speakers have oddities lurking in their pronunciation repertoires, tho they usually occur with words that are mainly familiar from people’s reading rather than their daily conversations. This matter was touched on in my Blog 49 on Prime Ministerial Pronunciations and Blog 30 on a Pronunciation Preference Survey. One comes across examples most weeks. The most recent example I remember noticing was the distinguished BBC Newsnight interviewer Jeremy Paxman’s saying thesaurus as /`θesərəs/.
 

Accentual Puzzlement

John Maidment’s blog of the 29th of September sed
Jack ... comments on my post on One(s) and says he is in strong disagreement with me for labelling the sentence
You’ve given me the WRONG one
as unacceptable. Well, it is unacceptable to me, but I am quite prepared to accept that others think it’s OK for their speech.

I’m inclined to suspect that this view of his is pretty unique to him.
Be that as it may, he continued
However, what Jack’s post does not do is dispel my puzzlement of long standing as to why certain adjectives allow a following one or ones to carry an accent, and usually a nuclear accent at that. In my original post I mentioned wrong, same, last. I have since remembered another — only.
A: Can I have a bigger apple?
B: No, this is the only ONE.

I think John’s approaching the problem in a way which, in focussing on the probably inessential matter of what adjective precedes the accented one, overlooks the important matter of the semantic value of that word. I dont think it’s right to presume that the stress values we’re concerned with hinge atall on the particular adjective applied to the word one. Any number of sentences can contain the final word one accented at least when they are responses to triggering sentences containing some sort of comparative. The response merely has to embody something which is some sort of superlative. There’s no necessary requirement of the presence of such a specific very limited set of adjectives as wrong, same or only. These particular adjectives merely commonly occur in such sentences. This kind of sequence is to be seen in John’s example about apples.
Other examples include
A. I want a `larger / `better one of those.
B. I’ve given you the biggest / best `one.

The crux of this matter is that here the word one has simultaneously two different functions. Altho a very common use of the pronominal one is purely anaphoric-replacive ie as a substitute for repeating a noun (or noun phrase) which has recently cropped up in a discourse, the word also may retain something of its primary meaning as a numeral. It’s this additional co-existing sense, I suggest, which prompts the speaker to accent it in these situations.

I’ve often given my EFL students a rule-of-thumb that the numeral one can be accented but the anaphoric pronominal one shd best not be so treated. In the cases we’ve been considering my rule simply doesnt work because that distinction is blurred.

BlogRoll

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