Spelling reform
Preparation for the Spelling Congress is underway. The more people will send in their proposals, the better. On the other hand (or so it seems to me), the fewer people participate in this event and the less it costs in terms of labor/labour and money, the more successful it will turn out to be. The fate of English spelling has been discussed in passionate terms since at least the 1840s. As early as 1848 Alexander J. Ellis wrote that spelling English is the most difficult of human attainments, that English spelling is the most foul, strange, and unnatural there is (his emphasis), and many other equally soothing things. Ellis’s scholarship arouses nothing but admiration, but one may agree with the verdict of The Westminster Review that “if on the subject phonetics he be a little mad, verily much learning hath made him so.”
I am addressing those who, at least provisionally, agree that English spelling should be reformed. The chances that such people will reach consensus are slim, as the fierce storms/tempests in the reformers’ teacup show. The fights were severe in the 1840s, in the 1880s, in the 1910s, and much later. Several hundred and sometimes several thousand people follow this blog. Could I appeal to them to express their opinion on the following propositions that are mine and only mine?
- The reform should be instituted in several steps and be as painless as possible. For instance, no one will suffer if scatter, unscathed, scoop, etc. reemerge as skatter, unskathed, and skoop. The distinction between sc– and sk– is artificial and easy to abolish. Along the same lines, dunghill will smell as sweet if it loses the last letter. So my proposal resolves itself into the following: begin with the most innocuous changes and stop for about ten or fifteen years. Then go on. The proposed reform should resemble a multilayer missile, because the public will not accept a revolution. For starters, bild for build is fine, but bery for bury is not. Consequently, I propose to draw up a list of changes, beginning with the ones that will not irritate too many opponents, go on slowly, and then (possibly!) strike where the change looks truly radical.
- Forget all plans that suggest an introduction of diacritics and new letters.
- However time-consuming the task may be, before the congress is convened, the Society needs a plan thought out from beginning to end; it should include all the words whose spelling will be changed.
- The congress will achieve nothing unless at the preparatory stage it ensures the support of those in whose power it is to implement the reform.
The less we talk about general principles, the better. A brief reminder of what I mean.
- “History will be effaced.” The etymology of English words will not fall victim to the reform. To begin with, etymology is not worth rescuing when it comes to the spelling of a modern language, and, which is more important, the spelling of numerous words that pretends to reflect the past is a monument to ignorance and does not preserve the true history of English.
- “Many homographs will appear.” There are already hundreds of them, and there are hundreds homophones, but context disambiguates them.
- “People will be confused.” Any change confuses people, but nothing can be more confusing than the spelling we now have.
- “Millions of books will have to be reprinted.” Not at all! Especially if the reform takes time, readers will gradually get used to the novelty. Anyway, most of us read book in new editions and reprints. Consult Shakespeare’s folios (facsimile editions are easily available) and compare them with what stands on your shelf. English spelling has changed dramatically since the days of the early seventeenth century, but English culture has not suffered as a result.
- Spelling reform has been implemented in several major European languages. Protest has always been vociferous, but there have been no casualties.
However, let me repeat: if the congress takes place, the focus should be on practical matters more than on the defence/defense of the reform, though some introduction on the predictable objections is naturally needed.
Finally, what do you think of the passage I have abstracted from a letter published by The Academy, July 17, 1915, p. 44, at the height of World War I?
“I want to see the grait and needless waste ov time and muny on teeching children to reed ended, and that speedily, espeshaly when there iz such an outcry for economy. Let us proov our sinserity in this crusade for thrift and carefulness by permitting teecherz to uze a sistem of speling which duz not waste time, duz not reezon at defians, nor disregard both history and etimolojy. Adults may spel, doutless will continue to spel, acording to the old fashon until the end ov the chapter. As Lord Haldane says—they ar too slugish to change, unles they wer taxt for every redundant letter emploid, the history, etimolojy, fashon, rithm, poetry ov wurds, wud fly to the windz. It is to the yung and the future, we need to giv attenshon.”
Smaller issues
Mainly Romance
Where is d in French aventure, as opposed to Engl. adventure? It died. In Middle French, in a combination of two consonants, the first one was regularly lost.
Is the etymology of crêpe given in the OED correct? Yes, it is. The Old French adjective was crespe. In the groups of two consonants the first one was lost (see above), and in the stressed syllable the vowel was lengthened (so-called compensatory lengthening). The circumflex over ê is the reminder of the lengthening (obviously, today no one needs it, but spelling is needlessly archaic not only in English). The Latin etymon was crispus (cf. Engl. crisp). To prove the validity of this etymology, it is enough to observe that in Italian, Spanish, and Portuguese the word still sounds as crespo. Crêpe ~ crape “fabric” is an etymological doublet of crêpe “cake.” In Modern French, the two homonyms are grammatically distinct: the first is feminine, while the second is masculine.
There was a question about the etymology of conscience and awareness. Neither word poses problems. Conscience, a thirteenth-century borrowing from French, is clearly con + science, so it refers to one’s inner knowledge of things, one’s inmost thoughts. The present day sense is late. By contrast, aware is of Germanic origin (compare German gewahr, the same meaning, as well as Engl. beware and wary). Aware is related to neither war, which is of French origin, nor the root of awaken.
Can Engl. fox be related to Latin focus? Since both f and k coincide in the two words, fox cannot be a cognate of the Latin noun, while the borrowing of focus “hearth, fireplace,” with reference to the color of fire, as an animal name is quite improbable.
Germanic and Celtic
Question: “I came across the following in Henry David Thoreau’s ‘A Plea for Captain Brown’: ‘Their great game is the game of straws, or rather that universal aboriginal game of the platter, at which the Indians cried hub, bub!’ Could the North American aboriginal expression be the origin of the word hubbub?” No. the English word was recorded in the sixteenth century, and its derivation from an Irish war cry is, most probably, correct. But the coincidence is not fortuitous: both exclamations are the product of the same impulse, and such words are sometimes universal.
A colleague called my attention to the fact that in my brief answer about the origin of OK, I should have mentioned the playful spelling oll korrect (1839), which is allegedly the true source of OK. The association with Old Kinderhook goes back to 1840, for only Van Buren’s 1840 presidential campaign made the word famous. Oll korrect has often been mentioned in the investigation of OK, and everything may have begun with this facetious misspelling. See Allan Metcalf’s book OK: The Improbable Story of America’s Greatest Word.
To conclude
The remarks about Latvian are apt! Most etymologists, unless they are specialists, cite words found in dictionaries and secondary sources. They rarely know the “non-major” languages to which they refer. This is also true of my recourse to Latvian. I can read linguistic literature in Lithuanian and, to a certain extent, in Latvian, but of course have no “feeling” for either.
Many thanks for pointing to Gąsirowski’s articles. I happen to know both (and many of his other works). My Bibliography of English Etymology features only one of them, but since its appearance the database has increased by about 4000 titles. With regard to books and articles, all advice is precious, for nothing is easier than to miss important publications on the subject in hand.
Image credits: (1) First Folio in the Folger Shakespeare Library, Washington, DC. Photo by Daderot. CC0 via Wikimedia Commons. (2) Crêpes con la Nutella. Photo by Dawid Skalec. CC BY-SA 3.0 via Wikimedia Commons. (3) Stags Fighting 3. Photo by MrT HK. CC BY 2.0 via mrthk Flickr.
Too late for spelling reform. As a retired computer programmer I can tell you that there are a huge number of programs that expect English to be spelled as it is. Remember the fuss over Y2K? It’s not going to happen.
The best time would have the 19th century. I’m not sure why it failed then: too hard to get the USA and UK to agree? Especially since pronunciation ihaf drifted apart?
Sorry about the incorrect autocorrect on my comment (think of the vast effort that has gone into spell checkers and autocorrect).
I agree with nearly all your views on spelling reform, except on what to change. I think that unless some of the early changes make a noticeable difference to the long time it takes to learn to read and write English, and start to reduce spelling errors, there won’t ever be any further reforms. I would therefore suggest to start with improvements to one or two areas that are chiefly responsible for making learning to read and write English exceptionally time-consuming.
By analysing the 7,000 most used English words, I have established that although 75 of the 83 main English spelling patterns have some exceptions (see my EnglishSpellingProblems blog), the main absorbers of learning time are seven areas which have no dominant spelling pattern:
Unpredictable consonant doubling after short, stressed vowels which is used
systematically in 405 words (e.g. merry),
omitted in 554 (e.g. very) and redundant in 195 (e.g. serrated)
Ea /ee/e-e… – eat, eel, even, ceiling, field, police, people, me, key, ski, debris, quay
(152 ea – 304 other spellings)
Er /ur/ir… – her, turn, third, learn, word, journey (70 er – 124 other)
o-e /oa/o… – mole, bowl, roll, soul; old, mould, most, boast, goes, mauve (143 o-e – 105)
-o/ -oe/ -ow … – no, toe, dough, sew, cocoa, pharaoh, oh, depot (47 –o – 60)
oo (long) – food, rude, shrewd, move, group, fruit, truth, tomb,
blue, do, shoe, through, manoeuvre (94 oo – 103)
au/ aw … – autumn, caught, bought, always, tall, crawl (39 au – 87)
Another 5 spellings have clear main patterns, but are disobeyed in more than 50 words each. Because many of them occur in some of the most used English words, they are especially injurious to early literacy progress:
e: end– head, any, said, Wednesday, friend, leopard, bury, heifer (301 e – 67 not)
i: ink– mystery, pretty, sieve, women, busy, build (421 i – 53)
u: up– front, some, couple, blood (308 u – 68)
a-e: late – wait, weight, straight, great, table dahlia, fete (338 a-e – 69)
i-e: bite – might, style, kind, climb, sign, island, indict, eider, height (278 i-e – 76)
The –ve ending and –te at the end of longer words (relative, delicate) have very few exceptions, but because they are used irrespective of whether the preceding vowel is long or short, they make learning to read more difficult than need be: survive relative, inflate frigate – cf. debate democrat, save chav.
Most of the surplus –e endings were added by printers to earn more money, because they were paid by the line.
If spelling reform is to make learning to read and write noticeably easier, it needs to address at least some of the above irregularities. Perhaps the Congress should concentrate on deciding which ones to start with?
Please note that
missing and surplus consonant doubling, redundant –e endings,
and the many exceptions to e, i, u, a-e and i-e
combine to undermine and make nonsense of the English short and long vowel system, as in ‘let, delete, letter’, ‘bit, bite, bitten’, ‘cut, cute, cutter’.
Perhaps one of the main aims of reform should be to declutter this area?
I agree with nearly all your views on spelling reform, except on what to reform first. I believe that unless some of the early changes noticeably reduce the time it takes to learn to read and write English, as well as spelling errors, there won’t ever be any further reforms. I think that reform has to improve at least one or some of the 14 spellings from the 3 groups of irregularities which are chiefly responsible for making learning to read and write exceptionally slow.
1. The 7 worst absorbers of learning time and sources of spelling errors are seven spellings which have no dominant pattern:
CONSONANT DOUBLING after short, stressed vowels in root words of more than one syllable is used in 405 words (e.g. merry), omitted in 554 (e.g. very) and redundant in 195 (e.g. serrated).
EA /EE/E-E… – eat, eel, even, ceiling, field, police, people, me, key, ski, debris …
has 456 unpredictable spellings (152 ea, but 304 others)
ER /UR/IR… – her, turn, third, learn, word, journey (70 er – 124 other)
O-E /OA/O… – mole, bowl, roll, soul; old, mould, boast, goes, mauve (143 o-e – 105)
-o/ -oe/ -ow … – no, toe, dough, sew, cocoa, pharaoh, oh, depot (47 –o – 60)
OO (long) – food, rude, shrewd, move, group, fruit, truth, tomb, blue, do, shoe, through, manoeuvre (94 oo – 103)
AU/ AW … – autumn, awful, caught, bought, always … (39 au – 87)
2. Another 5 spellings have clear main patterns, but are disobeyed in more than 50 words each, and because many of them occur in some of the most used English words, they impede literacy progress most seriously in the early school years:
e: end– head, any, said, Wednesday, friend, leopard, bury, heifer (301 e – 67 not)
i: ink– mystery, pretty, sieve, women, busy, build (421 i – 53)
u: up– front, some, couple, blood (308 u – 68)
a-e: late – wait, weight, straight, great, table dahlia, fete (338 a-e – 69)
i-e: bite – might, style, kind, climb, sign, island, indict, eider, height (278 i-e – 76)
3. The two endings –ve and –te (at the end of longer words) have very few exceptions, but because they are used irrespective of whether the preceding vowel is long or short, they make learning to read more difficult than need be: survive relative, inflate frigate (cf. debate democrat, save chav).
Oh dear!
When I first tried to comment, nothing appeared. Not even ‘Your comment is awaiting moderation’. So I tried again, with a slightly shorter one. Now both are there!
I would like to say to Anne that it’s never too late to do a good thing.
The amount of time English-speaking children have to spend on learning to read and write is ridiculous – 10 times longer than in Finland. This should be reduced. And the invention of computers makes spelling reform much more feasible than before. Without them it was even too difficult to establish exactly what’s wrong with English spelling.
What has prevented modernisation of English spelling more than anything was Sam Johnson’s pompous pronouncement in his 1755 dictionary that irregular spellings were “spots of barbarity impressed so deep in the English language, that criticism can never wash them away: these, therefore, must be permitted to remain untouched. Every language has its anomalies, which, though inconvenient, and in themselves once unnecessary, must be tolerated among the imperfections of human things… being once incorporated, can never be afterward dismissed or reformed”.
He would say so, because he was one of the main wreckers of the English spelling system. He made nonsense of English consonant doubling by exempting most Latinate words from it (arable, arid)) and decorating some others with pointless doubles (arrange, arrive) – cf. ‘arrow, barren – arise, aroma’.
He also wilfully and pointlessly chose to disregard the alphabetic principle in 335 words by linking their different meanings to different spellings (feet/feat). Over 2,000 homophones get by perfectly well with just one spellings (e.g. found, sound, ground).
English spelling could be made much more sensible by merely undoing the deliberate damage inflicted on it by thoughtless wreckers like Johnson.
Short u, short e and e-e ended up with 67, 68 and 456 irregular spelling each (head said; some front; seek, speak, shriek) because people who gave no thought to ease of learning messed them up, including printers who added hundreds of surplus letters (See History on my EnglishSpellingProblems blog for a fuller explanation).
It seems in cases like draught – draft spelling reform is already on the way. I don’t see any problem with writing eight and ate as “eit”. Moreover, it seems something like that has already happened in Middle Age France when octo changed to (h)uit. Angles have just to pluck up the courage.
“…the public will not accept a revolution.”
A kumpleetly noo speling sistum kan be intrudoosd nou, if we kan get dhv noo tekst pvblishd—and if we hav a speling sistum dhat wvrks wel. It woent hav to be intrudoosd in dhv skoolz until it bikvmz popyulur, which wil taek a long tiem. No wvn wil hav to uez it until dhae wvnt to. By dhat tiem, moest uv dhe “oeld skool” wil hav left dhv seen, and taekun dheir ubjekshunz with dhem.
A staejuz uproech wil be a shwr wae to nowher. At evry chaenj it wil be chalunjd and bikauz uv lak uv an end goel, dhv chaenjuz wil liekly not fit wel tugedhur.
I ekspekt that it wil be difikult to get eny noo speling sistum pvblishd regyulurly, no matur hou gwd it iz. Dhis aulsoe wil uplie to smaul chaenjuz.