Dr. Goodword’s Language Blog

Will English Become Unreadable?

Four months ago, Kate Gladstone wrote me an interesting letter that deserved a better fate than falling between the cracks of everything else we do around here—but it did. Fortunately, it resurfaced recently and here is the dialogue between Kate and me thus far:

As you know (and have ably demonstrated to the public), the pronunciation of the English language has dramatically changed over the centuries of the language’s documented existence, creating much of the present bad fit between English spelling and English pronunciation” (see The Chaos).

Since English spelling will foreseeably stay the same while English pronunciation will continue its long history of change, after several more centuries or millennia will there ever come a day when the sounds of English have changed far enough to destroy all useful remnants of a fit between spelling and sound?

In other words, could it ever happen that the English language would eventually get so far out of step with the writing system that hardly any words (or no words at all) had a phonemically transparent spelling, and reading had to rely 99+% (or even100%) on the memorization of words as wholes, even though the alphabet still existed and still putatively represented sounds?

After all, imagine what we would face if present-day English had standardized its spelling 1000+ years ago instead of only a few centuries ago: so that the word daily, pronounced /DEY-lee/, still required the ancient spelling “gedaeghwamlice”, and so on because nobody wanted to change the traditional spelling that showed how the word USED to sound a long, long time ago).

Yours for better letters, Kate Gladstone
Handwriting Repair and the World Handwriting Contest

Even though this letter fits neatly in the “as-if-we-didn’t-have-enough-to-worry-about” category, it raises an important issue that goes to the very heart of alphaDictionary’s existence, an isssue that was very close to the heart of Bernard Shaw and other literary notables. So, I have to respond.

When I taught writing at Bucknell, I assured all my students that the spelling errors in their writing was not their fault: it was the fault of the English spelling system (orthography) itself and the publishing houses that oppose reform. I still think that is generally a true representation of facts even though I realize opposition to reform lurks in other quarters, too.

I don’t think Kate’s scenario will play out, though, for several reasons. First, the relation  between orthography and pronunciation works both ways. Not only does the writing system (mal)adapt to speech, speech adapts itself to the spelling system (cf. the pronunciation of the T in “often”). Once a writing system is in place, phonological change slows down since language change is dependent on speakers never seeing images of words and having no concept of “correct spelling” (= ortho-graphy).

Second, the Web will eventually break the publishers’ control over what we read. In fact, we are probably undergoing spelling change now. If you watch the Web, you will see adaptations galore. Thru already appears over 100 million times on the current web (through 2 billion). Lite is a respectable spelling of light in some contexts. We can only pray that Imglish does not replace the current spelling system (LOL) and that we are not guided by pure frivolity in making changes.

If worse comes to worst, we really don’t need a close correlation between sound and meaning. In Chinese the relation is far looser than in English. Chinese uses pictures that represent meanings more often than sounds and Chinese society is burgeoning. So we can work around even a total disconnect betweeen sound and letter.

I would be much more frenetic in my activities at if I thought my grandchildren would face greater challenges in spelling than I face. Until this problem at least reaches the level of global warming or the war addiction of our federal government, our efforts would best be directed to trying to bring our children’s spelling abilities up to the level of ours, maybe push them a bit beyond. To help in that task is the reason I founded

2 Responses to “Will English Become Unreadable?”

  1. Kate Gladstone Says:

    Dr. Goodwords’ thoughtful letter deserved an equally thoughtful answer, so I decided to spend a couple of weeks before replying.

    True, spelling sometimes motivates a change in pronunciation (as in “often”) but those changes happen too seldom and too inconsistently to solve the problem (for example: the people who say a /t/ in “often” don’t say a “t” in “listen” or “castle” or “whistle,” and show no signs of beginning to do so).

    Therefore, phonological change slows down very little as a result of having standardized spelling, printing presses, and so on. The printing press mainly slows down change in writing-habits, far more than it ever slows down change in speaking-habits.

    For example:
    the Tibetan language has had an alphabet, a standardized orthography, and schools for about 1000 years — and has had printing presses and even dictionaries for most of that time: certainly for longer than we English-speakers have had printing presses or dictionaries — yet Tibetan pronunciation has changed throughout the past millennium even more than the pronunciation of English has changed in the same time.
    Despite their long literate tradition (which you’d imagine must have kept the language from changing), the once-clear relation between Tibetan speech and Tibetan spelling has therefore grown even hazier than the relation between English-language speech and spelling: for example,

    to write the Tibetan words pronounced
    “tulku” [‘spirit’] and “ngak” [‘mantra’],
    you must spell them
    “sprulsku” and “sngags.”

    To write the Tibetan place-names pronounced
    “Shigatse” and “Drepung,”
    you must spell them
    “Gzhiskartse” and “Brasspung” —

    and so on (from all I can find out) for every word in the language.

    Reportedly (judging by the comments of Tibetans as well as outsiders learning the language), by now — after a millennium of linguistic change that the written word has NOT prevented — learning the Tibetan spelling rules, their exceptions, the exceptions to their exceptions, the exceptions to the exceptions to those exceptions (and so on, and so forth) now takes even longer than it would take to ignore the alphabet altogether and just memorize each written word as an inexplicable, unanalyzable, unspellable whole (Chinese-fashion).

    Result: it takes Tibetan schoolkids about three times longer than English-speaking schoolkids to learn to read and spell even “easy” words in their native language (just as learning the rudiments of reading and spelling takes English-speaking schoolkids about three times longer than it takes schoolkids who speak Italian or Spanish or Finnish or any of the thousands of other languages worldwide that have figured out how to keep their standard spelling reasonably in sync with the language it represents.)

    Plainly, having a writing system, printing presses, and standardized orthography — even for about a millennium — does not help anywhere near as much as you’d like to think it should. Your guess that “language change is dependent on speakers never seeing images of words and having no concept of ‘correct spelling'” simply does not describe the real world — linguists have identified many other causes of language change (among literate as well as illiterate peoples), and have documented that language change (including change in pronunciation) proceeds apace even among the most impressibly literate populations.


    If literacy and spelling standards prevented pronunciation from changing, the 19th century would not have seen English change the pronunciation of “vision” from “VIZ-ee-un” to “VIZH-un”: among the thousands of other words which changed — in the usage of literate as well as illiterate folks — from /s/ or /z/ sounds to /sh/ or /zh/ sounds during that highly literate century (“Sure” and “sugar” used to sound like “syoor” and “syugar” in the mouths of people who spelled them just as we do today. We know this because some of them wrote dictionaries and indicated these pronunciations.)

    The French language — which has not only literacy and spelling standards, but a centuries-old national academy to enforce them — nevertheless found its speakers inexorably changing the pronunciation of many French words throughout the eighteenth and nineteenth century: changes that remain in effect today.
    For instance:
    French households used to have an item of furniture called a “chaire” (meaning “chair,” as you’ve probably guessed) that they pronounced /SHERR/ — the fact that millions of literate Frenchmen in the eighteenth and early nineteenth centuries knew how to spell this common, everyday word didn’t stop them from changing its pronunciation to /SHEZZ/ and eventually changing its spelling to the modern-day French “chaise” despite the complaints of purists.

    Similarly — and at about the same time — speakers of French changed thousands of words from having the sounds /WEH/ to having the sounds /WAH/. This happened in literally all the words that French spelled (and still spells) with the letters “oi” or “oy”:
    “moi/roi/royal/loi/foi/gloire/oie” and literally thousands of other common words changed in this way, between the middle of the eighteenth century and the beginning of the nineteenth century, among literates and illiterates alike, despite their stable spelling.

    Returning to English: most of the largest and farthest-reaching changes in the pronunciation of our language appeared or became universal sometime after — not before — printing presses came to England. These vast changes (which the existence of printing did not prevent) include the removal/alteration of the sound that the letters “gh” once represented, as well as a very thorough-going group of vowel-sound changes that linguists call the Great Vowel Shift (a phrase you might like to Google).
    If printing had prevented sound-changes in English — as you seem to imagine — then today the word “knife” would still sound like /K-NEE-feh/.


    “We can only pray that Imglish does not replace the current spelling system” —

    What do you mean by “Imglish,” sir? I can understand that we well might re-spell the name of our language as “Inglish,” but I cannot see why we would ever decide to spell it with an “m.”

    We might pray that “Inglish” and other similarly transparent spellings wouldn’t replace the current system — but our prayers would likely have no more effect than the prayers of some modern-day English lexicographer that the residents of the United States would put the letter “u” into “color” and change “z” to “s” in “realize.” (Few of us, after all, pray when we pass a “donut shop” that the owner will take down the sign and paint in the letters “ugh” … )


    “If worse comes to worst, we really don’t need a close correlation between sound and meaning.”

    Correlation between sound and meaning doesn’t enter into the matter at all. The difficulties involve correlating sound and appearance, not sound and meaning. (For a language to closely correlate sound and meaning, the word “inaudible” could not contain any sounds … and the word “invisible” could not contain any letters, but only blank space.)

  2. CrydayCeapy Says:

    Спасибо за пост! Добавил блог в RSS-ридер, теперь читать буду регулярно..

Leave a Reply