Thursday, November 19, 2009

Irreformable Spelling

A while ago, I posted a whimsical set of spelling reform "edicts." Clearly, I wasn't in earnest about them. But I admit that I've always been the type of guy who can't help thinking about the possibility of reforming English spelling. Many people have brought proposals forward. None of them have turned out to be practical. And I have decided to resign my hopes as well.

Theoretically, there are a lot of ways you could approach the reform of English spelling. In practice, however, each way has problems.

First, you could assign a different letter to each phoneme, which would mean introducing letters that don't currently exist in our alphabet. (This would give us an excuse to ransack other writing systems, such as Cyrillic, for handy characters not found in the Roman system). Then each consonant and vowel sound in English would be represented by exactly one letter, which would represent no other sound. Second, you could develop a consistent system of multiple-letter combinations, such as ch, dh, sh, th, zh, etc. to represent certain sounds, so that we could get by with even fewer letters than we now use. Or, theoretically (as my facetious edicts demonstrated), you could simply repurpose the letters of the existing alphabet to serve our needs more efficiently. And finally, short of inventing an entirely new writing system - and surprisingly few people in history have really had the genius for such a feat - you could throw some diacritical marks over and under the existing letters to clarify which of their multiple pronunciations is in use at a given point - especially when it comes to vowels.

The main problem with all of these options, a problem I am by no means first to point out, is that each regional variation in pronouncing a given word would result in a different spelling of that word. If one goal of spelling reform is to standardize spelling throughout the anglophone world, this one-to-one, phonetic approach would be self-defeating. Worse, it would probably result in more variant spellings than just British vs. American; or, if a standard spelling was somehow enforced, in simply another case where, for most speakers, words were pronounced otherwise than as spelled.

Besides, I have come to appreciate the charm of the differences between British and American spelling. You can spot which part of the anglophone world a writer comes from, often simply by observing which way he spells such words as color/colour, fiber/fibre, draft/draught, jail/gaol, etc. The most fun discoveries, for fans of regional spelling quirks, are the inarticulate grunts and sound effects that people casually sprinkle into their speech, such as uh/er, duh/der, nyah/nyer, ew/eurgh, and aah/aargh. It's enough to make an American pause and reconsider the letter R.

Then there are the homonyms, homophones, homographs, all those "homo" words, that make English such a risky and exciting linguistic track to race on. If the soundalike words were all spelled the same, how would we tell them apart? For literate people, isn't the difference in spelling one of the ways we mentally distinguish between homophones? Plays on words won't be so clever - nor so readily appeciated - when they've stopped being separate words and become different definitions of the same word. And fine distinctions like "effect" vs. "affect," troublesome though they may be, won't be as easy to make.

Another problem began growing on my mind after I posted my Spelling Edicts spoof. When you read Lincoln's Gettysburg Address in my "reformed" spelling, it really doesn't look like itself. Too much of the essential character of the language is tied up in its spelling. The history of its development is in there, too. Related words may appear more similar to each other in writing than they sound to the ear. Knowledge of that kinship between them is part of the context in which they are used, often in a clever way that cannot be appreciated if their spellings are forced to conform to strict, phonetic rules. Even forms of the same word lose contact with each other across the chasm of phonetic spelling, whereas their archaic, somewhat arbitrary, traditional spelling preserves a remnant of that relationship.

There's a genius in that, too, like the way German uses the umlaut to keep the same vowel letter in a verb stem even when (due to a sound shift that happened centuries ago) each tense is pronounced with a different vowel sound. I use the example from a sister tongue because it may be easier for us to perceive its aptness than if I tried to demonstrate it through English examples. By the same token, then, who would expect /froilain/ to be a diminutive form of /frau/? Yet two little dots enable Frau and Fräulein to stand side by side, their etymological relationship unmistakable.

There are many similar instances in English, groups of words that would appear wholly unrelated if we spelled them phonetically, but whose interrelatedness becomes apparent when you see them spelled. Words such as who and what, closely related pronouns that in a phonetic system might not have a single letter in common. Here and there, for another example, are a matched pair of adverbs that, in "Spelling Edictese," would be rendered hir and yer. And words like the, which (at least the way I was brought up to use it) is pronounced with one vowel-sound when followed by a consonant and another when followed by a vowel - entirely left up to the discretion of the speaker, of course - would appear in print as two different words whose morphology might be difficult to explain. I know people whose dual pronunciation of the word the runs exactly opposite to mine; it irritates me just a bit when I hear them, but I think it would irritate me even more if I had to fuss over how to spell our language's most frequently-used and inconsequential word, every single time it came up, and above all to have to take into account each individual speaker's proclivities.

Bottom line, I like English spelling. Warts and all. It has so many fine details of cultural background fossilized in it. We have borrowed many French terms, few of which we pronounce as the French do, but whose Frenchness is kept evergreen by the English language's easy-come, easy-go canons of spelling. A francophone struggling to read a page from an American newspaper might feel encouraged, now and then, by the random appearance of a word he knows and loves. Meanwhile, an American is hard put to decipher words the Japanese language has appropriated from English, even after they have been transliterated into Roman letters. A perfectly systematic written language, like Romaji, cannot tolerate the existence of foreign words. It must obliterate their foreignness. It casts a magic spell on them - the spell of spelling - and their outlandish origins disappear.

I am aware that some people, such as the composer Percy Grainger, hold that English speakers should use only words derived from Anglo-Saxon roots. Thus, instead of coining a new word by sticking a Greek prefix in front of a Latin stem, we ought to jam two Anglo-Saxon stems together and make a new word out of them. The German language does this a lot: Handschuh (glove), for example, is such a hippogriff word, made by suturing "hand" and "shoe" together. Maybe that works for German folks. But English language has a cosmopolitan streak. How could we say that our language had a certain je ne sais quoi without dipping into the treasury of another language? And if we insisted on spelling that term zhernersehkwah, what would we have? Nothing but a piece of random gibberish, anointed with a given meaning (I don't know what). Imagine that word dropped into a transcript of a cultured conversations. The perplexed speaker, reviewing the transcript, would swear he had said something in French - but not even a native French-speaker would be able to find it if he looked for it!

No, folks. English spelling is best left alone. It's not as if ours is the only language that hazes anyone who attempts to learn it as a second tongue. Pity any one coming to Chinese, Korean, Japanese, or Amharic as an adult - there are too many characters to learn. If you want to be able to read a newspaper in such a language, you need to start young and study hard. Or take Vietnamese and Gaelic: two languages that use Roman letters, but in such a complex and idosyncratic way that the adult learner must live each day on the edge of despair. Cyrillic, Greek, Hebrew, and Arabic have extremely learnable writing systems; I would love to have an opportunity to master Thai, Hindi, or similar scripts. Sooner that than figure out how to predict the spelling of a Gaelic sentence from its pronunciation, or vice versa.

And some languages clearly have the wrong alphabet; some have even changed alphabets, by law, within living memory. Some southern Slavic languages, for example, switched from Cyrillic to Roman writing systems during the last century. How I pity the folks who speak those languages - especially those whose lives cross cultural and linguistic borders into an area where a closely related spoken language is written in a vastly different alphabet. I think there are similar cases in and around the Subcontinent, where Sanskrit-based scripts have been exchanged for Arabic-based ones, or vice versa. Urdu, they say, is a language virtually identical to Hindi, yet their vocabularies and writing systems are mutually unintelligible. Maybe that doesn't seem like such a tragedy if you live in or around Pakistan, but it does give you pause when you hear that some bright-eyed know-all wants all English speakers to replace the Roman alphabet with some custom-designed hybrid of Sumerian and Hittite. Folks in Papua New Guinea, whose pidgin is based substantially on English, would need their own Rosetta stone to be able to decipher Australian missionaries' tracts. I can't imagine a worse tragedy.


Robbie F. said...

Before you jump on it, that last sentence was a joke.

Nigel said...

I think the fact you are missing out is that English spelling is virtually unteachable to 23+% of our children. Having such illiteracy is down to, in large part, to our spelling system being so antique. So I don't find it culturally rich but an oppressive burden blighting the lives of many. Many European languages have surmounted the dialect problem when modernizing their spellings and they are capable of great literature: Dante & Cervantes both wrote in a reformed spelling.

Robbie F. said...

Yes, well, but where would their societies be without absolutist monarchs, more or less benevolent despots, and utopian ideologues? Another thing enshrined in the English language, besides (heh) cultural diversity, is the struggle for liberty. Small wonder that disposing of the English language as we know it as at the top of the agenda for so many freedom-hating ideologies. As your kids struggle with dyslexia - as the "hook" on phonics tears right through their cheek and leaves them floundering - may they at least rejoice in knowing that the language that so stymies them does so out of a bullheaded resistance to being herded and dictated to.

Seriously, though, illiteracy and the inability to spell are not the same thing. Many highly literate people are middling-to-poor spellers. It isn't a matter of comprehension, only the ability to conform to accepted canons of style whose complexities, let's face it, go way beyond spelling. To the extent that a prejudice against weak spelling skills may effect their career prospects, maybe they have good reason to complain...but then again, there are spell-checkers now!

But beware the folks who preach the gospel of forcing everyone to adopt a sort of anglo-Esperanto. That sort of thing never happens without (A) blighting a culture's connection with its own literary history, and (B) lining up a lot of bright, promising people and shooting them. Pace mainland China, whose "Simplified" character system has gouged a chasm across the middle of the Chinese-reading/writing part of the world. Don't think it didn't (doesn't) happen without the shedding of blood.

steve said...

Robbie's claim that you have to shed blood in order to reform spelling overlooks the fact that most alphabetic writing systems have been updated several times. Even French had a set of reforms and at least one was fairly radical.

If you want to retain the link between the written and spoken language, you have to change the spelling when the pronunciation changes. Otherwise you begin to corrupt your alphabet.

Webster (1758-1843) in his dissertation wrote: "Letters, the most useful invention that ever blessed mankind, lose a part of their value by no longer being representatives of the sounds originally annexed to them." The effect is, "to destroy the benefits of the alphabet."

In answer to Nigel, if you allow people to spell each syllable an average of 4 differet ways, 85% of English spelling cn be taught.

It is going to take three times as long to teach ambiguous spelling as it does to teach dictionary key or pronunciation guide spelling, but it can be done.

Knowing 4 plausible ways to spell most words does not make you a good speller but it does enable you to communicate.

To be a good speller, you eventually have to memorize the dictionary.

steve said...

Webster Latin One (WLO) is an example of the refom mentioned. The purpose of this notation is simply to reduce the ambiguity of English spelling.

Danz féuu lýt and quick beige [bázh] zuu foxès wûr jûmping lông in dhè áìr óvèr éch 'thin litl dog. bût not with còlosàl ruum [rüm]. Gó luk hér nouu at mé, thañkful wen Í shout ènûf, fôr dhey, thè slý critèrs, jôyûsly fôild yü àgin.

vowels: á é ý ó yü
unstressed schwa à è ì ò ù

This transcription still contains a few surplus letters. To optimize it for texting, you might come up with somethin like QicRyt:

Dans fu lyt & qic bez su foxs wr jmpq loq in d er ovr ik tin litl dog, bt nt wid c'losl rum. Go luc hir nw at mi, tancfl wen i zwt in'f, 4 de, d sly critrs, joi'sli foild u agin.

For more discussion, go to

Robbie F. said...

Your example makes my point exactly. I wouldn't want to spell English the way WLO does. It doesn't have the look or feel of English. It doesn't have a connection with history. It doesn't appear to represent the thought of a literate anglophone.

To paraphrase Yogi Berra, all language is half mental - and that includes the written language, which in some cases must be recognized as distinct from the spoken language. I call Arabic to witness. The literary, written dialect of Arabic is not spoken anywhere, yet Arabic speakers throughout the world accept it as correctly written, even if they refuse to read something written in any spoken dialect other than their own. Or take Chinese, whose writing system transcends differences in vocab & even grammar between the various spoken "dialects" of Chinese, which are really distinct languages.

Of course, that makes reading an Arabic newspaper an exercise in studying dead languages; while two people reading the same Chinese newspaper aloud might be heard speaking entirely different languages. I'm not suggesting that the disparity between spoken and written English will or should go that far. But it isn't unprecedented. And, forgive me, but I think I would be willing to shed blood to resist having an unhistorical and unidiomatic spelling "reform" such as WLO forced on me & mine.

Nigel said...

Tyrants and despots are bad. Also illiberal is a writing system that excludes many. Very few of us read Shakespeare or Chaucer in the original - that is for graduates or post-graduates; my concern is for those who are struggling to learn to read and are never going to get anywhere near "the culture's connection with its own literary history". This notion of English as enshrining liberty is besides the point - and not very convincing to my mind. I am talking about the spelling system not the language.

Martín Rincón Botero said...

[DRE]: Sorry fòr commenting on such an óld póst. Maybé yoù aulreddy chánjed yoùr mínd on the subject, hoó knóws. But yoù cáme up with the anser yoùrself when yoù tauked about the jerman langwaj, which Í happen tu speak evry day (aulthó iz not mý môther tung). Yoù sed:

"There's a genius in that, too, like the way German uses the umlaut to keep the same vowel letter in a verb stem even when (due to a sound shift that happened centuries ago) each tense is pronounced with a different vowel sound. I use the example from a sister tongue because it may be easier for us to perceive its aptness than if I tried to demonstrate it through English examples. By the same token, then, who would expect /froilain/ to be a diminutive form of /frau/? Yet two little dots enable Frau and Fräulein to stand side by side, their etymological relationship unmistakable."

[DRE]: The way Jerman tríes tu acommodáte pronuncìátion and òrthography at the sáme tíme iz bý úzing díacritics. Ênglish woûdn't dù such níce thing. In Ênglish yoù woûd probably hav "Frau" and "Fraulein", without díèresis, and yoù woûd hav tu lern bý heàrt the pronuncìátion ov the tew wôrds. Thâre iz nó jénìus in that. The díèresis and ôther díacritics àr a practical concern. They'r prezent in Jerman, Spanish, French, Pòrtjugéze... Díacritics àr vèry "cozmopolitan", since yoù seem tu líke such terms...