Tuesday 31 August 2010

linking semivowels?

Lourdes Achard (blog, 27 Aug) also wrote
I am a Phonetics teacher and am teaching linking /w/ and /j/ at the moment. Have you got any text to practise this kind of simplification?
I found this request puzzling, and replied
I wonder what your students' first language is. Since native speakers don't normally use such linking semivowels, this is not a topic I would teach! And I don't understand how it could be seen as a "simplification". Perhaps I have misunderstood your question.

Peter Roach’s widely used textbook English Phonetics and Phonology (CUP, 4th ed. 2009) quite rightly makes no mention of this topic. Looking further afield, I wonder if Lourdes was influenced by Cruttenden’s formulation in his current (7th) edition of Gimson’s Pronunciation of English (Hodder Education, 2008).
(Click to enlarge this for better legibility.)

Cruttenden is being rather naughty here in his phonetic notation. The IPA symbols [ʲ, ʷ] are properly no more than diacritics, indicating palatalization and labialization respectively. He, though, is obviously using them to denote very short, transitional, non-phonemic glides.

I suppose he is right in saying that these not-quite-segments may sometimes be “heard”, since experience shows that some naïve transcribers are convinced that they exist. Personally, I take the line that they are figments of our imagination: the supposed “[ʲ]” in my arms merely represents the point of maximum upward excursus of the tongue body as it moves from [a] through [ɪ] towards [ɑ]. How could one possibly detect the presence vs. absence of this entity on a spectrogram?

If it meant anything, it would presumably mean that some small part of the is to be regarded as falling in the arms syllable rather than the my syllable. But we do not say *maɪ jɑːmz, as evidenced by minimal pairs such as Cruttenden’s I earn vs. I yearn: there is no phonological (phonemic, linguistic) j present in my arms or I earn.

And there’s no w in two evils (see below).

In this respect English differs from, for example, Serbian, which obligatorily inserts a phoneme j between i and a following vowel, as reflected in the spelling Srbija (not *Srbia): the syllables are sr . bi . ja.

Whatever we conclude from this discussion, I can certainly see no case for wasting any time in the EFL classroom on teaching such a dubious topic.

Monday 30 August 2010


When in Barbados recently I learnt a new word, speleothem ˈspiːliəʊθem. It refers to any underground rock formation, more precisely to “secondary mineral deposits formed from water in caves”. Examples include stalactites and stalagmites. According to the OED, this word was coined (from Greek) as recently as 1952.

On the island we visited an impressive series of limestone caverns, Harrison’s Cave, which has only recently been developed as a tourist attraction. You are driven around underground in a “tram” (wagons pulled by a tractor) through tunnels and caverns full of stalactites, stalagmites, and other, um, speleothems. Somewhat reminiscent of Postojna in Slovenia, it’s well worth a visit.

I was struck by the fact that our guide, presumably a native Barbadian, pronounced stəˈlæktaɪt and stəˈlæɡmaɪt (or with stæ-). These are what I take to be the American stressings of stalactite and stalagmite, in contrast to the British ˈstæləktaɪt and ˈstæləɡmaɪt. Given the generally British orientation of Barbadian English, I would have expected rather to hear initial (or perhaps final) stress.
I don’t think I have ever heard penultimate stress in these words in BrE. Nor, as far as I recall, have I heard a strong vowel in the penultimate syllable (ˈstælæktaɪt, ˈstælæɡmaɪt), despite the fact that the OED allows only for the strong vowel.

In fact the OED pronunciation entries are rather odd: they imply that BrE can have either stressing and always a strong medial vowel, both of which claims seem to me to be at best dubious.For what it’s worth, initial stress is what Chomsky and Halle’s SPE rules would predict for stalactite #stælækt+īt# and stalagmite #stælægm+īt#. Their Main Stress Rule imposes stress on the final heavy syllable, then the Alternating Stress Rule converts this to antepenultimate, preserving the strength of the final vowel. The American, penultimate, stress pattern, as used by our Bajan guide, would necessitate marking the pronunciation of these words as exceptional.

To remember which is which, stalactite or stalagmite, I regret to say that at school we learned the mnemonic "first the tights come down, then the mites will grow up".

Friday 27 August 2010

humanitarian, archaic

Lourdes Achard writes with two queries. One concerns the word humanitarian, transcribed in LPD as hjuˌmænɪˈteəriən but in the Cambridge EPD as hju:ˌmænɪˈteəriən. She asks
Why does the second one take a long U? If it takes a long U shouldn't the secondary stress fall on the first syllable? About the first one, why does it take the u sound instead of the ʊ one?

You can regard the first vowel of this word as either long or short. Indeed, in earlier editions of EPD you will find hju(:)- (Jones) or hju:-, hjʊ- (Gimson/Ramsaran). The current editors have chosen to ignore the lax-vowel possibility.

The line I take in LPD is that this is a weak vowel, in which the opposition between and ʊ is neutralized, just as in ˈthank you!. I write this neutralized close back vowel as u, which you can interpret as a cover symbol for both and ʊ. It is comparable to the i I write at the end of words such as happy ˈhæpi. Given that it is a weak vowel, it will not attract stress.

Thus the LPD entry corresponds to what EPD used to have.

However there is also an alternative possibility, much less common, in which the vowel is made strong and does attract secondary stress, thus ˌhjuːmænɪˈteəriən. [This is misprinted in LPD as ˌhju-. Thanks, Peter et al.]

Lourdes’s other question is:
Why is the word archaic transcribed with two ɪ sounds, ˌɑ:ˈkeɪɪk?

—to which one can only answer “because that is how it is pronounced”. It has three syllables, ˌɑ: . ˈkeɪ . ɪk. It does not end like cake keɪk. The first “ɪ” is part of the diphthong , the second is an independent vowel. We have the same thing in algebraic, fomulaic, mosaic, which have a similar morphological structure and similarly end -eɪɪk.

Both the spelling and the pronunciation of archaic go right back to the origin of this word in classical Greek ἀρχαϊκός arkhaïkós, from the stem of ἀρχαῖ-ος arkhaî-os ‘ancient’ plus the suffix -ικός -ikós ‘-ic’.

PS (for advanced students): Actually, it is also conceivable that speakers might sometimes compress the two last syllables into one, reducing the eɪɪ sequence to simple . But don’t tell EFL learners this: it will only confuse them.

Thursday 26 August 2010

forceful sports supporters

Russ Stygall asks
In the cases of the following FORCE words, do the compounds agree with the base forms in having the same sound [or]: force/forceful, sport/sporty and support/supporter?

If Russ is a native speaker of English, why is he asking? (He should say the words aloud and find out.) If he is not, the answer is very simple: yes, they agree with the base forms. The suffixes -ful, -y, -er do not normally trigger any vowel change in the stem to which they are attached.

Whether the sound is or or something else depends on the kind of English you speak. In RP and similar accents these words all have ɔː, thus fɔːs, ˈfɔːsfl̩, spɔːt, ˈspɔːti, səˈpɔːt, səˈpɔːtə. In Scottish English, and for some Americans and others, yes, they have or, thus fors, ˈforsfl̩ etc. For other Americans, of course, they have ɔr.

But there are certain other suffixes which do indeed trigger a vowel change (for some of us). Historically, this is the alternation that we see in pairs such as tone–tonic, episode–episodic, verbose–verbosity, where a stem with the GOAT vowel switches to the LOT vowel when the triggering suffix is attached. (We have the same “trisyllabic laxing” with other pairs of vowels in serene–serenity, divine–divinity, profane–profanity and so on, as students of Chomsky and Halle’s Sound Pattern of English know.)

Since historical GOAT plus r becomes FORCE, in stems ending in r the alternation is instead between FORCE and LOT. We see it in pairs such as historian–historical, floral–florist, explore–exploratory, in which in RP and most accents of England there is a vowel difference ɔː–ɒ, thus (h)ɪˈstɔːriən, (h)ɪˈstɒrɪkl̩, ˈflɔːrəl, ˈflɒrɪst, ɪkˈsplɔː, ɪkˈsplɒrətri. But others may have lost this alternation.

Wednesday 25 August 2010

some day my prints will come

Dan Schneider asked:
Do you happen to know what the difference is between /pænts/ and /pæns/ or /faɪ.næns/ and /faɪ.nænts/? Whenever I look up the IPA of of words like these, ending in either /-ns/ or /-nts/, they seem to fit with the actual spelling of the word. For example, finance doesn't have a T in it, so the IPA shows /faɪ.næns/, and pants does, so therefore you get /pænts/. In my mind, (I'm from Massachusetts, USA) I don't hear or make any sort of difference between the two. I feel like I always say /ns/, but I could be wrong. I can't seem to find any sound clip online that shows the difference between the two, and I can't really imagine what the difference would be. I feel like putting the /t/ in there would be to cumbersome to do in everyday speech.

In my reply I explained that essentially it’s a question of the timing of the articulatory movements. To go smoothly from [n] to [s] you have to release the tongue tip before creating the velic closure (= raising the soft palate). Conversely, if the soft palate movement is completed first (thus preventing nasal escape), before the tongue tip leaves the alveolar ridge, there is a moment during which no air can escape: this is then perceived as [t].
So it all depends on which moves first, the tongue tip or the soft palate. (The cessation of vocal cord vibration can take care of itself.)

I natively make a firm distinction between the two possibilities. My
prince has [ns], my prints has [nts] (or [nʔs]). But I know that many people don’t make any such distinction.

In my experience Americans usually pronounce both words like my prints.

Speakers who have this epenthetic plosive normally do so whenever a nasal is followed by a voiceless fricative within the same syllable. If the nasal and fricative are in different syllables (e.g. inside, uncertain, consider — you have to buy into my views on syllabification here), there is no epenthesis. So Dan can get a model for plain -ns- by considering his consider.

Tuesday 24 August 2010


Two correspondents have separately drawn my attention to mysterious goings-on in Nevada, where a local politician is pressing for official recognition of the pronunciation -ˈvædə, with the “alternate” -ˈvɑ(ː)də being “also acceptable”.
Ben Rutter says
I saw this today and thought it might have a place on your blog. Can you legislate for variation in pronunciations?!?

—while Jürgen Trouvain refers to an article in German:
[This] article from the online service of the "Tagesschau", the most popular news magazine in German television, highly reputed for decades, might be of interest.

(The banner illustrating the Tagesschau article, reproduced above, has got its IPA symbols slightly confused. No one is at all likely to say nəˈvədə. The deprecated version is nəˈvɑdə, or with length marks nəˈvɑːdə.)

Reading further in the Politics Daily report, you can see that something deeper is involved than a run-of-the-mill disagreement over how to pronounce some ordinary vocabulary item. The pronunciation that most British people use, -ˈvɑːdə, is evidently perceived by Americans as “the Spanish pronunciation”.
Mortenson says he's not asking Nevadans to change. He just wants the Spanish pronunciation recognized.

This accords with the common but inaccurate identification by Americans of Spanish a with their own LOT vowel. (The British, in contrast, are perfectly happy to identify the Spanish vowel with English TRAP, as in Málaga, Spanish ˈmalaɣa, anglicized as ˈmæləɡə).
So what lies behind Mortenson’s proposal seems to be the deep-laid but irrational fear among many Americans that English is somehow being displaced by Spanish.

The Tagesschau article refers also to a number of German placenames whose spelling does not indicate the pronunciation unambiguously: Itzehoe, Coesfeld, Stralsund, Bleckede, Varel, Grevenbroich, Troisdorf and Poing, which are respectively ɪtsəˈhoː, ˈkoːsfɛlt, ˈʃtraːlzʊnt, ˈbleːkədə, ˈfaːrəl, ɡreːvn̩ˈbroːx, ˈtroːsdɔɐf, ˈpoːɪŋ — to which we might also add Duisburg ˈdyːsbʊɐk (blog, 29 July).

Monday 23 August 2010


I enjoyed the recent Society for Caribbean Linguistics biennial conference in Barbados.
Amongst other things, it offered plenty of opportunities to hear the locals talking. The Bajan (Barbadian) accent is rather easily distinguished from other kinds of Caribbean English. As I put it in AofE,
The most striking characteristics of Barbadian pronunciation, when compared with that of other parts of the West Indies, are full rhoticity, the use of a glottal allophone of /t/, and the quality of the PRICE vowel.

At first you might take Bajan nice nʌis for Canadian. But whereas the so-called Canadian Raising affects the PRICE vowel only in the context of a following fortis consonant (including a t that has undergone tapping/voicing, as in writer ˈrʌiɾɚ), Bajan PRICE is ʌi (or əi) in all phonetic contexts: child tʃʌil. Historically, it is presumably an archaism, a failure to lower the starting point rather than an innovation of raising it.
The Cockney-style glottal stop for t, on the other hand, seems to be an independent local innovation. I heard someone from a different part of the West Indies comment on the (to her) confusing way in which an announcement at the airport referred to flight AA 88 as ʔeː ʔeː ʔeːʔ ʔeːʔ.
Other West Indian accents are either non-rhotic or variably/partially rhotic. But the Bajans pronounce historical r in all positions. For English people, they sound a bit like people from the west of England. Combined with other features, this makes Bajan speech reminiscent of the popular stereotype of pirate talk.

I wonder why the name of the island, bɑːˈbeɪdɒs (or the like), is usually pronounced in BrE with a short vowel and voiceless fricative in the final syllable. In comparable words such as tornado, desperado, avocado, torpedo we pluralize in the usual way with z and retain long əʊ.

Unfortunately, IDEA hasn’t got any sound clips of Barbadian. Plenty of websites offer to tell you about Bajan dialect, including this one and this one. But none seem to have audio samples. Does anyone know where you can hear Bajan on-line? (You can, however, listen to local radio here.)

Friday 6 August 2010

more mishearings

The silly season continues for Guardian letter writers. Here’s another one relating to the close NZ DRESS vowel, then Newcastle GOAT heard as NURSE. Then yet more NZ.

I think the last one goes too far, at least as far as Sally is concerned. You’d have to come from Glasgow or Belfast to mistake NZ TRAP for KIT.

But for a Londoner to mistake northern Irish SQUARE for NURSE is all too likely.
— — —
That’s it. There will now be a two-week break in this blog. Next posting: 23 August.

Meanwhile, to keep you entertained while I’m away, there’s more silly fun to be had from Jack Windsor Lewis and Karen Chung.

Thursday 5 August 2010

yː in Wales?

Roy Becker-Kristal writes
The plot of the film Plots with a View takes place in Wales, in the fictional town of Wrottin Powys. Brenda Blethyn plays the main character, who speaks English with a Welsh-like non-rhotic accent.
Having watched the film a couple of times, I noticed that, in addition to the pronunciation of [ø:] in NURSE words, Brenda Blethyn pronounces NEAR words (here, beard) with [y:], with visible lip-rounding. On one occasion she also pronounces the word 'girl' as [gy:l] instead of [gø:l]. I should perhaps mention that her pronunciation of GOOSE words in this film is a relatively retracted [u:]. (Sorry I can't provide a link to a video that shows all this, you'll have to trust me.)
I haven’t seen the film, so can’t comment directly. In my Accents of English book I said (p. 381) that in south Wales
In NEAR words there is usually /jɜː/, so that beer /bjɜː/ can be said to rhyme with fur /fɜː/ as well as with fear /fjɜː/, while year, ear, and (/h/-dropped) here are all identical as /jɜː/. Some speakers, though use a disyllabic sequence /iːə/, thus beer /ˈbiːə/ etc.,, and in the environment _rV simple /iː/ is usual, thus period /ˈpiːriːəd/…
We know that NURSE is often rounded in southeast Wales, “giving the effect of a centralized raised [œː] or lowered [øː]” (p. 381), with an “auditorily unusual quality” (p. 383). So what one would expect for here is something like (h)jøː. If Brenda Blethyn really says (h)yː, this could be seen as a kind of coalescence in which two segments, j and øː, are collapsed into a single segment that shares the features of both.

Roy has another theory.
Suppose that Brenda Blethyn normally pronounces NURSE as a slightly advanced schwa, and NEAR with a slightly advanced high-central-unrounded monophthong (as many speakers of Estuary English do). These two vowels may actually form a vertical series for her. Then, as part of the effort to sound Welsh, she pronounces this series as front-rounded, thus correctly pronouncing [ø:] in NURSE but incorrectly pronouncing [y:] in NEAR.
Ingenious! Do we buy this?

(Blethyn is her married name. She was born Brenda Bottle in Ramsgate, Kent, and as far as I know has no Welsh connection beyond the surname of her ex-husband, which she retains as her professional name.)

Wednesday 4 August 2010

cries of old London

The May–June 1910 issue of the m.f. (blog, 2 Aug) contains an unusual article by G. Noël-Armfield on the topic of London “street cries”.(I think the length mark at the end of the last word must be a mistake.)

You will see that a century ago Noël-Armfield was already bemoaning the decline in street cries. He was concerned chiefly with hawkers who walked the streets as they tried to sell their wares: his examples include sellers of coal, newspapers, watercress, milk, muffins, firewood and oysters. Apart from milk, which many Londoners still have delivered by a milkman, and coal, which no one uses any more, these are goods which you would nowadays buy in a supermarket. So nowadays there are no “cries” on the streets of London except possibly in the street markets. Even there I think market vendors at Portobello Road and Petticoat Lane are too sophisticated these days to shout their wares, though I suppose you might possibly find the odd stallholder at Brixton or Ridley Road calling out some special offer. And there are always the tricksters in Oxford Street.

Note the unexpected transcriptions kærəktristik and inʌnsjeiʃn (each with only four syllables). It’s interesting, too, to see ə in the endings -less and -ness, something DJ didn’t recognize for RP.

Noël-Armfield must be mistaken in his speculation that the glottal stop (“by no means uncommon … in the East End”) is due to German Jews. Apart from anything else, German/Yiddish glottal stop is initial (or as he would presumably have put it, “ʔim ʔanlɑut”), reinforcing a vowel, whereas the Cockney glottal stop is indeed typically medial or final and represents t.

At that time the Jews were the latest wave of immigrants in the East End. They were mainly from present-day Poland, Belarus or Ukraine rather than “German”, but of course Yiddish-speaking. A century later, their descendants have now moved out to Hendon or Chigwell and their place has been taken by the Bengalis and Somalis.

Noël-Armfield then launches into musical notation. I have space only for a small sample, covering the cries uttered by milkmen and chimney sweeps.
You’ll notice here another strange length mark in a NEAR word, ear-splitting in the last line. I wonder, too, whether “dʒenuin” is a mistake for the expected dʒenjuin, or whether it was intentional.

Noël-Armfield became Jones’s Assistant at UCL. He was the author of several books. One, called English Humour in Phonetic Transcript, you can view here. Another was General Phonetics for Missionaries and Students of Languages (read it here). Yet another was Un peu de rire français avec transcription phonétique, which I have never seen.

Tuesday 3 August 2010

Guardian mishearings

Over the past ten days letter writers to the Guardian have been entertaining us with various mishearings and misunderstandings supposedly caused by regional/social pronunciation differences.

They all depend on phonetic overlap: where a sound [x], intended by the speaker as a token of lexical set /A/, is perceived by the hearer as representing the different lexical set /B/.

It started with crux (blog, 22 July) and the joke about a northerner in a hotel asking for a double rum and being offered not a drink but a room. Then...
So English pɜːnəʊ was heard as French pryno pruneaux. A northern monophthongal GOAT vowel was heard as a southern NORTH vowel.

…Geordie NURSE as non-Geordie NORTH; provincial FACE as Cockney FLEECE; and the problem that the sequence fʊks is taboo in the north of England but not in the south.
…southern STRUT heard as northern TRAP (take courage, Spanish and Japanese learners of EFL!).

…Ulster KIT (which can be very open) heard as LOT; New Zealand raised DRESS heard as non-NZ KIT (this is different from the corresponding American southern confusion, which operates only before a nasal).

…U-RP MOUTH heard as PRICE, and back to New Zealand DRESS heard as KIT……where TRAP is evidently getting so close nowadays that it can even be heard as FLEECE (or perhaps rather just as DRESS, with an unusual pronunciation of semen).

Monday 2 August 2010

125 years

It’s 125 years since the International Phonetic Association was founded. I have been looking at my copy of its journal, lə mɛːtrə fɔnetik, for 1910. A century ago the Association’s quarter-centenary was celebrated in the form of an artiklə də fɔ̃ by Paul Passy entitled œ̃ kaːr də sjɛkl (‘A Quarter of a Century’). In it he records — in phonetically transcribed French — the Association’s “immense success” in the reform of language teaching, the development of phonetic research, and (in third place) the popularization of phonetic notation (la vylɡarizɑːsjɔ̃ d l ekrityːr fɔnetik).

Daniel Jones was to marry Passy’s niece Cyrille the following year, in 1911.

Meanwhile in 1910 we read [At the University of London, starting on the tenth of October, D. Jones will deliver a whole series of phonetics courses for various categories of student (English Phonetics, French Phonetics, Old English, Old French, Experimental Phonetics). These courses will end with three examinations, on English phonetics for foreigners, French phonetics, and English phonetics for the English.]

There’s an enthusiastic paragraph from DJ himself on the subject of Atkinson’s Mouth Measurer. This was before he became disillusioned with experimental phonetics. Mr Atkinson’s invention was later rendered obsolete by the development of x-ray imaging. It’s interesting to see that the overseas price is given with the then standard orthographic abbreviations for shillings (s.) and pence (d.). You’d think Jones might have written “8ʃ. 6p.”. Or even “eit n sikspəns”.