November 28, 2021
by Graham


The only pronunciation I’ve heard for this letter of the Greek alphabet since it was used for the name of the latest Coronavirus variant stresses the first letter, which is being pronounced to rhyme with both ohm and Tom indiscriminately.

And yet the name ‘means’ “little O” as opposed to “big O” which is omega.

If we split the names of both letters into their constituent parts, then the pronunciation is clearly ‘O micron’ and ‘O mega’, but when the two words are fused into one, the traditional English stress rule comes into play, and ‘O mega’, stressed (as two words) with a main stress on both halves, changes its stress to the first syllable of the three, because a short vowel in a penultimate syllable causes the stress to move back to the antepenulitmate, in this case initial, syllable, and we get OM-e-ga (capitals indicate the stressed syllable – I’m using a re-spelling rather than phonetic transcription to allow for variations in the vowel qualities). All three current English pronunciation dictionaries (Oxford, Cambridge and Longman) give this as the principal British English pronunciation, but allow that o-MEGGa is the American version.

When we come to omicron as a single word, the other part of the rule applies: a long vowel in the penultimate syllable retains the stress. As we all know, the word micron has a ‘long’ -i- (actually, of course, a diphthong) – cf microphone, micrometer, microscope, so the pronunciation of omicron ‘ought to be’ with the same vowel: o-MY-kron. This is the first pronunciation given by the same three dictionaries, but allowing OHM-i-kron or OMM-i-kron as less common alternatives.

In my view, the English stress rule is breaking down at the moment (changing might be a better word to use if I could be sure in which direction it is moving), and presumably by analogy with omega‘s initial stress, omicron is shifting to initial stress as well.

Boris Johnson prides himself on his Classical scholarship, but even he was using initial stress in his news conference on 27 November 2021, so I think that any attempt to rein this in, and return to the traditional second syllable stress with a ‘long’ I, is doomed.

November 6, 2021
by Graham

Two deliberately created alternative pronunciations?

During the past week, I lstened to a fascinating radio programme which included a contribution by the world expert on meerkat behaviour, Tim Clutton-Brock.

I was surprised that he pronounced the word for their mating behaviour as /pɒliˈgaɪnəs/. As always these days, not trusting my own memory and usage, I consulted all the pronouncing dictionaries I own, and not a single one gave this as a possible pronunciation.

How could a world expert mispronounce a word which he must have heard from his colleagues many hundreds if not thousands of times?

But then it occurred to me that he may be deliberately mispronoucing it in order for it not to be confused with the very similar, but much more frequently heard word, polygamous.

A similar deliberate ‘mispronunciation’ is used by foreign language teachers in order to distinguish between the two sorts of examination that test the production and understanding of an L2: aural and oral. The two words have identical pronunciations in my accent at least, and I suspect most others, but failing to distinguish between them could have consequences for students and teachers if they were confused. Hence the language teacher’s pronunciation /ˈaʊrəl/ for aural and /ˈɔːrəl/ for oral.

And for those still wondering what word Clutton-Block was using – polygynous, usually pronounced /pəˈlɪdʒɪnəs/.

PS I wonder how he pronounces misogyny and misogynous.

October 25, 2021
by Graham

Notify, mandate and sanction

Thomas West (@IntermarkLS) has tweeted:

“In BrE I keep seeing things like “the last address you notified to the company” and “the complaint notified to the police.” This sounds so wrong in AmE (we notify someone of something; we don’t notify something to someone).”

I refer him to the OED, which gives the following quotation from the Rolls of Parliament in 1433:

“The whiche offre and agrement..[was] notified and communed to all the Lordes.”

On this side of the Atlantic, we have recently acquired an American usage which sounds equally wrong to me, and was until this year very unusual in the UK: the verb “mandate” to mean “compel” (why not use compel, which is not ambiguous?). The problem here is that the normal meaning of the word in British English is “authorise”. The difference, I suppose, is that while both usages are transitive, the AmE usage may be to mandate something, while the BrE usage is only to mandate someone to do something – which AE can also have. So when we read that President Biden has mandated an organisation to enforce the wearing of masks, I wonder why this is so contentious, because to me, the meaning of this sentence is that he has authorised that organisation to enforce mask wearing if it wishes to. The use of notify, however, while it may sound wrong to AmE speakers, at least is not confusing – there is no other meaning of notify.

Sanction is another word with confusing uses. And in this case, the two uses are diametrically opposed. The earliest example quoted in OED is its use as a noun in 1570, where it refers to an ecclesiastical decree (with no sense of either condemnation or approval). Within a hundred years this had become “the penalty enacted in order to enforce obedience to a law”, but at the same time a sanction could now also be the reward for obeying a law (quotations given from 1696 and 1692 respectively).

Since at least the 18th century, a further extension of its meaning has become “something which serves to support, authorize, or confirm an action, procedure, etc.”, with an example from 1727. It is an easy step from there to “permission”.

Interestingly, OED has a quotation from Richard Baxter (1651): “The Law hath two parts, the mandate and the sanction.” Interpret that as you wish!

October 14, 2021
by Graham

The Commonwealth of Poland and Lithuania

The latest edition of In Our Time on Radio 4 (first broadcast on 14 October 2021), the programme that tries to deal in depth with a narrow topic, was about the medieval elective kingdom of Lithuania and Poland (or the other way round).

One of the speakers, as he introduced the topic, described the kingdom as covering an area of “one million kilometres squared”.

Has the difference between “square kilometres” and “kilometres squared” changed since I was at school? (or similarly square miles and miles squared).

To me the phrases “one kilometre square(d)” and “one square kilometre” indicate identical areas, but for any larger amount, the difference between the two will only increase as the numbers get bigger. So, “two kilometres square(d)” is equal to “four square kilometres” (because “2 kilometres square(d)” is equivalent to saying “an area which stretches 2 kilometres along the x axis and 2 kilometres along the y axis”). By my calculation, using this logic, “1 million kilometres squared” is the same as “1 million million square kilometres”. The land surface area of the earth is said to be about 150,000,000 sq km according to Science Desk Reference American Scientific, so presumably the expert speaking this morning meant to say “one million square kilometres”. If an expert can get a figure wrong by this magnitude (which makes the area of Poland and Lithuania about 6,660 times bigger than the total land area of the earth), what hope have we for the rest of the population?

Unless, of course, the meanings of “square kilometres” and “kilometres square(d)” are now identical. In which case, how do we know how to convert an areal measure in miles to its equivalent in kilometres? (4 square miles is 2 miles in each direction, but as 2 miles is equivalent to ca 3.2 kilometres, 4 square miles is 10.24 square kilometres, while 4 miles squared is 16 square miles, or 40.96 square kilometres.)

If these two expressions have now coalesced, then the language has lost a valuable distinction, and may well cause confusion to budding mathematicians, to say nothing of architects and land surveyors.

September 2, 2021
by Graham

/s/ – apical or laminal

I’ve been asked about the pronunciation of /s/ in initial strings of /str/.

I know that it is common to pronounce it with a post-alveolar, apical articulation in Glasgow and London (David Abercrombie was talking about its occurrence in Glasgow at least forty years ago, and I have heard it from many Londoners myself). I’m now being asked how prevalent it is in English as a whole, and as none of my pronouncing dictionaries mention it (why should they?), and John Wells’ Accents of English doesn’t either (unless I’m just not seeing it), I’m hoping that someone can throw more light on it. It’s not a variant I use myself, but that might also be a feature of my age.

Presumably, those who have the apical allophone use it also in their pronunciation of words such as syringe, when said in a single syllable: /srɪndʒ/.

August 26, 2021
by Graham

A new development in English phonology?

With the increasing number of BBC journalists who have South Asian heritage, we are hearing an ever greater number of examples of a non-traditional pronunciation of the orthographic symbol {t} in words of South Asian origin.

I’m thinking in particular, but not solely, of the word Taleban. The initial T is pronounced with a dental articulation rather than alveolar by such renowned presenters/reporters/correspondents as Mishal Husain and Rita Chakrabarti, as well as those whose accents betray a South Asian upbringing. Interestingly, some Afghans who speak extremely good English are using the traditional alveolar place of articulation in the same context. How long before those of another tradition take to imitating them (if they can both recognise the difference, and reproduce it without special phonetic training)?

Will this come to initiate a change to English phonology, with the two articulations beginning to develop minimal pairs, or will it disappear again as the present generation is replaced by their children who will probably have less connexion with the sub-continent?

Time will tell.

August 13, 2021
by Graham

BBC news writing

Are BBC news scriptwriters using a system with predictive spelling? In the last few days, I’ve been hearing nonsensical statements from reputable, and usually reliable Radio 4 newsreaders, and either they are uncharacteristically mis-reading their script, or the script is mis-spelt in a way that I can only attribute to predictive spelling.

First, Alan Smith read out a business piece which included a claim that there were increasing vaccines in several industries (vacancies?), and today, Arlene Fleming has twice (at both one o’clock and six o’clock) talked about events in Suburbiton near Kingston upon Thames.

What is going on? Does nobody check what they’ve written before delivering it to the newsreader?

July 10, 2021
by Graham

A chasm in education

It never ceases to amaze me how people can find new ways in which to mispronounce words.

I have just heard a writer (a writer, and so one would assume a person who is interested in words) on The Radio 4 programme Saturday Live, pronounce, in all seriousness, the word chasm with the same initial sound as in the word church. Yes, I have heard it said like this as a joke, poking fun at English spelling conventions, but never as what was clearly the speaker’s belief that it was the usual way to say the word.

It isn’t even such a rare word. He must have heard ‘kazzm’ many times in the course of his life. Has he never connected the two?

April 30, 2021
by Graham


This is the Scots Gaelic name for Scotland, and has been chosen by Alex Salmond, former First Minister of Scotland for the Scottish National Party, as the name of his new party, set up in opposition to the SNP after his ‘disagreements’, to put it politely, with his former colleagues.

I have been very surprised to hear some of the BBC’s news presenters – although not all – use a pseudo-Gaelic pronunciation for the name. They appear to be trying to follow the pronunciation given by Wikipedia, which uses this IPA transcription for the Scots Gaelic: [ˈal̪ˠapə]. Such a pronunciation will be very hard for non-Gaelic speakers to get their tongues around. Wikipedia does also give the usual English pronunciation as [ˈælbə]. Most Scots – the vast majority of Scots in fact – do not speak a single word of Gaelic, and it has not been spoken natively by Scots in Edinburgh and points south for hundreds of years. It therefore seems to me perverse to want to inflict a pronunciation which is not intuitive to most of the listeners and viewers, who will simply wonder what is going on.

Is this a subtle way of trying to influence the audience in its attitude to the party?

Will pressure be put on the Pronunciation Unit to make the ‘Gaelic’ version its official recommendation?

April 4, 2021
by Graham

Richard III – The Reunion

I have always thought the Radio 4 programme The Reunion as being reliable and well presented. When Sue MacGregor was leading it, I never had any quibbles with it, but this morning, the programme about the discovery of Richard III’s body in a car park in Leicester (they didn’t ask why he was buried in a car park, fortunately), perpetuated the canard that there are at least three descendants of Richard, including Benedict Cumberbatch. I wrote about this some years ago, when the body was first discovered (see post entitled Genealogy).

Kirsty Wark also perpetrated two pronunciations which I am sure I should never have heard from Sue M: she used ‘lay’ as the past tense of the verb ‘lay’ (while Richard’s coffin was lying in the cathedral at Leicester, people ‘lay’ flowers there) instead of ‘laid’, and she pronounced genealogy as if it were spelt ‘geneology’. To paraphrase Maureen Lipman, family historians have an ‘alogy’, not an ‘ology’.

Scots will always tell you that they speak the ‘best’ English. Don’t you believe it – they slip as often as the rest of us.