The General Central Northern

Non-Dialectal Pronunciation of England

1. At p. 349 in Volume 2 of his invaluable Accents of English (1982) Professor J. C. Wells remarked that "about half of the English speak with some degree of northern accent". The geographical region which is the subject of the present account is identified geographically by Wells (ibid. p. 350) thus: "The middle north includes the densely populated industrial belt straddling the Pennines from Manchester through Huddersfield, Bradford and Leeds, to Sheffield". Thus the speech of this central northern region of England is notable as having a particularly large number of speakers compared with the other accents of the British Isles. It will be noticed that this excludes peripheral locations such as the conurbations of Birmingham, Liverpool and Newcastle which Cruttenden 2008 at § 7.6.4 naturally includes under the heading 'Northern English'.

2. Most of the linguistic attention to this area has been given to the demotic 'broad' accents of speakers of the least sophisticated kinds notably by the Leeds University 'Survey of English Dialects' (carried out from 1950 to 1961). On the present occasion it is proposed to consider the usages of the relatively sophisticated speakers of this central northern area. It is observable that very many such people exhibit only fairly moderate differences of pronunciation from the equivalent groups in other, including the more southerly, parts of England. In such places many of the more sophisticated speakers employ the General British (hereafter 'GB') English accent of Great Britain which by definition exhibits no very readily identifiable regional features.

3. In the title the General Central Northern Standard Pronunciation of English (hereafter 'GCN'), the term 'central' is used to indicate the exclusion of the far northern areas of Tyneside and Tees-side and the peripheral northern ones of Merseyside, East Yorkshire and the northerly midlands. The expression 'sophisticated' refers to the exclusion of consideration of markedly demotic or strongly localised usages. It may be taken that in daily conversation the speakers in question make use essentially only of standard English grammar and vocabulary. It is therefore what should be designated as a Standard Accent of English.  The observations upon which the descriptions have been based have been made in the course of 35 years or so spent residing in the area (within the Wakefield and Leeds postal districts) and while listening, to a considerable degree systematically, to a fair amount of regional broadcasting and of national broadcasting by GCN speakers. Much, but by no means all, of what we have to say here about GCN is also applicable to the peripheral northern districts mentioned and indeed even to some more southerly ie north midland areas. GCN speakers include many who can not be distinguished from GB speakers immediately one hears them but reveal their GCN affiliations almost only in their treatment of the vowels described in the next paragraph and their preference for what is popularly referred to as the "flat a". Others are more instantly observable as GCN speakers by certain voice quality and rhythmic features that are quite hard to define, notably varieties of nasality that differ slightly from the ones common among GB speakers. On the other hand one quite often comes across (unknown) speakers whose accents seem completely indistinguishable from General British ones in every way — segment qualities, intonations, rhythms and voice qualities — who have been speaking for perhaps some minutes (though more often a minute or so) until one suddenly notices either preference for a strong vowel in the prefix of a word like contain or for the use in a word like bath of the short vowel that both they and GB speakers employ in a word like maths. This is indicative of how very close to GB usages those of many GCN speakers are.

Strong vowels in unstressed pretonic syllables

4. The most pervasive GCN characteristic that contrasts most obviously with the whole of the rest of the English-speaking world is the preference in ordinary unselfconscious speech for a rhythmically strong vowel in unstressed but strong immediately-pretonic monosyllabic prefixes and similar syllables that are 'closed'.

Wells at Accents of English pp 362/3 said “Northern speech tends to retain strong vowels in certain environments where ... other accents show weakening ... Latin prefixes such as ad-. con-, ex- when pretonic." He gives seven examples but doesn’t point out explicitly that the reason why whereas eg concern has the strong vowel but connect has predictably the weak is because Northern initial syllables ending with a strong consonant (ie non-approximant non-liquid) sound immediately followed by another strong consonant regularly maintain a strong vowel .

Thus in Northern English the vowel of an unstressed initial syllable is usually not converted to an ordinary schwa or to a 'front schwa' / ɪ / or eliminated, as generally happens in all types of English in familiar vocabulary items, but exhibits the rhythmically strong value that a vowel has in English as a rule if it bears any degree of stress as eg in the first syllable of `expert or the first and second syllables of 'expec`tation. Words embodying this feature include many very common ones such as absorb, abstain, admit, advise, complain, computer, concise, confuse, contain, employ, enclose, endure, enjoy, entire, entwine, example, expect, explain, object, obtain, observe, etc. Some such words have common GB mainly subvariant versions with strong vowels eg accept, acknowledge, agnostic, antagonise, Antarctic and various other relatively uncommon and learned words.

5. Where the unstressed initial syllable ends with a vowel sound, GCN usage is generally no different from that of practically all of the rest of the English-speaking world eg in words like abrupt, abuse, accuse, acquire, address, affect, afraid, agree, amuse, apply, attract, collect, commercial, commit, community, connect, correct, oblige, offend, oppose, produce, promote, suggest, suppress, etc. It will be noted that the weak vowel in these initial syllables is followed either by a simple consonant or by a cluster ending with one of the approximant consonants / l, r, j / and / w /.

6. The foregoing comments are not to be understood as suggesting that GCN speakers invariably use the non-GB strong-vowel forms but that such forms are extremely common and for most GCN speakers the predominant usage in most words. However, in a small number of words, eg companion, expect etc, for reasons that are difficult to decide about conclusively, a strong vowel may well not be used by a considerable number of GCN speakers who in most other such words have the usual GCN preference.

7. Occasionally but not at all regularly a GCN speaker may be heard to employ a strong vowel in initial syllables not in the strong category described above. This could possibly in some cases be a product of more than usual influence of the orthography but it is a kind of variation that can occur among GB speakers. The especially heavy prefix trans- as eg in translate which may be heard from a small minority of GB speakers in a few words as /tr(ə)nz-/ may well represent among GCN (or former GCN) speakers a hyper-adaptational avoidance of the perceivedly less prestigious GCN strong-vowel-in-prefix pattern.

8. Sometimes, in a departure from their usual pattern of using weak vowels in weak prefixes, GCN speakers pronounce words like suspect, sustain etc as /sʌ`spekt, sʌ`steɪn/ compared with GB /sə`spekt, sə`steɪn/. 

9. Unfamiliar learned or technical words may well not conform to the patterns found with everyday items. For example words like the medical terms ectopic and endemic seem ordinarily to be begun with a strong vowel in practically all forms of English. On the other hand many, it seems particularly Londonish (popularly 'Estuary') speakers, may opt for a strong-vowel version of various common words beginning with the ex-.

10. Although most regularly observable in monosyllabic pretonic closed syllables which are prefixes, there are indications of such a GCN tendency to use a strong-behaving vowel in some other similarly structured words eg harmonious, historical, July, leukemia, mosaics, museum, musician, united, Victorian etc which are not so treated in most other accents of English.

11. It can be said that the vowel / ɪ / is ambiguous in terms of strong or weak rhythmic value so that its feeling for the speaker on any occasion has to be extrapolated from their general practice and therefore can be differently interpreted for a GB speaker and a GCN speaker when exactly the same prosody has been used by each. For instance ˈIg ˎnore it or ˈVicˎtorian can be quite emphatic for a GB speaker but represent a neutral prosody for a GCN speaker.

12. A somewhat related circumstance is that the word / ɪn`kwaɪə / is probably no less often so pronounced by GCN speakers than by GB speakers. Since it is in ordinary orthography, despite its traditional listing in the OED etc as a mere subvariant spelling of inquire, more often written and printed (at least in everyday non-official applications) as enquire, one might have expected GCN speakers to pronounce generally as / en`kwaɪə / but they don't necessarily do so. However, the fact that the spelling inquiry is more often preferred in the north than in the south of England in everyday applications as opposed to legal etc tradition is probably a reflection of spoken usage.

13. A consequence of the preference for the strong vowels of the kinds we have been noting is that in certain prosodic contexts some GCN speakers quite often use accentuations such as a 'com(')binedˎforce, the 'ex(')istingˎsnow, on 'ex(')posed ˎcoasts, 'im(')proved offers, the 'Con(')servative ˎParty etc with or without the second (lower) level tone or with a single pretonic tone which is a descending type.

The bath and chance types of vowel

15. The next most notably characteristic GCN feature is the use of the / æ / phoneme (actually in most GCN and much GB a short almost open and almost front vowel) in words like bath, laugh, glass and chance (though not in items like half, can’t, banana and palm). As Wells remarked (ibid. p. 354) such a usage "extends much further up the social scale than does" use of the put vowel in words like cut, adding the lively comment: "There are many educated northerners who would not be caught dead doing something so vulgar as to pronounce" cut-type words with [ʊ], but who "would feel it to be a denial of their identity as northerners to say BATH words with anything other than short [a]". The strong prefixes preference is comparable to the bath vowel preference in sociological value.

16. He also offered the perhaps less confidently endorsable opinion that one or two such words depart from this pattern as a result of "school-inspired standards of correctness" an instance being "master, which, he added, is often pronounced / 'maːstə / by northerners who would nevertheless pronounce plaster and disaster with short / a /". Oddly enough, though I haven't particularly noticed master so pronounced, I have observed plaster with the long vowel from speakers from whom it was mildly surprising. (An OED note mentions that N.E.D. (1907) recorded a late ME form plaaster.)

17. The fact that even the most speech-conscious GCN speakers so generally resist any temptation to adopt the prestigious long vowels in almost all such words may owe something to an understandable uncertainty regarding which words in GB receive the "broadening" treatment. The fact is that a very large proportion of the words with such spellings have the short vowel in GB (more than twice as many as have the long vowel) and that a substantial group (of perhaps 50 common words) vary from speaker to speaker within GB over whether the short or the long vowel is favoured.

18. It may well be further evidence of the perhaps surprising high status of this GCN vowel choice that the Scot James A. H. Murray should have decided to adopt a special "avowedly ambiguous" phonetic symbol for the original Oxford Dictionary to deal with words like pass and command because he considered that they were "variously identified by different speakers with the a in man and the a in father". Present GB usage was only established in the latter nineteenth century. It is likewise noteworthy that subsequent authorities including Robert Burchfield, the great lexicographer of the latter twentieth century, should have maintained such a custom despite declaring the intention of representing only the "educated speech of southern England". The OED second edition of 1989 regularly added the ash symbol (æ) after such words in turning over the notation from Murray's set of symbols to one employing the International Phonetic Association's alphabet which provides no such ambiguous symbols. A similar policy persists in the current OED.

19. A curious usage which relates to the northern preference for a short vowel in a word like bath can be seen in the speech of some individuals who, apparently rejecting the alien "southern" way of saying a word which they seem to imagine (having presumably heard it mainly from GB speakers) to be spelt "*athritis" give it the pronunciation /æθ`raɪtəs/. For a number of years this could be heard in a much-repeated tv advertisement in which a dear little old lady expressed her satisfaction with a particular brand of armchair. She was not at all unique in adopting this way of saying the word.

Words with the vowel of cup, cut, luck etc

20. Probably the most immediately recognised north-versus-south-of-England contrast is the absence from the speech of typical at least demotic northerners of the vowel contrast heard in pairs like could and cud, look and luck, put and putt. This absence is fairly complete amongst the least sophisticated central northern speakers but the GCN speaker most of the time maintains a fairly clear contrast either by using in words like cut a variety of schwa or having something not very noticeably if at all different from the mainstream GB (General British) value. As O'Connor (1973) remarked "In educated Yorkshire and much other Northern speech ... the vowel is central and above half open". The more front-tending varieties of southern / ʌ / are not to be heard in GCN but quite retracted values may occur even alternating in the same speaker with types of schwa either randomly or according to context, so much so in some cases that southern listeners may have the impression of / ɒ /. The Leeds-born dramatist Alan Bennett is an example of this. GCN speakers may quote the usage of demotic speakers in very characteristically northern expressions such as "/ mʊʃɪ / ie mushy peas".

21. Speakers of high social standing who maintain / ʊ / in at least some cup-type words may be met with sporadically. Some of them, but not nearly all of them, in doing so can tend to sound rather eccentrically or aggressively assertive of their northern background. Incidentally I have never observed any GCN use of such striking hyperadaptive forms as / `ʃʌgə / for sugar or / `bʌʧə / for butcher. The remark by Wells (ibid. p. 353) that the word someone as ['sʌmwən] is in effect a sort of shibboleth for GCN doesn't chime with my impressions because such a version strikes me as well within the range of pronunciations used by GB speakers, certainly as a weakform in a phrase like ˈsomeone `else.

22. A notable consequence of the fact that GCN speakers often employ a schwa or something nearer to it than an ordinary GB-type / ʌ / is that certain ambiguities arise where weakforms of an, and and of are involved. Wells 1984 (p. 67) gives the examples in northern speech of an ending vs unending and a large and tidy room vs a large untidy room. This kind of thing occurs also where the word other from a northern speaker may correspond to a southerner's of the as in some of the time versus some other time. I have, for instance, momentarily misheard what other book she likes for what are the books she likes. Examples of such ambiguities are very frequent. They include many rather misunderstanding-causing ones of the type involving contrast of and and un-, eg labelled and classified versus labelled unclassified, considered and approved versus considered “unapproved”, the tomb was found and opened versus the tomb was found unopened and home from work and well versus home from work unwell.

In Leeds Studies in English (1968) Charles L. Houck reported a randomly sampled survey of over 100 Leeds speakers one conclusion from which was that 'the vocalic segmental phoneme they used in "bud" seemed to indicate that the linguistic behaviour in Leeds is a function of socio-economic class'.

The unstressed word-final -y vowel

23. The other most interesting vowel of GCN to compare between GB, GCN and traditional central northern demotic dialect is the one Wells referred to as 'the happY vowel'. He remarked (ibid. p. 362): "in the heartland of the north – places such as Manchester and Leeds – this vowel is most often [ɪ] clearly to be identified phonemically with the /ɪ/ of KIT". This no doubt was and still is perfectly true of demotic speech but at least from the seventies onward it has become steadily clearer that for GCN it applies to only a certain proportion of speakers despite the social unimpeachability of the / ɪ / vowel choice. The preference is for many rather more sophisticated speakers now not for the rhythmically strong value of many southern, especially London-influenced, but for a weak /i/ of a closeness which relates to the speaker's value for stressed /iː/ and accords with mainstream GB usage. GCN speakers today are less likely than formerly to use for example exactly the same vowel in each of the syllables of a word like mimicry or visibility. Which choice predominates it is hardly feasible to say. Cf my comment on "Open-y values" viz "one often hears quite markedly close values higher up the social scale" in Studies in the Pronunciation of English (ed.) Susan Ramsaran. London: Routledge (p.161, 1990) a version of which is Item 2 of Section 3 on this website.

Other vowels

24. GCN speakers do not in general very noticeably if at all lengthen their value for /iː/ as in street or for /uː/ as in boot or for the initial elements of diphthongs like near and cure which are sometimes ascribed such lengthening in the most traditional dialect speech of the area. Nor is their much noticeable use of monophthongs rather than diphthongs in words like face and coat. Both the cap and palm vowels are usually heard in versions which in themselves do not clearly counter-indicate a GB accent being usually quite within the range of the later-twentieth-century drawing nearer together of GB / æ / and / ɑː/. The use of /uː/ in words like book and look has never come to my notice from GCN speakers.

25. It is probable that GCN speakers use slightly fewer weakforms than GB speakers possibly making slightly less use of reductions of Sir and Saint to / sə / and /snt/. On the other hand weakening of non-auxiliary uses of parts of the verb to have seem less usual than they possibly once were. For example 'He's a dog' was once a familiar way of saying 'He has a dog' that seems now quite recessive.


26. The use of / ŋ / and the / ŋg / sequence matches GB at all social levels. So does the value of / r / in GCN though non-GCN speakers may often be heard to flap / r / quite vigorously especially after labial consonants in words like bright etc. Realisation of / t / as approximant [ ɹ ] eg in Shut up ("Shurrup") is mainly classifiable as deliberately rough or casual from GCN speakers. Wright in 1906 at EDG §294 remarked "The change of final t to r in monosyllables with short stem-vowel occurs sporadically in most parts of England, when the next word begins with  a vowel, as ger əm get them, ler ɪt biː, ser ɪt, wɒr ɪz ɪt. It occurs far more frequently in Yorkshire, Lancashire and the north Midlands than elsewhere".

27. GCN speakers usually show little or no difference from GB usage as regards aspiration of / p, t, k / or / ʧ /. The same can be said of their use of light and dark varieties of / l /. Dropping of aitches is regarded by GCN speakers as unacceptably 'uneducated' to just the same degree as by GB speakers.

Pronunciations of individual words

28. In the cases of certain individual words listed below the Wells LPD provides corroboration for our GCN items as not dialectal describing them as being either GB subvariants or as "widespread in England among educated speakers, but ... nevertheless judged to fall outside RP" (identified with §).

almond /`ɒlmənd/ LPD "§"

among /ə`mɒŋ/ LPD "§"

assume / ə`ʃu:m/

ate /eɪt/ : once rather a Northernism this pronunciation is now spreading ever more widely in GB

Atlantic / æt`læntɪk / LPD "§"

basin /`beɪsɪn/

because / bɪ`kɒs, bɪ`kɔːs, -z / LPD "§"

breakfast / `breɪkfəst / LPD "§"

comfortable / `kʌmfətəbəl / in LPD as a GB subvariant

constable / `kɒnstəbl / in LPD as a GB subvariant

difficult / `dɪfɪkʌlt / is a GCN subvariant shown in LPD as "§"

disrupt /dɪz`rʌpt/ LPD has only /dɪs-/

eighteen / eɪt`tiːn / LPD "§"

eighty / `eɪtti / LPD "§"

`exam / `egzam / LPD "§"

`exhaust (noun) / `egzɔːst / LPD "§"

gooseberry / `guːsbri / LPD "§"

horrible / `hɒrɪbl / in LPD as a GB subvariant

industry / `ɪndʌstri / is a GCN subvariant shown in LPD as "§"

`magazine in LPD as a GB subvariant

mischief / `mɪsʧiːf / LPD "§"

The next four items are commonly heard but within limited GCN areas. Speakers who employ them sometimes exhibit hyper-adaptive forms such as / nʌn / for the first sylable of non-smoking etc.

none / nɒn / LPD "§"

nothing / `nɒθɪŋ / LPD "§"

once / wɒns / LPD "§"

one / wɒn / LPD "§"

perfect (adj) / `pɜːfekt / in LPD as a GB subvariant

possible / `pɒsɪbl / in LPD as a GB subvariant

presume /prɪ`ʒuːm/

raisin /`reɪzɪn/

raspberry / `rasbri / not in LPD though it has as variant / `rɑːs- / and "§" / `ræz- /

says / seɪz / LPD "§"

terrible / `terɪbl / in LPD as a GB subvariant

us /əz / and / ʌz / LPD "§"

version / `vɜːʒn / in LPD as a GB subvariant

with / wɪθ / LPD "§"

yester`day LPD "§"

Not included by LPD as "§" variants but to be heard occasionally from some people with very largely GCN profiles are

aggravate / `ægrɪveɪt /

basin / `beɪsɪn /

fundamental / fʌndɪ `mentl /

ornamental / ɔːnɪ `mentl /

Until relatively recently, the noun `dispute was so stressed only in the north: currently (since the seventies) it is probably very often so stressed by younger southerners too.

Asia with / ʒ / during the later twentieth century moved from being a predominant choice only in the north to being so in General British usage.

(If a GCN speaker uses owt or nowt it’s jocular linguistic slumming.)

The special Yorkshire assimilation

29. In Yorkshire alone there is widely to be heard the unique phenomenon within the English-speaking world in which the final soft consonants /b, d, g, ʤ, v, ð , z, ʒ / of a preceding syllable are regularly converted to their corresponding sharp equivalents /p, t, k, ʧ, f, θ, s, ʃ/ in rhythmically close-knit sequences under the influence of a following sharp consonant. This phenomenon can be observed sporadically in more general forms of English in variant versions of the words breadth, width and length and in of course, used-to, have-to and supposed-to, but in non-GCN Yorkshire it happens wholesale. Thus Bradford is often pronounced as if it were spelt Bratford and broadcast similarly. However GCN speakers rarely produce obvious examples of this feature in eg Good Friday, Grand Theatre etc. Although traces of the tendency are occasionally to be heard from GCN Yorkshire speakers, extensive use of it is more characteristic of uncompromisingly demotic-style speakers like Arthur Scargill, the miners’ union leader who has been heard to say eg dock population for dog population. The making of, in a Wells example (ibid. p. 367), white trousers identical with wide trousers is unlikely because the word white is, unlike as in demotic Yorkshire speech, not usually stretched in its vowel value by GCN speakers. Although placenames like Bardsey and Pudsey may receive / t / for their d from some GCN speakers largely unnoticed from time to time, their recognised forms have / d /. However the BBC Pronouncing Dictionary of British Names of 1983 admitted as sole recommendation to its staff the use of /`empseɪ/, which embodied this assimilation, for the North Yorkshire place name Embsay.

Rhythms and Intonations

30. It is difficult to draw lines between ascriptions of some observers of extra length given to certain vowels and what have been describe as northern English “drawled” tones as they are referred to by Roger Kingdon at p. 225 of his Groundwork of English Intonation.

It is probably fair to say that many GCN speakers often favour intonation patterns like high-level-plus-low rise, as eg in We pre 'fer ˏcoffee, in situations where southerners would much more often be likely to select a falling-rising tone.

Similarly Kingdon (ibid. p 234) observed that in an expression like ˏCome with me to the ˏstation, that although the same tone choices can occur in the north and the south a "slight difference is ... observable" between them in that the range covered by the rises is greater" in the north.

None of these features is very marked in GCN.

A valuable record of mid-twentieth-century GCN.

31. In 1964 the journal of the International Phonetic Association, then known as Le Maître Phonétique, published a short article by a GCN speaker of the highest linguistic sophistication the well-known professional dialectologist Stanley Ellis. He was a native of Bradford who was the principal fieldwork researcher of the Leeds University 'Survey of English Dialect'. In the then customary requirement of contributors to that journal he wrote using symbols of the Association's International Phonetic Alphabet by doing which he naturally gave precise detailed information on his speech. This article on 'Hypercorrect Dialectal Pronunciations' clearly illustrated many of the points we have been discussing above. He was born in 1926 and died in 2009. An obituary on him may be seen on this website at Section 10 Item 5.

It showed him as generally employing the typical GCN strong prefixes in obtained, expansion (twice), enforce, examination, enthusiasts, accept (twice) and admittedly. But, in a way that should surprise no careful student of variation in speaker's performance, he recorded his use of an initial weak syllable in the words completion, conformity, considerable, continue, examination, observing and example (twice). Regarding the fact that examination occurs in both values it is interesting to note that the strong-initial-syllable version appears at the beginning of a sentence while the weak one occurs within a word group ( ... future examination of ... ).

In the bath and chance group of words he showed himself as using the traditional short vowel in answer and example though with the symbol ash (æ) suggesting a value less open and less retracted than the more conservative GCN style. Similarly for other words in this group he preferred a symbol for a less advanced value than many GCN speakers use viz ɑ (which he employed with a length mark) in the words after, ask, asked, staff, and task.

Like most speakers his usages showed a mixture of newer and more conservative values. He for instance favoured the more traditional diphthong / oə / for more; among was / ə`mɒŋ /, because was / bɪ `kɔːs /; possible was / `pɒsɪbl /. He showed no tendency to use an / i / quality in the final syllable of happy words.

However, his general leaning was, as we have seen, to more modern GCN values: he didn't use / θ / in with ; his schwa in the second syllable of education sounded rather modern; he employed / ʌ / wherever it corresponded to GB, even in no-one.

The kind of "hypercorrection" with which his article had been concerned referred to developments in the deeply dialectal speech of informants born decades back into the nineteenth century, but an interesting more modern example of his topic was exemplfied by Ellis's transcription of his own version of the word difficult. The vowel he showed in its last syllable was / ʌ /.


Ellis, Stanley (1964) Hypercorrect Dialectal Pronunciation Le Maître Phonétique pp 2 & 3

Gimson, A. C. (1984) The RP Accent. In Peter Trudgill (ed.) Language in the British Isles. Cambridge:CUP

Kingdon, Roger (1958) The Groundwork of English Intonation. London: Longman

Pointon, G. E. (1983) BBC Pronouncing Dictionary of British Names. Oxford: OUP

Wells, J. C. (1982) Accents of English. (3 vols). Cambridge: CUP

Wells, J. C.(1984) English Accents in England. In Trudgill...

Wells, J. C. (1990, 2/e 2000) Longman Pronunciation Dictionary. London: Longman (“LPD”)