3

In Accents of English (1982), John C. Wells came up with a useful notation for English vowels that allows easy comparison of the pronunciation of English vowels in varieties of this language. This notation has been widely adopted, and it is now common to talk about the STRUT vowel to refer to the vowels in cub, rub, hum. In standard British and American English this vowel is pronounced as [ʌ], but in Northern England as [ʊ]. What makes this notation extremely useful is that mergers of two vowels can be easily referred to by shorthand, such as the NEAR - SQUARE merger.

Q: Has anyone suggested an equivalent notation for consonants? Such as the TOOTH consonant when referring to what in British and American English is a voiceless interdental fricative? I couldn't find anything online or in the literature. I get why the lexical set notation was first devised for vowels since English spelling is particularly inconsistent when it comes to vowels. But I feel it's still inconsistent enough for consonants to make a lexical set notation in this area useful.

Q: If not, what would be criteria by which to choose keywords to represent lexical sets for consonants?

Improve this question
5
  • 5
    Consonants across English varieties don't vary as much as the vowels. And consonants can be perfectly well specified using IPA, unlike vowels (eg your example of [ʌ], which is not actually standard IPA but a traditional usage for English).
    – Gaston Ümlaut
    Commented Oct 10, 2013 at 21:26
  • 1
    Yes, I know they don't vary as much, but I would still find it useful. I don't really find IPA adequate for this task. /θ/ is pronounced /θ/ in Standard British English - see how it's circular?
    – robert
    Commented Oct 10, 2013 at 21:31
  • 4
    It should be: /θ/ is pronounced [θ] in Standard British English, that way it's not circular.
    – Gaston Ümlaut
    Commented Oct 10, 2013 at 23:58
  • 1
    It's not that I don't get your point, and I do use this kind of notation at the moment. But it' still circular because the phoneme character is /θ/ because in British English is commonly pronounced [θ]. In a variety where it is realised as [t] there would be a merger with /t/, so that only one phoneme remains. It would be useful to refer to sth. like this as, say, a TRUST - THRUST merger.
    – robert
    Commented Oct 11, 2013 at 14:15
  • Expressing it as a 'TRUST-THRUST merge' doesn't give much information: which one is used? or are [t] and [θ] in free variation? or is it conditional on context? I think that no-one uses consonant equivalents to Wells' lexical sets because vowels may be varied in several dimensions (height, backness, length, rounding...) so it wouldn't be as useful, and would in fact obscure what's actually happening with dialectal variation.
    – Gaston Ümlaut
    Commented Oct 11, 2013 at 23:39

2 Answers 2

Reset to default
1

One of the benefits of lexical sets is that they make communication possible across dialects and IPA conventions without positing a single reference dialect which would also have to include an idealised speaker of that dialect.

If you say, dialect X has pronunciation Y for vowel X, you may not know what you're talking about because of the way IPA is used as well as the great variation across dialects.The lexical sets also don't just correspond to individual sounds but rather to collections of words that have a particular sound which can generally be identified. The distribution of the vowels is also not uniform. So you cannot say, in dialect X, replace ɑː with æ but you can say that BATH and TRAP vowels have merged. Or you could say, that some speakers pronounce the i in the HAPPY lexical set as schwa without saying anything about the pronunciation of i in general.

None of these things are an issue with consonants. The IPA is consistent across all dialects, the variation is less and the distribution is more uniform. So you can say things like in context Z, consonant X is pronounced Y. (E.g. intervocalically, voiceless stops are pronounced as glottal stops). Having a lexical set here would only confuse matters.

It seems that what you want is simply examples of consonants in English. And those, of course, are readily available.

Improve this answer
2
  • Of course, this applies in the English context, which has quite low consonant variation across its dialects. Mergers of e.g. THRUST into TRUST or THIRST into FIRST can be expressed more succinctly as replacing /θ/ with /t/ or /f/. But where there is a more complex consonant phonology and where inter-dialectal variation affects it, e.g. the palatalisation in Irish, lexical sets can be useful.
    – Michaelyus
    Commented Dec 17, 2014 at 15:30
  • 1
    You're absolutely right. This is all very much language specific.
    – Dominik Lukes
    Commented Dec 17, 2014 at 22:43
-2

It would still be handy to have keywords for consonants. I tend to think of it the other way round, though, going from spelling to pronunciation. Here is my current list:

B - BAD, DUMB

C - CAT, CITY, OCEAN

CH - CHURCH, MACHINE, CHAOS

D - DOG, BAKED

DG - EDGE

F - FUN

G - BIG, AGE, POIGNANT, FOREIGN

GG - EGG, EXAGGERATE

GH - GHOST, LAUGH, THOUGH

H - HAT, OH

J - JOIN, JUNTA

L - LIME, COLONEL, COULD/TALK

NG - LONG, FINGER, ONGOING, STRANGE

P - POP, PSALM

PH - PHYSICS, HAPHAZARD

Q - QUEEN, MARQUIS, IRAQ

R - RUN

RH - RHYME

S - SAME, CAUSE, BIRDS, SUGAR, FUSION, MARQUIS

SH - SHAPE, MISHAP

SS - PASS, MISSION, POSSESS

T - TIME, EDITION, EIGHTEEN

TH - THIN, THAT

V - VEIN

W - WAS, TWO

WH - WHAT, WHO

X - BOX, EXACT, BORDEAUX

Y - YARD

Z - ZOO

Improve this answer
3
  • 3
    The problem with your list is that you combined orthography and phonetics. E.g. the ss in mission is pronounced the same as sh in shock. Whereas the sh in mishap is pronounced as two separate consonants s and h. The bottom line is - as others pointed out - that in English you don't need lexical sets for consonants.
    – Dominik Lukes
    Commented Dec 17, 2014 at 12:22
  • It's not a problem; it's the whole point. I'm deliberately trying to indicate both spelling and pronunciation.
    – Steven Lytle
    Commented Dec 17, 2014 at 14:40
  • 2
    It's a useful list but you're not in any way replicating lexical sets. Also, you're missing lots of other ways consonants can be spelled: en.wikipedia.org/wiki/English_orthography#Consonants
    – Dominik Lukes
    Commented Dec 17, 2014 at 22:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.