Thursday 18 February 2010

going on twur

Ian Bekker writes from Rhodes University in Grahamstown, South Africa.
A student of mine is working on the realisation of CURE in South African English. We are coming to the tentative conclusion that a current locus of change is the replacement of CURE by NURSE (with a few exceptions like ‘sure’ which are often said with /o:/ and a few others which mutate into GOOSE).
I’m aware that there are similar developments elsewhere; in particular parts of the US and East Anglia. What seems different, however, is that the use of NURSE extends (we think) to non-post-palatal or palato-alveolar contexts. Thus while it’s not particularly surprising to find ‘pure’ pronounced as [pjɞː] (using ɞː [— my symbol, JCW, replacing Ian’s @: —] to indicate the rounded SAE NURSE) we seem to be finding pronunciations like [twɞː] for ‘tour’ and [pwɞː] for ‘poor’.
He asks whether I have come across anything similar elsewhere, to which the answer is no.

If these pronunciations are verified, they would represent a development comparable to the change in the NEAR diphthong from ɪə to jɜː that is familiar from South Welsh English (and treated by Daniel Jones as an RP possibility in EPD, an option dropped by Gimson when he took over as editor).
ɪə → jɜː
This represents merely a switch from a falling (diminuendo) to a rising (crescendo) diphthong, and is thus a very natural kind of development. The syllabicity moves from the first, close segment to the second, mid one. A diphthong is recast as a semivowel plus strong vowel.
The putative South African change is structurally identical, but back rather than front.
ʊə → wɜː
I wonder if it extends to words such as jury. Presumably not: I imagine it cannot apply after velar or palatoalveolar consonants.
_ _ _

This blog will now take a month’s break. Next posting: 19 March.

Wednesday 17 February 2010


We’re used to the Japanese habit of decorating t-shirts and the like with meaningless combinations of words of English. The purpose is evidently not so much to convey any overt slogan or message as rather to give a general impression of being fashionable, modern, international and in tune with the globalized times.
Now we may be seeing the same thing in reverse, and I keep noticing it on the streets in London. A British clothing company, Superdry, has devised a logo that contains what appears to be Japanese text above the company name. There are four kanji (Chinese characters) followed by four parenthesized hiragana (Japanese syllabic symbols). The hiragana read shinasai.
You see this logo in a prominent place on the clothes and accessories sold by this company.

Is it real Japanese, or is it fake?

I asked Masaki Taniguchi.
He tells me that 極度 kyokudo ˈkjokudo means ‘extremely’ and 乾燥 kansō ˈkansoː ‘dry’, so the kanji are meaningful and say the same as the English name. (He thinks, though, that 超 chō tɕoː would be a better translation for ‘super’ than 極度.)

But the hiragana bit, しなさい ɕinasai, seems pointless, because it is no more than the imperative ‘do!’.

Tuesday 16 February 2010

scolding water

Harry Campbell draws my attention to a newspaper report to the effect that
The company which owns a care home where an Oxford teenager died after being bathed in scolding water will be sentenced today.

He comments,
It’s my impression there are people for whom “scalding” and “scolding” are homophones, both having (I think) /ɒ/.

I think this is probably correct. There are certainly varieties of English English in which there is great confusion among back vowels before dark l.

It is well known that words like salt can have either ɔː or ɒ. Although I have continued to prioritize ɔː in LPD I have to confess that the most recent poll I did on salt showed a sharp trend of change over time in the direction of ɒ. The youngest age group voted 71% for the LOT vowel. Only 34% of my own age group voted that way, which accounts for my bias in favour of THOUGHT in this and similar words (i.e. words in which the vowel is followed by l and a voiceless consonant — alter, false, fault, waltz etc.).
Where the consonant after the lateral is voiced I believe there is less variation. Nevertheless I recall that my mother pronounced scald as skɒld, which was odd because she had the expected ɔː in bald, alder etc. (Why, even as a child, did I think her pronunciation of scald was odd?)
In this position before l and a consonant some people also have ɒ where RP traditionally has GOAT (əʊ, including for some its positional variant ɒʊ). That is presumably how scold can come to be a homophone of scald for some. This can apply to the vowel before final l, too. One sometimes hears -ˈtɒl for traditional -ˈtəʊl (-ˈtɒʊl) in extol, and similarly in bolster.

I play a musical instrument called a melodeon. One of the well-known designs of melodeon is Hohner’s poker-work design, with side panels imitating wood decorated by burning the motif in with a hot poker. On the melodeon website I have seen this called polka-work (and even polkerwork). I wonder if anyone really has polka and poker as homophones, as this seems to suggest. For me they’re ˈpɒlkə and ˈpəʊkə respectively. I think there may be a different explanation: contamination from the spelling of folk fəʊk, plus of course the fact that the melodeon is a very suitable folk instrument for playing polkas on.

Monday 15 February 2010

prioritizing the unimportant

I have been continuing to look through the “Languages for the 21st century” booklets included with last week’s issues of the Guardian newspaper.
Of all the things one might possibly say about Portuguese pronunciation, which are the most important?
Opinions can legitimately differ on this point. But I would claim that among the very least important is the fact that the writing system uses the letters k, w, and y only in foreign words, despite their being recognized as “official letters” of the Portuguese alphabet. Yet that is what the treatment in the Guardian’s Brazilian Portuguese booklet starts with and wastes valuable space on.
If we have to focus on spelling rather than pronunciation, you might expect some comment on the accented vowel letters that Portuguese orthography does include, and what (other than stress) they imply in terms of reading rules: ã, ê, õ; á, à, â, é, ó, ô. There’s nothing in the booklet beyond the misleading statement that ã is “as in ‘rang’”. (Actually it represents [ɐ̃], nasalized [ɐ], and the closest model in English would be the first vowel in money ˈmʌni.)
What would I prioritize if I had to write one short paragraph on Brazilian Portuguese pronunciation? I would say that in very general terms it is similar to that of Spanish or Italian (since the reader is likely to have some knowledge of these). However, I would say that it is important to note that vowels in unstressed syllables undergo reduction, posttonic e becoming i and o becoming u. (This makes the general auditory impression very different from that of Spanish.) I would draw attention to the nasalized vowels and diphthongs and do my best to describe what they sound like: e.g. vão is like English vow, but with nasal escape. Yes, I would tell the reader about the spellings lh, nh, x, standing for [ʎ, ɲ, ʃ] respectively, and would attempt to describe the first two. I would say that in Brazil the spellings t and d before i and e usually stand for the affricates [tʃ] and [dʒ].
One of the differences between European and Brazilian Portuguese is the pronunciation of initial r (and double rr). In (some kinds of) Brazilian it is a voiceless fricative, glottal [h], uvular [χ], or velar [x]. As we discovered with an informant from Rio at UCL, it can even be pharyngeal [ħ]: our informant called her native city [ˈħiu], though others would say [ˈhiu] or [ˈχiu]. How does the Guardian describe this phenomenon? “A little bit like a hard ‘h’.”
Neither the layman nor the phonetician can extract a meaningful description from the expression “hard ‘h’”. What would I say if I were writing this booklet? Perhaps that initial r and double rr are voiceless fricatives, like the ch in loch.

And I don’t even know Portuguese.

(I do, however, know the difference between a train and a bus, a distinction that the author of the booklet seems to have problems with.)

If you can’t say anything useful, say nothing. If the reader has no phonetic knowledge there is not a great deal you can do. So you ought to cater for the intelligent layman who does know a little phonetics and can understand such terms as “voiceless” and “fricative”.
If you were writing about archaeology and wanted to discuss bronze, would you explain it as being a mixture of two kinds of melted stony stuff? No, you would say that it is an alloy of copper and tin.
The technical term “voiceless” ought to be no more frightening to the general reader than the technical term “alloy”. “Velar” ought to be no more offputting than “tin” (the chemical element, not the container).
Wikipedia doesn’t hesitate to use appropriate technical terms and technical descriptions. Why should a newspaper catering for an educated and literate readership be more timid?

Friday 12 February 2010

more Guardian phrasebooks

The Guardian is currently running another series of introductory language phrasebooks, now “Languages for the 21st century”. (See comments on an earlier series: blog, 8 and 14 July 2009.) Parts of them are excellent.
You can’t expect depth in a 24-page A6 booklet, but we can at least demand clarity.

(1) The Japanese booklet mentions that Japanese has a handful of geminated plosives, as in もっと motto ˈmot:o ‘more’. They are like those of Italian or Finnish: articulated just like ordinary plosives, but with a noticeably longer hold phase (compression stage). Thus the duration of the plosive is approximately twice as long as usual. To its credit, the Guardian does not ignore the question of how to pronounce them, though it could have used better wording for its explanation.
A double consonant indicates that you should pause slightly before saying it, as you would in these English examples (say them out loud):
headdress (pause after ‘hea’ - not ‘head’)
bookcase (pause after ‘boo’)
In my judgment, to refer to a pause “after” the vowel is misleading, no matter whether you interpret “pause” in the musical sense of sustaining a note (fermata) or in the usual phonetic sense of a brief interruption in speaking. The pause surely comes during the consonant, not before it. ‘Headdress’ is not an apposite example, because Japanese does not have doubled d; but ‘bookcase’ is a good model for geminated [].

(2) Full marks to the Hindi booklet for a valiant attempt to explain retroflex versus dental. However, I’m not sure that the novice would understand ‘unaspirated’ vs ‘aspirated’ when the explanation given is just that “the first is much less breathy than the second”. The crucial difference is a matter of timing: whether or not there is a delay between the release of the closure and the start of voicing. If there is a delay, you get an audible puff of air; if there isn’t, you don’t.

(3) I found the Russian booklet unimpressive. How does Russian work?
You'll be glad to know that Russian is a logical language and that all you need to make up your own sentences is an understanding of the patterns of the language.
Yeah, right.

Russian words and phrases are given only in romanization, not in Cyrillic. Unfortunately the romanization is not a transliteration or a proper transcription, but a respelling (e.g. “ee as in street”), with all the inaccuracies and inadequacies that entails. Both х (IPA x) and ч (IPA ) are represented as ch.
How do we pronounce the vowel transliterated y? “As in toy”. Try that on the pronoun ‘you’, romanized as vy. (You and I know it’s actually вы .) Worse, try “Let’s drink”, presented as ”Davai vyp’yem”. If I were learning Russian I’d much prefer to be given the proper spelling and phonetic transcription, which is давай выпьем dʌˈvaj ˈvɨpʲjɪm. (I don’t actually know Russian. Thanks to commentators for corrections now implemented. By the way, please let’s not repeat the debate over how to represent Russian a in pretonic syllables: I’ve followed Jones & Ward in writing ʌ.)At least there’s an audio clip.

Thursday 11 February 2010

buttered peas

There’s an English folk dance tune called Buttered Peas. You can watch and listen to it here, played on the Northumbrian small pipes.
On Tuesday evening I took part in a folk music session in the depths of rural Walthamstow (London E17), and this was one of the tunes we played. (I was on the melodeon, and five other people were also playing melodeon; in addition we had several fiddlers, guitarists and so on.) One of the other musicians, a fiddler, commented that the name of the tune was actually a “corruption” of a Welsh name Pwt ar y Bys, which he claimed meant “fingering exercise”. He said the tune was originally a fingering exercise for harpists.
I had never heard this explanation of the name Buttered Peas before. The name does perhaps need some explaining, since we have no tunes called Buttered Parsnips, Buttered Cabbage, Buttered Potatoes or the like, nor for that matter Mushy Peas or Peas in White Sauce: so why Buttered Peas?
I didn’t say anything beyond expressing polite interest, because I felt I had better check. As I suspected, Welsh pwt pʊt means not “exercise” but “stump, something short”. Ar y bys ar ə biːs, ar ə bɨːs does indeed mean “on the finger” (though that’s finger in the singular: the plural is bysedd ˈbəseð).
It’s certainly true that the Welsh name for the tune is Pwt ar y Bys (here), and it means literally “something short on the finger”.

Here is someone playing it on the harp.

The folk music traditions of the various parts of the British Isles are hopelessly entangled together — in my experience, English folk musicians at least have lots of Welsh, Scottish and Irish stuff in their repertoire. You can see how folk etymology could easily turn pwt ar y bys ˈpʊtarəˈbiːs into a north-of-England buttered peas ˈbʊtə(r)dˈpiːz, ˈbʊtəb ˈpiːz.
I think that’s more likely than that Welsh folk etymology would turn buttered peas into pwt ar y bys.

Wednesday 10 February 2010


The word synod is back in the news with the meeting of the Anglican General Synod in London.
In LPD I prioritize the traditional pronunciation with a weakened second vowel, ˈsɪnəd. This was the only pronunciation recognized in EPD during Daniel Jones’s editorship. But nowadays one often hears this word said with a strong vowel in the unstressed syllable, ˈsɪnɒd.

Classicists will recognize that synod is parallel in its morphology with method and period, words whose final syllable is unquestionably always weakened to -əd. All three consist etymologically of a Greek prepositional prefix plus the stem of ὁδός hodos ‘way, travel’ (cognate with Russian ходить ‘go’).

We have two other words in English that have this structure: anode (etymologically ‘way up’) and cathode (‘way down’). We know that they were coined by William Whewell in 1834 at the request of his friend Michael Faraday. Another friend, one Dr Nicholl, coined electrode (‘amber/electric way’). Strange, then, that Whewell and Nicholl chose to write them with a final unetymological (or French-style) e and to say them with the corresponding strong pronunciation -əʊd. This was then also applied to the later coinings diode, triode, pentode etc.

Tuesday 9 February 2010

lamb-flavoured ice cream

It is striking that Japanese users of English have difficulty not only in hearing and making the differences between the sounds r and l and between the sounds æ and ʌ, but also in distinguishing the corresponding spellings. They tend to confuse not only the sounds but also the letters r and l, a and u.
Because Japanese has only one liquid and only one open vowel, lamb, ram, and rum are all mapped onto Japanese ラム ramu. And we get confusions like this.It’s meant to be rum and raisin.

- - -

The inspiration for the book with the Russian sunflower illustration that I reproduced a few days ago was John Trim’s English Pronunciation Illustrated (CUP 1975), with illustrations by Peter Kneebone. Here are some of their items for practising æ and ʌ. Surprisingly, they offer none for practising r and l.
It was John Trim who first initiated me into phonetics at Cambridge University, and he is someone whom I have always held in great admiration. I think it’s a little sad that the main thing he is known for in the wider world is this perfectly respectable but, alas, not very deeply intellectual book. Relatively few people know of his work as Director of CILT (the UK’s Centre for Information on Language Teaching) or on defining threshold levels in language teaching for the Council of Europe.

Monday 8 February 2010

bunched/molar r

When I was still a schoolboy I found Daniel Jones’s Outline of English Phonetics in the local public library and devoured it with great interest and excitement.
It all made perfect sense to me, with one exception. That concerned the English r sound. I couldn’t square Daniel Jones’s description with what I could feel myself doing when I pronounced this consonant.
Obviously I was not using a “rolled” r (trill) or a “flapped” ɾ (tap). And I wasn’t making any kind of uvular sound. So I must be using the “most usual English r”, the “fricative lingual sound” ɹ (which we would today call a postalveolar approximant).
But I didn’t seem to be articulating it with the tip of my tongue so much as with the body. And the combinations kr (as in crown) and ɡr (as in green seemed to have some sort of affricated pronunciation.
My puzzlement was resolved only several years later when John Trim explained to me that I use a “molar r”. The strange thing is that it doesn’t sound any different from the postalveolar r sounds that other people use. People don’t react by saying “that’s a funny kind of r”.
The VASTA listserv has been having a discussion about “bunched r” recently (another name for the same thing), and I said I would write something about it here.
Here’s Catford, talking about the AmE NURSE vowel:
This vowel was formerly described as ‘retroflexed’ but this is not a correct description. It does not usually have the upward curling of the tongue that is characteristic of retrofexion. Instead, the main body of the tongue is bunched up into a kind of half-close-central position, but with two peculiar modifications: one modification is a moderate degree of deep pharyngalization: the root of the tongue
is drawn back into the pharynx just above the larynx. The second modification is a fairly deep depression in the surface of the tongue opposite the uvular zone. This sub-uvular concavity can be acquired as follows. Produce a uvular trill. Note that in order to do this you have to form a longitudinal furrow in the tongue within which the uvula vibrates. Now move the whole body of the tongue slightly forward, while retaining precisely that deeply furrowed configuration. The result should be a close approximation to the typical American ‘bird vowel’, for which the phonetic symbols [ɜ˞] and [ə˞] have been used — both representing a central vowel with an r-like modification.
As we saw, this very strange American vowel involves not only a concavity — or ‘sulcalization’ (from the Latin sulcus `a furrow, or trench') — of the tongue in the neighborhood of the uvula, but also some slight degree of pharyngalization. It is because of this that a series of vowel-sounds with modification of this rhotacized type in some languages spoken in the Caucasus area of Russia, notably Tsakhur and Udi, are known as ‘pharyngalized’ vowels.

Like Erik Singer, who found this quote from Catford, I'm not sure I agree with the claim of pharyngalization. It can be there, no doubt, but I don’t think it's a necessary accompaniment.
My own r is consonantal, of course, while the American vowel is syllabic (a vowel). But the articulation used appears to be the same.
In my experience, when someone has claimed that there is an audible difference between the molar and the postalveolar kinds, I find that there is also some kind of difference in secondary articulations (pharyngalization, labialization etc), but no audible difference when this is stripped out. That’s why I have some sympathy with the argument that if we can’t hear the difference we don’t need a special IPA symbol for the molar r. It will be sufficient to symbolize the secondary articulation(s), e.g. ɹʕ or ɚˁ.
But the consensus on the VASTA list seems to be that we need a special symbol. Erik Singer said
I very much agree that we need a symbol for the “bunched,” “braced,” or “molar” /r/. If it has to go in the set of IPA extensions for disordered speech, so be it. At least we’ll have a symbol. I wonder, though, should there be separate symbols for the vowel and consonant? It’s really a single physical action (varying only, I think, in the degree of the bracing), and according to the rules of the IPA, there should therefore be only one symbol corresponding to it. […] There simply must be a unique symbol. This is my awkwardly bitmapped attempt to create a symbol that takes the turned-r and adds a another wing to it, stretching out to the right. We can recognize this as r-ish, but it is sufficiently different from the alveolar-approximant symbol to avoid confusion with that placement. This symbol could also be modified to specify degree of bracing, with one or two additional “wings” stretching to the right, at mid-level and at the top.

Amy Stoller says she associates this sound (a "hard-R" sound) particularly with an Oklahoma or Texas accent, though it is also found elsewhere. ‘Michael’ added
It is not impossible to make a "lighter" sound with this placement —though it becomes a bit more of a mid-tongue bunching rather than the hard retraction of Oklahoma.
Can people actually hear a difference based purely on place of articulation? Or is there always something else involved? Let’s have some sound files.

Friday 5 February 2010

Russian illustrated

No time for a proper blog today, but here are scans of the two pictures mentioned yesterday by David Crosbie, taken from Dennis Ward's Russian Pronunciation Illustrated (CUP 1966), p. 81. The drawings are by George Him.

Thursday 4 February 2010


Dominik Lukeš writes
I've always assumed that [ts] is one of the lesser affricates of English on the cusp of phoneme status given the single minimal pair in pizza/Pisa. I've never studied English phonetics (only general [phonetics] having Czech as the starting point) so when I started preparing some talks on pronunciation for literacy teachers, I was amazed to find out that [tr] and [dr] are considered affricates and [ts] is at best affricate-like. All the samples on the pronouncing dictionaries' CD-ROMs are distinctly pronounced as affricates and yet the Cambridge one even splits the syllables of pizza as /piːt.sə/.
Is there a phonetic reason for this? Am I being misled by my Central European ears? Or is this just a result of convention? I've asked a few non-expert native speakers to contrast pizza/Pisa and Nazi/*nasi and they were all able to do so effortlessly. When asked to reflect on what sound they were pronouncing, one speaker said: 'it's a sort of squished soft sound' - surely a folk definition of an affricate.

The definition of “affricate” has always been a bit tricky when we are discussing English phonetics. (That’s why we set it as a topic for student essays.) Do we define it in purely articulatory terms, interested only in the manner of articulation? Or are we concerned with English-specific phonology? (E.g., do these entities have phonemic status?)
I don’t understand why Dominik adduces Pisa to go with pizza: the consonant(s) in the middle of the latter are voiceless (ˈpiːtsə), while that in the former is voiced (ˈpiːzə). If we’re looking for minimal or near-minimal pairs, there’s more mileage in blitz blɪts vs bliss blɪs, curtsey ˈkɜːtsi vs cursing ˈkɜːsɪŋ, Betsy ˈbetsi vs Bessie ˈbesi. But then who has ever doubted that ts was in contrast to s in English?

Except for a handful of proper names and recent loanwords and the single item curtsey (derived from trisyllabic courtesy), English ts always involves a morpheme boundary between the plosive and the fricative: cat#s, put#s, hint#s, drift#s, list#s, bat#s#man, eight#some. (Please don’t start quibbling about flotsam.) No word begins with ts (ditto: no quibbling about zeitgeist or tsetse). No single-morpheme word ends with ts (except recent loanwords).

In this respect ts is like , just an adventitious consonant sequence straddling a morpheme boundary. On the other hand occurs readily in initial, medial, and final position in single-morpheme words: chin, kitchen, touch. It even contrasts with the corresponding plosive-fricative sequence: ratchet vs rat shit. The same is true of tr, at least in initial and medial position.

I would even say that to English ears (or brains) ks is probably more of a single unit than ts (because of box, six, exit etc). (Articulatorily, of course, ks is obviously not an affricate, because the articulation is not homorganic.)

Wednesday 3 February 2010

glottal opera

An interesting YouTube video has been going the rounds recently, marrying a choral performance of a song with laryngoscopic shots of the singers’ glottises.
See each glottis opening, closing, and vibrating! Admire those arytenoids! Marvel how the epiglottis swings back and forth, sometimes obscuring the view of the vocal folds!
Once upon a time (such as when I was a postgraduate) there were only rare and specialized occasions to view a laryngosocopic film clip and view the vocal folds directly. Now here it is for everyone to enjoy (if that is the word).
I think we have to take it on trust that the film clips are properly synchronized with the sound.
I don’t know anything about “the all-girl vocal group Kaya” — does anyone? I have to say, though, that when they sing lips it sounds remarkably like leaps liːps. It must be because they’re Australian, with raised KIT (and diphthongal FLEECE).
Thanks, anyhow, Australian colleagues.

Tuesday 2 February 2010

mnemonic rhymes

Nowadays we tend to have a low opinion of learning by rote. But I have to say that for language learning it has its uses. Things we learn off by heart at a tender age tend to remain in our memory for ever.
When I was being taught Latin (from age 9 onwards) we were encouraged to learn various little rhymes intended to help us remember the gender of nouns.
Many nouns in is we find
to the masculine assigned:
amnis, axis, caulis, collis,
clūnis, crīnis, fascis, follis,
fūstis, ignis, orbis, ēnsis,
pānis, piscis, postis, mēnsis…

Latin nouns that fluctuate between masculine and feminine are said to be of ‘common’ gender.
Diēs in the singular
common we define:
but its plural cases are
always masculine.

Aequor, marmor, cor decline
neuter; arbor feminine.

(It didn’t worry us to pronounce ˈmæskjulaɪn and ˈfemɪnaɪn here for the sake of the rhyme, instead of the usual ˈmæskjʊlɪn and ˈfemənɪn. Sometimes we said ˈsɪŋɡjulɑː for the same reason.)

We had our little parodies, of course. We were supposed to recite
Common are: sacerdōs, dux,
vātēs, parēns, et coniūnx…

but we actually took delight in saying
Common are: sacerdōs, dux,
Hoover and Electrolux.

You’ll find these and other “memorial lines” in an appendix at the end of Kennedy’s Revised Latin Primer.

Do today’s language learners ever memorize mnemonic verses? Here’s one I’ve just made up.
Come and go, but came and went;
tend and tended, bend and bent.
Try and tried, but buy and bought;
reach and reached, but teach and taught.

Monday 1 February 2010

lexical sets

Amy Stoller writes
I use your lexical sets in preparing my materials […] and I do give you credit for them (albeit obliquely in some cases). I don't publish my materials, but I do distribute them to clients, and there may be publication in the not-too-distant future. Am I going to run into copyright difficulties? […] I don't want to infringe on your rights, and I don't want to get into trouble with Cambridge University Press!

Don’t worry, Amy: I claim no copyright in lexical sets. Everyone is free to make whatever use of them they wish. I am delighted that they have been taken up by many other authors.
Of course it’s nice to be acknowledged from time to time as their originator, but even that’s not necessary.
I sometimes think that a century from now my lexical sets will be the one thing I shall be remembered for. Yet I dreamt them up over a weekend, frustrated with the incoherent mess of symbols used in such contemporary publications as Weinreich’s ‘Is a structural dialectology possible?’. I did not try them out anywhere before publishing them in Accents of English, vol. 1.
For the uninitiated, in my 1982 book I proposed a system of “standard lexical sets”,
a set of keywords, each of which […] stands for a large number of words which behave in the same way in respect of the incidence of vowels in different accents.
Thus KIT is the lexical set associated with the (strong) vowel ɪ of RP and GenAm, DRESS the set containing e (or, if you prefer, ɛ), TRAP the set containing æ, etc.
I called them “standard” lexical sets because they were based on my two ‘reference’ accents of English, RP and GenAm. The sets were defined by the intersection of vowel incidence in these two varieties (conservatively defined).
So NURSE is treated as a single set, because the reference accents have merged the formerly distinct vowels of verse, serve etc as against nurse, curve etc. If the sets had been defined by a wider range of accents, it would have been necessary to split the NURSE set to take account of the speakers of Scottish and Irish English who make the distinction. People dealing with varieties that make such further distinctions in other sets have quite rightly proposed and defined further lexical sets or subsets.
On the other hand FORCE and NORTH are distinct sets in my system, because of the fact that GenAm (at least as set out in then current reference books) allows for the contrast of vowels in sport vs short, hoarse vs horse etc. It is then a bonus that we can use these keywords to discuss the corresponding contrast in Scottish English, West Indian English etc.
I took great care in the choice of suitable keywords. I wanted words that could never be mistaken for other words, no matter what accent you pronounced them in.
Although FLEECE is not the commonest of words, it cannot be mistaken for a word with some other vowel; whereas beat, say, if we had chosen it instead, would have been subject to the drawback that one man’s pronunciation of beat may sound like another’s pronunciation of bait or bit. As far as possible the keywords have been chosen so as to end in a voiceless alveolar or dental consonant…
though that was not always possible.
The least satisfactory keyword is PALM, and its set is also fairly incoherent. Amy says she prefers to replace it with FATHER, which is fine up to a point: but not if we are discussing Hiberno-English, where father often has not the expected of Armagh, Karachi, Java etc but the ɔː of THOUGHT.
The choice of the keyword DRESS has proved awkward for people dealing with New Zealand English, where the upward shifting of the vowel in question has led them to have to refer to DRESS Raising.