English accents
and their implications for spelling reform

J.C. Wells, University College London

This is an edited version of a talk given to the Simplified Spelling Society on 25 January 1986. — John Wells, 2003.


1.1 The alphabet

An ideal spelling system, we all know, will have one symbol for one sound, one grapheme for one phoneme. But this principle throws up certain difficulties in practice. If we confine ourselves to consideration of the Latin alphabet, one major difficulty is that it is an arbitrary list of 26 letters which do not necessarily correspond to the sound systems of the languages which have to use it. In particular, they do not correspond to the sound system of English. On the one hand, the Latin alphabet provides us with no unambiguous way of spelling English sounds that Latin lacked (e.g. the sound we often spell sh, the two sounds we spell th, and many of our vowels and diphthongs); on the other, it contains at least two letters, q and x, that were unnecessary even from the point of view of Latin. In this article, however, I am concerned not so much with the deficiencies of the alphabet and how we might remedy them (the ‘grapheme’ part) as with the problems arising from the fact that we English speakers do not all pronounce our language in the same way (the ‘phoneme’ part).

As my eminent predecessor Daniel Jones pointed out in his article about phonetics and spelling reform (1944),

people in different parts of the country speak differently [...;] what is a phonetic representation of a word for one person is not necessarily phonetic for another.

In raising these problems I do not want to detract from the fact that there are large numbers of words in our language where they do not arise. All speakers of English, no matter where they come from, pronounce friend so that it rhymes with bend, send, tend. So a reformed spelling frend ought to be uncontroversial. Everyone pronounces sight, site and cite identically, so it is absurd (except for advocates of etymological spelling) that we have to learn to spell them all differently. Everyone distinguishes the verb to advise from the noun the advice, so we can see the justification for distinguishing them in spelling — yet we all make the same pronunciation difference between to use, to house, to excuse and the nouns the use, the house, the excuse where we make no spelling distinction.

1.2 Danger of parochialism

The sounds of any language can be viewed as a system of contrasting phonemes. The pronunciation of any word can be specified in terms of the string of phonemes that represent it, together perhaps with information about relevant prosodic features (in the case of English, about stress placement). In designing a scheme of spelling reform, we face a certain danger of insularity or parochialism, of assuming that everybody has the same set of phonemes, and uses the same phonemes in particular words as we do ourselves. Unfortunately this is not the case. What seems obvious and normal to one speaker may be exotic, unusual, subtle and strange to another. There are all sorts of little facts about how English is pronounced round the world by native speakers which may give us pause in our reforming zeal. Here is a simple example. The traditional spelling of the words any and many conflicts with the way most of us say them. It may seem obvious to most of us that they rhyme with penny and so ought to be spelt in the same way, perhaps as enny and menny. In making such an assumption, however, we are ignoring the awkward fact that many southern Irish people pronounce them to rhyme with nanny, so that they would see nothing strange about writing them with the letter a. Maybe they would want to write anny and manny rather than any and many, but that is not my point. I concede that in English as a whole the preferences of the southern Irish may have to give way before those of the vast majority of other English speakers — but we should be aware of what our proposals imply.


2.1 Spelling the past tense

Ought mist and missed to be spelt identically because they are pronounced identically? Or should we give the past tense a consistent spelling shape with d, even when, as in missed, it is pronounced /t/? In deciding this issue, we should perhaps consider the Nigerians, who do not usually pronounce missed like mist. This is because — under the influence of traditional orthography — they typically use a /d/ sound in missed, and in fact usually assimilate the /s/ sound to a /z/, so saying /mizd/, with voicing throughout. For them kicked, likewise, tends to rhyme with rigged rather than with strict. I am not necessarily saying that we have to let our reform proposals be determined by how Nigerians pronounce English, even though they do constitute a substantial body of users of English. But I am saying that we should at least be aware that a reform that makes spelling more logical for one group of speakers may make it less logical for another.

2.2 Spell or omit r?

From New Spelling onwards the importance of catering for accents other than Received Pronunciation has been clear from the treatment of historical r (Ripman and Archer, 1948). Like most English people, in my speech I don't distinguish stork and stalk. If spelling reform proposals do make a distinction, as they usually do, then the reason is (a) historical and (b) because they are pronounced differently from one another in other accents. Historically, stork had /r/, and stalk did not. In many varieties of English (Scottish, Irish, west of England, most American, Canadian — the rhotic accents) the distinction in is still made in speech. Similarly pairs such as larva - lava, rotor - rota, homophonous for English people like me, are distinct in the rhotic accents. This justifies our keeping the distinction in spelling, even though the task oflearning which words to write with r and which without will impose some burden on those of us whose English is non-rhotic. And those of us who pronounce intrusive /r/, saying perhaps rotar of duties, will have to remember not to write r in some positions where we pronounce it, as well as sometimes writing it where we do not pronounce it. Faced with this problem, spelling reform has little alternative to accommodating the rhotic speakers, even if the consequence is that we non-rhotic speakers must learn by rote when to write r and when not.

2.3 Singer and finger

A similar problem arises with ng. Consider the pair singer:finger. For most speakers these words do not rhyme exactly, because finger has a /g/ sound after the nasal. It seems logical to write singer but fingger (Ripman 1941). The trouble here is that people in the trapezium linking Birmingham-Manchester-Liverpool make these words rhyme, with /g/ in both. So if we show a difference in spelling, some Midlanders and Northerners will have to learn an extra arbitrary distinction. Alternatively, I suggest, it is a distinction we might well decide to ignore — so incidentally also simplifying the spelling of the comparative and superlative of long, strong, young, whose irregular pronunciation in most accents would otherwise be reflect in reformed spelling as longger, longgest etc.


3.1 Greater problems with vowels

Such variations in pronunciation mean we may have to violate the principle of one sound per letter and one letter per sound in quite obvious ways, ways that probably everyone can accept. Greater difficulties perhaps arise with vowel-sounds and sets of vowel-contrasts, where I think the danger is particularly strong of wrongly assuming that everybody makes the same contrasts.

In what follows I make use of the concept of standard lexical sets, as proposed in Wells 1982: 2.2. Each keyword, shown in capitals, stands for perhaps hundreds or thousands of words containing the vowel sound in question. The keywords are chosen so as to maximize clarity: whatever accent of English we use, they can hardly be mistaken for any other word.

3.2 The Sam - psalm contrast

New Spelling makes special provision for the words I shall refer to as the lexical set BATH — words such as pass, path, chance — by allowing either a or aa. There is an assumption behind this permissive solution, namely that everyone distinguishes the vowel sound in gather from that in father. However, this is not the case. In parts of the west of England and certainly in Scotland and Northern Ireland some people have no such contrast in their phoneme system. They use the same vowel in Sam as in psalm, so that these two words are homophones for them. It would actually have been consistent with this fact for New Spelling to ignore the difference between these two vowel sounds. It has very low functional load, which is to say that there are very few word-pairs that are distinguished as Sam and psalm are. So we might prefer, in a reformed spelling scheme, to ignore the contrast that RP makes between the vowel sounds of mass and pass, and abandon the New Spelling aa entirely. On the other hand RP speakers and others who make pass rhyme with farce must still remember the spelling difference reflecting the historical r in farce (fars) but not in pass (pas). By taking accent variability into account one lays oneself open to the objection that one has abandoned the principle of one letter per sound. The southern English will protest that mass and pass differ in sound, while pass and farce do not — yet we would be proposing the same spelling for the first pair and different spellings for the second. I do not think we can avoid this difficulty.


Less well known is the very similar situation affecting the lexical set CLOTH, namely words such as cross, cough, lost, where even within RP in this century we had a rival pronunciation which might be respelt as clawth, crawss, cawf, lawst. This situation has now resolved: the clawth variant is now very much aminority form, if it indeed still exists. But the problem here is that American pronunciation really corresponds to crawss, cawf, clawth, lawst rather than to cross, cof, cloth, lost (to the extent that Americans distinguish the two vowel sounds at all, which many do not). If we follow New Spelling and keep lot as lot while changing thought to thawt, my point is that most people in England would logically not change the spelling of cloth and other words like it. But Americans generally speaking identify the CLOTH set with the THOUGHT set, not the LOT set, and might therefore logically want to write clawth, etc. This would also apply to words such as long (lawng), since for them it too belongs with THOUGHT, not with LOT.

3.4 Ignore such contrasts?

If pressed, I would propose the same radical solution here as with TRAP, BATH and PALM, namely to ignore the whole set of contrasts and write LOT and THOUGHT identically, probably as o. This would also suit Scottish and Northern Irish pronunciation, many Scots having the same vowel for LOT and THOUGHT, as do also most Canadians and many Americans. We would then have to forget the distinction we English (and Australians, New Zealanders and South Africans) make between cot and caught, don and dawn. In fact those of us who make a sharp difference between these two vowels are perhaps a minority; if we were to insist on distinguishing them in reformed spelling we would be imposing a real burden on the many people who make no such distinction and who would have to learn what for them would be an arbitrary difference in spelling.

3.5 Misconceptions about r

In this area RP speakers, and many others, would again face the same problem of r, as in stalk (stok) but stork (stork). In reformed spelling we should have to write kot for cot, caught, but kort for court. Inevitably, some people would have to accept that they should write certain words identically that they pronounce differently, and other words differently that they pronounce the same. This could lead to difficulties. Wherever the spelling depends on what happens in other people’s accents rather than in one’s own, one is liable to be misled because of mistaken ideas of what happens in other accents. We can see this in the difficulty we English people have if we try to imitate American or Scottish accents — even talented English actors or impressionists attempting American or Scottish speech quite often make mistakes. Peter Sellers, in his spoof travelogue Balham, Gateway to the South, talking lyrically in his pseudo-American accent about morning coming, says “and now at last we see the [dorn] approach”. But Americans don't say an r-sound in dawn. Its reformed spelling will have to be don, not dorn. Clearly even for highly literate people sound in a sense dominates spelling in their mental picture of words; so when imitating another accent they set up correspondences between sounds in their own accent and what they imagine are the equivalents in the other accent, rather than be guided to the spelling. (After all, the presence/absence of r in traditional spelling is a pretty reliable indicator of whether rhotic speakers pronounce /r/ in a particular word.) English actors playing Scottish parts likewise make many errors with /r/, pronouncing words such as comma and China with final /r/. This is because in many words English final schwa does correspond to Scottish final /r/, as in father; but in many other words there is no such correspondence.

4. Further problems with a and o

4.1 Bother and father

A difficulty with the simplification that I have just been advocating is that while I propose the reformed spelling a for PALM and o for LOT, most Americans pronounce these two lexical sets with the same vowel-sound. Thus in a typical American pronunciation father and bother rhyme perfectly. Americans make puns that don't work for the British: a Saab car sounds like sob for them, and they put up posters for the painter Salvador Dali saying “Hello Dali”, punning on Hello Dolly, with which it is for them homophonous. This means that the Americans, the most numerous and influential part of the English-speaking community, will be confronted with the uncertainty of how to spell words in which they use this vowel sound. They are going to have to write it o as in lot in most words, but presumably — unless we allow both possibilities in reformed spelling — as a in a minority of cases such as father and palm. Words like pasta will also come under this heading: Americans will have to remember to write pasta rather than the posta that might seem to them more logical. Hey, we all have to make compromises.


Most of us nowadays use the same vowel sound in NORTH as in FORCE. New Spelling, however, was inclined to allow for two distinct sounds here. Ripman 1941 marks FORCE words, but not NORTH words, with an asterisk:

A considerable number of words written here with or are pronounced by many speakers with oer, and may be written so if desired. (p. 6)

This distinction is a historical one which is now mostly lost in England, North America and the southern hemisphere, though still made in Scotland and Ireland and to some extent in Wales, the West Indies, and the United States. Speakers who have it make a difference, for example, between horse and hoarse, and do not rhyme short and sport, fork and pork. Common sense may tell us to ignore the distinction, since only a minority now make it. But this does mean that the Scots, etc., would risk misspelling FORCE words, in which they use their GOAT vowel, not their THOUGHT vowel.


Not only have NORTH and FORCE sets merged for the English: the CURE words (e.g. sure, poor, tour have joined them too for many speakers. Many English people now pronounce Shaw, shore, sure as homophones, and likewise paw, pore-pour, poor. I would suggest all the forms with r could be spelt in the same way, and that we ignore the Scottish or Irish distinction between war, shore, sure, writing them all perhaps with or.


5.1 Maximalism and minimalist

Logically there are two extreme positions one can adopt towards phonemic contrasts. The maximalist position would say our orthography must reflect all the contrasts that anyone makes, and the minimalist position would say we should reflect only those that everyone makes.

5.2 Problems of maximal contrast

The maximalist position would lead to very undesirable consequences, as the following examples show. Many southern English make a longer vowel in bad than in lad and may even have minimal pairs between the name Sally and the verb to sally, or between shandy the long drink and brandy the short drink; obviously spelling should ignore this distinction. Likewise some Scots distinguish tide:tied. The point here is not the quality of the vowels as such, but whether contrasts are made between vowels in different sets of words. In fact for these Scots the difference between the diphthong in tie#d and that in tide reflects the presence of a grammatical boundary before the suffix in the first but not the second. They make a similar distinction between the diphthong in Fife and that in five, but here it depends on the identity of the following consonant. So this distinction is on the whole predictable, and can accordingly be ignored. In Northern Ireland and a few other places they distinguish days and daze; we are obviously going to have to ignore that, too, as we must the contrast between late and eight or mane and main made in various parts of Britain, for historical reasons reflected in traditional orthography.

5.3 Who needs the distinctions?

Then there is a contrast within the lexical set NURSE, where English and Americans make no distinction; but the Scots and Irish may distinguish pearl and curl, for example, contrasting them in the same way as perry and curry; and they may well make a similar distinction between Hertz and hurts, or fir and fur. That is a kind of justification for present spellings, which accurately reflect this distinction (though sometimes writing the first vowel ear, as in pearl, sometimes er, as in defer). But if, like New Spelling, we abolish that distinction, then we are ignoring that contrast, which is a real one for millions of Scots and Irish. English or Americans will typically say these distinctions are subtleties they can't possibly cope with. The only point I would make is that those who make the distinction are going to feel that logically it ought to be retained in a reformed spelling, and they are going to have to be convinced otherwise if we want to abolish it.

5.4 Contrasts matter, not sounds

In all of this we must remember that it doesn't matter what the actual phonetic realization is; what is important is the network of contrasts, because that is what the orthography must reflect. If we take, say, the word soap where we all use our long-o vowel, it doesn’t matter what precise quality of vowel or diphthong an individual speaker uses, provided he or she uses the same sound in rope, goat, coat and all other words in the same lexical set.


6.1 A Jamaican merger

The minimalist position, spelling only the contrasts everybody makes, has difficulties too. One might think everybody contrasts pat and pot (the lexical sets TRAP and LOT), but that is not so. Jamaicans, for example, typically say /pat/ for both. Now people may object that there are not many Jamaicans, so they can be ignored; but as a group they have very special problems with English spelling. Whereas other English-speakers mainly follow their pronunciation in spelling the a - o contrast (ignoring one or two exceptional words like wash), for Jamaicans it is hard to decide which spelling is right and what therefore is also the posh pronunciation. Ordinary Jamaicans say /rat/ both for the animal and for putrefaction, and have to learn which to spell rat and which rot. A social factor is also involved: to speak educated Jamaican, they have to learn to distinguish rat and rot in pronunciation, too, as other English-speakers do. But if we are aiming to lighten the burden of arbitrary spelling distinctions, no reform project I have ever seen solves this difficulty for an important group of West Indians.

6.2 A southern US merger

Then there is the START-NORTH distinction, exemplified in pairs such as farm and form. It is not only the Jamaicans that tend to pronounce these identically; many Americans speaking popular accents in the south do so too. They have a test phrase about being born in a barn, and it is well known that some people confuse or reverse the two: country bumpkins in the southern states are ridiculed as being barn in a born, with the typical confusion of people trying to introduce a contrast into their speech that they don’t natively have. Again, I think on balance we must retain the spelling distinction in those sets of words. But it will constitute a difficulty for some.

6.3 Another southern US merger

Many American Southerners, of all social classes, and including most Californians, don’t distinguish the vowels in KIT and DRESS before nasals, and so make no difference between pin and pen (hence the terms writing-pen and stick-pin, to make the distinction clear). Listen to Jim Reeves singing Lord, give me strength, where the strength begins just like string. It would be rather drastic, I think, to abolish this contrast and write pin and pen identically in a reformed orthography, but unless we did, it would cause many American Southerners a big spelling problem. For them the logical reformed spelling of many might be minny rather than the menny that seems logical to the rest of us.


7.1 FOOT and GOOSE

New Spelling was criticized for distinguishing the vowels of the lexical sets FOOT and GOOSE, and this distinction, I think, could indeed be dispensed with. Consider the pair good and mood, which for most of us do not rhyme perfectly. However in Scotland they do, and also in Ulster. Forcing a contrast of spelling with good - muud, as New Spelling does, is therefore an arbitrary extra distinction from the point of view of the Scots and the people of Northern Ireland. As there are very few word-pairs distinguished in this way (only pull - pool, full - fool, look - Luke, and a few pairs involving inflected forms, such as wood - wooed, could - cooed), I think we wouldn’t suffer too seriously if we ignore the distinction.

7.2 Functional load of sh - zh

What is involved here is the important question of what is known technically as functional load, that is, the number of words that are distinguished by a given contrast. When the functional load of a contrast is low, then the contrast can be ignored, whereas when the functional load is rather high, then presumably it ought to be reflected in the spelling. Using this criterion, one might well decide not to distinguish the consonant sounds sh and zh, because there are very few word-pairs distinguished by this contrast. There are indeed non-rhyming pairs like pressure - measure, mission - vision, there are no everyday words directly distinguished by this contrast. (We can safely ignore such rarities as Aleutian - allusion, Confucian - confusion.) Rather than impose the non-English-looking zh to furnish a consistent spelling for the voiced member of this pair of sounds, we might write sh for both. This would have the welcome side-effect of remopving the uncertainty that would otherwise arise in words such as Asian, where some speakers use one sound and some the other.


There is another contrast with very low functional load which nobody has proposed abolishing as far as I know: that exemplified by my lexical sets STRUT and FOOT, respectively, which contains pairs like cut - put. We really need to consider three sets here, though, STRUT, FOOT and GOOSE, because the Scots, as we have seen, do not contrast the latter two. People from the north of England, however, do not contrast STRUT and FOOT, although they do distinguish these from GOOSE. So we face a kind of chain, on which everybody contrasts the extreme points (e.g. luck as against Luke), but not everybody contrasts both extremes and the middle (luck - look, look - Luke. The situation with the BATH words is parallel to this. I suppose the logical solution is again to allow the intermediate group, in this case the FOOT words, to be written with either vowel. So words like put, foot could be spelt either like strut, cut (which would seem logical for people in northern England), or like boot, shoot, which would seem logical the Scots and northern Irish and would probably be acceptable in the south of England and everywhere else. A factor to bear in mind here is that the lexical set FOOT is very small, containing only about 40 words (Wells 1982: 132), although several are quite common (good, should, foot, put). So not very many words would be affected if we were to allow them to be spelt in two competing ways. If diacritics were used in a reformed spelling, a solution might be to spell all the vowels in the STRUT-FOOT-GOOSE chain with u, but to minimize the visual differences by perhaps optional use of diacritics according to pronunciation.

7.4 Further contrasts?

The Scots do have an extra possibility of contrast here that other accents lack, between for example brood and brewed, where they have a longer vowel in the second word to mark the stem-suffix boundary within it, brew#ed. Distinctions like this may just have to be ignored. Similarly, in Northern Ireland people tend to distinguish in pronunciation between a tenor and a tenner (longer vowel because of the stem-suffix boundary). Again, there are very few pairs of words involved. Speakers who make such distinctions will have to be reconciled to the fact that most people do not.


8.1 Accents as social labels

A complication with the northern pronunciation which merges the vowels of cut and put is that it bears considerable sociolinguistic value. As everyone knows (in England), the vowel sounds you use in the STRUT set tend to flag your social class and to symbolize educated versus uneducated speech. The same is true of many other pronunciation variables. A reformed spelling that seems to buttress a low-prestige pronunciation will encounter resistance.

8.2 Hypercorrection

Aware of social prejudice, northerners not uncommonly attempt to use a southern- or RP-style vowel in STRUT words — but may do the same thing in FOOT words (since they do not natively distinguish the STRUT and FOOT vowels). Hence the phenomenon of northerners pronouncing sugar to rhyme with RP rugger, and pudding like RP budding. Phrases like good luck are particularly problematic: northerners attempting to sound posh may easily change the first word as well as the second, or indeed sometimes the first word but not the second. The word gas-mask gives rise to the same problem: if your basic pronunciation is with a short vowel in each word, and you later discover that it is considered better to use a long vowel in mask and grass, you might well lengthen the vowel in gas as well. If people have these problems in hitting the intended target in pronunciation, they would obviously have similar problems in reformed spelling if it were to follow RP too slavishly.

8.3 Accent prejudice and spelling reform

Spelling reformers have to confront sociolinguistic facts of this kind. Many ways of pronouncing are liable to be condemned as ugly and uneducated and not to be encouraged. Though this may well depend on the unfavourable stereotyping of the social groups who pronounce in these ways, we have to recognize that such stigmatization exists. If in a spelling reform we make provision for the such stigmatized pronunciations, we could be seen as bolstering vulgarity and ignorance. The objective, scientific observer of course discounts these social views and refuses to make such value judgments, but a reforming movement does have to take such prejudices into account.

8.4 H-dropping

A case in point is h-dropping. Millions of English people do not pronounce /h/ consistently: they omit it most or all of the time. We can still understand them. So it might seem logical to omit the letter h from our reformed spelling, and write pairs such as harm and arm identically, in line with that pronunciation. But no! that would go against the social attitude that it is incorrect to drop /h/ and that therefore the spelling ought to reflect its presence; and it would shock all the Scots, Irish, and Americans who are strangers to h-dropping. Obviously we should not continue to write h in the words honest and hour; but it would certainly be wise to continue to write it in harm and house, to reflect the prestige pronunciation that does distinguish harm from arm — even if this is going to constitute a spelling problem for h-droppers.

8.5 Northernisms

So with some of the other phonetic variables we have considered. These prejudices might well say that we must retain the difference between the STRUT and FOOT vowels in our reformed spelling, perhaps by writing u and oo respectively. This is not the only vowel-sound contrast which some people don't make, despite a widespread feeling that it would be better if they did so. Another example is the vowel contrast between the lexical sets SQUARE and NURSE. Liverpudlians, for example, typically have these vowel sounds merged, so that fair and fur are not distinguished, and the name Mary is pronounced to rhyme with furry. Again, perhaps we ought to keep the distinction in a reformed orthography, despite the problems that Liverpudlians will then face in remembering the correct spelling. That is to say, we ought to reflect the vowel-sound contrasts that everybody makes except northerners. This is hard on the northerners, but maybe that’s life as it is — unless we can somehow remove these prejudices about accents.

8.6 The -ing ending

Similar considerations apply to the ending -ing. Almost everywhere where English is spoken there is a rivalry between a relatively high-status pronunciation with a velar nasal (as in sing) and a relatively low-status pronunciation with an alveolar nasal (as in sin). The low-status variant is reflected in our current orthographic conventions by writing n’, thus runnin’ rather than running. Again, I think there would be general agreement that we have to keep the ing spelling, to reflect the prestige pronunciation. Nevertheless there are words where it is a source of uncertainty for many people, who may produce hypercorrections like a brazing (brazen) hussy or indeed (to take an authentic historical example) a mountain lark for what was originally a poetical mounting lark.


9.1 Declining status of RP

I have the impression that reform proposals this century — those originating in Britain, at least — have been very firmly based upon RP, together with some nods in the direction of archaizing tendencies (which is why historical r is reflected). It is clear that in the last quarter of a century in England the position of RP has been very seriously eroded, in that RP no longer enjoys the unquestioned status that it previously did. There are now many people who not only don’t speak it — that was always the case — but who also don’t aspire to it, and who would regard it as quite unrealistic to aspire towards it. In fact I think what has changed is the perceived model of beautiful or ideal speech, which is for many people no longer RP. This can be seen in all sorts of ways. Teachers of English as a foreign language, for example, get increasingly dissatisfied with the transcriptions the phoneticians offer them as the models for foreigners to imitate. This is what lies behind the change in the phonetic transcription of the final vowel in words like happy. Until the 1980s it was identified with the vowel of bit. But now [2003] the many people who use a final beat-like feel no longer feel it as lacking the prestige that formerly attached to using only a bit-like vowel. Current pronouncing dictionaries (EPD 1997; LPD 1990, 2000) write it with a compromise symbol, to accommodate the many speakers whose vowel is more similar to that of beat.

9.2 Allophonic variation

I have yet to discuss various technical phonological questions like the phonemic principle. It is clear that where we have allophonic differences, that is, realizational differences within a phoneme, we can ignore them. This means that essentially where two sounds are used in such a way that we can predict from the surrounding sounds which will be used, then we can ignore any such difference. This is why we can ignore the difference between an ordinary t-sound and a glottal stop: whichever way you say that is, not only but also, the meaning is the same and your choice of pronunciation should not influence your choice of spelling. Contrary to popular belief, there are certain positions in a word where a glottal stop is by now the norm, as in department or atmosphere, and many other where it is very widely used, as in network, football. Another example is the really a rather sharp difference between the o-sound that many people use in most cases (go, show) and the o-sound they use before /l/ (goal, shoulder. As long as we can set up a rule that our long o-sound has a special pronunciation before /l/, there is no problem: the two sounds are just allophonic variants of the same phoneme. They may sound a bit different but the difference is predictable, and so it may be ignored in an orthography.

9.3 American intervocalic t

American intervocalic t is an interesting case, because it is moving from being allophonic to involving a neutralization and therefore becoming phonemic. As you know, in words such as city, waited, and in phrases such as right away, Americans tend to use a d-like sound. Indeed, increasingly it is identical with their d-sound, so that atom and Adam are pronounced identically. Or there may be a subtle distinction, perhaps more in the mind of the speaker than perceptible for the hearer. In 1961 Webster's Third International was the first American dictionary to transcribe these words with /d/. For this it incurred considerable criticism: the /d/ pronunciation was said to be slovenly speech which should not be admitted to the dictionary. Nevertheless, it is a fact; and I have even encountered reverse spellings: I read an American novel in which somebody gave an ‘involuntary shutter’, shutter and shudder for the author clearly not being distinct. But we shall not want to admit this to a reformed spelling scheme. Americans will have to learn by rote which words are written with t and which with d. That would accord with their prejudices in many cases anyhow, so is not yet a problem; but it may be so in a hundred years’ time, particularly if this sound-change spreads, as seems likely, to all other accents of English. Already it occurs in Australia, South Africa and England, being heard as a stylistic variant even in RP. I don't see a major problem, particularly since in most cases the pronunciation of related words (wait, atomic) will make it clear whether the spelling should be t or d.


What I hope I have done is to highlight the dangers of parochialism in designing a reformed orthography for English, of being unaware of the varying patterns of contrast in different accents. But even with this awareness, it is impossible to satisfy all of the speakers all of the time; the best that can be hoped is that a proposed reform will satisfy most of the speakers most of the time.


Placed on the web 2003 03 19. Last modified 2006 02 27
Author's home page