The topic of nasality has come up on the Facebook page of Pronunciation Best Practice. A question asked there by Daniel Schulte made me check on what Wikipedia has to say on the topic, and this has raised some other issues concerning the WP material.
 
Here's the original question:

I've got a question about nasality (not only the m n ing sounds)

as far as I know: The air from the lung can flow through your mouth and nose ok. when the air flows a lot through your nose the voice sounds nasal. but why does a blocked nose (a cold or pressing the nose together with your fingers) lead to a nasal voice when its actually causing that the air is only flowing through your mouth?

also: is the fact that some accents (standard general american compared to standard southern british, or portuguese compared to spanish) sound more nasal, a direct consequence of the exact tongue/lips/jaw configuration within the vowel chart or is that a separate, independent aspect? thanks.

The first response posted to this question was from Anthony Lauder:

Before you can release air through your nose you have to first build up air pressure in the nose. The more "blocked" the nose, the more pressure you can build, and the more "nasal" the sound will be. When you have a cold, then, it blocks the nose and helps create that air pressure leading to nasal sounds.

This was followed by a contribution from Petr Roesel.

When you block your nose, the velum usually hangs down, i.e. the nose as a resonator is added to the oral cavity. This enhances certain frequency bands, which cause the impression of nasality


The comment I added to this Facebook discussion was this:

In answer to Daniel Schulte, it's a fallacy that a blocked nose leads to a "nasal voice". Read the Wikipedia article on Nasalization, and note "Denasalization". See also the WP article on Hypernasal speech. If American or Portuguese speech sounds more nasalised, that must be because the speech contains more nasalised vowels (phonemes or allophones) than the accents you are comparing them with. Nothing to do with tongue/lips/jaw configurations. In answer to Anthony Lauder, I'm afraid you've got it completely wrong. It's nothing to do with pressure, and if anything the more blocked the nose is, the LESS nasalised the sound will be. And Petr Roesel, please can you tell me where you got the evidence that the velum is lowered when the nose is blocked?


So far, nobody else has joined in the discussion. Since I recommended reading relevant bits of Wikipedia on the subject, I thought I ought to check on what there is. There's a good article on Nasal vowel and another on Nasal consonant (though editors have flagged problems with references in this latter article). In addition there is an article on Nasalization, which contains quite a lot of information relevant to the question, and contains a brief section on Denasalization, with a link to a separate article on this topic. The article on Denasalization contains a few questionable points. It states "When one speaks with a cold, the nasal passages still function as a resonant cavity so a denasalized nasal [m͊] does not sound like a voiced oral stop [b], and a denasalized vowel [a͊] does not sound like an oral vowel [a]." It seems to me that the nasal passages are only likely to function as a resonant cavity if the soft palate (velum) is lowered, as would be the case in the two examples given. When the soft palate is raised, it seems to me that the nasal cavity must be effectively sealed off from the vocal tract, hence my query to Petr Roesel.

The other worrying thing in the Denasalization article is the statement that "[Denasalization] ... occurs when the sinuses are blocked from a common cold, when it is called a nasal voice". The article adds the caution that this "is not a linguistic term". It turns out there is a Wikipedia article on Nasal voice, and I must say I don't feel very happy about it from the phonetic point of view. The lead section says "A nasal voice is a type of speaking voice characterized by speech with a 'nasal' quality", and a comment has been added on the Talk page of this article, correctly pointing out that this "definition" is "self-referential and therefore not helpful". A WP editor has added a [clarification needed] note. The lead continues "It can also occur naturally because of genetic variation.", which seems to me meaningless as it stands. The article then states that the term Nasal voice may be used to refer both to hyponasal and to hypernasal speech (i.e. speech with too little nasality or too much), thus referring to two opposite entities. This view would, if correct, support the implication in the original question by Daniel Schulte that a blocked nose (causing hyponasality) would lead to "nasal voice", but there seems little or no justification for using the term "nasal voice" in this way. The article's citations to justify the use of the term are two: one is a brief note in the website of an American speech therapy clinic, which talks about Nasal sounding voice and warns that "Nasal voice is an incorrect term as the voice itself is normal". The other citation is of a short section of a textbook called Basic Clinical Radiobiology , and the article seems to be largely based on this work. I need to do some more reading about this topic in the speech pathology literature, to see how widely used the term nasal voice is there, but my current feeling is that from the phonetic point of view the article is of doubtful value.

For once, here is a bit of criticism that doesn't concern Wikipedia (from which I have been taking a rest). I am bothered by an article about the French people and their language, and in particular their supposed fondness for saying "no". You can read the article, entitled "Why the French Love to Say No" by Sylvia Sabes, here:
http://www.bbc.com/travel/story/20190804-why-the-french-love-to-say-no?ocid=ww.social.link.email and you can see that it was published recently on a website called BBC Travel (which I had never heard of before). I wouldn't usually comment on something so far off my field, but the article came to my notice via Facebook when a teacher of English pronunciation recommended it to a group of pronunciation teachers because it " .... tells us about the complexities of teaching our students to communicate in a new language - the subtleties that each culture brings to dialogue. Not just a matter of articulation and prosody. Not to mention the nuancing of tone in English that can turn a statement into its own opposite". I completely agree with the sentiment, and I realize the article is not meant to be taken too seriously, but the article itself seems so riddled with unsubstantiated generalizations, wrong facts and a condescending attitude towards the French that I feel I want to challenge some of what is said.

There is a story running through the article that recounts how the author is on the phone in Paris trying to get an airline to change a booking, and the airline employee keeps saying that this is not possible. The author, we are told, is wise to this Parisian ploy, and she will not take "no" for an answer. In the end her persistence pays off and she gets the flight changed. I am reminded of the condescending tone of Peter Mayle's book A Year in Provence, where our British hero outwits the shifty French workmen who are supposed to be restoring his house but are always behind schedule; he tricks them into finishing on time.

Apart from this anecdote, we get quite a lot of amateur linguistics and sociology, much of it quoted from various publications. There is a quote from something written by comedian Olivier Giraud: “Answering ‘non’ gives you the option to say ‘oui’ [yes] later; [it’s] the opposite when you say ‘oui’, you can no longer say ‘non’! We must not forget that the French are a people of protest, and a protest always starts with a ‘non’.” I find that many people are fond of explaining things about French character and culture in terms of the French Revolution, and this article goes this way, on up to the Gilets Jaunes protests. This all apparently has to do with French people preferring to say "no".

Sylvia Sabes goes on to quote extensively from a book called "The Bonjour Effect: the Secret Codes of French Conversation Revealed". Apparently “The French Revolution was about the irrevocable right of all citizens to refuse, and ‘non’ has a quality of ‘revanche des petits contre les grands’ [revenge of the underclasses] that seems to satisfy the inner peasant or proletarian in every French person, of any class".

Next we hear the views of "Business School professor Erin Meyer", whose book The Culture Map: Through the Invisible Boundaries of Global Business" is quoted at length. The quality of the material in this book (which I have not and probably will not read) can be judged by the breathtaking insight that "there are 500,000 words in the English language, but only 70,000 in French. This means that Anglophones are more likely to have the exact word to say what they want, whereas Francophones must often string together a series of words to communicate their message." It goes on "This not only forces the French to be more creative with language, it also allows them to be more ambiguous with what they want to say. As a result, ‘non’ in France does not always mean ‘no’." You should try presenting this insight to some first-year Linguistics students. There is a long explanation of different types of French "no" (the different "no"'s, including the "flirtatious no", seem to me to be exactly like different types of "no" in any other language I know about), passing through points such as the French terror of being proved wrong in public (this is to do with their educational system, it seems). If this is based on something like a scientific study it could have some merit, but I suspect it is made up from armchair speculation and remembered anecdote.

According to Sylvia Sabes, the French may require a handwritten application for an official document because "While an online document to be reviewed in advance would seem more effective, this lack of an official list empowers the clerk to say no at various stages throughout the application process. In fact, organised French citizens don’t need a list to know that for any administrative issue, they had best arrive at their appointment with copies of their birth certificate and proof of permanent address issued in the last three months, as well as proof of their identity and banking information, all photocopied in triplicate. While this sheath (sic) of documents does not guarantee success, it is considered a reliable shield against countless nos.

What leads me to doubt this stuff? Well, I have done occasional work for the French government since 1962, and lived part-time in my house in France for the last fifteen years, and the attitudes described simply don't fit with my experience. We spent yesterday evening dining in our village square with lots of our French friends and I spent this morning hiking with a group of French fellow-walkers. I can't think of any way in which any of those friends (including the Mayor of the village) ever show a tendency to say "non". I have spent lots of time in French offices dealing with such things as building permission, land boundaries, local taxes or car importing, and never found the people who work in them obstructive or negative. It's a bit depressing to read a condescending piece based on a dubious national stereotype, though I admit there is a long tradition of Brits making jokes about the French. I can't help feeling that such stereotypes are partly responsible for British people having been persuaded to vote to leave the EU.

I feel in need of some expert advice on the pronunciation of a couple of names that are in the news. The first is Beira (city in Mozambique, currently badly damaged by storm). On radio and TV, there seems to be a 50/50 divide over whether the first syllable has /ai/ or /ei/. All the pron dictionaries I have looked at have the /ai/ diphthong for British English, including CEPD, but if the name was given by Portuguese-speaking colonisers I'd have thought /ei/ was more likely. Wikipedia gives no pronunciation for the city in Mozambique,  but Beira in Portugal is given as ˈbɐjɾɐ. 

The other name is the surname of the Prime Minister of New Zealand, Jacinda Ardern. The 'r' in the second syllable seems to be causing confusion. Several broadcasters produce a stress on the second syllable, as in the word 'astern'. Others give the name initial stress but a long vowel in the second syllable (sometimes the second syllable is pronounced so emphatically that it sounds like Speaker John Bercow's way of shouting "order"). Wikipedia gives the pronunciation as  ˈɑ:rdɜ:rn. I would have expected stress on the first syllable and a final syllable with schwa, as in 'Malvern', but I don't hear that much.

I have recently done some editing of the "Notable speakers" section of the Wikipedia article on Received Pronunciation. As you may have read in an earlier post, I added some YouTube links so that readers could listen to an example of each speaker. These were removed by an editor, who has now gently explained to me why such links are not appropriate for WP articles. I have deleted the complaints I wrote in this blog about the deletion.

The YouTube links can still be found here. I hope that they may be useful.

I would be interested to get any comments on this material. To me, it does little to support the idea that there is a single identifiable accent that could be called RP, even if you divide the speakers into different sub-groups and label their accents "Conservative RP", "Mainstream RP", "Advanced RP", "Juvenile RP", "Luvvie RP" or whatever.

[NOTE: I HAVE NOW EDITED THE WIKIPEDIA PAGES ON 'VOWEL LENGTH', 'ENGLISH PHONOLOGY' AND 'RECEIVED PRONUNCIATION' TO CORRECT THE VOWEL LENGTHENING/VOWEL SHORTENING PROBLEM DESCRIBED BELOW]

​I have noticed that an old fallacy survives in a couple of Wikipedia articles. In Vowel length and in English phonology you find the statement (made without any citation to support it) that English vowels are lengthened when followed by voiced (lenis) consonants or in an open syllable. The correct statement is that vowels are shortened when they are followed by a voiceless (fortis) consonant. You can see this in the descriptions in Cruttenden’s Gimson, section 8.4.1. The basic scientific work was published in 1970 by Matthew Chen in an interesting article entitled ‘Vowel length variation as a function of the consonant environment’ (Phonetica, 22, 129-59). I have written about this issue very briefly in ‘English Phonetics and Phonology’ (page 29 in the latest edition, in “problems and further reading”, 4.4.)  A simple “do-it-yourself” test is to measure your pronunciation of ‘ride’, ‘rye’ and ‘right’. The duration of the diphthong in ‘rye’, being in a zero-coda syllable, may be taken as the baseline duration. ‘Ride’ will show a very similar duration, but ‘right’ will have a markedly shorter duration. I will edit the two articles (and any others I find with the same error) and also remove the “generative rule” included in the Vowel length article (which is not only ad hoc and unnecessary, but also wrong).

Just a follow-up to my recent post about the unsatisfactory section headed "Prosody and intonation" in Wikipedia's article on Vowels. I've removed quite a lot of this section and rewritten the rest, but kept the title Prosody and intonation. I hope that has improved it a bit. At least it now has a couple of references.

Some time ago I grumbled that the Wikipedia article on Articulatory Phonetics had plenty of information about consonants but only a couple of sketchy notes about vowels. I've made a start on improving the situation by editing in a summary of the main articulatory variables affecting vowels, but it's pretty sketchy, like its predecessor. I hope that at least the framework is more appropriate. I have made no attempt to provide a detailed treatment of each of the variables - they are already amply covered in the huge network of overlapping phonetics articles in WP.

I've had to read the material in the corresponding WP article on Vowel as I went along. Something that has puzzled me is the statement there, in the section "Front, raised and retracted", that The conception of the tongue moving in two directions, high–low and front–back, is not supported by articulatory evidence and does not clarify how articulation affects vowel quality. Vowels may instead be characterized by the three directions of movement of the tongue from its neutral position: front, raised, and retracted. There is a reference to a paper by John Esling at the end of this section, so I suppose this comes from him (I haven't been able to get hold of the article yet). I'm afraid I can't see why the traditional open/close and front/back dimensions are inapplicable, as long as one doesn't treat them as absolute determinants of vowel quality.

The Wikipedia coverage of vowels is on the whole pretty comprehensive. In the article on Vowels all the variables that characterize vowels are listed one by one, with the odd exception of vowel length or quantity. To find information on that topic, you have to look further down in the article, to a section entitled Prosody and intonation. Something that immediately looks dodgy about that section is that it carries no references to published material, something that in general Wikipedia does not approve of. A lot of the section is, as you might expect, about stress, tone and intonation, all of which are aspects of speech that I think of as syllable-related, not specifically vowel-related, and all of which are covered comprehensively in other WP articles. It seems to me that basic information on vowel length is relevant and should be retained, but the rest of the material should be removed. However, a sentence in the lead paragraph might perhaps mention that vowels are involved in prosodic variables. What’s left should be re-titled “Vowel quantity” and moved up to be appended to the list of vowel variables.
There is, in fact, a separate WP article on Vowel length, about which I will write some comments later. Meanwhile I will try to tidy up the Vowel article.

Wikipedia has a rather cursory treatment of the term Phoneme, and I noticed that there was almost no mention of the possibility of choosing alternative analyses of phonemic systems such as the English vowel system, in spite of the fact that there was a great debate during the early and middle 20th century (before Generative Phonology) around this topic. It was usually known as the “God’s Truth vs. hocus-pocus” debate. A key paper was Yuen Ren Chao’s catchily-titled ‘The non-uniqueness of phonemic solutions of phonetic systems’. When I first taught in the Linguistic Science Department at Reading University we used to get the first year Linguistics students to read the various papers on the subject in Martin Joos’s collection of papers Readings in Linguistics 1 (I never liked the sequel, Readings in Linguistics 2). To be honest, I don’t think the students really enjoyed reading them. Still, it seems odd that so many publications on the phonemes of English still seem stuck in the God’s Truth camp. It seems to be commonplace to state without reservation that English (well, RP) has twenty or more vowel phonemes. Sometimes writers note in passing that this is a bigger total of vowel phonemes than one finds in the world’s other languages. It is, indeed, massively bigger, considering that most other languages have only a handful. Yet it is perfectly possible (as I have pointed out before) to analyse the vowel phonemes of English in a different (perhaps hocus-pocus) way that reduces their number to seven or six, if one should wish to.
​
I have now added a section to the Phoneme article called “The non-uniqueness of phonemic solutions”, and I am waiting to see if anyone objects to it. So far, it has just received one attack of vandalism!

A happy New Year to all.

[NOTE: I HAVE NOW ADDED NAMES OF SOME WOMEN WHO HAVE BEEN DESCRIBED AS RP SPEAKERS. NOT EVERYONE WILL AGREE WITH THE CHOICES, AND THE SOURCES ARE NOT IMPECCABLE]


I don't want to keep coming back to the Wikipedia article on Received Pronunciation, but it's one of WP's most controversial topics and it gets a lot of attention (some of it unwelcome). What I am complaining about at the moment is the section on "Notable RP speakers", which has been there for years. The problem with this section is that all the speakers listed are male, with the exception that the reference to The British Royal Family necessarily includes some royal female speakers.

I could easily come up with a list of female speakers that I would consider to be speakers of RP as defined in this article, but I'm not going to. For a start, I don't believe in RP any more than I believe in Father Christmas. If I did believe in it, I wouldn't include the Royal Family as RP speakers. Finally, it's not possible just to add new names to the list of notables, because only people who have been cited by name in a published source will be acceptable to Wikipedia (Dame Vera Lynn used to be on the list, but I removed her name because no reference was given). What all the listed speakers have in common is that they have been classed as RP speakers by John Wells in his blog, which is acceptable as a citation. Phoneticians other than John Wells are acceptable sources.

So here's the challenge: if you would like to see a proper representation of women speakers and you care about what WP says about RP, publish your list of female speakers somewhere other than WP (a blog will do), and the names can then be added to the present list. If you prefer, you could send me your list and I'll publish it here.