Thursday 16 April 2009


Edward Aveyard asked
Would you happen to know how Towton (as in the Battle of Towton, the largest battle on English soil) is pronounced? I am not sure whether this would be ˈtəʊtn or ˈtaʊtn.

I didn’t know, but I knew someone who would: the BBC Pronunciation Unit (blog, 13 April).

My former student Jo Kim, now working for the Unit, replied
Martha [Figueroa-Clark] and I were reading your blog yesterday and were especially grateful for your kind (and informed!) words in our defence. You are quite right, we have been recommending the pronunciation for L'Aquila as LACK-will-uh since 1971 and with a special note to draw attention to first syllable stress. We did all we could to disseminate the information but with the resources we have and being an advisory service, we cannot enforce pronunciations, as you very well know.
As for Towton, we have an entry from 2007, which gives the pronunciation TOW-tuhn (-ow as in now). This is the pronunciation given to us by Dr Buckberry of Bradford University's Towton Mass Grave Project.

So there you have it, Edward: it’s ˈtaʊtn.
The Battle of Towton (1461), fought between the Lancastrians and the Yorkists in the Wars of the Roses, was indeed the bloodiest in English history. It is reported that over 80,000 people were killed there, which would represent about 1% of the entire English population at the time.
* * *
This blog will now take an extended post-Easter break. Next entry: 24 May.

Wednesday 15 April 2009

VOT is more

Yesterday’s account of aspiration and VOT only covered initial consonants and initial consonant clusters: more precisely, those at the beginning of a stressed syllable. It covered the English aspiration of [p] in words such as page, pick, appear, spin, price, pure, display, of [t] in words such as time, top, attack, stand, tree, and of [k] in words such as coffee, key, account, scanner, cross, clean, accuse.
But what about [p t k] in other positions? In those positions that I (but not everyone) would consider syllable-final?
It is unfortunately the case that English aspiration is not a matter of all or nothing. In some positions voiceless plosives may have a certain amount of aspiration, but not enough to call them fully aspirated. The VOT in such cases is intermediate between that of “aspirated” voiceless plosives and that of “unaspirated” ones.
As I put it in LPD, English [p t k] are aspirated
• when they occur at the beginning of a syllable in which the vowel is strong.
They are unaspirated
• when preceded by s at the beginning of a syllable
• when followed by any FRICATIVE, as in lapse læps, depth depθ
• if immediately followed by another plosive as with the k in doctor ˈdɒktə || ˈdɑːktər. The release stage of the first plosive is then usually inaudible (‘masked’).
Otherwise, they are unaspirated or just slightly aspirated. For example, ripe raɪp, shut ʃʌt, lake leɪk; happy ˈhæpi, writer ˈraɪtə (BrE), lucky ˈlʌki; wasp wɒsp || wɑːsp, restinɡ ˈrestɪŋ, Oscar ˈɒskə || ˈɑːskər, lifted ˈlɪftɪd, today təˈdeɪ.
Peter Roach suggests the useful word potato pəˈteɪtəʊ (BrE). The initial p is unaspirated (because a weak vowel follows). The first t is aspirated (at the beginning of a stressed syllable). The second one is weakly aspirated (at the end of a syllable, I would say; or you could alternatively say because followed by an unstressed vowel).

Spanish [p t k] are always unaspirated. They have zero VOT.

Another phonological feature sometimes claimed to be involved is [spread glottis]. Aspirated stops and voiceless sonorants are then said to share a common feature of aspiration (or [spread glottis]). This helps account for the fact that in some languages unaspirated and aspirated voiceless plosives contrast in word-final position (see the waveforms for Armenian final [k] and [kh], taken from Ladefoged and Maddieson’s The Sounds of the World’s Languages). It’s no use talking about VOT there: in sentence-final position both possibilities have trivially infinite VOT (there’s no VOT because there’s no voice onset).

Tuesday 14 April 2009

VOT is that?

Carlos Leiva writes
Could you speak about VOT in your blog next time? Drawing the differences between English and Spanish VOTs?
While smoking a cigarette, I think I found a difference when pronouncing some words. I mean, the smoke came out from my mouth differently, I think.

As a non-smoker, I cannot comment on the second paragraph. And it’s not altogether clear what Mr Leiva is asking for in the first.
Yes, the [p] in English pair is aspirated, i.e. has a long voice onset time (VOT). There is a substantial delay after the lips separate before the vocal folds kick in with voicing. During this time air from the lungs escapes unobstructed through the oral cavity, sounding “like a little [h]”.
The Spanish [p] in perro is unaspirated, i.e. has zero VOT (or a very short VOT). There is very little or no delay between the labial release and the onset of voicing, so no “little [h]”.
Unlike pair, the [p] in English spare is unaspirated.
In words like pray, play, twin, cure the aspiration is manifested in the voicelessness of the liquid or semivowel; but in spray, splay, obscure in principle not (because the [s] inhibits aspiration).
Aspiration affects all three English voiceless plosives, [p t k].

Spanish [b d g] are voiced throughout. English [b d g] are only partially voiced, unless surrounded by voiced sounds.
Here is a schematic VOT diagram from Cruttenden, Gimson’s Pronunciation of English (7th ed., Hodder, 2008). (Click to enlarge.)So far, this is the information given in all textbooks of phonetics. But there’s more that can be said. We’ll return to the subject tomorrow.
If you want measurements of VOT in milliseconds, consult a textbook or measure a waveform. The usual value given for English VOT in words like pair is of the order of 40-75 ms.

Monday 13 April 2009


In last Friday’s Guardian, Alexander Chancellor commented:
It took the BBC all week to master the correct pronunciation of L'Aquila, the Italian city devastated by the earthquake. Broadcasters as famous as John Humphrys and Fiona Bruce started out by calling it la-QUEE-la, but more surprisingly, even reporters on the spot got it wrong.
Duncan Kennedy, who was described by the corporation as its Rome correspondent, also called it la-QUEE-la, while George Alagiah, who had hastened to Italy to address us from among the ruins, gave it the more rarefied pronunciation of la-KEE-la, as if it were a Mexican liqueur. It should, of course, be pronounced LA-qui-la (with the stress on the first syllable).
The BBC's pronunciation research unit is sadly not succeeding in its proclaimed purpose "to ensure that pronunciations used on the BBC are accurate and consistent".

The fault lies not with the BBC Pronunciation Unit, but with newsreaders and reporters who do not bother to consult the database the Unit provides. The Pron Unit will certainly have the correct pronunciation on file, and as far as I am aware always responds promptly and accurately to enquiries.
The name L’Aquila is not in British pronunciation dictionaries (perhaps it ought to be), but it is in the Duden Aussprachewörterbuch, as ˈlaːku̯ila.
Once upon a time any educated person had some knowledge of Latin. Not any more. But classicists will tell you that the Latin word for ‘eagle’, the origin of this name, is ăquĭla (with a short i). As readers of this blog will know by now, a Latin word in which the penultimate vowel is short and not followed by a consonant cluster has its stress on the antepenultimate. This Latin rule still works for Italian, as long as you know the Latin vowel quantity, lost as such in Italian.
And by the way, the Latin word for ‘songs’ is carmĭna, so this word too was/is stressed on the antepenultimate. As far as Orff’s Carmina Burana goes, insisting on this stressing seems to be becoming a lost cause.

Sunday 12 April 2009

ˈmeidʒər | ənˈmainə | \toungruːps | in\iŋgliʃ

In his comment on my posting “punctuation in transcription” (9 April), Peter Roach referred to John Trim’s excellent 1959 article in m.f. 112:26-29.
Although when I vacated my room at UCL I had to get rid of a large number of books (lacking the space to keep them all at home), I nevertheless couldn’t bear to dispose of my collected issues of Le Maître Phonétique. That means I am able to offer you scans of excerpts from the article, to give you a flavour of it.

Let’s hope that one day soon the entire run of mf/JIPA will be digitized and made available to all.

Thursday 9 April 2009

punctuation in transcription

Apropos writing an essay in transcription (blog, 6 April), Eric Armstrong asked
And what do you do about things like question marks? Use phonemic transcription plus punctuation? What is acceptable?
For three-quarters of a century the journal of the International Phonetic Association, then called Le Maître Phonétique (lə mɛːtrə fɔnetik), was written entirely in transcription. Here is an extract from one hundred years ago, from the issue dated vɛ̃tkatriɛm ane 5-6 mɛ - ʒɥœ̃ [sic] 1909.You will see here that the custom was to ignore stress and to use ordinary punctuation, including quotation marks. The asterisk denotes a proper name. This is only what one would expect, given that the early ambition of the IPA was to supplant traditional orthography, at least for language teaching.
And here is an extract from a more recent article, the youthful David Crystal’s ə pəspɛktɪv fə parəlaŋɡwɪdʒ (m.f. 1963, 120: 25-29).Here, too, stress is ignored and ordinary punctuation is used. Additionally we find cited terms in italicized orthography.

Wednesday 8 April 2009


Martin Ball writes
I have a Speech Pathology colleague (of many years standing) who I have heard several times recently pronounce obstruent as /əbˈstr.../
I was wondering whether this is just because it is a word she had not heard said often (though in this same meeting I had used it several times), or whether this is a recognized US variant (not in LPD).
I share Martin’s bemusement. I have never heard anything but initial stress for this word, ˈɒbstruənt.
All other English words ending in -uent have antepenultimate stress for most of us, too: affluent, effluent, congruent, constituent. However, some Americans do say conˈgruent with penultimate stress, and some do the same with affluent and effluent, so perhaps this is what provided the model for the obˈstruent produced by Martin’s colleague. (But I don’t think even Americans ever say constiˈtuent.)
Classicists will know that the Latin etymon has a short vowel, obstrŭ-ō, -ent-, and therefore initial stress.
Words with other vowel stems plus -ant ~ -ent include brilliant, radiant, valiant, variant, gradient, lenient, prurient (all antepenultimate), but on the other hand compliant and defiant (penultimate, because of the verbs comply and defy).
A word I don’t think I’ve ever encountered outside early nineteenth century poetry is reboant. The OED says it has initial stress, ˈrebəʊənt. Merriam-Webster agrees (sound clip).

Tuesday 7 April 2009

[h]: fricative or approximant?

María Alicia Maldonado wrote to say that she had come across a textbook in which English /h/ was described as a “voiceless glottal approximant”. Given that in LPD I classify it as a “voiceless glottal fricative”, she wonders which is right.
First, I am not alone in classifying [h] as a fricative. That is how it is categorized on the IPA Chart; that is how Cruttenden, Roach, Collins, Ashby & Maidment and other respected writers on English phonetics categorize it.
Secondly, however, it is clear that [h] is different from the other English voiceless fricatives [f, θ, s, ʃ], in that it does not involve a constriction within the mouth cavity. Conventionally we classify it as a glottal fricative, but the constriction giving rise to the turbulence that we hear as friction may be not so much located at the glottis itself as distributed throughout the whole of the upper vocal tract.
That is why Ladefoged and Maddison in The Sounds of the World’s Languages (Blackwell 1996), at the beginning of their chapter on fricatives, comment
Forms of h, ɦ in which a turbulent airstream is produced at the glottis are also sometimes classed as fricatives […], but it is more appropriate to consider them in the chapter on vowels.
At the end of the vowels chapter they mention the possible description of [h] as the voiceless counterpart of the vowel that follows.
In such cases it is more appropriate to regard h and ɦ as segments that have only a laryngeal specification, and are unmarked for all other features.
But in some languages, including Hebrew and Arabic, a glottal constriction is observable during the production of these sounds.

One problem with classifying [h] as an approximant is that voiceless approximants are by definition inaudible. (Or by one definition, at least. Approximants used to be known as “frictionless continuants”.) If there’s no friction and no voicing, there’s nothing to hear. Anything you can hear during a voiceless [h] must be some sort of weak friction, resulting from some sort of weak turbulence, which means that [h] is some sort of weak fricative — but still a fricative.
The English /h/ phoneme does not behave like a vowel. We say a house, not *an house. We say ðə house, not ði house. You get a linking r in you’re out, but not in your house — except among non-standard speakers who drop h.
The tradition in generative phonology is to class [h] as a glide, along with [j] and [w]. That’s fine phonologically, but not very helpful phonetically.
For practical teaching, it’s convenient to call [h] a fricative. But you do have to emphasize that there is no friction at the uvular or velar place (of the sort you get in [x] and [χ]). Many EFL learners can be helped by thinking of [h] as just a voiceless onset to the following vowel.
Advanced students can be asked to write an essay on the problem of defining the terms fricative and approximant.

Monday 6 April 2009

Regulations for postgraduates

Here, from Speculative Grammarian clvi:2, are some Guidelines for the Behavior of Graduate Students, attributed to the author Felicity Conditions “for distribution by the International Phonetic Association”. I think this may have been published at the beginning of April. (Thanks to Paul Carter for bringing it to my attention.)
ʘ ʘ ʘ ʘ ʘ
1. The IPA symbol for a bilabial click is not called “the cervix,” even if it really looks like one.
2. Students must not giggle every time someone says “labial.”
3. Students are not allowed to start a letter-writing campaign demanding enfranchisement for voiceless phonemes.
4. Students must remember that the IPA only covers the sounds the human mouth can make.
5. Students are not allowed to ask the “cute” TA to produce difficult phonemes for them... slowly... over and over again.
6. “Only prats use Praat” was never funny, and still isn’t. Especially if the faculty use Praat.
ʄ ʄ ʄ ʄ ʄ
7. There is absolutely no mystical significance to the fact that the IPA symbol for a voiced palatal implosive slightly resembles the helix symbol from the TV show Heroes.
8. Students are not allowed to erase everything on the vowel chart besides /a, e, i, o, and u/ and insist that the TA “teach the controversy.”
9. Students are not allowed to turn in papers written entirely in IPA.
10. Students may not be excused from discussions of tonal languages because they are tone-deaf.
11. There is no diacritic for “drunken voice.”
12. Students are not allowed to color in the “blank” areas on the vowel spectrograms.
13. Students are not allowed to convince wide-eyed, trusting first-years that the nasal ingressive voiceless velar trill will be officially added to the IPA chart next year.
aɪ æktʃəli fiːl ə bɪt sæd əbaʊt kəndɪʃn nʌmbə naɪn. ɑːɡjuəbli, ɪt maɪt bi ə ɡʊd test əv stjuːdnts əbɪləti ɪf ɪt wəz tʃeɪndʒd tu
9'. Every student must turn in at least one paper written entirely in IPA.

Sunday 5 April 2009

Vista on-screen keyboard

Jay, commenting on my blog entry for 24 March (Unicode Phonetic Keyboard), mentioned that Windows Vista also has an on-screen keyboard. You find it by going to All Programs | Accessories | Ease of Access | On-Screen Keyboard.
The display is context-sensitive. If you install and engage Mark Huckvale’s Unicode Phonetic Keyboard and then launch the on-screen keyboard you see this:
(It’s about half as big again as shown here.)
If you click on Shift, the display changes to

Click on the righthand Alt (actually called AltGr on a UK keyboard), and you get

Clicking on one of the virtual keys on the on-screen keyboard has the same effect as typing the key itself. This constitutes a very convenient way of entering phonetic symbols.

Thursday 2 April 2009

How to sound American

My former colleague Geoff Lindsey is a film maker and writer of screenplays and TV scripts. But now he is supplementing his income by offering his services as a dialect coach, along with accent reduction, English tuition and forensic analysis.
Some of you may remember him from his days as a tutor on the UCL Summer Course in English Phonetics. In his c.v., for some reason, he keeps quiet about this part of his career, and about the fact that he not only got his BA in Linguistics from UCL but was also a lecturer in phonetics with us for several years.
I’m happy to give him a bit of publicity in this new line of work, because on his website there is an excellent video clip in which he shows an Australian Hollywood lookalike how to sound American by changing just two words in her vocabulary. This is the sort of thing that is sadly lacking from my Accents of English.

Wednesday 1 April 2009

Ask Professor Wells

The CD-ROM accompanying the third edition of my Longman Pronunciation Dictionary comes with a button for emailing questions to me. There’s also one on my home page. I do my best to reply helpfully. But sometimes what a reader sends is not a question but a complaint.
Dear Professor Wells:
I expected the dictionary to show the actual pronunciation of the entries. However, on checking a few examples, the dictionary did not live up to my expectations.
For example, I expected the word train begin with an affricated t and similarly, the word drain to begin with affricated d. There are many other aspects that do not reflect real English (American or British) pronunciation. EFL users of the dictionary do not have authentic pronunciation guide.
Professor XYZ (from an address in Jordan)

[I do find it a bit surprising that a professor, presumably of English, can send me a message with several grammatical errors:
• a dangling participle “on checking”;
• missing “to” before “begin”;
• the countable noun “guide” treated as a mass noun (cf. “guidance”).
If I were writing to a professor, a native speaker of a language I was teaching at university, particularly if I were writing to criticize his work, I would have my letter checked by a native speaker beforehand if there were any doubt at all in my mind. Personally, when I write even a short informal message in French, German, or Welsh, I check everything very carefully in dictionaries and grammars (and am mortified if I do nevertheless make a mistake). Why can’t EFL learners/experts do the same? In spoken communication the odd slip can be tolerated, but anything you write ought to be as far as possible error-free.]
Anyhow, here’s what I replied.
Dear Prof. XYZ,
I am sorry that you are disappointed. On page 465 of LPD (third edition) there is a clear statement that "in a consonant cluster after t or d, r is made FRICATIVE instead of approximant. The result is that tr and dr form AFFRICATES..."
In my view it is better to make general statements such as this rather than to clutter the transcription at each entry with allophonic diacritics to show frication, aspiration, clipping, etc.
John Wells
—a sentiment with which I hope you, my readers, agree.